From 28a48de72b876af794853593cc1412119ada9efc Mon Sep 17 00:00:00 2001 From: "David A. Marlin" Date: Mon, 17 Jan 2005 18:29:21 +0000 Subject: [MTD] NAND extended commands, badb block table autorefresh Added extended commands for AG-AND device and added option for BBT_AUTO_REFRESH. Signed-off-by: David A. Marlin Signed-off-by: Thomas Gleixner --- include/linux/mtd/nand.h | 27 ++++++++++++++++++++++++++- 1 file changed, 26 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index 9a19c65abd74..0118128ae384 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -5,7 +5,7 @@ * Steven J. Hill * Thomas Gleixner * - * $Id: nand.h,v 1.68 2004/11/12 10:40:37 gleixner Exp $ + * $Id: nand.h,v 1.69 2005/01/17 18:29:18 dmarlin Exp $ * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -48,6 +48,8 @@ * 02-08-2004 tglx added option field to nand structure for chip anomalities * 05-25-2004 tglx added bad block table support, ST-MICRO manufacturer id * update of nand_chip structure description + * 01-17-2005 dmarlin added extended commands for AG-AND device and added option + * for BBT_AUTO_REFRESH. */ #ifndef __LINUX_MTD_NAND_H #define __LINUX_MTD_NAND_H @@ -115,6 +117,25 @@ extern int nand_read_raw (struct mtd_info *mtd, uint8_t *buf, loff_t from, size_ #define NAND_CMD_READSTART 0x30 #define NAND_CMD_CACHEDPROG 0x15 +/* Extended commands for AG-AND device */ +/* + * Note: the command for NAND_CMD_DEPLETE1 is really 0x00 but + * there is no way to distinguish that from NAND_CMD_READ0 + * until the remaining sequence of commands has been completed + * so add a high order bit and mask it off in the command. + */ +#define NAND_CMD_DEPLETE1 0x100 +#define NAND_CMD_DEPLETE2 0x38 +#define NAND_CMD_STATUS_MULTI 0x71 +#define NAND_CMD_STATUS_ERROR 0x72 +/* multi-bank error status (banks 0-3) */ +#define NAND_CMD_STATUS_ERROR0 0x73 +#define NAND_CMD_STATUS_ERROR1 0x74 +#define NAND_CMD_STATUS_ERROR2 0x75 +#define NAND_CMD_STATUS_ERROR3 0x76 +#define NAND_CMD_STATUS_RESET 0x7f +#define NAND_CMD_STATUS_CLEAR 0xff + /* Status bits */ #define NAND_STATUS_FAIL 0x01 #define NAND_STATUS_FAIL_N1 0x02 @@ -170,6 +191,10 @@ extern int nand_read_raw (struct mtd_info *mtd, uint8_t *buf, loff_t from, size_ /* Chip has a array of 4 pages which can be read without * additional ready /busy waits */ #define NAND_4PAGE_ARRAY 0x00000040 +/* Chip requires that BBT is periodically rewritten to prevent + * bits from adjacent blocks from 'leaking' in altering data. + * This happens with the Renesas AG-AND chips, possibly others. */ +#define BBT_AUTO_REFRESH 0x00000080 /* Options valid for Samsung large page devices */ #define NAND_SAMSUNG_LP_OPTIONS \ -- cgit v1.2.1 From 99f2a8aea18c9779c141050c6f95a8f1da63bbe4 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Mon, 24 Jan 2005 00:37:04 +0000 Subject: [MTD] Platform RAM Driver Driver for generic RAM blocks which are exported by an platform_device from the device driver system. Signed-off-by: Ben Dooks Signed-off-by: Thomas Gleixner --- include/linux/mtd/plat-ram.h | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) create mode 100644 include/linux/mtd/plat-ram.h (limited to 'include/linux') diff --git a/include/linux/mtd/plat-ram.h b/include/linux/mtd/plat-ram.h new file mode 100644 index 000000000000..2332eda07e0e --- /dev/null +++ b/include/linux/mtd/plat-ram.h @@ -0,0 +1,35 @@ +/* linux/include/mtd/plat-ram.h + * + * (c) 2004 Simtec Electronics + * http://www.simtec.co.uk/products/SWLINUX/ + * Ben Dooks + * + * Generic platform device based RAM map + * + * $Id: plat-ram.h,v 1.2 2005/01/24 00:37:40 bjd Exp $ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#ifndef __LINUX_MTD_PLATRAM_H +#define __LINUX_MTD_PLATRAM_H __FILE__ + +#define PLATRAM_RO (0) +#define PLATRAM_RW (1) + +struct platdata_mtd_ram { + char *mapname; + char **probes; + struct mtd_partition *partitions; + int nr_partitions; + int bankwidth; + + /* control callbacks */ + + void (*set_rw)(struct device *dev, int to); +}; + +#endif /* __LINUX_MTD_PLATRAM_H */ -- cgit v1.2.1 From 068e3c0a002c79a5e3cc7c42cb749c4bb126288c Mon Sep 17 00:00:00 2001 From: "David A. Marlin" Date: Mon, 24 Jan 2005 03:07:46 +0000 Subject: [MTD] NAND Add optional ECC status check callback Add optional hardware specific callback routine to perform extra error status checks on erase and write failures for devices with hardware ECC. Signed-off-by: David A. Marlin Signed-off-by: Thomas Gleixner --- include/linux/mtd/nand.h | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index 0118128ae384..cf52f20c6de2 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -5,7 +5,7 @@ * Steven J. Hill * Thomas Gleixner * - * $Id: nand.h,v 1.69 2005/01/17 18:29:18 dmarlin Exp $ + * $Id: nand.h,v 1.70 2005/01/24 03:07:42 dmarlin Exp $ * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -50,6 +50,8 @@ * update of nand_chip structure description * 01-17-2005 dmarlin added extended commands for AG-AND device and added option * for BBT_AUTO_REFRESH. + * 01-20-2005 dmarlin added optional pointer to hardware specific callback for + * extra error status checks. */ #ifndef __LINUX_MTD_NAND_H #define __LINUX_MTD_NAND_H @@ -164,7 +166,7 @@ extern int nand_read_raw (struct mtd_info *mtd, uint8_t *buf, loff_t from, size_ /* * Constants for Hardware ECC -*/ + */ /* Reset Hardware ECC for read */ #define NAND_ECC_READ 0 /* Reset Hardware ECC for write */ @@ -172,6 +174,10 @@ extern int nand_read_raw (struct mtd_info *mtd, uint8_t *buf, loff_t from, size_ /* Enable Hardware ECC before syndrom is read back from flash */ #define NAND_ECC_READSYN 2 +/* Bit mask for flags passed to do_nand_read_ecc */ +#define NAND_GET_DEVICE 0x80 + + /* Option constants for bizarre disfunctionality and real * features */ @@ -308,6 +314,8 @@ struct nand_hw_control { * @badblock_pattern: [REPLACEABLE] bad block scan pattern used for initial bad block scan * @controller: [OPTIONAL] a pointer to a hardware controller structure which is shared among multiple independend devices * @priv: [OPTIONAL] pointer to private chip date + * @errstat: [OPTIONAL] hardware specific function to perform additional error status checks + * (determine if errors are correctable) */ struct nand_chip { @@ -363,6 +371,7 @@ struct nand_chip { struct nand_bbt_descr *badblock_pattern; struct nand_hw_control *controller; void *priv; + int (*errstat)(struct mtd_info *mtd, struct nand_chip *this, int state, int status, int page); }; /* @@ -484,6 +493,9 @@ extern int nand_update_bbt (struct mtd_info *mtd, loff_t offs); extern int nand_default_bbt (struct mtd_info *mtd); extern int nand_isbad_bbt (struct mtd_info *mtd, loff_t offs, int allowbbt); extern int nand_erase_nand (struct mtd_info *mtd, struct erase_info *instr, int allowbbt); +extern int nand_do_read_ecc (struct mtd_info *mtd, loff_t from, size_t len, + size_t * retlen, u_char * buf, u_char * oob_buf, + struct nand_oobinfo *oobsel, int flags); /* * Constants for oob configuration -- cgit v1.2.1 From 72b56a2d7dccd9ea90f34f6ddb653086a3f3bd2e Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Sat, 5 Feb 2005 02:06:19 +0000 Subject: [MTD] Add OTP basisc add structure definition for OTP region info Signed-off-by: Nicolas Pitre Signed-off-by: Thomas Gleixner --- include/linux/mtd/cfi.h | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mtd/cfi.h b/include/linux/mtd/cfi.h index 2ed8c585021e..d87dc3fbd4ba 100644 --- a/include/linux/mtd/cfi.h +++ b/include/linux/mtd/cfi.h @@ -1,7 +1,7 @@ /* Common Flash Interface structures * See http://support.intel.com/design/flash/technote/index.htm - * $Id: cfi.h,v 1.50 2004/11/20 12:46:51 dwmw2 Exp $ + * $Id: cfi.h,v 1.51 2005/02/05 02:06:16 nico Exp $ */ #ifndef __MTD_CFI_H__ @@ -148,6 +148,14 @@ struct cfi_pri_intelext { uint8_t extra[0]; } __attribute__((packed)); +struct cfi_intelext_otpinfo { + uint32_t ProtRegAddr; + uint16_t FactGroups; + uint8_t FactProtRegSize; + uint16_t UserGroups; + uint8_t UserProtRegSize; +} __attribute__((packed)); + struct cfi_intelext_blockinfo { uint16_t NumIdentBlocks; uint16_t BlockSize; -- cgit v1.2.1 From f77814dd5728edaf1239d19755d2aa0d8c33d861 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Tue, 8 Feb 2005 17:11:19 +0000 Subject: [MTD] Support for protection register support on Intel FLASH chips This enables support for reading, writing and locking so called "Protection Registers" present on some flash chips. A subset of them are pre-programmed at the factory with a unique set of values. The rest is user-programmable. Signed-off-by: Nicolas Pitre Signed-off-by: Thomas Gleixner --- include/linux/mtd/cfi.h | 4 ++-- include/linux/mtd/flashchip.h | 3 ++- include/linux/mtd/map.h | 15 ++++++++++++++- include/linux/mtd/mtd.h | 10 +++++----- 4 files changed, 23 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/cfi.h b/include/linux/mtd/cfi.h index d87dc3fbd4ba..76255474a27c 100644 --- a/include/linux/mtd/cfi.h +++ b/include/linux/mtd/cfi.h @@ -1,7 +1,7 @@ /* Common Flash Interface structures * See http://support.intel.com/design/flash/technote/index.htm - * $Id: cfi.h,v 1.51 2005/02/05 02:06:16 nico Exp $ + * $Id: cfi.h,v 1.52 2005/02/08 17:11:15 nico Exp $ */ #ifndef __MTD_CFI_H__ @@ -252,7 +252,7 @@ static inline uint32_t cfi_build_cmd_addr(uint32_t cmd_ofs, int interleave, int * It looks too long to be inline, but in the common case it should almost all * get optimised away. */ -static inline map_word cfi_build_cmd(u_char cmd, struct map_info *map, struct cfi_private *cfi) +static inline map_word cfi_build_cmd(u_long cmd, struct map_info *map, struct cfi_private *cfi) { map_word val = { {0} }; int wordwidth, words_per_bus, chip_mode, chips_per_word; diff --git a/include/linux/mtd/flashchip.h b/include/linux/mtd/flashchip.h index c66ba812bf90..e778a1ab23c4 100644 --- a/include/linux/mtd/flashchip.h +++ b/include/linux/mtd/flashchip.h @@ -6,7 +6,7 @@ * * (C) 2000 Red Hat. GPLd. * - * $Id: flashchip.h,v 1.15 2004/11/05 22:41:06 nico Exp $ + * $Id: flashchip.h,v 1.16 2005/02/08 17:11:15 nico Exp $ * */ @@ -29,6 +29,7 @@ typedef enum { FL_ERASE_SUSPENDED, FL_WRITING, FL_WRITING_TO_BUFFER, + FL_OTP_WRITE, FL_WRITE_SUSPENDING, FL_WRITE_SUSPENDED, FL_PM_SUSPENDED, diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h index f0268b99c900..8fc6679aa9b1 100644 --- a/include/linux/mtd/map.h +++ b/include/linux/mtd/map.h @@ -1,6 +1,6 @@ /* Overhauled routines for dealing with different mmap regions of flash */ -/* $Id: map.h,v 1.46 2005/01/05 17:09:44 dwmw2 Exp $ */ +/* $Id: map.h,v 1.47 2005/02/08 17:11:15 nico Exp $ */ #ifndef __LINUX_MTD_MAP_H__ #define __LINUX_MTD_MAP_H__ @@ -263,6 +263,17 @@ static inline map_word map_word_and(struct map_info *map, map_word val1, map_wor return r; } +static inline map_word map_word_clr(struct map_info *map, map_word val1, map_word val2) +{ + map_word r; + int i; + + for (i=0; i et al. * @@ -113,12 +113,12 @@ struct mtd_info { * flash devices. The user data is one time programmable but the * factory data is read only. */ - int (*read_user_prot_reg) (struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char *buf); - + int (*get_fact_prot_info) (struct mtd_info *mtd, struct otp_info *buf, size_t len); int (*read_fact_prot_reg) (struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char *buf); - - /* This function is not yet implemented */ + int (*get_user_prot_info) (struct mtd_info *mtd, struct otp_info *buf, size_t len); + int (*read_user_prot_reg) (struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char *buf); int (*write_user_prot_reg) (struct mtd_info *mtd, loff_t from, size_t len, size_t *retlen, u_char *buf); + int (*lock_user_prot_reg) (struct mtd_info *mtd, loff_t from, size_t len); /* kvec-based read/write methods. We need these especially for NAND flash, with its limited number of write cycles per erase. -- cgit v1.2.1 From 8f15fd55f9bf266139b10850947e19c4e3f4e9b7 Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Wed, 9 Feb 2005 09:17:45 +0000 Subject: [JFFS2] Add support for JFFS2-on-Dataflash devices. For Dataflash, can_mark_obsolete = false and the NAND write buffering code (wbuf.c) is used. Since the DataFlash chip will automatically erase pages when writing, the cleanmarkers are not needed - so cleanmarker_oob = false and cleanmarker_size = 0 DataFlash page-sizes are not a power of two (they're multiples of 528 bytes). The SECTOR_ADDR macro (added in the previous core patch) is replaced with a (slower) div/mod version if CONFIG_JFFS2_FS_DATAFLASH is selected. Signed-off-by: Andrew Victor Signed-off-by: Thomas Gleixner --- include/linux/jffs2_fs_sb.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jffs2_fs_sb.h b/include/linux/jffs2_fs_sb.h index 4afc8d8c2e9e..faec29559fed 100644 --- a/include/linux/jffs2_fs_sb.h +++ b/include/linux/jffs2_fs_sb.h @@ -1,4 +1,4 @@ -/* $Id: jffs2_fs_sb.h,v 1.48 2004/11/20 10:41:12 dwmw2 Exp $ */ +/* $Id: jffs2_fs_sb.h,v 1.49 2005/02/09 09:17:41 pavlov Exp $ */ #ifndef _JFFS2_FS_SB #define _JFFS2_FS_SB @@ -94,7 +94,7 @@ struct jffs2_sb_info { to an obsoleted node. I don't like this. Alternatives welcomed. */ struct semaphore erase_free_sem; -#if defined CONFIG_JFFS2_FS_NAND || defined CONFIG_JFFS2_FS_NOR_ECC +#if defined CONFIG_JFFS2_FS_NAND || defined CONFIG_JFFS2_FS_NOR_ECC || defined CONFIG_JFFS2_FS_DATAFLASH /* Write-behind buffer for NAND flash */ unsigned char *wbuf; uint32_t wbuf_ofs; -- cgit v1.2.1 From 2f82ce1eb637c06dfc60f095cd1891ae0ba4894c Mon Sep 17 00:00:00 2001 From: Andrew Victor Date: Wed, 9 Feb 2005 09:24:26 +0000 Subject: [JFFS2] Use a single config option for write buffer support This patch replaces the current CONFIG_JFFS2_FS_NAND, CONFIG_JFFS2_FS_NOR_ECC and CONFIG_JFFS2_FS_DATAFLASH with a single configuration option - CONFIG_JFFS2_FS_WRITEBUFFER. The only functional change of this patch is that the slower div/mod calculations for SECTOR_ADDR(), PAGE_DIV() and PAGE_MOD() are now always used when CONFIG_JFFS2_FS_WRITEBUFFER is enabled. Signed-off-by: Andrew Victor Signed-off-by: Thomas Gleixner --- include/linux/jffs2_fs_sb.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jffs2_fs_sb.h b/include/linux/jffs2_fs_sb.h index faec29559fed..1bd6cdfb7d78 100644 --- a/include/linux/jffs2_fs_sb.h +++ b/include/linux/jffs2_fs_sb.h @@ -1,4 +1,4 @@ -/* $Id: jffs2_fs_sb.h,v 1.49 2005/02/09 09:17:41 pavlov Exp $ */ +/* $Id: jffs2_fs_sb.h,v 1.50 2005/02/09 09:23:55 pavlov Exp $ */ #ifndef _JFFS2_FS_SB #define _JFFS2_FS_SB @@ -94,7 +94,7 @@ struct jffs2_sb_info { to an obsoleted node. I don't like this. Alternatives welcomed. */ struct semaphore erase_free_sem; -#if defined CONFIG_JFFS2_FS_NAND || defined CONFIG_JFFS2_FS_NOR_ECC || defined CONFIG_JFFS2_FS_DATAFLASH +#ifdef CONFIG_JFFS2_FS_WRITEBUFFER /* Write-behind buffer for NAND flash */ unsigned char *wbuf; uint32_t wbuf_ofs; -- cgit v1.2.1 From 0040bf382c77414739c933e4d2ee35ff817d0b99 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 9 Feb 2005 12:20:00 +0000 Subject: [MTD] NAND: Skip bad block table scan on request Signed-off-by: Thomas Gleixner --- include/linux/mtd/nand.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index cf52f20c6de2..cf25c7cfd0ba 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -5,7 +5,7 @@ * Steven J. Hill * Thomas Gleixner * - * $Id: nand.h,v 1.70 2005/01/24 03:07:42 dmarlin Exp $ + * $Id: nand.h,v 1.71 2005/02/09 12:12:59 gleixner Exp $ * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -223,7 +223,8 @@ extern int nand_read_raw (struct mtd_info *mtd, uint8_t *buf, loff_t from, size_ * This can only work if we have the ecc bytes directly behind the * data bytes. Applies for DOC and AG-AND Renesas HW Reed Solomon generators */ #define NAND_HWECC_SYNDROME 0x00020000 - +/* This option skips the bbt scan during initialization. */ +#define NAND_SKIP_BBTSCAN 0x00040000 /* Options set by nand scan */ /* Nand scan has allocated oob_buf */ -- cgit v1.2.1 From f16407d73effc59e1e9f88e45a3dc53cacbb8264 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Wed, 16 Feb 2005 15:55:03 +0000 Subject: [MTD] Quiet unused variable warning Signed-off-by: Nioclas Pitre Signed-off-by: Thomas Gleixner --- include/linux/mtd/map.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h index 8fc6679aa9b1..115b14a634da 100644 --- a/include/linux/mtd/map.h +++ b/include/linux/mtd/map.h @@ -1,6 +1,6 @@ /* Overhauled routines for dealing with different mmap regions of flash */ -/* $Id: map.h,v 1.47 2005/02/08 17:11:15 nico Exp $ */ +/* $Id: map.h,v 1.48 2005/02/16 15:54:59 nico Exp $ */ #ifndef __LINUX_MTD_MAP_H__ #define __LINUX_MTD_MAP_H__ @@ -418,7 +418,7 @@ extern void simple_map_init(struct map_info *); #define simple_map_init(map) BUG_ON(!map_bankwidth_supported((map)->bankwidth)) -#define map_is_linear(map) (1) +#define map_is_linear(map) ({ (void)(map); 1; }) #endif /* !CONFIG_MTD_COMPLEX_MAPPINGS */ -- cgit v1.2.1 From 31fbdf7aa5aac8a2a34f180a25deb157297a10c9 Mon Sep 17 00:00:00 2001 From: "Artem B. Bityuckiy" Date: Mon, 28 Feb 2005 08:21:09 +0000 Subject: [JFFS2] Fix NOR specific scan BUG Fix fairly sad NOR-specific bug - during FS building ic->scan_dents isn't zero, but jffs2_mark_node_obsolete() migt be called it tries to finde the ic corresponding to ref - this requires ic->scan_dents = 0. Signed-off-by: Artem B. Bityuckiy Signed-off-by: Thomas Gleixner --- include/linux/jffs2_fs_sb.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jffs2_fs_sb.h b/include/linux/jffs2_fs_sb.h index 1bd6cdfb7d78..350b82bd6529 100644 --- a/include/linux/jffs2_fs_sb.h +++ b/include/linux/jffs2_fs_sb.h @@ -1,4 +1,4 @@ -/* $Id: jffs2_fs_sb.h,v 1.50 2005/02/09 09:23:55 pavlov Exp $ */ +/* $Id: jffs2_fs_sb.h,v 1.51 2005/02/28 08:21:06 dedekind Exp $ */ #ifndef _JFFS2_FS_SB #define _JFFS2_FS_SB @@ -14,7 +14,8 @@ #include #define JFFS2_SB_FLAG_RO 1 -#define JFFS2_SB_FLAG_MOUNTING 2 +#define JFFS2_SB_FLAG_SCANNING 2 /* Flash scanning is in progress */ +#define JFFS2_SB_FLAG_BUILDING 4 /* File system building is in progress */ struct jffs2_inodirty; -- cgit v1.2.1 From 0514cd938009de1d6b3239d98c3cf2a67b620103 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Mon, 14 Mar 2005 18:27:18 +0000 Subject: [MTD] Fixed signed 1bit bitfield Signed-off-by: Ben Dooks Signed-off-by: Thomas Gleixner --- include/linux/mtd/flashchip.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/flashchip.h b/include/linux/mtd/flashchip.h index e778a1ab23c4..675776fa3e27 100644 --- a/include/linux/mtd/flashchip.h +++ b/include/linux/mtd/flashchip.h @@ -6,7 +6,7 @@ * * (C) 2000 Red Hat. GPLd. * - * $Id: flashchip.h,v 1.16 2005/02/08 17:11:15 nico Exp $ + * $Id: flashchip.h,v 1.17 2005/03/14 18:27:15 bjd Exp $ * */ @@ -63,8 +63,8 @@ struct flchip { flstate_t state; flstate_t oldstate; - int write_suspended:1; - int erase_suspended:1; + unsigned int write_suspended:1; + unsigned int erase_suspended:1; unsigned long in_progress_block_addr; spinlock_t *mutex; -- cgit v1.2.1 From c927cd3a226bed5cf063cdf04de13cef51144cef Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 15 Mar 2005 19:03:16 +0000 Subject: [MTD] Add the reverse operation of cfi_build_cmd() This is necessary to fix the broken status check in cfi_cmdset_0001 Signed-off-by: Thomas Gleixner --- include/linux/mtd/cfi.h | 65 ++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 64 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mtd/cfi.h b/include/linux/mtd/cfi.h index 76255474a27c..66e0a32efbac 100644 --- a/include/linux/mtd/cfi.h +++ b/include/linux/mtd/cfi.h @@ -1,7 +1,7 @@ /* Common Flash Interface structures * See http://support.intel.com/design/flash/technote/index.htm - * $Id: cfi.h,v 1.52 2005/02/08 17:11:15 nico Exp $ + * $Id: cfi.h,v 1.53 2005/03/15 19:03:13 gleixner Exp $ */ #ifndef __MTD_CFI_H__ @@ -315,6 +315,69 @@ static inline map_word cfi_build_cmd(u_long cmd, struct map_info *map, struct cf } #define CMD(x) cfi_build_cmd((x), map, cfi) + +static inline unsigned char cfi_merge_status(map_word val, struct map_info *map, + struct cfi_private *cfi) +{ + int wordwidth, words_per_bus, chip_mode, chips_per_word; + unsigned long onestat, res = 0; + int i; + + /* We do it this way to give the compiler a fighting chance + of optimising away all the crap for 'bankwidth' larger than + an unsigned long, in the common case where that support is + disabled */ + if (map_bankwidth_is_large(map)) { + wordwidth = sizeof(unsigned long); + words_per_bus = (map_bankwidth(map)) / wordwidth; // i.e. normally 1 + } else { + wordwidth = map_bankwidth(map); + words_per_bus = 1; + } + + chip_mode = map_bankwidth(map) / cfi_interleave(cfi); + chips_per_word = wordwidth * cfi_interleave(cfi) / map_bankwidth(map); + + onestat = val.x[0]; + /* Or all status words together */ + for (i=1; i < words_per_bus; i++) { + onestat |= val.x[i]; + } + + res = onestat; + switch(chips_per_word) { + default: BUG(); +#if BITS_PER_LONG >= 64 + case 8: + res |= (onestat >> (chip_mode * 32)); +#endif + case 4: + res |= (onestat >> (chip_mode * 16)); + case 2: + res |= (onestat >> (chip_mode * 8)); + case 1: + ; + } + + /* Last, determine what the bit-pattern should be for a single + device, according to chip mode and endianness... */ + switch (chip_mode) { + case 1: + break; + case 2: + res = cfi16_to_cpu(res); + break; + case 4: + res = cfi32_to_cpu(res); + break; + default: BUG(); + } + return res; +} + +#define MERGESTATUS(x) cfi_merge_status((x), map, cfi) + + /* * Sends a CFI command to a bank of flash for the given geometry. * -- cgit v1.2.1 From 963a6fb0a0d336d0513083b7e4b5c3ff9d6d2061 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Fri, 1 Apr 2005 02:59:56 +0100 Subject: [MTD] Add reboot notifier to Intel NOR flash driver to make sure the flash is in array mode whenever we're about to reboot. This is especially useful to allow "soft" reboot to work which consists of branching back into the bootloader. Signed-off-by: Nicolas Pitre Signed-off-by: Thomas Gleixner --- include/linux/mtd/mtd.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h index 3aab1b8729e0..f574cd498816 100644 --- a/include/linux/mtd/mtd.h +++ b/include/linux/mtd/mtd.h @@ -1,5 +1,5 @@ /* - * $Id: mtd.h,v 1.57 2005/02/08 17:11:15 nico Exp $ + * $Id: mtd.h,v 1.58 2005/04/01 01:59:54 nico Exp $ * * Copyright (C) 1999-2003 David Woodhouse et al. * @@ -18,6 +18,7 @@ #include #include #include +#include #include #include @@ -147,6 +148,8 @@ struct mtd_info { int (*block_isbad) (struct mtd_info *mtd, loff_t ofs); int (*block_markbad) (struct mtd_info *mtd, loff_t ofs); + struct notifier_block reboot_notifier; /* default mode before reboot */ + void *priv; struct module *owner; -- cgit v1.2.1 From 65c6e0a657012d104fe42be5f01a7b9b451b687c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 11 Apr 2005 11:19:05 +0100 Subject: [MTD] Fix broken user ABI Move kernel data where it belongs. Previous change broke user abi. Signed-off-by: Thomas Gleixner --- include/linux/mtd/mtd.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h index f574cd498816..c50c3f3927d9 100644 --- a/include/linux/mtd/mtd.h +++ b/include/linux/mtd/mtd.h @@ -1,5 +1,5 @@ /* - * $Id: mtd.h,v 1.58 2005/04/01 01:59:54 nico Exp $ + * $Id: mtd.h,v 1.59 2005/04/11 10:19:02 gleixner Exp $ * * Copyright (C) 1999-2003 David Woodhouse et al. * @@ -70,7 +70,6 @@ struct mtd_info { u_int32_t oobblock; // Size of OOB blocks (e.g. 512) u_int32_t oobsize; // Amount of OOB data per block (e.g. 16) - u_int32_t oobavail; // Number of bytes in OOB area available for fs u_int32_t ecctype; u_int32_t eccsize; @@ -81,6 +80,7 @@ struct mtd_info { // oobinfo is a nand_oobinfo structure, which can be set by iotcl (MEMSETOOBINFO) struct nand_oobinfo oobinfo; + u_int32_t oobavail; // Number of bytes in OOB area available for fs /* Data for variable erase regions. If numeraseregions is zero, * it means that the whole device has erasesize as given above. -- cgit v1.2.1 From fff7afd791f6a685b3ddedb8cfb152aed85f3cf8 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 19 May 2005 17:18:11 +0100 Subject: [JFFS2] Convert thread start semaphore to completion Signed-off-by: Thomas Gleixner --- include/linux/jffs2_fs_sb.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/jffs2_fs_sb.h b/include/linux/jffs2_fs_sb.h index 350b82bd6529..1e21546622de 100644 --- a/include/linux/jffs2_fs_sb.h +++ b/include/linux/jffs2_fs_sb.h @@ -1,4 +1,4 @@ -/* $Id: jffs2_fs_sb.h,v 1.51 2005/02/28 08:21:06 dedekind Exp $ */ +/* $Id: jffs2_fs_sb.h,v 1.52 2005/05/19 16:12:17 gleixner Exp $ */ #ifndef _JFFS2_FS_SB #define _JFFS2_FS_SB @@ -32,7 +32,7 @@ struct jffs2_sb_info { unsigned int flags; struct task_struct *gc_task; /* GC task struct */ - struct semaphore gc_thread_start; /* GC thread start mutex */ + struct completion gc_thread_start; /* GC thread start completion */ struct completion gc_thread_exit; /* GC thread exit completion port */ struct semaphore alloc_sem; /* Used to protect all the following -- cgit v1.2.1 From 3a3ab48c68de656736f091c6ed768fa8c110a7ab Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 24 May 2005 20:50:18 +0200 Subject: [MTD] Make map_word_ff ware of the flash buswidth map_word_ff() was setting the mapword to ~0UL regardless of the buswidth of the mapped flash chip. The read_map functions are buswidth aware and therefor the map_word_equal function failed. Signed-off-by: Thomas Gleixner --- include/linux/mtd/map.h | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h index 115b14a634da..dd36d9433f00 100644 --- a/include/linux/mtd/map.h +++ b/include/linux/mtd/map.h @@ -1,6 +1,6 @@ /* Overhauled routines for dealing with different mmap regions of flash */ -/* $Id: map.h,v 1.48 2005/02/16 15:54:59 nico Exp $ */ +/* $Id: map.h,v 1.49 2005/05/24 18:45:15 gleixner Exp $ */ #ifndef __LINUX_MTD_MAP_H__ #define __LINUX_MTD_MAP_H__ @@ -340,13 +340,22 @@ static inline map_word map_word_load_partial(struct map_info *map, map_word orig return orig; } +#if BITS_PER_LONG < 64 +#define MAP_FF_LIMIT 4 +#else +#define MAP_FF_LIMIT 8 +#endif + static inline map_word map_word_ff(struct map_info *map) { map_word r; int i; - - for (i=0; i Date: Wed, 25 May 2005 12:20:29 +0200 Subject: [MTD] map.h Use the correct macro and fix the resulting compiler warning Signed-off-by: Thomas Gleixner --- include/linux/mtd/map.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h index dd36d9433f00..dbd7b9b510d3 100644 --- a/include/linux/mtd/map.h +++ b/include/linux/mtd/map.h @@ -1,6 +1,6 @@ /* Overhauled routines for dealing with different mmap regions of flash */ -/* $Id: map.h,v 1.49 2005/05/24 18:45:15 gleixner Exp $ */ +/* $Id: map.h,v 1.51 2005/05/25 10:15:29 gleixner Exp $ */ #ifndef __LINUX_MTD_MAP_H__ #define __LINUX_MTD_MAP_H__ @@ -351,8 +351,9 @@ static inline map_word map_word_ff(struct map_info *map) map_word r; int i; - if (map_bank_width(map) < MAP_FF_LIMIT) { - r.x[0] = (1 << (8*map_bank_width(map))) - 1; + if (map_bankwidth(map) < MAP_FF_LIMIT) { + int bw = 8 * map_bankwidth; + r.x[0] = (1 << bw) - 1; } else { for (i=0; i Date: Wed, 25 May 2005 12:32:37 +0200 Subject: [MTD] Fix it really tglx declares him self to be the idiot of the day. Signed-off-by: Thomas Gleixner --- include/linux/mtd/map.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h index dbd7b9b510d3..142963f01d29 100644 --- a/include/linux/mtd/map.h +++ b/include/linux/mtd/map.h @@ -1,6 +1,6 @@ /* Overhauled routines for dealing with different mmap regions of flash */ -/* $Id: map.h,v 1.51 2005/05/25 10:15:29 gleixner Exp $ */ +/* $Id: map.h,v 1.52 2005/05/25 10:29:41 gleixner Exp $ */ #ifndef __LINUX_MTD_MAP_H__ #define __LINUX_MTD_MAP_H__ @@ -352,7 +352,7 @@ static inline map_word map_word_ff(struct map_info *map) int i; if (map_bankwidth(map) < MAP_FF_LIMIT) { - int bw = 8 * map_bankwidth; + int bw = 8 * map_bankwidth(map); r.x[0] = (1 << bw) - 1; } else { for (i=0; i Date: Tue, 24 May 2005 01:46:34 -0700 Subject: [MTD] NAND: Add Hynix to manufacturer list Signed-off-by: Nicolas S. Dade Signed-off-by: Thomas Gleixner --- include/linux/mtd/nand.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index cf25c7cfd0ba..bee78969cb21 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -384,6 +384,7 @@ struct nand_chip { #define NAND_MFR_NATIONAL 0x8f #define NAND_MFR_RENESAS 0x07 #define NAND_MFR_STMICRO 0x20 +#define NAND_MFR_HYNIX 0xad /** * struct nand_flash_dev - NAND Flash Device ID Structure -- cgit v1.2.1 From 024ac44c701d43f5e2d34bd6a35b2813a36e6010 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Sun, 29 May 2005 02:26:31 -0500 Subject: Input: This patch implements compat_ioctl for joydev. I've tested it with a Logitech WingMan Rumblepad on an x86-64 machine, and on an ia32 machine to make sure I didn't break anything. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Andrew Morton Signed-off-by: Vojtech Pavlik Signed-off-by: Dmitry Torokhov --- include/linux/joystick.h | 33 +++++++++++++++++++++++++-------- 1 file changed, 25 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/joystick.h b/include/linux/joystick.h index b7e0ab622cd7..06b9af77eb7f 100644 --- a/include/linux/joystick.h +++ b/include/linux/joystick.h @@ -111,18 +111,35 @@ struct js_corr { #define JS_SET_ALL 8 struct JS_DATA_TYPE { - int buttons; - int x; - int y; + __s32 buttons; + __s32 x; + __s32 y; }; -struct JS_DATA_SAVE_TYPE { - int JS_TIMEOUT; - int BUSY; - long JS_EXPIRETIME; - long JS_TIMELIMIT; +struct JS_DATA_SAVE_TYPE_32 { + __s32 JS_TIMEOUT; + __s32 BUSY; + __s32 JS_EXPIRETIME; + __s32 JS_TIMELIMIT; struct JS_DATA_TYPE JS_SAVE; struct JS_DATA_TYPE JS_CORR; }; +struct JS_DATA_SAVE_TYPE_64 { + __s32 JS_TIMEOUT; + __s32 BUSY; + __s64 JS_EXPIRETIME; + __s64 JS_TIMELIMIT; + struct JS_DATA_TYPE JS_SAVE; + struct JS_DATA_TYPE JS_CORR; +}; + +#if BITS_PER_LONG == 64 +#define JS_DATA_SAVE_TYPE JS_DATA_SAVE_TYPE_64 +#elif BITS_PER_LONG == 32 +#define JS_DATA_SAVE_TYPE JS_DATA_SAVE_TYPE_32 +#else +#error Unexpected BITS_PER_LONG +#endif + #endif /* _LINUX_JOYSTICK_H */ -- cgit v1.2.1 From 0fbf87caf70acec0c435233fbc39c7bd0aca3ca6 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Sun, 29 May 2005 02:29:25 -0500 Subject: Input: add semaphore and user count to input_dev structure; serialize open and close calls and ensure that device's open and close methods are only called when first user opens it or last user closes it. Signed-off-by: Dmitry Torokhov --- include/linux/input.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/input.h b/include/linux/input.h index 72731d7d189e..43e8ecec602b 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -859,6 +859,10 @@ struct input_dev { int (*erase_effect)(struct input_dev *dev, int effect_id); struct input_handle *grab; + + struct semaphore sem; /* serializes open and close operations */ + unsigned int users; + struct device *dev; struct list_head h_list; -- cgit v1.2.1 From c611763d048990de5cdf848d97af6392f8fa7430 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 1 Jun 2005 02:39:51 -0500 Subject: Input: add ps2_drain() to libps2 to allow reading and discarding given number of bytes from device. Change ps2_command to allow using 0 as command ID and actually pass it to the device instead of working as a drain. Signed-off-by: Dmitry Torokhov --- include/linux/libps2.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/libps2.h b/include/linux/libps2.h index 923bdbc6d9e4..a710bddda4eb 100644 --- a/include/linux/libps2.h +++ b/include/linux/libps2.h @@ -41,6 +41,7 @@ struct ps2dev { void ps2_init(struct ps2dev *ps2dev, struct serio *serio); int ps2_sendbyte(struct ps2dev *ps2dev, unsigned char byte, int timeout); +void ps2_drain(struct ps2dev *ps2dev, int maxbytes, int timeout); int ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command); int ps2_schedule_command(struct ps2dev *ps2dev, unsigned char *param, int command); int ps2_handle_ack(struct ps2dev *ps2dev, unsigned char data); -- cgit v1.2.1 From dbf4ccd6043e58ed32fbf253fb3f0a9991e4c13a Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 1 Jun 2005 02:40:01 -0500 Subject: Input: psmouse - export protocol as a sysfs per-device attribute to allow easy switching at run-time. Signed-off-by: Dmitry Torokhov --- include/linux/serio.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/serio.h b/include/linux/serio.h index a2d3b9ae06f4..aa4d6493a034 100644 --- a/include/linux/serio.h +++ b/include/linux/serio.h @@ -83,6 +83,7 @@ static inline void serio_register_port(struct serio *serio) } void serio_unregister_port(struct serio *serio); +void serio_unregister_child_port(struct serio *serio); void __serio_unregister_port_delayed(struct serio *serio, struct module *owner); static inline void serio_unregister_port_delayed(struct serio *serio) { @@ -153,6 +154,11 @@ static inline int serio_pin_driver(struct serio *serio) return down_interruptible(&serio->drv_sem); } +static inline void serio_pin_driver_uninterruptible(struct serio *serio) +{ + down(&serio->drv_sem); +} + static inline void serio_unpin_driver(struct serio *serio) { up(&serio->drv_sem); -- cgit v1.2.1 From 986a80d5c154808cc78170584670324a22fd8219 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 16 Jun 2005 15:14:00 -0700 Subject: [PATCH] avoid signed vs unsigned comparison in efi_range_is_wc() warning when building with gcc -W : include/linux/efi.h: In function `efi_range_is_wc': include/linux/efi.h:320: warning: comparison between signed and unsigned It looks to me like a significantly large 'len' passed in could cause the loop to never end. Isn't it safer to make 'i' an unsigned long as well? Like this little patch below (which of course also kills the warning) : Signed-off-by: Jesper Juhl Signed-off-by: Tony Luck --- include/linux/efi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/efi.h b/include/linux/efi.h index 047e7222df7a..73781ec165b4 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -315,7 +315,7 @@ extern struct efi_memory_map memmap; */ static inline int efi_range_is_wc(unsigned long start, unsigned long len) { - int i; + unsigned long i; for (i = 0; i < len; i += (1UL << EFI_PAGE_SHIFT)) { unsigned long paddr = __pa(start + i); -- cgit v1.2.1 From b3d5496ea5915fa4848fe307af9f7097f312e932 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Sat, 2 Apr 2005 20:31:02 +0200 Subject: [PATCH] I2C: Kill address ranges in non-sensors i2c chip drivers Some months ago, you killed the address ranges mechanism from all sensors i2c chip drivers (both the module parameters and the in-code address lists). I think it was a very good move, as the ranges can easily be replaced by individual addresses, and this allowed for significant cleanups in the i2c core (let alone the impressive size shrink for all these drivers). Unfortunately you did not do the same for non-sensors i2c chip drivers. These need the address ranges even less, so we could get rid of the ranges here as well for another significant i2c core cleanup. Here comes a patch which does just that. Since the process is exactly the same as what you did for the other drivers set already, I did not split this one in parts. A documentation update is included. The change saves 308 bytes in the i2c core, and an average 1382 bytes for chip drivers which use I2C_CLIENT_INSMOD, 126 bytes for those which do not. This change is required if we want to merge the sensors and non-sensors i2c code (and we want to do this). Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman Index: gregkh-2.6/Documentation/i2c/writing-clients =================================================================== --- include/linux/i2c.h | 12 ------------ 1 file changed, 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2c.h b/include/linux/i2c.h index ebcd745f4cd6..be837b13f297 100644 --- a/include/linux/i2c.h +++ b/include/linux/i2c.h @@ -290,11 +290,8 @@ static inline void i2c_set_adapdata (struct i2c_adapter *dev, void *data) */ struct i2c_client_address_data { unsigned short *normal_i2c; - unsigned short *normal_i2c_range; unsigned short *probe; - unsigned short *probe_range; unsigned short *ignore; - unsigned short *ignore_range; unsigned short *force; }; @@ -563,24 +560,15 @@ union i2c_smbus_data { #define I2C_CLIENT_INSMOD \ I2C_CLIENT_MODULE_PARM(probe, \ "List of adapter,address pairs to scan additionally"); \ - I2C_CLIENT_MODULE_PARM(probe_range, \ - "List of adapter,start-addr,end-addr triples to scan " \ - "additionally"); \ I2C_CLIENT_MODULE_PARM(ignore, \ "List of adapter,address pairs not to scan"); \ - I2C_CLIENT_MODULE_PARM(ignore_range, \ - "List of adapter,start-addr,end-addr triples not to " \ - "scan"); \ I2C_CLIENT_MODULE_PARM(force, \ "List of adapter,address pairs to boldly assume " \ "to be present"); \ static struct i2c_client_address_data addr_data = { \ .normal_i2c = normal_i2c, \ - .normal_i2c_range = normal_i2c_range, \ .probe = probe, \ - .probe_range = probe_range, \ .ignore = ignore, \ - .ignore_range = ignore_range, \ .force = force, \ } -- cgit v1.2.1 From 3886246a257e828248ce1e72ced00408a3557f0d Mon Sep 17 00:00:00 2001 From: Sebastian Witt Date: Wed, 13 Apr 2005 22:25:39 +0200 Subject: [PATCH] I2C: i2c-vid.h: Support for VID to reg conversion Adds conversion from VID (mV) to register value. Used by the atxp1 I2C module. Removed uneeded switch case. Signed-off-by: Sebastian Witt Signed-off-by: Greg Kroah-Hartman --- include/linux/i2c-vid.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/linux') diff --git a/include/linux/i2c-vid.h b/include/linux/i2c-vid.h index 974835e3530f..41d0635e0ba9 100644 --- a/include/linux/i2c-vid.h +++ b/include/linux/i2c-vid.h @@ -97,3 +97,15 @@ static inline int vid_from_reg(int val, int vrm) 2050 - (val) * 50); } } + +static inline int vid_to_reg(int val, int vrm) +{ + switch (vrm) { + case 91: /* VRM 9.1 */ + case 90: /* VRM 9.0 */ + return ((val >= 1100) && (val <= 1850) ? + ((18499 - val * 10) / 25 + 5) / 10 : -1); + default: + return -1; + } +} -- cgit v1.2.1 From 10c08f8100ee2c4d27b862635574cdf4ef439e67 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Mon, 6 Jun 2005 19:34:45 +0200 Subject: [PATCH] I2C: rename i2c-sysfs.h to hwmon-sysfs.h This patch renames the new linux/i2c-sysfs.h header file to linux/hwmon-sysfs.h. This names seems to be more appropriate since this file defines macros and structures not related to i2c but to hardware monitoring drivers. The patch also updates the five hardware monitoring driver which include that header file already. Signed-off-by: Jean Delvare Signed-off-by: Greg Kroah-Hartman --- include/linux/hwmon-sysfs.h | 36 ++++++++++++++++++++++++++++++++++++ include/linux/i2c-sysfs.h | 36 ------------------------------------ 2 files changed, 36 insertions(+), 36 deletions(-) create mode 100644 include/linux/hwmon-sysfs.h delete mode 100644 include/linux/i2c-sysfs.h (limited to 'include/linux') diff --git a/include/linux/hwmon-sysfs.h b/include/linux/hwmon-sysfs.h new file mode 100644 index 000000000000..1b5018a965f5 --- /dev/null +++ b/include/linux/hwmon-sysfs.h @@ -0,0 +1,36 @@ +/* + * hwmon-sysfs.h - hardware monitoring chip driver sysfs defines + * + * Copyright (C) 2005 Yani Ioannou + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. + */ +#ifndef _LINUX_HWMON_SYSFS_H +#define _LINUX_HWMON_SYSFS_H + +struct sensor_device_attribute{ + struct device_attribute dev_attr; + int index; +}; +#define to_sensor_dev_attr(_dev_attr) \ + container_of(_dev_attr, struct sensor_device_attribute, dev_attr) + +#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index) \ +struct sensor_device_attribute sensor_dev_attr_##_name = { \ + .dev_attr = __ATTR(_name,_mode,_show,_store), \ + .index = _index, \ +} + +#endif /* _LINUX_HWMON_SYSFS_H */ diff --git a/include/linux/i2c-sysfs.h b/include/linux/i2c-sysfs.h deleted file mode 100644 index d7bf6ce11679..000000000000 --- a/include/linux/i2c-sysfs.h +++ /dev/null @@ -1,36 +0,0 @@ -/* - * i2c-sysfs.h - i2c chip driver sysfs defines - * - * Copyright (C) 2005 Yani Ioannou - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - */ -#ifndef _LINUX_I2C_SYSFS_H -#define _LINUX_I2C_SYSFS_H - -struct sensor_device_attribute{ - struct device_attribute dev_attr; - int index; -}; -#define to_sensor_dev_attr(_dev_attr) \ - container_of(_dev_attr, struct sensor_device_attribute, dev_attr) - -#define SENSOR_DEVICE_ATTR(_name,_mode,_show,_store,_index) \ -struct sensor_device_attribute sensor_dev_attr_##_name = { \ - .dev_attr = __ATTR(_name,_mode,_show,_store), \ - .index = _index, \ -} - -#endif /* _LINUX_I2C_SYSFS_H */ -- cgit v1.2.1 From c124a78d8c7475ecc43f385f34112b638c4228d9 Mon Sep 17 00:00:00 2001 From: Randy Vinson Date: Fri, 3 Jun 2005 14:36:06 -0700 Subject: [PATCH] I2C: Add support for Maxim/Dallas DS1374 Real-Time Clock Chip (1/2) Add support for Maxim/Dallas DS1374 Real-Time Clock Chip This change adds support for the Maxim/Dallas DS1374 RTC chip. This chip is an I2C-based RTC that maintains a simple 32-bit binary seconds count with battery backup support. Signed-off-by: Randy Vinson Signed-off-by: Greg Kroah-Hartman --- include/linux/i2c-id.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/i2c-id.h b/include/linux/i2c-id.h index 89270ce51470..33f08258f22b 100644 --- a/include/linux/i2c-id.h +++ b/include/linux/i2c-id.h @@ -108,6 +108,7 @@ #define I2C_DRIVERID_TDA7313 62 /* TDA7313 audio processor */ #define I2C_DRIVERID_MAX6900 63 /* MAX6900 real-time clock */ #define I2C_DRIVERID_SAA7114H 64 /* video decoder */ +#define I2C_DRIVERID_DS1374 65 /* DS1374 real time clock */ #define I2C_DRIVERID_EXP0 0xF0 /* experimental use id's */ -- cgit v1.2.1 From dd7f0b80926befc8c70a873b5b0c0c7b5fd1e7b9 Mon Sep 17 00:00:00 2001 From: Harald Welte Date: Wed, 22 Jun 2005 12:38:33 -0700 Subject: [NETFILTER]: Fix "iptables -D" rule deletion with ipt_CLUSTERIP target. The patch just changes the order of structure members. Signed-off-by: Harald Welte Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ipt_CLUSTERIP.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h b/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h index baa83e757156..d9bceedfb3dc 100644 --- a/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h +++ b/include/linux/netfilter_ipv4/ipt_CLUSTERIP.h @@ -18,7 +18,6 @@ struct clusterip_config; struct ipt_clusterip_tgt_info { u_int32_t flags; - struct clusterip_config *config; /* only relevant for new ones */ u_int8_t clustermac[6]; @@ -27,6 +26,8 @@ struct ipt_clusterip_tgt_info { u_int16_t local_nodes[CLUSTERIP_MAX_NODES]; enum clusterip_hashmode hash_mode; u_int32_t hash_initval; + + struct clusterip_config *config; }; #endif /*_IPT_CLUSTERIP_H_target*/ -- cgit v1.2.1 From 5ee0ed7d3ab620a764740fb018f469d45f561931 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:20 +0000 Subject: [PATCH] RPC: Make rpc_create_client() probe server for RPC program+version support Ensure that we don't create an RPC client without checking that the server does indeed support the RPC program + version that we are trying to set up. This enables us to immediately return an error to "mount" if it turns out that the server is only supporting NFSv2, when we requested NFSv3 or NFSv4. Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index 2709caf4d128..d25e80f77ff5 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -111,6 +111,9 @@ struct rpc_procinfo { struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname, struct rpc_program *info, u32 version, rpc_authflavor_t authflavor); +struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname, + struct rpc_program *info, + u32 version, rpc_authflavor_t authflavor); struct rpc_clnt *rpc_clone_client(struct rpc_clnt *); int rpc_shutdown_client(struct rpc_clnt *); int rpc_destroy_client(struct rpc_clnt *); @@ -129,6 +132,7 @@ void rpc_clnt_sigmask(struct rpc_clnt *clnt, sigset_t *oldset); void rpc_clnt_sigunmask(struct rpc_clnt *clnt, sigset_t *oldset); void rpc_setbufsize(struct rpc_clnt *, unsigned int, unsigned int); size_t rpc_max_payload(struct rpc_clnt *); +int rpc_ping(struct rpc_clnt *clnt, int flags); static __inline__ int rpc_call(struct rpc_clnt *clnt, u32 proc, void *argp, void *resp, int flags) -- cgit v1.2.1 From 4ce79717ce32a9f88c1ddce4b9658556cb59d37a Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:21 +0000 Subject: [PATCH] NFS: Header file cleanup... - Move NFSv4 state definitions into a private header file. - Clean up gunk in nfs_fs.h Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 241 ------------------------------------------------- 1 file changed, 241 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index dbac7f363e5d..fb33e7655cfa 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -15,7 +15,6 @@ #include #include #include -#include #include @@ -29,7 +28,6 @@ #include #include #include -#include #include /* @@ -43,13 +41,6 @@ #define NFS_MAX_FILE_IO_BUFFER_SIZE 32768 #define NFS_DEF_FILE_IO_BUFFER_SIZE 4096 -/* - * The upper limit on timeouts for the exponential backoff algorithm. - */ -#define NFS_WRITEBACK_DELAY (5*HZ) -#define NFS_WRITEBACK_LOCKDELAY (60*HZ) -#define NFS_COMMIT_DELAY (5*HZ) - /* * superblock magic number for NFS */ @@ -60,9 +51,6 @@ */ #define NFS_RPC_SWAPFLAGS (RPC_TASK_SWAPPER|RPC_TASK_ROOTCREDS) -#define NFS_RW_SYNC 0x0001 /* O_SYNC handling */ -#define NFS_RW_SWAP 0x0002 /* This is a swap request */ - /* * When flushing a cluster of dirty pages, there can be different * strategies: @@ -434,11 +422,6 @@ static inline void nfs_writedata_free(struct nfs_write_data *p) mempool_free(p, nfs_wdata_mempool); } -/* Hack for future NFS swap support */ -#ifndef IS_SWAPFILE -# define IS_SWAPFILE(inode) (0) -#endif - /* * linux/fs/nfs/read.c */ @@ -515,230 +498,6 @@ extern void * nfs_root_data(void); #define NFS_JUKEBOX_RETRY_TIME (5 * HZ) -#ifdef CONFIG_NFS_V4 - -struct idmap; - -/* - * In a seqid-mutating op, this macro controls which error return - * values trigger incrementation of the seqid. - * - * from rfc 3010: - * The client MUST monotonically increment the sequence number for the - * CLOSE, LOCK, LOCKU, OPEN, OPEN_CONFIRM, and OPEN_DOWNGRADE - * operations. This is true even in the event that the previous - * operation that used the sequence number received an error. The only - * exception to this rule is if the previous operation received one of - * the following errors: NFSERR_STALE_CLIENTID, NFSERR_STALE_STATEID, - * NFSERR_BAD_STATEID, NFSERR_BAD_SEQID, NFSERR_BADXDR, - * NFSERR_RESOURCE, NFSERR_NOFILEHANDLE. - * - */ -#define seqid_mutating_err(err) \ -(((err) != NFSERR_STALE_CLIENTID) && \ - ((err) != NFSERR_STALE_STATEID) && \ - ((err) != NFSERR_BAD_STATEID) && \ - ((err) != NFSERR_BAD_SEQID) && \ - ((err) != NFSERR_BAD_XDR) && \ - ((err) != NFSERR_RESOURCE) && \ - ((err) != NFSERR_NOFILEHANDLE)) - -enum nfs4_client_state { - NFS4CLNT_OK = 0, -}; - -/* - * The nfs4_client identifies our client state to the server. - */ -struct nfs4_client { - struct list_head cl_servers; /* Global list of servers */ - struct in_addr cl_addr; /* Server identifier */ - u64 cl_clientid; /* constant */ - nfs4_verifier cl_confirm; - unsigned long cl_state; - - u32 cl_lockowner_id; - - /* - * The following rwsem ensures exclusive access to the server - * while we recover the state following a lease expiration. - */ - struct rw_semaphore cl_sem; - - struct list_head cl_delegations; - struct list_head cl_state_owners; - struct list_head cl_unused; - int cl_nunused; - spinlock_t cl_lock; - atomic_t cl_count; - - struct rpc_clnt * cl_rpcclient; - struct rpc_cred * cl_cred; - - struct list_head cl_superblocks; /* List of nfs_server structs */ - - unsigned long cl_lease_time; - unsigned long cl_last_renewal; - struct work_struct cl_renewd; - struct work_struct cl_recoverd; - - wait_queue_head_t cl_waitq; - struct rpc_wait_queue cl_rpcwaitq; - - /* used for the setclientid verifier */ - struct timespec cl_boot_time; - - /* idmapper */ - struct idmap * cl_idmap; - - /* Our own IP address, as a null-terminated string. - * This is used to generate the clientid, and the callback address. - */ - char cl_ipaddr[16]; - unsigned char cl_id_uniquifier; -}; - -/* - * NFS4 state_owners and lock_owners are simply labels for ordered - * sequences of RPC calls. Their sole purpose is to provide once-only - * semantics by allowing the server to identify replayed requests. - * - * The ->so_sema is held during all state_owner seqid-mutating operations: - * OPEN, OPEN_DOWNGRADE, and CLOSE. Its purpose is to properly serialize - * so_seqid. - */ -struct nfs4_state_owner { - struct list_head so_list; /* per-clientid list of state_owners */ - struct nfs4_client *so_client; - u32 so_id; /* 32-bit identifier, unique */ - struct semaphore so_sema; - u32 so_seqid; /* protected by so_sema */ - atomic_t so_count; - - struct rpc_cred *so_cred; /* Associated cred */ - struct list_head so_states; - struct list_head so_delegations; -}; - -/* - * struct nfs4_state maintains the client-side state for a given - * (state_owner,inode) tuple (OPEN) or state_owner (LOCK). - * - * OPEN: - * In order to know when to OPEN_DOWNGRADE or CLOSE the state on the server, - * we need to know how many files are open for reading or writing on a - * given inode. This information too is stored here. - * - * LOCK: one nfs4_state (LOCK) to hold the lock stateid nfs4_state(OPEN) - */ - -struct nfs4_lock_state { - struct list_head ls_locks; /* Other lock stateids */ - fl_owner_t ls_owner; /* POSIX lock owner */ -#define NFS_LOCK_INITIALIZED 1 - int ls_flags; - u32 ls_seqid; - u32 ls_id; - nfs4_stateid ls_stateid; - atomic_t ls_count; -}; - -/* bits for nfs4_state->flags */ -enum { - LK_STATE_IN_USE, - NFS_DELEGATED_STATE, -}; - -struct nfs4_state { - struct list_head open_states; /* List of states for the same state_owner */ - struct list_head inode_states; /* List of states for the same inode */ - struct list_head lock_states; /* List of subservient lock stateids */ - - struct nfs4_state_owner *owner; /* Pointer to the open owner */ - struct inode *inode; /* Pointer to the inode */ - - unsigned long flags; /* Do we hold any locks? */ - struct semaphore lock_sema; /* Serializes file locking operations */ - rwlock_t state_lock; /* Protects the lock_states list */ - - nfs4_stateid stateid; - - unsigned int nreaders; - unsigned int nwriters; - int state; /* State on the server (R,W, or RW) */ - atomic_t count; -}; - - -struct nfs4_exception { - long timeout; - int retry; -}; - -struct nfs4_state_recovery_ops { - int (*recover_open)(struct nfs4_state_owner *, struct nfs4_state *); - int (*recover_lock)(struct nfs4_state *, struct file_lock *); -}; - -extern struct dentry_operations nfs4_dentry_operations; -extern struct inode_operations nfs4_dir_inode_operations; - -/* nfs4proc.c */ -extern int nfs4_map_errors(int err); -extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short); -extern int nfs4_proc_setclientid_confirm(struct nfs4_client *); -extern int nfs4_proc_async_renew(struct nfs4_client *); -extern int nfs4_proc_renew(struct nfs4_client *); -extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode); -extern struct inode *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *); -extern int nfs4_open_revalidate(struct inode *, struct dentry *, int); - -extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops; -extern struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops; - -/* nfs4renewd.c */ -extern void nfs4_schedule_state_renewal(struct nfs4_client *); -extern void nfs4_renewd_prepare_shutdown(struct nfs_server *); -extern void nfs4_kill_renewd(struct nfs4_client *); - -/* nfs4state.c */ -extern void init_nfsv4_state(struct nfs_server *); -extern void destroy_nfsv4_state(struct nfs_server *); -extern struct nfs4_client *nfs4_get_client(struct in_addr *); -extern void nfs4_put_client(struct nfs4_client *clp); -extern int nfs4_init_client(struct nfs4_client *clp); -extern struct nfs4_client *nfs4_find_client(struct in_addr *); -extern u32 nfs4_alloc_lockowner_id(struct nfs4_client *); - -extern struct nfs4_state_owner * nfs4_get_state_owner(struct nfs_server *, struct rpc_cred *); -extern void nfs4_put_state_owner(struct nfs4_state_owner *); -extern void nfs4_drop_state_owner(struct nfs4_state_owner *); -extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *); -extern void nfs4_put_open_state(struct nfs4_state *); -extern void nfs4_close_state(struct nfs4_state *, mode_t); -extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode); -extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp); -extern void nfs4_schedule_state_recovery(struct nfs4_client *); -extern struct nfs4_lock_state *nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t); -extern struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t); -extern void nfs4_put_lock_state(struct nfs4_lock_state *state); -extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls); -extern void nfs4_notify_setlk(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *); -extern void nfs4_notify_unlck(struct nfs4_state *, struct file_lock *, struct nfs4_lock_state *); -extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t); - - - -struct nfs4_mount_data; -#else -#define init_nfsv4_state(server) do { } while (0) -#define destroy_nfsv4_state(server) do { } while (0) -#define nfs4_put_state_owner(inode, owner) do { } while (0) -#define nfs4_put_open_state(state) do { } while (0) -#define nfs4_close_state(a, b) do { } while (0) -#define nfs4_renewd_prepare_shutdown(server) do { } while (0) -#endif - #endif /* __KERNEL__ */ /* -- cgit v1.2.1 From a656db998785324a818005bcf71bae6dcbbb3cf5 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:21 +0000 Subject: [PATCH] NFS: Remove unused NFS inode field readdir_timestamp. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index fb33e7655cfa..68d5aae89972 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -128,7 +128,6 @@ struct nfs_inode { * * mtime != read_cache_mtime */ - unsigned long readdir_timestamp; unsigned long read_cache_jiffies; unsigned long attrtimeo; unsigned long attrtimeo_timestamp; -- cgit v1.2.1 From 96651ab341cde0fee940ec837f323d711cbfa7d5 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:21 +0000 Subject: [PATCH] RPC: Shrink struct rpc_task by switching to wait_on_bit() Signed-off-by: Trond Myklebust --- include/linux/sunrpc/sched.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index 99d17ed7cebb..4d77e90d0b30 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -31,7 +31,6 @@ struct rpc_wait_queue; struct rpc_wait { struct list_head list; /* wait queue links */ struct list_head links; /* Links to related tasks */ - wait_queue_head_t waitq; /* sync: sleep on this q */ struct rpc_wait_queue * rpc_waitq; /* RPC wait queue we're on */ }; -- cgit v1.2.1 From 464a98bd70bae8c559cfc82af799faf44824ce64 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:21 +0000 Subject: [PATCH] NFS: cleanup: shrink struct nfs_open_context Remove the wait queue, and replace the functions that depended on it with wait_on_bit(). Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 68d5aae89972..0b01b96337f8 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -84,7 +84,6 @@ struct nfs_open_context { int error; struct list_head list; - wait_queue_head_t waitq; }; /* -- cgit v1.2.1 From 92cfc62cb8412c9563860b1bf70cd4701f03092e Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 22 Jun 2005 17:16:22 +0000 Subject: [PATCH] NFS: Allow NFS versions to support different sets of inode operations. ACL support will require supporting additional inode operations in v4 (getxattr, setxattr, listxattr). This patch allows different protocol versions to support different inode operations by adding a file_inode_ops to the nfs_rpc_ops (to match the existing dir_inode_ops). Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/nfs_xdr.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 47037d9521cb..5b45bafd9db5 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -667,6 +667,7 @@ struct nfs_rpc_ops { int version; /* Protocol version */ struct dentry_operations *dentry_ops; struct inode_operations *dir_inode_ops; + struct inode_operations *file_inode_ops; int (*getroot) (struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); -- cgit v1.2.1 From ada70d9425bcc5e376fef8591e4e76e204c0834c Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:22 +0000 Subject: [PATCH] NFS: Add hooks to allow common NFS attribute code to clear cached acls Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 + include/linux/nfs_xdr.h | 1 + 2 files changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 0b01b96337f8..140bdf489f71 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -189,6 +189,7 @@ struct nfs_inode { #define NFS_INO_INVALID_DATA 0x0010 /* cached data is invalid */ #define NFS_INO_INVALID_ATIME 0x0020 /* cached atime is invalid */ #define NFS_INO_INVALID_ACCESS 0x0040 /* cached access cred invalid */ +#define NFS_INO_INVALID_ACL 0x0080 /* cached acls are invalid */ static inline struct nfs_inode *NFS_I(struct inode *inode) { diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 5b45bafd9db5..cf38db59f347 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -714,6 +714,7 @@ struct nfs_rpc_ops { int (*file_open) (struct inode *, struct file *); int (*file_release) (struct inode *, struct file *); int (*lock)(struct file *, int, struct file_lock *); + void (*clear_acl_cache)(struct inode *); }; /* -- cgit v1.2.1 From 029d105e66e5a90850d5a09dad76815d0bcfcaa3 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 22 Jun 2005 17:16:22 +0000 Subject: [PATCH] NFSv4: Client-side xdr for reading NFSv4 acls Client-side support for NFSv4 acls: xdr encoding and decoding routines for reading acls Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/nfs4.h | 1 + include/linux/nfs_xdr.h | 7 +++++++ 2 files changed, 8 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 5ca8a8d8ccdf..6ee7e2585af5 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -382,6 +382,7 @@ enum { NFSPROC4_CLNT_READDIR, NFSPROC4_CLNT_SERVER_CAPS, NFSPROC4_CLNT_DELEGRETURN, + NFSPROC4_CLNT_GETACL, }; #endif diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index cf38db59f347..9f5e1d407c7b 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -326,6 +326,13 @@ struct nfs_setattrargs { const u32 * bitmask; }; +struct nfs_getaclargs { + struct nfs_fh * fh; + size_t acl_len; + unsigned int acl_pgbase; + struct page ** acl_pages; +}; + struct nfs_setattrres { struct nfs_fattr * fattr; const struct nfs_server * server; -- cgit v1.2.1 From 23ec6965c20db96bc8ea7af0ec178f074dd31c40 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 22 Jun 2005 17:16:22 +0000 Subject: [PATCH] NFSv4: Client-side xdr for writing NFSv4 acls Client-side support for NFSv4 acls: xdr encoding and decoding routines for writing acls Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/nfs4.h | 1 + include/linux/nfs_xdr.h | 7 +++++++ 2 files changed, 8 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 6ee7e2585af5..5bb5b2fd7ba2 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -383,6 +383,7 @@ enum { NFSPROC4_CLNT_SERVER_CAPS, NFSPROC4_CLNT_DELEGRETURN, NFSPROC4_CLNT_GETACL, + NFSPROC4_CLNT_SETACL, }; #endif diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 9f5e1d407c7b..46b206b460c0 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -326,6 +326,13 @@ struct nfs_setattrargs { const u32 * bitmask; }; +struct nfs_setaclargs { + struct nfs_fh * fh; + size_t acl_len; + unsigned int acl_pgbase; + struct page ** acl_pages; +}; + struct nfs_getaclargs { struct nfs_fh * fh; size_t acl_len; -- cgit v1.2.1 From e50a1c2e1f816c81eed6a589019052cb44189267 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Wed, 22 Jun 2005 17:16:23 +0000 Subject: [PATCH] NFSv4: client-side caching NFSv4 ACLs Add nfs4_acl field to the nfs_inode, and use it to cache acls. Only cache acls of size up to a page. Also prepare for up to a page of acl data even when the user doesn't pass in a buffer, as when they want to get the acl length to decide what size buffer to allocate. Signed-off-by: J. Bruce Fields Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 140bdf489f71..d2b5d7e0e85a 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -169,13 +169,13 @@ struct nfs_inode { wait_queue_head_t nfs_i_wait; #ifdef CONFIG_NFS_V4 + struct nfs4_cached_acl *nfs4_acl; /* NFSv4 state */ struct list_head open_states; struct nfs_delegation *delegation; int delegation_state; struct rw_semaphore rwsem; #endif /* CONFIG_NFS_V4*/ - struct inode vfs_inode; }; -- cgit v1.2.1 From 007e251f2b2760f738c92adc8c80cbae0bed3ce5 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:23 +0000 Subject: [PATCH] RPC: Allow multiple RPC client programs to share the same transport Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Trond Myklebust --- include/linux/sunrpc/clnt.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/clnt.h b/include/linux/sunrpc/clnt.h index d25e80f77ff5..ab151bbb66df 100644 --- a/include/linux/sunrpc/clnt.h +++ b/include/linux/sunrpc/clnt.h @@ -114,6 +114,8 @@ struct rpc_clnt *rpc_create_client(struct rpc_xprt *xprt, char *servname, struct rpc_clnt *rpc_new_client(struct rpc_xprt *xprt, char *servname, struct rpc_program *info, u32 version, rpc_authflavor_t authflavor); +struct rpc_clnt *rpc_bind_new_program(struct rpc_clnt *, + struct rpc_program *, int); struct rpc_clnt *rpc_clone_client(struct rpc_clnt *); int rpc_shutdown_client(struct rpc_clnt *); int rpc_destroy_client(struct rpc_clnt *); -- cgit v1.2.1 From e053d1ab62c8ef0eff3dd4c95448cad3c6d2fbf4 Mon Sep 17 00:00:00 2001 From: Olaf Kirch Date: Wed, 22 Jun 2005 17:16:24 +0000 Subject: [PATCH] RPC: Lazy RPC receive buffer allocation Signed-off-by: Olaf Kirch Signed-off-by: Andreas Gruenbacher Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 541dcf838abf..0f5b7a5a7432 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -160,7 +160,7 @@ typedef struct { typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); -extern void xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, +extern int xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, skb_reader_t *, skb_read_actor_t); struct socket; -- cgit v1.2.1 From 7e06b53d796a3740307b54aa2799077f8a0c84e7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:24 +0000 Subject: [PATCH] RPC: fix accounting bug in the case of a truncated RPC message Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 0f5b7a5a7432..5d1eed2b58a1 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -160,7 +160,7 @@ typedef struct { typedef size_t (*skb_read_actor_t)(skb_reader_t *desc, void *to, size_t len); -extern int xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, +extern ssize_t xdr_partial_copy_from_skb(struct xdr_buf *, unsigned int, skb_reader_t *, skb_read_actor_t); struct socket; -- cgit v1.2.1 From bd8100e7eda87507649c6ba4cb32173b34e49986 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:24 +0000 Subject: [PATCH] RPC: Encode and decode arbitrary XDR arrays Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Trond Myklebust --- include/linux/sunrpc/xdr.h | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 5d1eed2b58a1..34ec3e8d99b3 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -146,7 +146,8 @@ extern void xdr_shift_buf(struct xdr_buf *, size_t); extern void xdr_buf_from_iov(struct kvec *, struct xdr_buf *); extern int xdr_buf_subsegment(struct xdr_buf *, struct xdr_buf *, int, int); extern int xdr_buf_read_netobj(struct xdr_buf *, struct xdr_netobj *, int); -extern int read_bytes_from_xdr_buf(struct xdr_buf *buf, int base, void *obj, int len); +extern int read_bytes_from_xdr_buf(struct xdr_buf *, int, void *, int); +extern int write_bytes_to_xdr_buf(struct xdr_buf *, int, void *, int); /* * Helper structure for copying from an sk_buff. @@ -168,6 +169,22 @@ struct sockaddr; extern int xdr_sendpages(struct socket *, struct sockaddr *, int, struct xdr_buf *, unsigned int, int); +extern int xdr_encode_word(struct xdr_buf *, int, u32); +extern int xdr_decode_word(struct xdr_buf *, int, u32 *); + +struct xdr_array2_desc; +typedef int (*xdr_xcode_elem_t)(struct xdr_array2_desc *desc, void *elem); +struct xdr_array2_desc { + unsigned int elem_size; + unsigned int array_len; + xdr_xcode_elem_t xcode; +}; + +extern int xdr_decode_array2(struct xdr_buf *buf, unsigned int base, + struct xdr_array2_desc *desc); +extern int xdr_encode_array2(struct xdr_buf *buf, unsigned int base, + struct xdr_array2_desc *desc); + /* * Provide some simple tools for XDR buffer overflow-checking etc. */ -- cgit v1.2.1 From 9ba02638e4be28dd4ff724202a640264427c62d1 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:24 +0000 Subject: [PATCH] RPC: Allow the sunrpc server to multiplex serveral programs on a single port The NFS and NFSACL programs run on the same RPC transport. This patch adds support for this by converting svc_program into a chained list of programs (server-side). Signed-off-by: Andreas Gruenbacher Signed-off-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- include/linux/sunrpc/svc.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index 37003970cf2e..facb94488bb1 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -240,9 +240,10 @@ struct svc_deferred_req { }; /* - * RPC program + * List of RPC programs on the same transport endpoint */ struct svc_program { + struct svc_program * pg_next; /* other programs (same xprt) */ u32 pg_prog; /* program number */ unsigned int pg_lovers; /* lowest version */ unsigned int pg_hivers; /* lowest version */ -- cgit v1.2.1 From a257cdd0e2179630d3201c32ba14d7fcb3c3a055 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:26 +0000 Subject: [PATCH] NFSD: Add server support for NFSv3 ACLs. This adds functions for encoding and decoding POSIX ACLs for the NFSACL protocol extension, and the GETACL and SETACL RPCs. The implementation is compatible with NFSACL in Solaris. Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- include/linux/nfsacl.h | 58 ++++++++++++++++++++++++++++++++++++++++++++++ include/linux/nfsd/nfsd.h | 16 +++++++++++++ include/linux/nfsd/xdr.h | 4 ++++ include/linux/nfsd/xdr3.h | 26 +++++++++++++++++++++ include/linux/sunrpc/svc.h | 11 +++++++++ 5 files changed, 115 insertions(+) create mode 100644 include/linux/nfsacl.h (limited to 'include/linux') diff --git a/include/linux/nfsacl.h b/include/linux/nfsacl.h new file mode 100644 index 000000000000..54487a99beb8 --- /dev/null +++ b/include/linux/nfsacl.h @@ -0,0 +1,58 @@ +/* + * File: linux/nfsacl.h + * + * (C) 2003 Andreas Gruenbacher + */ +#ifndef __LINUX_NFSACL_H +#define __LINUX_NFSACL_H + +#define NFS_ACL_PROGRAM 100227 + +#define ACLPROC2_GETACL 1 +#define ACLPROC2_SETACL 2 +#define ACLPROC2_GETATTR 3 +#define ACLPROC2_ACCESS 4 + +#define ACLPROC3_GETACL 1 +#define ACLPROC3_SETACL 2 + + +/* Flags for the getacl/setacl mode */ +#define NFS_ACL 0x0001 +#define NFS_ACLCNT 0x0002 +#define NFS_DFACL 0x0004 +#define NFS_DFACLCNT 0x0008 + +/* Flag for Default ACL entries */ +#define NFS_ACL_DEFAULT 0x1000 + +#ifdef __KERNEL__ + +#include + +/* Maximum number of ACL entries over NFS */ +#define NFS_ACL_MAX_ENTRIES 1024 + +#define NFSACL_MAXWORDS (2*(2+3*NFS_ACL_MAX_ENTRIES)) +#define NFSACL_MAXPAGES ((2*(8+12*NFS_ACL_MAX_ENTRIES) + PAGE_SIZE-1) \ + >> PAGE_SHIFT) + +static inline unsigned int +nfsacl_size(struct posix_acl *acl_access, struct posix_acl *acl_default) +{ + unsigned int w = 16; + w += max(acl_access ? (int)acl_access->a_count : 3, 4) * 12; + if (acl_default) + w += max((int)acl_default->a_count, 4) * 12; + return w; +} + +extern unsigned int +nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, + struct posix_acl *acl, int encode_entries, int typeflag); +extern unsigned int +nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt, + struct posix_acl **pacl); + +#endif /* __KERNEL__ */ +#endif /* __LINUX_NFSACL_H */ diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 8f85d9a59607..4bf931d5ff56 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -124,6 +125,21 @@ int nfsd_statfs(struct svc_rqst *, struct svc_fh *, int nfsd_notify_change(struct inode *, struct iattr *); int nfsd_permission(struct svc_export *, struct dentry *, int); +#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) +#ifdef CONFIG_NFSD_V2_ACL +extern struct svc_version nfsd_acl_version2; +#else +#define nfsd_acl_version2 NULL +#endif +#ifdef CONFIG_NFSD_V3_ACL +extern struct svc_version nfsd_acl_version3; +#else +#define nfsd_acl_version3 NULL +#endif +struct posix_acl *nfsd_get_posix_acl(struct svc_fh *, int); +int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *); +#endif + /* * NFSv4 State diff --git a/include/linux/nfsd/xdr.h b/include/linux/nfsd/xdr.h index ecccef777dae..130d4f588a37 100644 --- a/include/linux/nfsd/xdr.h +++ b/include/linux/nfsd/xdr.h @@ -169,4 +169,8 @@ int nfssvc_encode_entry(struct readdir_cd *, const char *name, int nfssvc_release_fhandle(struct svc_rqst *, u32 *, struct nfsd_fhandle *); +/* Helper functions for NFSv2 ACL code */ +u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp); +u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp); + #endif /* LINUX_NFSD_H */ diff --git a/include/linux/nfsd/xdr3.h b/include/linux/nfsd/xdr3.h index 0ae9e0ef5f68..21e18ce7ca63 100644 --- a/include/linux/nfsd/xdr3.h +++ b/include/linux/nfsd/xdr3.h @@ -110,6 +110,19 @@ struct nfsd3_commitargs { __u32 count; }; +struct nfsd3_getaclargs { + struct svc_fh fh; + int mask; +}; + +struct posix_acl; +struct nfsd3_setaclargs { + struct svc_fh fh; + int mask; + struct posix_acl *acl_access; + struct posix_acl *acl_default; +}; + struct nfsd3_attrstat { __u32 status; struct svc_fh fh; @@ -209,6 +222,14 @@ struct nfsd3_commitres { struct svc_fh fh; }; +struct nfsd3_getaclres { + __u32 status; + struct svc_fh fh; + int mask; + struct posix_acl *acl_access; + struct posix_acl *acl_default; +}; + /* dummy type for release */ struct nfsd3_fhandle_pair { __u32 dummy; @@ -241,6 +262,7 @@ union nfsd3_xdrstore { struct nfsd3_fsinfores fsinfores; struct nfsd3_pathconfres pathconfres; struct nfsd3_commitres commitres; + struct nfsd3_getaclres getaclres; }; #define NFS3_SVC_XDRSIZE sizeof(union nfsd3_xdrstore) @@ -316,6 +338,10 @@ int nfs3svc_encode_entry(struct readdir_cd *, const char *name, int nfs3svc_encode_entry_plus(struct readdir_cd *, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int); +/* Helper functions for NFSv3 ACL code */ +u32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, + struct svc_fh *fhp); +u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp); #endif /* _LINUX_NFSD_XDR3_H */ diff --git a/include/linux/sunrpc/svc.h b/include/linux/sunrpc/svc.h index facb94488bb1..5af8800e0ce3 100644 --- a/include/linux/sunrpc/svc.h +++ b/include/linux/sunrpc/svc.h @@ -185,6 +185,17 @@ xdr_ressize_check(struct svc_rqst *rqstp, u32 *p) return vec->iov_len <= PAGE_SIZE; } +static inline struct page * +svc_take_res_page(struct svc_rqst *rqstp) +{ + if (rqstp->rq_arghi <= rqstp->rq_argused) + return NULL; + rqstp->rq_arghi--; + rqstp->rq_respages[rqstp->rq_resused] = + rqstp->rq_argpages[rqstp->rq_arghi]; + return rqstp->rq_respages[rqstp->rq_resused++]; +} + static inline int svc_take_page(struct svc_rqst *rqstp) { if (rqstp->rq_arghi <= rqstp->rq_argused) -- cgit v1.2.1 From b7fa0554cf1ba6d6895cd0a5b02989a26e0bc704 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:27 +0000 Subject: [PATCH] NFS: Add support for NFSv3 ACLs This adds acl support fo nfs clients via the NFSACL protocol extension, by implementing the getxattr, listxattr, setxattr, and removexattr iops for the system.posix_acl_access and system.posix_acl_default attributes. This patch implements a dumb version that uses no caching (and thus adds some overhead). (Another patch in this patchset adds caching as well.) Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 31 +++++++++++++++++++++++++++++++ include/linux/nfs_fs_sb.h | 1 + include/linux/nfs_mount.h | 1 + include/linux/nfs_xdr.h | 27 +++++++++++++++++++++++++++ 4 files changed, 60 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index d2b5d7e0e85a..3a5e442ac776 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -301,6 +301,9 @@ extern u32 root_nfs_parse_addr(char *name); /*__init*/ * linux/fs/nfs/file.c */ extern struct inode_operations nfs_file_inode_operations; +#ifdef CONFIG_NFS_V3 +extern struct inode_operations nfs3_file_inode_operations; +#endif /* CONFIG_NFS_V3 */ extern struct file_operations nfs_file_operations; extern struct address_space_operations nfs_file_aops; @@ -315,6 +318,22 @@ static inline struct rpc_cred *nfs_file_cred(struct file *file) return NULL; } +/* + * linux/fs/nfs/xattr.c + */ +#ifdef CONFIG_NFS_V3_ACL +extern ssize_t nfs3_listxattr(struct dentry *, char *, size_t); +extern ssize_t nfs3_getxattr(struct dentry *, const char *, void *, size_t); +extern int nfs3_setxattr(struct dentry *, const char *, + const void *, size_t, int); +extern int nfs3_removexattr (struct dentry *, const char *name); +#else +# define nfs3_listxattr NULL +# define nfs3_getxattr NULL +# define nfs3_setxattr NULL +# define nfs3_removexattr NULL +#endif + /* * linux/fs/nfs/direct.c */ @@ -329,6 +348,9 @@ extern ssize_t nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, * linux/fs/nfs/dir.c */ extern struct inode_operations nfs_dir_inode_operations; +#ifdef CONFIG_NFS_V3 +extern struct inode_operations nfs3_dir_inode_operations; +#endif /* CONFIG_NFS_V3 */ extern struct file_operations nfs_dir_operations; extern struct dentry_operations nfs_dentry_operations; @@ -449,6 +471,15 @@ static inline void nfs_readdata_free(struct nfs_read_data *p) extern void nfs_readdata_release(struct rpc_task *task); +/* + * linux/fs/nfs3proc.c + */ +#ifdef CONFIG_NFS_V3_ACL +extern struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type); +extern int nfs3_proc_setacl(struct inode *inode, int type, + struct posix_acl *acl); +#endif /* CONFIG_NFS_V3_ACL */ + /* * linux/fs/mount_clnt.c * (Used only by nfsroot module) diff --git a/include/linux/nfs_fs_sb.h b/include/linux/nfs_fs_sb.h index fc51645d61ee..3d3a305488cf 100644 --- a/include/linux/nfs_fs_sb.h +++ b/include/linux/nfs_fs_sb.h @@ -10,6 +10,7 @@ struct nfs_server { struct rpc_clnt * client; /* RPC client handle */ struct rpc_clnt * client_sys; /* 2nd handle for FSINFO */ + struct rpc_clnt * client_acl; /* ACL RPC client handle */ struct nfs_rpc_ops * rpc_ops; /* NFS protocol vector */ struct backing_dev_info backing_dev_info; int flags; /* various flags */ diff --git a/include/linux/nfs_mount.h b/include/linux/nfs_mount.h index 0071428231f9..659c75438454 100644 --- a/include/linux/nfs_mount.h +++ b/include/linux/nfs_mount.h @@ -58,6 +58,7 @@ struct nfs_mount_data { #define NFS_MOUNT_KERBEROS 0x0100 /* 3 */ #define NFS_MOUNT_NONLM 0x0200 /* 3 */ #define NFS_MOUNT_BROKEN_SUID 0x0400 /* 4 */ +#define NFS_MOUNT_NOACL 0x0800 /* 4 */ #define NFS_MOUNT_STRICTLOCK 0x1000 /* reserved for NFSv4 */ #define NFS_MOUNT_SECFLAVOUR 0x2000 /* 5 */ #define NFS_MOUNT_FLAGMASK 0xFFFF diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 46b206b460c0..a2bf6914ff1b 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -2,6 +2,7 @@ #define _LINUX_NFS_XDR_H #include +#include struct nfs4_fsid { __u64 major; @@ -368,6 +369,20 @@ struct nfs_readdirargs { struct page ** pages; }; +struct nfs3_getaclargs { + struct nfs_fh * fh; + int mask; + struct page ** pages; +}; + +struct nfs3_setaclargs { + struct inode * inode; + int mask; + struct posix_acl * acl_access; + struct posix_acl * acl_default; + struct page ** pages; +}; + struct nfs_diropok { struct nfs_fh * fh; struct nfs_fattr * fattr; @@ -491,6 +506,15 @@ struct nfs3_readdirres { int plus; }; +struct nfs3_getaclres { + struct nfs_fattr * fattr; + int mask; + unsigned int acl_access_count; + unsigned int acl_default_count; + struct posix_acl * acl_access; + struct posix_acl * acl_default; +}; + #ifdef CONFIG_NFS_V4 typedef u64 clientid4; @@ -748,4 +772,7 @@ extern struct rpc_version nfs_version2; extern struct rpc_version nfs_version3; extern struct rpc_version nfs_version4; +extern struct rpc_version nfsacl_version3; +extern struct rpc_program nfsacl_program; + #endif -- cgit v1.2.1 From 055ffbea0596942579b0dae71d5dab78de8135f6 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:27 +0000 Subject: [PATCH] NFS: Fix handling of the umask when an NFSv3 default acl is present. NFSv3 has no concept of a umask on the server side: The client applies the umask locally, and sends the effective permissions to the server. This behavior is wrong when files are created in a directory that has a default ACL. In this case, the umask is supposed to be ignored, and only the default ACL determines the file's effective permissions. Usually its the server's task to conditionally apply the umask. But since the server knows nothing about the umask, we have to do it on the client side. This patch tries to fetch the parent directory's default ACL before creating a new file, computes the appropriate create mode to send to the server, and finally sets the new file's access and default acl appropriately. Many thanks to Buck Huppmann for sending the initial version of this patch, as well as for arguing why we need this change. Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 3a5e442ac776..7662c5131b47 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -478,6 +478,15 @@ extern void nfs_readdata_release(struct rpc_task *task); extern struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type); extern int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl); +extern int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, + mode_t mode); +#else +static inline int nfs3_proc_set_default_acl(struct inode *dir, + struct inode *inode, + mode_t mode) +{ + return 0; +} #endif /* CONFIG_NFS_V3_ACL */ /* -- cgit v1.2.1 From 5c6a9f7d92291c832d47e792ed1fafa44acb066e Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 22 Jun 2005 17:16:27 +0000 Subject: [PATCH] NFS: Cache the NFSv3 acls. Attach acls to inodes in the icache to avoid unnecessary GETACL RPC round-trips. As long as the client doesn't retrieve any acls itself, only the default acls of exiting directories and the default and access acls of new directories will end up in the cache, which preserves some memory compared to always caching the access and default acl of all files. Signed-off-by: Andreas Gruenbacher Acked-by: Olaf Kirch Signed-off-by: Andrew Morton Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 7662c5131b47..4ceac9ddac93 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -91,6 +91,8 @@ struct nfs_open_context { */ struct nfs_delegation; +struct posix_acl; + /* * nfs fs inode data in memory */ @@ -144,6 +146,10 @@ struct nfs_inode { atomic_t data_updates; struct nfs_access_entry cache_access; +#ifdef CONFIG_NFS_V3_ACL + struct posix_acl *acl_access; + struct posix_acl *acl_default; +#endif /* * This is the cookie verifier used for NFSv3 readdir @@ -480,6 +486,7 @@ extern int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl); extern int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, mode_t mode); +extern void nfs3_forget_cached_acls(struct inode *inode); #else static inline int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, @@ -487,6 +494,10 @@ static inline int nfs3_proc_set_default_acl(struct inode *dir, { return 0; } + +static inline void nfs3_forget_cached_acls(struct inode *inode) +{ +} #endif /* CONFIG_NFS_V3_ACL */ /* -- cgit v1.2.1 From 00a926422765064cb28e218d4837411c88bf6a3e Mon Sep 17 00:00:00 2001 From: Olivier Galibert Date: Wed, 22 Jun 2005 17:16:29 +0000 Subject: [PATCH] NFS: Hide NFS server-generated readdir cookies from userland NFSv3 currently returns the unsigned 64-bit cookie directly to userspace. The following patch causes the kernel to generate loff_t offsets for the benefit of userland. The current server-generated READDIR cookie is cached in the nfs_open_context instead of in filp->f_pos, so we still end up work correctly under directory insertions/deletion. Signed-off-by: Olivier Galibert Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 4ceac9ddac93..f810195ef7ad 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -84,6 +84,9 @@ struct nfs_open_context { int error; struct list_head list; + + int dir_pos; /* Directory cookie cache */ + __u64 dir_cookie; }; /* -- cgit v1.2.1 From f0dd2136da6d2070e12bfa6d199b136318e666c7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:29 +0000 Subject: [PATCH] NFS: Clean up readdir changes. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index f810195ef7ad..c90313bfa435 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -85,7 +85,6 @@ struct nfs_open_context { struct list_head list; - int dir_pos; /* Directory cookie cache */ __u64 dir_cookie; }; -- cgit v1.2.1 From 951a143b3fcf15cfa9d38250b7462f821db241db Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:30 +0000 Subject: [PATCH] NFS: Fix the file size revalidation Instead of looking at whether or not the file is open for writes before we accept to update the length using the server value, we should rather be looking at whether or not we are currently caching any writes. Failure to do so means in particular that we're not updating the file length correctly after obtaining a POSIX or BSD lock. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index c90313bfa435..211266c56ce5 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -294,7 +294,6 @@ extern void nfs_begin_attr_update(struct inode *); extern void nfs_end_attr_update(struct inode *); extern void nfs_begin_data_update(struct inode *); extern void nfs_end_data_update(struct inode *); -extern void nfs_end_data_update_defer(struct inode *); extern struct nfs_open_context *alloc_nfs_open_context(struct dentry *dentry, struct rpc_cred *cred); extern struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx); extern void put_nfs_open_context(struct nfs_open_context *ctx); -- cgit v1.2.1 From 7d52e86274e09fce8ac8f963e3605a84d0a305a7 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:30 +0000 Subject: [PATCH] NFS: Cleanup of caching code, and slight optimization of writes. Unless we're doing O_APPEND writes, we really don't care about revalidating the file length. Just make sure that we catch any page cache invalidations. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 211266c56ce5..443103c13e53 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -289,6 +289,7 @@ extern int nfs_release(struct inode *, struct file *); extern int nfs_attribute_timeout(struct inode *inode); extern int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode); extern int __nfs_revalidate_inode(struct nfs_server *, struct inode *); +extern void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping); extern int nfs_setattr(struct dentry *, struct iattr *); extern void nfs_begin_attr_update(struct inode *); extern void nfs_end_attr_update(struct inode *); -- cgit v1.2.1 From fe51beecc55d0b0dce289e4758e7c529a642f63e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:30 +0000 Subject: [PATCH] NFS: Ensure that fstat() always returns the correct mtime Even if the file is open for writes. Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 443103c13e53..2954e44ed498 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -198,6 +198,7 @@ struct nfs_inode { #define NFS_INO_INVALID_ATIME 0x0020 /* cached atime is invalid */ #define NFS_INO_INVALID_ACCESS 0x0040 /* cached access cred invalid */ #define NFS_INO_INVALID_ACL 0x0080 /* cached acls are invalid */ +#define NFS_INO_REVAL_PAGECACHE 0x1000 /* must revalidate pagecache */ static inline struct nfs_inode *NFS_I(struct inode *inode) { -- cgit v1.2.1 From c6a556b88adfacd2af90be84357c8165d716c27d Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:30 +0000 Subject: [PATCH] NFS: Make searching and waiting on busy writeback requests more efficient. Basically copies the VFS's method for tracking writebacks and applies it to the struct nfs_page. Signed-off-by: Trond Myklebust --- include/linux/nfs_page.h | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index 39e4895bcdb4..db40e4590ba2 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -19,6 +19,11 @@ #include +/* + * Valid flags for the radix tree + */ +#define NFS_PAGE_TAG_WRITEBACK 1 + /* * Valid flags for a dirty buffer */ @@ -62,6 +67,9 @@ extern int nfs_coalesce_requests(struct list_head *, struct list_head *, unsigned int); extern int nfs_wait_on_request(struct nfs_page *); extern void nfs_unlock_request(struct nfs_page *req); +extern int nfs_set_page_writeback_locked(struct nfs_page *req); +extern void nfs_clear_page_writeback(struct nfs_page *req); + /* * Lock the page of an asynchronous request without incrementing the wb_count @@ -96,10 +104,6 @@ nfs_list_remove_request(struct nfs_page *req) { if (list_empty(&req->wb_list)) return; - if (!NFS_WBACK_BUSY(req)) { - printk(KERN_ERR "NFS: unlocked request attempted removed from list!\n"); - BUG(); - } list_del_init(&req->wb_list); req->wb_list_head = NULL; } -- cgit v1.2.1 From 3da28eb1c6545fe73263a24eba0996217490e1eb Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:31 +0000 Subject: [PATCH] NFS: Replace nfs_page insertion sort with a radix sort Signed-off-by: Trond Myklebust --- include/linux/nfs_fs.h | 4 ++-- include/linux/nfs_page.h | 18 ++++++++++++++++-- 2 files changed, 18 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 2954e44ed498..8ea249110fb0 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -395,10 +395,10 @@ extern void nfs_commit_done(struct rpc_task *); */ extern int nfs_sync_inode(struct inode *, unsigned long, unsigned int, int); #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4) -extern int nfs_commit_inode(struct inode *, unsigned long, unsigned int, int); +extern int nfs_commit_inode(struct inode *, int); #else static inline int -nfs_commit_inode(struct inode *inode, unsigned long idx_start, unsigned int npages, int how) +nfs_commit_inode(struct inode *inode, int how) { return 0; } diff --git a/include/linux/nfs_page.h b/include/linux/nfs_page.h index db40e4590ba2..da2e077b65e2 100644 --- a/include/linux/nfs_page.h +++ b/include/linux/nfs_page.h @@ -22,6 +22,7 @@ /* * Valid flags for the radix tree */ +#define NFS_PAGE_TAG_DIRTY 0 #define NFS_PAGE_TAG_WRITEBACK 1 /* @@ -31,6 +32,7 @@ #define PG_NEED_COMMIT 1 #define PG_NEED_RESCHED 2 +struct nfs_inode; struct nfs_page { struct list_head wb_list, /* Defines state of page: */ *wb_list_head; /* read/write/commit */ @@ -59,8 +61,8 @@ extern void nfs_clear_request(struct nfs_page *req); extern void nfs_release_request(struct nfs_page *req); -extern void nfs_list_add_request(struct nfs_page *, struct list_head *); - +extern int nfs_scan_lock_dirty(struct nfs_inode *nfsi, struct list_head *dst, + unsigned long idx_start, unsigned int npages); extern int nfs_scan_list(struct list_head *, struct list_head *, unsigned long, unsigned int); extern int nfs_coalesce_requests(struct list_head *, struct list_head *, @@ -94,6 +96,18 @@ nfs_lock_request(struct nfs_page *req) return 1; } +/** + * nfs_list_add_request - Insert a request into a list + * @req: request + * @head: head of list into which to insert the request. + */ +static inline void +nfs_list_add_request(struct nfs_page *req, struct list_head *head) +{ + list_add_tail(&req->wb_list, head); + req->wb_list_head = head; +} + /** * nfs_list_remove_request - Remove a request from its wb_list -- cgit v1.2.1 From ecdbf769b2cb8903e07cd482334c714d89fd1146 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:31 +0000 Subject: [PATCH] NLM: fix a client-side race on blocking locks. If the lock blocks, the server may send us a GRANTED message that races with the reply to our LOCK request. Make sure that we catch the GRANTED by queueing up our request on the nlm_blocked list before we send off the first LOCK rpc call. Signed-off-by: Trond Myklebust --- include/linux/lockd/lockd.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h index 0d9d22578212..16d4e5a08e1d 100644 --- a/include/linux/lockd/lockd.h +++ b/include/linux/lockd/lockd.h @@ -72,6 +72,8 @@ struct nlm_lockowner { uint32_t pid; }; +struct nlm_wait; + /* * Memory chunk for NLM client RPC request. */ @@ -81,6 +83,7 @@ struct nlm_rqst { struct nlm_host * a_host; /* host handle */ struct nlm_args a_args; /* arguments */ struct nlm_res a_res; /* result */ + struct nlm_wait * a_block; char a_owner[NLMCLNT_OHSIZE]; }; @@ -142,7 +145,9 @@ extern unsigned long nlmsvc_timeout; * Lockd client functions */ struct nlm_rqst * nlmclnt_alloc_call(void); -int nlmclnt_block(struct nlm_host *, struct file_lock *, u32 *); +int nlmclnt_prepare_block(struct nlm_rqst *req, struct nlm_host *host, struct file_lock *fl); +void nlmclnt_finish_block(struct nlm_rqst *req); +long nlmclnt_block(struct nlm_rqst *req, long timeout); int nlmclnt_cancel(struct nlm_host *, struct file_lock *); u32 nlmclnt_grant(struct nlm_lock *); void nlmclnt_recovery(struct nlm_host *, u32); -- cgit v1.2.1 From 8d0a8a9d0ec790086c64d210af413ac351d89e35 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 22 Jun 2005 17:16:32 +0000 Subject: [PATCH] NFSv4: Clean up nfs4 lock state accounting Ensure that lock owner structures are not released prematurely. Signed-off-by: Trond Myklebust --- include/linux/fs.h | 1 + include/linux/nfs_fs_i.h | 5 +++++ 2 files changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 9b8b696d4f15..e5a8db00df29 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -674,6 +674,7 @@ struct file_lock { struct lock_manager_operations *fl_lmops; /* Callbacks for lockmanagers */ union { struct nfs_lock_info nfs_fl; + struct nfs4_lock_info nfs4_fl; } fl_u; }; diff --git a/include/linux/nfs_fs_i.h b/include/linux/nfs_fs_i.h index e9a749588a7b..e2c18dabff86 100644 --- a/include/linux/nfs_fs_i.h +++ b/include/linux/nfs_fs_i.h @@ -16,6 +16,11 @@ struct nfs_lock_info { struct nlm_lockowner *owner; }; +struct nfs4_lock_state; +struct nfs4_lock_info { + struct nfs4_lock_state *owner; +}; + /* * Lock flag values */ -- cgit v1.2.1 From 6ca4f65e6b390d09e1de7280cf9fd4f5d8e4b48b Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Wed, 22 Jun 2005 22:04:55 -0700 Subject: [NETPOLL]: Set poll_owner to -1 before unlocking in netpoll_poll_unlock() This trivial patch moves the assignment of poll_owner to -1 inside of the lock. This fixes a potential SMP race in the code. Signed-off-by: Jeff Moyer Signed-off-by: David S. Miller --- include/linux/netpoll.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index c0d8b90c5202..449a4fde6587 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -53,8 +53,8 @@ static inline void netpoll_poll_lock(struct net_device *dev) static inline void netpoll_poll_unlock(struct net_device *dev) { if (dev->np) { - spin_unlock(&dev->np->poll_lock); dev->np->poll_owner = -1; + spin_unlock(&dev->np->poll_lock); } } -- cgit v1.2.1 From 115c1d6e61b70851d9a363328c3b8d4c2559a1d3 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Wed, 22 Jun 2005 22:05:31 -0700 Subject: [NETPOLL]: Introduce a netpoll_info struct This patch introduces a netpoll_info structure, which the struct net_device will now point to instead of pointing to a struct netpoll. The reason for this is two-fold: 1) fields such as the rx_flags, poll_owner, and poll_lock should be maintained per net_device, not per netpoll; and 2) this is a first step in providing support for multiple netpoll clients to register against the same net_device. The struct netpoll is now pointed to by the netpoll_info structure. As such, the previous behaviour of the code is preserved. Signed-off-by: Jeff Moyer Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 ++-- include/linux/netpoll.h | 25 +++++++++++++++++-------- 2 files changed, 19 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ba5d1236aa17..d6afd440cf7b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -41,7 +41,7 @@ struct divert_blk; struct vlan_group; struct ethtool_ops; -struct netpoll; +struct netpoll_info; /* source back-compat hooks */ #define SET_ETHTOOL_OPS(netdev,ops) \ ( (netdev)->ethtool_ops = (ops) ) @@ -468,7 +468,7 @@ struct net_device unsigned char *haddr); int (*neigh_setup)(struct net_device *dev, struct neigh_parms *); #ifdef CONFIG_NETPOLL - struct netpoll *np; + struct netpoll_info *npinfo; #endif #ifdef CONFIG_NET_POLL_CONTROLLER void (*poll_controller)(struct net_device *dev); diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 449a4fde6587..388cd91bc7a6 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -16,14 +16,18 @@ struct netpoll; struct netpoll { struct net_device *dev; char dev_name[16], *name; - int rx_flags; void (*rx_hook)(struct netpoll *, int, char *, int); void (*drop)(struct sk_buff *skb); u32 local_ip, remote_ip; u16 local_port, remote_port; unsigned char local_mac[6], remote_mac[6]; +}; + +struct netpoll_info { spinlock_t poll_lock; int poll_owner; + int rx_flags; + struct netpoll *np; }; void netpoll_poll(struct netpoll *np); @@ -39,22 +43,27 @@ void netpoll_queue(struct sk_buff *skb); #ifdef CONFIG_NETPOLL static inline int netpoll_rx(struct sk_buff *skb) { - return skb->dev->np && skb->dev->np->rx_flags && __netpoll_rx(skb); + struct netpoll_info *npinfo = skb->dev->npinfo; + + if (!npinfo || !npinfo->rx_flags) + return 0; + + return npinfo->np && __netpoll_rx(skb); } static inline void netpoll_poll_lock(struct net_device *dev) { - if (dev->np) { - spin_lock(&dev->np->poll_lock); - dev->np->poll_owner = smp_processor_id(); + if (dev->npinfo) { + spin_lock(&dev->npinfo->poll_lock); + dev->npinfo->poll_owner = smp_processor_id(); } } static inline void netpoll_poll_unlock(struct net_device *dev) { - if (dev->np) { - dev->np->poll_owner = -1; - spin_unlock(&dev->np->poll_lock); + if (dev->npinfo) { + dev->npinfo->poll_owner = -1; + spin_unlock(&dev->npinfo->poll_lock); } } -- cgit v1.2.1 From fbeec2e1552949002065435c9829dc244ad85407 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Wed, 22 Jun 2005 22:05:59 -0700 Subject: [NETPOLL]: allow multiple netpoll_clients to register against one interface This patch provides support for registering multiple netpoll clients to the same network device. Only one of these clients may register an rx_hook, however. In practice, this restriction has not been problematic. It is worth mentioning, though, that the current design can be easily extended to allow for the registration of multiple rx_hooks. The basic idea of the patch is that the rx_np pointer in the netpoll_info structure points to the struct netpoll that has rx_hook filled in. Aside from this one case, there is no need for a pointer from the struct net_device to an individual struct netpoll. A lock is introduced to protect the setting and clearing of the np_rx pointer. The pointer will only be cleared upon netpoll client module removal, and the lock should be uncontested. Signed-off-by: Jeff Moyer Signed-off-by: David S. Miller --- include/linux/netpoll.h | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index 388cd91bc7a6..bcd0ac33f592 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -27,7 +27,8 @@ struct netpoll_info { spinlock_t poll_lock; int poll_owner; int rx_flags; - struct netpoll *np; + spinlock_t rx_lock; + struct netpoll *rx_np; /* netpoll that registered an rx_hook */ }; void netpoll_poll(struct netpoll *np); @@ -44,11 +45,19 @@ void netpoll_queue(struct sk_buff *skb); static inline int netpoll_rx(struct sk_buff *skb) { struct netpoll_info *npinfo = skb->dev->npinfo; + unsigned long flags; + int ret = 0; - if (!npinfo || !npinfo->rx_flags) + if (!npinfo || (!npinfo->rx_np && !npinfo->rx_flags)) return 0; - return npinfo->np && __netpoll_rx(skb); + spin_lock_irqsave(&npinfo->rx_lock, flags); + /* check rx_flags again with the lock held */ + if (npinfo->rx_flags && __netpoll_rx(skb)) + ret = 1; + spin_unlock_irqrestore(&npinfo->rx_lock, flags); + + return ret; } static inline void netpoll_poll_lock(struct net_device *dev) -- cgit v1.2.1 From cb65d506c34c86df5bcef939ce5a8666a451bd8b Mon Sep 17 00:00:00 2001 From: Shaun Pereira Date: Wed, 22 Jun 2005 22:15:01 -0700 Subject: [X25]: Selective sub-address matching with call user data. From: Shaun Pereira This is the first (independent of the second) patch of two that I am working on with x25 on linux (tested with xot on a cisco router). Details are as follows. Current state of module: A server using the current implementation (2.6.11.7) of the x25 module will accept a call request/ incoming call packet at the listening x.25 address, from all callers to that address, as long as NO call user data is present in the packet header. If the server needs to choose to accept a particular call request/ incoming call packet arriving at its listening x25 address, then the kernel has to allow a match of call user data present in the call request packet with its own. This is required when multiple servers listen at the same x25 address and device interface. The kernel currently matches ALL call user data, if present. Current Changes: This patch is a follow up to the patch submitted previously by Andrew Hendry, and allows the user to selectively control the number of octets of call user data in the call request packet, that the kernel will match. By default no call user data is matched, even if call user data is present. To allow call user data matching, a cudmatchlength > 0 has to be passed into the kernel after which the passed number of octets will be matched. Otherwise the kernel behavior is exactly as the original implementation. This patch also ensures that as is normally the case, no call user data will be present in the Call accepted / call connected packet sent back to the caller Future Changes on next patch: There are cases however when call user data may be present in the call accepted packet. According to the X.25 recommendation (ITU-T 10/96) section 5.2.3.2 call user data may be present in the call accepted packet provided the fast select facility is used. My next patch will include this fast select utility and the ability to send up to 128 octets call user data in the call accepted packet provided the fast select facility is used. I am currently testing this, again with xot on linux and cisco. Signed-off-by: Shaun Pereira (With a fix from Alexey Dobriyan ) Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- include/linux/x25.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/linux') diff --git a/include/linux/x25.h b/include/linux/x25.h index 7531cfed5885..6f43b3d20248 100644 --- a/include/linux/x25.h +++ b/include/linux/x25.h @@ -4,6 +4,8 @@ * History * mar/20/00 Daniela Squassoni Disabling/enabling of facilities * negotiation. + * apr/02/05 Shaun Pereira Selective sub address matching with + * call user data */ #ifndef X25_KERNEL_H @@ -16,6 +18,7 @@ #define SIOCX25GCALLUSERDATA (SIOCPROTOPRIVATE + 4) #define SIOCX25SCALLUSERDATA (SIOCPROTOPRIVATE + 5) #define SIOCX25GCAUSEDIAG (SIOCPROTOPRIVATE + 6) +#define SIOCX25SCUDMATCHLEN (SIOCPROTOPRIVATE + 7) /* * Values for {get,set}sockopt. @@ -109,4 +112,11 @@ struct x25_causediag { unsigned char diagnostic; }; +/* + * Further optional call user data match length selection + */ +struct x25_subaddr { + unsigned int cudmatchlength; +}; + #endif -- cgit v1.2.1 From ebc3f64b864fc16a594c2e63bf55a55c7d42084b Mon Sep 17 00:00:00 2001 From: Shaun Pereira Date: Wed, 22 Jun 2005 22:16:17 -0700 Subject: [X25]: Fast select with no restriction on response This patch is a follow up to patch 1 regarding "Selective Sub Address matching with call user data". It allows use of the Fast-Select-Acceptance optional user facility for X.25. This patch just implements fast select with no restriction on response (NRR). What this means (according to ITU-T Recomendation 10/96 section 6.16) is that if in an incoming call packet, the relevant facility bits are set for fast-select-NRR, then the called DTE can issue a direct response to the incoming packet using a call-accepted packet that contains call-user-data. This patch allows such a response. The called DTE can also respond with a clear-request packet that contains call-user-data. However, this feature is currently not implemented by the patch. How is Fast Select Acceptance used? By default, the system does not allow fast select acceptance (as before). To enable a response to fast select acceptance, After a listen socket in created and bound as follows socket(AF_X25, SOCK_SEQPACKET, 0); bind(call_soc, (struct sockaddr *)&locl_addr, sizeof(locl_addr)); but before a listen system call is made, the following ioctl should be used. ioctl(call_soc,SIOCX25CALLACCPTAPPRV); Now the listen system call can be made listen(call_soc, 4); After this, an incoming-call packet will be accepted, but no call-accepted packet will be sent back until the following system call is made on the socket that accepts the call ioctl(vc_soc,SIOCX25SENDCALLACCPT); The network (or cisco xot router used for testing here) will allow the application server's call-user-data in the call-accepted packet, provided the call-request was made with Fast-select NRR. Signed-off-by: Shaun Pereira Signed-off-by: Andrew Morton Signed-off-by: David S. Miller --- include/linux/x25.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/x25.h b/include/linux/x25.h index 6f43b3d20248..16d44931afa0 100644 --- a/include/linux/x25.h +++ b/include/linux/x25.h @@ -19,6 +19,8 @@ #define SIOCX25SCALLUSERDATA (SIOCPROTOPRIVATE + 5) #define SIOCX25GCAUSEDIAG (SIOCPROTOPRIVATE + 6) #define SIOCX25SCUDMATCHLEN (SIOCPROTOPRIVATE + 7) +#define SIOCX25CALLACCPTAPPRV (SIOCPROTOPRIVATE + 8) +#define SIOCX25SENDCALLACCPT (SIOCPROTOPRIVATE + 9) /* * Values for {get,set}sockopt. -- cgit v1.2.1 From 408fde81c1bff15c875a3618481e93a01dcc79ea Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:37 -0700 Subject: [PATCH] remove non-DISCONTIG use of pgdat->node_mem_map This patch effectively eliminates direct use of pgdat->node_mem_map outside of the DISCONTIG code. On a flat memory system, these fields aren't currently used, neither are they on a sparsemem system. There was also a node_mem_map(nid) macro on many architectures. Its use along with the use of ->node_mem_map itself was not consistent. It has been removed in favor of two new, more explicit, arch-independent macros: pgdat_page_nr(pgdat, pagenr) nid_page_nr(nid, pagenr) I called them "pgdat" and "nid" because we overload the term "node" to mean "NUMA node", "DISCONTIG node" or "pg_data_t" in very confusing ways. I believe the newer names are much clearer. These macros can be overridden in the sparsemem case with a theoretically slower operation using node_start_pfn and pfn_to_page(), instead. We could make this the only behavior if people want, but I don't want to change too much at once. One thing at a time. This patch removes more code than it adds. Compile tested on alpha, alpha discontig, arm, arm-discontig, i386, i386 generic, NUMAQ, Summit, ppc64, ppc64 discontig, and x86_64. Full list here: http://sr71.net/patches/2.6.12/2.6.12-rc1-mhp2/configs/ Boot tested on NUMAQ, x86 SMP and ppc64 power4/5 LPARs. Signed-off-by: Dave Hansen Signed-off-by: Martin J. Bligh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 4733d35d8223..b79633d3a97b 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -284,6 +284,8 @@ typedef struct pglist_data { #define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages) #define node_spanned_pages(nid) (NODE_DATA(nid)->node_spanned_pages) +#define pgdat_page_nr(pgdat, pagenr) ((pgdat)->node_mem_map + (pagenr)) +#define nid_page_nr(nid, pagenr) pgdat_page_nr(NODE_DATA(nid),(pagenr)) extern struct pglist_data *pgdat_list; -- cgit v1.2.1 From 6f167ec721108c9282d54424516a12c805e3c306 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:39 -0700 Subject: [PATCH] sparsemem base: simple NUMA remap space allocator Introduce a simple allocator for the NUMA remap space. This space is very scarce, used for structures which are best allocated node local. This mechanism is also used on non-NUMA ia64 systems with a vmem_map to keep the pgdat->node_mem_map initialized in a consistent place for all architectures. Issues: o alloc_remap takes a node_id where we might expect a pgdat which was intended to allow us to allocate the pgdat's using this mechanism; which we do not yet do. Could have alloc_remap_node() and alloc_remap_nid() for this purpose. Signed-off-by: Andy Whitcroft Signed-off-by: Dave Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bootmem.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 0dd8ca1a3d5a..500f451ce0c0 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -67,6 +67,15 @@ extern void * __init __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, __alloc_bootmem_node((pgdat), (x), PAGE_SIZE, 0) #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */ +#ifdef CONFIG_HAVE_ARCH_ALLOC_REMAP +extern void *alloc_remap(int nid, unsigned long size); +#else +static inline void *alloc_remap(int nid, unsigned long size) +{ + return NULL; +} +#endif + extern unsigned long __initdata nr_kernel_pages; extern unsigned long __initdata nr_all_pages; -- cgit v1.2.1 From 348f8b6c4837a07304d2f72b11ce8d96588065e0 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:40 -0700 Subject: [PATCH] sparsemem base: reorganize page->flags bit operations Generify the value fields in the page_flags. The aim is to allow the location and size of these fields to be varied. Additionally we want to move away from fixed allocations per field whilst still enforcing the overall bit utilisation limits. We rely on the compiler to spot and optimise the accessor functions. Signed-off-by: Andy Whitcroft Signed-off-by: Dave Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 53 +++++++++++++++++++++++++++++++++++++++++--------- include/linux/mmzone.h | 19 +++++++----------- 2 files changed, 51 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 1813b162b0a8..57b2ead51dba 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -395,19 +395,41 @@ static inline void put_page(struct page *page) /* * The zone field is never updated after free_area_init_core() * sets it, so none of the operations on it need to be atomic. - * We'll have up to (MAX_NUMNODES * MAX_NR_ZONES) zones total, - * so we use (MAX_NODES_SHIFT + MAX_ZONES_SHIFT) here to get enough bits. */ -#define NODEZONE_SHIFT (sizeof(page_flags_t)*8 - MAX_NODES_SHIFT - MAX_ZONES_SHIFT) + +/* Page flags: | NODE | ZONE | ... | FLAGS | */ +#define NODES_PGOFF ((sizeof(page_flags_t)*8) - NODES_SHIFT) +#define ZONES_PGOFF (NODES_PGOFF - ZONES_SHIFT) + +/* + * Define the bit shifts to access each section. For non-existant + * sections we define the shift as 0; that plus a 0 mask ensures + * the compiler will optimise away reference to them. + */ +#define NODES_PGSHIFT (NODES_PGOFF * (NODES_SHIFT != 0)) +#define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_SHIFT != 0)) + +/* NODE:ZONE is used to lookup the zone from a page. */ +#define ZONETABLE_SHIFT (NODES_SHIFT + ZONES_SHIFT) +#define ZONETABLE_PGSHIFT ZONES_PGSHIFT + +#if NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED +#error NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED +#endif + #define NODEZONE(node, zone) ((node << ZONES_SHIFT) | zone) +#define ZONES_MASK ((1UL << ZONES_SHIFT) - 1) +#define NODES_MASK ((1UL << NODES_SHIFT) - 1) +#define ZONETABLE_MASK ((1UL << ZONETABLE_SHIFT) - 1) + static inline unsigned long page_zonenum(struct page *page) { - return (page->flags >> NODEZONE_SHIFT) & (~(~0UL << ZONES_SHIFT)); + return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK; } static inline unsigned long page_to_nid(struct page *page) { - return (page->flags >> (NODEZONE_SHIFT + ZONES_SHIFT)); + return (page->flags >> NODES_PGSHIFT) & NODES_MASK; } struct zone; @@ -415,13 +437,26 @@ extern struct zone *zone_table[]; static inline struct zone *page_zone(struct page *page) { - return zone_table[page->flags >> NODEZONE_SHIFT]; + return zone_table[(page->flags >> ZONETABLE_PGSHIFT) & + ZONETABLE_MASK]; +} + +static inline void set_page_zone(struct page *page, unsigned long zone) +{ + page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT); + page->flags |= (zone & ZONES_MASK) << ZONES_PGSHIFT; +} +static inline void set_page_node(struct page *page, unsigned long node) +{ + page->flags &= ~(NODES_MASK << NODES_PGSHIFT); + page->flags |= (node & NODES_MASK) << NODES_PGSHIFT; } -static inline void set_page_zone(struct page *page, unsigned long nodezone_num) +static inline void set_page_links(struct page *page, unsigned long zone, + unsigned long node) { - page->flags &= ~(~0UL << NODEZONE_SHIFT); - page->flags |= nodezone_num << NODEZONE_SHIFT; + set_page_zone(page, zone); + set_page_node(page, node); } #ifndef CONFIG_DISCONTIGMEM diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index b79633d3a97b..39e912708e2a 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -414,30 +414,25 @@ extern struct pglist_data contig_page_data; #include +#endif /* !CONFIG_DISCONTIGMEM */ + #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED) /* * with 32 bit page->flags field, we reserve 8 bits for node/zone info. * there are 3 zones (2 bits) and this leaves 8-2=6 bits for nodes. */ -#define MAX_NODES_SHIFT 6 +#define FLAGS_RESERVED 8 + #elif BITS_PER_LONG == 64 /* * with 64 bit flags field, there's plenty of room. */ -#define MAX_NODES_SHIFT 10 -#endif +#define FLAGS_RESERVED 32 -#endif /* !CONFIG_DISCONTIGMEM */ - -#if NODES_SHIFT > MAX_NODES_SHIFT -#error NODES_SHIFT > MAX_NODES_SHIFT -#endif +#else -/* There are currently 3 zones: DMA, Normal & Highmem, thus we need 2 bits */ -#define MAX_ZONES_SHIFT 2 +#error BITS_PER_LONG not defined -#if ZONES_SHIFT > MAX_ZONES_SHIFT -#error ZONES_SHIFT > MAX_ZONES_SHIFT #endif #endif /* !__ASSEMBLY__ */ -- cgit v1.2.1 From 93b7504e3e6c1d98586854806e51bea329ea3aa9 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jun 2005 00:07:47 -0700 Subject: [PATCH] Introduce new Kconfig option for NUMA or DISCONTIG There is some confusion that arose when working on SPARSEMEM patch between what is needed for DISCONTIG vs. NUMA. Multiple pg_data_t's are needed for DISCONTIGMEM or NUMA, independently. All of the current NUMA implementations require an implementation of DISCONTIG. Because of this, quite a lot of code which is really needed for NUMA is actually under DISCONTIG #ifdefs. For SPARSEMEM, we changed some of these #ifdefs to CONFIG_NUMA, but that broke the DISCONTIG=y and NUMA=n case. Introducing this new NEED_MULTIPLE_NODES config option allows code that is needed for both NUMA or DISCONTIG to be separated out from code that is specific to DISCONTIG. One great advantage of this approach is that it doesn't require every architecture to be converted over. All of the current implementations should "just work", only the ones implementing SPARSEMEM will have to be fixed up. The change to free_area_init() makes it work inside, or out of the new config option. Signed-off-by: Dave Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 39e912708e2a..95f4a780ea66 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -402,7 +402,7 @@ int lowmem_reserve_ratio_sysctl_handler(struct ctl_table *, int, struct file *, /* Returns the number of the current Node. */ #define numa_node_id() (cpu_to_node(raw_smp_processor_id())) -#ifndef CONFIG_DISCONTIGMEM +#ifndef CONFIG_NEED_MULTIPLE_NODES extern struct pglist_data contig_page_data; #define NODE_DATA(nid) (&contig_page_data) @@ -410,11 +410,11 @@ extern struct pglist_data contig_page_data; #define MAX_NODES_SHIFT 1 #define pfn_to_nid(pfn) (0) -#else /* CONFIG_DISCONTIGMEM */ +#else /* CONFIG_NEED_MULTIPLE_NODES */ #include -#endif /* !CONFIG_DISCONTIGMEM */ +#endif /* !CONFIG_NEED_MULTIPLE_NODES */ #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED) /* -- cgit v1.2.1 From b159d43fbf7eaaac6ecc647f51cf4257332db47b Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:07:52 -0700 Subject: [PATCH] generify early_pfn_to_nid Provide a default implementation for early_pfn_to_nid returning node 0. Allow architectures to override this with their own implementation out of asm/mmzone.h. Signed-off-by: Andy Whitcroft Signed-off-by: Dave Hansen Signed-off-by: Martin Bligh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 95f4a780ea66..6ef07de98d69 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -435,6 +435,10 @@ extern struct pglist_data contig_page_data; #endif +#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID +#define early_pfn_to_nid(nid) (0UL) +#endif + #endif /* !__ASSEMBLY__ */ #endif /* __KERNEL__ */ #endif /* _LINUX_MMZONE_H */ -- cgit v1.2.1 From d41dee369bff3b9dcb6328d4d822926c28cc2594 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:07:54 -0700 Subject: [PATCH] sparsemem memory model Sparsemem abstracts the use of discontiguous mem_maps[]. This kind of mem_map[] is needed by discontiguous memory machines (like in the old CONFIG_DISCONTIGMEM case) as well as memory hotplug systems. Sparsemem replaces DISCONTIGMEM when enabled, and it is hoped that it can eventually become a complete replacement. A significant advantage over DISCONTIGMEM is that it's completely separated from CONFIG_NUMA. When producing this patch, it became apparent in that NUMA and DISCONTIG are often confused. Another advantage is that sparse doesn't require each NUMA node's ranges to be contiguous. It can handle overlapping ranges between nodes with no problems, where DISCONTIGMEM currently throws away that memory. Sparsemem uses an array to provide different pfn_to_page() translations for each SECTION_SIZE area of physical memory. This is what allows the mem_map[] to be chopped up. In order to do quick pfn_to_page() operations, the section number of the page is encoded in page->flags. Part of the sparsemem infrastructure enables sharing of these bits more dynamically (at compile-time) between the page_zone() and sparsemem operations. However, on 32-bit architectures, the number of bits is quite limited, and may require growing the size of the page->flags type in certain conditions. Several things might force this to occur: a decrease in the SECTION_SIZE (if you want to hotplug smaller areas of memory), an increase in the physical address space, or an increase in the number of used page->flags. One thing to note is that, once sparsemem is present, the NUMA node information no longer needs to be stored in the page->flags. It might provide speed increases on certain platforms and will be stored there if there is room. But, if out of room, an alternate (theoretically slower) mechanism is used. This patch introduces CONFIG_FLATMEM. It is used in almost all cases where there used to be an #ifndef DISCONTIG, because SPARSEMEM and DISCONTIGMEM often have to compile out the same areas of code. Signed-off-by: Andy Whitcroft Signed-off-by: Dave Hansen Signed-off-by: Martin Bligh Signed-off-by: Adrian Bunk Signed-off-by: Yasunori Goto Signed-off-by: Bob Picco Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 92 ++++++++++++++++++++++++++++++++++++++--------- include/linux/mmzone.h | 96 ++++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/numa.h | 2 +- 3 files changed, 172 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 57b2ead51dba..6eb7f48317f8 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -397,40 +397,80 @@ static inline void put_page(struct page *page) * sets it, so none of the operations on it need to be atomic. */ -/* Page flags: | NODE | ZONE | ... | FLAGS | */ -#define NODES_PGOFF ((sizeof(page_flags_t)*8) - NODES_SHIFT) -#define ZONES_PGOFF (NODES_PGOFF - ZONES_SHIFT) + +/* + * page->flags layout: + * + * There are three possibilities for how page->flags get + * laid out. The first is for the normal case, without + * sparsemem. The second is for sparsemem when there is + * plenty of space for node and section. The last is when + * we have run out of space and have to fall back to an + * alternate (slower) way of determining the node. + * + * No sparsemem: | NODE | ZONE | ... | FLAGS | + * with space for node: | SECTION | NODE | ZONE | ... | FLAGS | + * no space for node: | SECTION | ZONE | ... | FLAGS | + */ +#ifdef CONFIG_SPARSEMEM +#define SECTIONS_WIDTH SECTIONS_SHIFT +#else +#define SECTIONS_WIDTH 0 +#endif + +#define ZONES_WIDTH ZONES_SHIFT + +#if SECTIONS_WIDTH+ZONES_WIDTH+NODES_SHIFT <= FLAGS_RESERVED +#define NODES_WIDTH NODES_SHIFT +#else +#define NODES_WIDTH 0 +#endif + +/* Page flags: | [SECTION] | [NODE] | ZONE | ... | FLAGS | */ +#define SECTIONS_PGOFF ((sizeof(page_flags_t)*8) - SECTIONS_WIDTH) +#define NODES_PGOFF (SECTIONS_PGOFF - NODES_WIDTH) +#define ZONES_PGOFF (NODES_PGOFF - ZONES_WIDTH) + +/* + * We are going to use the flags for the page to node mapping if its in + * there. This includes the case where there is no node, so it is implicit. + */ +#define FLAGS_HAS_NODE (NODES_WIDTH > 0 || NODES_SHIFT == 0) + +#ifndef PFN_SECTION_SHIFT +#define PFN_SECTION_SHIFT 0 +#endif /* * Define the bit shifts to access each section. For non-existant * sections we define the shift as 0; that plus a 0 mask ensures * the compiler will optimise away reference to them. */ -#define NODES_PGSHIFT (NODES_PGOFF * (NODES_SHIFT != 0)) -#define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_SHIFT != 0)) +#define SECTIONS_PGSHIFT (SECTIONS_PGOFF * (SECTIONS_WIDTH != 0)) +#define NODES_PGSHIFT (NODES_PGOFF * (NODES_WIDTH != 0)) +#define ZONES_PGSHIFT (ZONES_PGOFF * (ZONES_WIDTH != 0)) -/* NODE:ZONE is used to lookup the zone from a page. */ +/* NODE:ZONE or SECTION:ZONE is used to lookup the zone from a page. */ +#if FLAGS_HAS_NODE #define ZONETABLE_SHIFT (NODES_SHIFT + ZONES_SHIFT) +#else +#define ZONETABLE_SHIFT (SECTIONS_SHIFT + ZONES_SHIFT) +#endif #define ZONETABLE_PGSHIFT ZONES_PGSHIFT -#if NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED -#error NODES_SHIFT+ZONES_SHIFT > FLAGS_RESERVED +#if SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > FLAGS_RESERVED +#error SECTIONS_WIDTH+NODES_WIDTH+ZONES_WIDTH > FLAGS_RESERVED #endif -#define NODEZONE(node, zone) ((node << ZONES_SHIFT) | zone) - -#define ZONES_MASK ((1UL << ZONES_SHIFT) - 1) -#define NODES_MASK ((1UL << NODES_SHIFT) - 1) +#define ZONES_MASK ((1UL << ZONES_WIDTH) - 1) +#define NODES_MASK ((1UL << NODES_WIDTH) - 1) +#define SECTIONS_MASK ((1UL << SECTIONS_WIDTH) - 1) #define ZONETABLE_MASK ((1UL << ZONETABLE_SHIFT) - 1) static inline unsigned long page_zonenum(struct page *page) { return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK; } -static inline unsigned long page_to_nid(struct page *page) -{ - return (page->flags >> NODES_PGSHIFT) & NODES_MASK; -} struct zone; extern struct zone *zone_table[]; @@ -441,6 +481,18 @@ static inline struct zone *page_zone(struct page *page) ZONETABLE_MASK]; } +static inline unsigned long page_to_nid(struct page *page) +{ + if (FLAGS_HAS_NODE) + return (page->flags >> NODES_PGSHIFT) & NODES_MASK; + else + return page_zone(page)->zone_pgdat->node_id; +} +static inline unsigned long page_to_section(struct page *page) +{ + return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK; +} + static inline void set_page_zone(struct page *page, unsigned long zone) { page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT); @@ -451,12 +503,18 @@ static inline void set_page_node(struct page *page, unsigned long node) page->flags &= ~(NODES_MASK << NODES_PGSHIFT); page->flags |= (node & NODES_MASK) << NODES_PGSHIFT; } +static inline void set_page_section(struct page *page, unsigned long section) +{ + page->flags &= ~(SECTIONS_MASK << SECTIONS_PGSHIFT); + page->flags |= (section & SECTIONS_MASK) << SECTIONS_PGSHIFT; +} static inline void set_page_links(struct page *page, unsigned long zone, - unsigned long node) + unsigned long node, unsigned long pfn) { set_page_zone(page, zone); set_page_node(page, node); + set_page_section(page, pfn_to_section_nr(pfn)); } #ifndef CONFIG_DISCONTIGMEM diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 6ef07de98d69..19860d317ec2 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -269,7 +269,9 @@ typedef struct pglist_data { struct zone node_zones[MAX_NR_ZONES]; struct zonelist node_zonelists[GFP_ZONETYPES]; int nr_zones; +#ifdef CONFIG_FLAT_NODE_MEM_MAP struct page *node_mem_map; +#endif struct bootmem_data *bdata; unsigned long node_start_pfn; unsigned long node_present_pages; /* total number of physical pages */ @@ -284,7 +286,11 @@ typedef struct pglist_data { #define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages) #define node_spanned_pages(nid) (NODE_DATA(nid)->node_spanned_pages) +#ifdef CONFIG_FLAT_NODE_MEM_MAP #define pgdat_page_nr(pgdat, pagenr) ((pgdat)->node_mem_map + (pagenr)) +#else +#define pgdat_page_nr(pgdat, pagenr) pfn_to_page((pgdat)->node_start_pfn + (pagenr)) +#endif #define nid_page_nr(nid, pagenr) pgdat_page_nr(NODE_DATA(nid),(pagenr)) extern struct pglist_data *pgdat_list; @@ -416,6 +422,10 @@ extern struct pglist_data contig_page_data; #endif /* !CONFIG_NEED_MULTIPLE_NODES */ +#ifdef CONFIG_SPARSEMEM +#include +#endif + #if BITS_PER_LONG == 32 || defined(ARCH_HAS_ATOMIC_UNSIGNED) /* * with 32 bit page->flags field, we reserve 8 bits for node/zone info. @@ -439,6 +449,92 @@ extern struct pglist_data contig_page_data; #define early_pfn_to_nid(nid) (0UL) #endif +#define pfn_to_section_nr(pfn) ((pfn) >> PFN_SECTION_SHIFT) +#define section_nr_to_pfn(sec) ((sec) << PFN_SECTION_SHIFT) + +#ifdef CONFIG_SPARSEMEM + +/* + * SECTION_SHIFT #bits space required to store a section # + * + * PA_SECTION_SHIFT physical address to/from section number + * PFN_SECTION_SHIFT pfn to/from section number + */ +#define SECTIONS_SHIFT (MAX_PHYSMEM_BITS - SECTION_SIZE_BITS) + +#define PA_SECTION_SHIFT (SECTION_SIZE_BITS) +#define PFN_SECTION_SHIFT (SECTION_SIZE_BITS - PAGE_SHIFT) + +#define NR_MEM_SECTIONS (1UL << SECTIONS_SHIFT) + +#define PAGES_PER_SECTION (1UL << PFN_SECTION_SHIFT) +#define PAGE_SECTION_MASK (~(PAGES_PER_SECTION-1)) + +#if (MAX_ORDER - 1 + PAGE_SHIFT) > SECTION_SIZE_BITS +#error Allocator MAX_ORDER exceeds SECTION_SIZE +#endif + +struct page; +struct mem_section { + struct page *section_mem_map; +}; + +extern struct mem_section mem_section[NR_MEM_SECTIONS]; + +/* + * Given a kernel address, find the home node of the underlying memory. + */ +#define kvaddr_to_nid(kaddr) pfn_to_nid(__pa(kaddr) >> PAGE_SHIFT) + +static inline struct mem_section *__pfn_to_section(unsigned long pfn) +{ + return &mem_section[pfn_to_section_nr(pfn)]; +} + +#define pfn_to_page(pfn) \ +({ \ + unsigned long __pfn = (pfn); \ + __pfn_to_section(__pfn)->section_mem_map + __pfn; \ +}) +#define page_to_pfn(page) \ +({ \ + page - mem_section[page_to_section(page)].section_mem_map; \ +}) + +static inline int pfn_valid(unsigned long pfn) +{ + if (pfn_to_section_nr(pfn) >= NR_MEM_SECTIONS) + return 0; + return mem_section[pfn_to_section_nr(pfn)].section_mem_map != 0; +} + +/* + * These are _only_ used during initialisation, therefore they + * can use __initdata ... They could have names to indicate + * this restriction. + */ +#ifdef CONFIG_NUMA +#define pfn_to_nid early_pfn_to_nid +#endif + +#define pfn_to_pgdat(pfn) \ +({ \ + NODE_DATA(pfn_to_nid(pfn)); \ +}) + +#define early_pfn_valid(pfn) pfn_valid(pfn) +void sparse_init(void); +#else +#define sparse_init() do {} while (0) +#endif /* CONFIG_SPARSEMEM */ + +#ifndef early_pfn_valid +#define early_pfn_valid(pfn) (1) +#endif + +void memory_present(int nid, unsigned long start, unsigned long end); +unsigned long __init node_memmap_size_bytes(int, unsigned long, unsigned long); + #endif /* !__ASSEMBLY__ */ #endif /* __KERNEL__ */ #endif /* _LINUX_MMZONE_H */ diff --git a/include/linux/numa.h b/include/linux/numa.h index bd0c8c4e9a95..f0c539bd3cfc 100644 --- a/include/linux/numa.h +++ b/include/linux/numa.h @@ -3,7 +3,7 @@ #include -#ifdef CONFIG_DISCONTIGMEM +#ifndef CONFIG_FLATMEM #include #endif -- cgit v1.2.1 From 641c767389b19859a45e6de46d8e18cd935bdb60 Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:07:59 -0700 Subject: [PATCH] sparsemem swiss cheese numa layouts The part of the sparsemem patch which modifies memmap_init_zone() has recently become a problem. It changes behavior so that there is a call to pfn_to_page() for each individual page inside of a node's range: node_start_pfn through node_end_pfn. It used to simply do this once, at the beginning of the node, but having sparsemem's non-contiguous mem_map[]s inside of a node made it necessary to change. Mike Kravetz recently wrote a patch which made the NUMA code accept some new kinds of layouts. The system's memory was laid out like this, with node 0's memory in two pieces: one before and one after node 1's memory: Node 0: +++++ +++++ Node 1: +++++ Previous behavior before Mike's patch was to assign nodes like this: Node 0: 00000 XXXXX Node 1: 11111 Where the 'X' areas were simply thrown away. The new behavior was to make the pg_data_t span node 0 across all of its areas, including areas that are really node 1's: Node 0: 000000000000000 Node 1: 11111 This wastes a little bit of mem_map space, but ends up being OK, and more fully utilizes the system's memory. memmap_init_zone() initializes all of the "struct page"s for node 0, even for the "hole", but those never get used, because there is no pfn_to_page() that resolves to those pages. However, only calling pfn_to_page() once, memmap_init_zone() always uses the pages that were allocated for node0->node_mem_map because: struct page *start = pfn_to_page(start_pfn); // effectively start = &node->node_mem_map[0] for (page = start; page < (start + size); page++) { init_page_here();... page++; } Slow, and wasteful, but generally harmless. But, modify that to call pfn_to_page() for each loop iteration (like sparsemem does): for (pfn = start_pfn; pfn < < (start_pfn + size); pfn++++) { page = pfn_to_page(pfn); } And you end up trying to initialize node 1's pages too early, along with bogus data from node 0. This patch checks for those weird layouts and declines to touch the pages, making the more frequent pfn_to_page() calls OK to do. Signed-off-by: Dave Hansen Signed-off-by: Andy Whitcroft Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 19860d317ec2..746b57e3d370 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -528,6 +528,12 @@ void sparse_init(void); #define sparse_init() do {} while (0) #endif /* CONFIG_SPARSEMEM */ +#ifdef CONFIG_NODES_SPAN_OTHER_NODES +#define early_pfn_in_nid(pfn, nid) (early_pfn_to_nid(pfn) == (nid)) +#else +#define early_pfn_in_nid(pfn, nid) (1) +#endif + #ifndef early_pfn_valid #define early_pfn_valid(pfn) (1) #endif -- cgit v1.2.1 From 29751f6991e845f7d002a6ae520bf996b38c8dcd Mon Sep 17 00:00:00 2001 From: Andy Whitcroft Date: Thu, 23 Jun 2005 00:08:00 -0700 Subject: [PATCH] sparsemem hotplug base Make sparse's initalization be accessible at runtime. This allows sparse mappings to be created after boot in a hotplug situation. This patch is separated from the previous one just to give an indication how much of the sparse infrastructure is *just* for hotplug memory. The section_mem_map doesn't really store a pointer. It stores something that is convenient to do some math against to get a pointer. It isn't valid to just do *section_mem_map, so I don't think it should be stored as a pointer. There are a couple of things I'd like to store about a section. First of all, the fact that it is !NULL does not mean that it is present. There could be such a combination where section_mem_map *is* NULL, but the math gets you properly to a real mem_map. So, I don't think that check is safe. Since we're storing 32-bit-aligned structures, we have a few bits in the bottom of the pointer to play with. Use one bit to encode whether there's really a mem_map there, and the other one to tell whether there's a valid section there. We need to distinguish between the two because sometimes there's a gap between when a section is discovered to be present and when we can get the mem_map for it. Signed-off-by: Dave Hansen Signed-off-by: Andy Whitcroft Signed-off-by: Jack Steiner Signed-off-by: Bob Picco Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 56 +++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 51 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 746b57e3d370..6c90461ed99f 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -476,11 +476,56 @@ extern struct pglist_data contig_page_data; struct page; struct mem_section { - struct page *section_mem_map; + /* + * This is, logically, a pointer to an array of struct + * pages. However, it is stored with some other magic. + * (see sparse.c::sparse_init_one_section()) + * + * Making it a UL at least makes someone do a cast + * before using it wrong. + */ + unsigned long section_mem_map; }; extern struct mem_section mem_section[NR_MEM_SECTIONS]; +static inline struct mem_section *__nr_to_section(unsigned long nr) +{ + return &mem_section[nr]; +} + +/* + * We use the lower bits of the mem_map pointer to store + * a little bit of information. There should be at least + * 3 bits here due to 32-bit alignment. + */ +#define SECTION_MARKED_PRESENT (1UL<<0) +#define SECTION_HAS_MEM_MAP (1UL<<1) +#define SECTION_MAP_LAST_BIT (1UL<<2) +#define SECTION_MAP_MASK (~(SECTION_MAP_LAST_BIT-1)) + +static inline struct page *__section_mem_map_addr(struct mem_section *section) +{ + unsigned long map = section->section_mem_map; + map &= SECTION_MAP_MASK; + return (struct page *)map; +} + +static inline int valid_section(struct mem_section *section) +{ + return (section->section_mem_map & SECTION_MARKED_PRESENT); +} + +static inline int section_has_mem_map(struct mem_section *section) +{ + return (section->section_mem_map & SECTION_HAS_MEM_MAP); +} + +static inline int valid_section_nr(unsigned long nr) +{ + return valid_section(__nr_to_section(nr)); +} + /* * Given a kernel address, find the home node of the underlying memory. */ @@ -488,24 +533,25 @@ extern struct mem_section mem_section[NR_MEM_SECTIONS]; static inline struct mem_section *__pfn_to_section(unsigned long pfn) { - return &mem_section[pfn_to_section_nr(pfn)]; + return __nr_to_section(pfn_to_section_nr(pfn)); } #define pfn_to_page(pfn) \ ({ \ unsigned long __pfn = (pfn); \ - __pfn_to_section(__pfn)->section_mem_map + __pfn; \ + __section_mem_map_addr(__pfn_to_section(__pfn)) + __pfn; \ }) #define page_to_pfn(page) \ ({ \ - page - mem_section[page_to_section(page)].section_mem_map; \ + page - __section_mem_map_addr(__nr_to_section( \ + page_to_section(page))); \ }) static inline int pfn_valid(unsigned long pfn) { if (pfn_to_section_nr(pfn) >= NR_MEM_SECTIONS) return 0; - return mem_section[pfn_to_section_nr(pfn)].section_mem_map != 0; + return valid_section(__nr_to_section(pfn_to_section_nr(pfn))); } /* -- cgit v1.2.1 From 1946089a109251655c5438d92c539bd2930e71ea Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 23 Jun 2005 00:08:19 -0700 Subject: [PATCH] NUMA aware block device control structure allocation Patch to allocate the control structures for for ide devices on the node of the device itself (for NUMA systems). The patch depends on the Slab API change patch by Manfred and me (in mm) and the pcidev_to_node patch that I posted today. Does some realignment too. Signed-off-by: Justin M. Forbes Signed-off-by: Christoph Lameter Signed-off-by: Pravin Shelar Signed-off-by: Shobhit Dayal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/blkdev.h | 6 +++++- include/linux/genhd.h | 1 + include/linux/ide.h | 2 +- include/linux/mempool.h | 11 ++++++++--- 4 files changed, 15 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 4a99b76c5a33..235c3414d268 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -396,6 +396,7 @@ struct request_queue */ unsigned int sg_timeout; unsigned int sg_reserved_size; + int node; struct list_head drain_list; @@ -615,6 +616,8 @@ static inline void blkdev_dequeue_request(struct request *req) /* * Access functions for manipulating queue properties */ +extern request_queue_t *blk_init_queue_node(request_fn_proc *rfn, + spinlock_t *lock, int node_id); extern request_queue_t *blk_init_queue(request_fn_proc *, spinlock_t *); extern void blk_cleanup_queue(request_queue_t *); extern void blk_queue_make_request(request_queue_t *, make_request_fn *); @@ -646,7 +649,8 @@ extern void blk_wait_queue_drained(request_queue_t *, int); extern void blk_finish_queue_drain(request_queue_t *); int blk_get_queue(request_queue_t *); -request_queue_t *blk_alloc_queue(int); +request_queue_t *blk_alloc_queue(int gfp_mask); +request_queue_t *blk_alloc_queue_node(int,int); #define blk_put_queue(q) blk_cleanup_queue((q)) /* diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 47dedaf971d6..af26dc718ef6 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -403,6 +403,7 @@ extern int rescan_partitions(struct gendisk *disk, struct block_device *bdev); extern void add_partition(struct gendisk *, int, sector_t, sector_t); extern void delete_partition(struct gendisk *, int); +extern struct gendisk *alloc_disk_node(int minors, int node_id); extern struct gendisk *alloc_disk(int minors); extern struct kobject *get_disk(struct gendisk *disk); extern void put_disk(struct gendisk *disk); diff --git a/include/linux/ide.h b/include/linux/ide.h index 336d6e509f59..92129078d4f3 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -917,7 +917,7 @@ typedef struct hwif_s { unsigned dma; void (*led_act)(void *data, int rw); -} ide_hwif_t; +} ____cacheline_maxaligned_in_smp ide_hwif_t; /* * internal ide interrupt handler type diff --git a/include/linux/mempool.h b/include/linux/mempool.h index 4a36edf1c974..796220ce47cc 100644 --- a/include/linux/mempool.h +++ b/include/linux/mempool.h @@ -20,9 +20,14 @@ typedef struct mempool_s { mempool_free_t *free; wait_queue_head_t wait; } mempool_t; -extern mempool_t * mempool_create(int min_nr, mempool_alloc_t *alloc_fn, - mempool_free_t *free_fn, void *pool_data); -extern int mempool_resize(mempool_t *pool, int new_min_nr, unsigned int __nocast gfp_mask); + +extern mempool_t *mempool_create(int min_nr, mempool_alloc_t *alloc_fn, + mempool_free_t *free_fn, void *pool_data); +extern mempool_t *mempool_create_node(int min_nr, mempool_alloc_t *alloc_fn, + mempool_free_t *free_fn, void *pool_data, int nid); + +extern int mempool_resize(mempool_t *pool, int new_min_nr, + unsigned int __nocast gfp_mask); extern void mempool_destroy(mempool_t *pool); extern void * mempool_alloc(mempool_t *pool, unsigned int __nocast gfp_mask); extern void mempool_free(void *element, mempool_t *pool); -- cgit v1.2.1 From fa72b903f75e4f0f0b2c2feed093005167da4023 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 23 Jun 2005 00:08:49 -0700 Subject: [PATCH] blk: remove blk_queue_tag->real_max_depth optimization blk_queue_tag->real_max_depth was used to optimize out unnecessary allocations/frees on tag resize. However, the whole thing was very broken - tag_map was never allocated to real_max_depth resulting in access beyond the end of the map, bits in [max_depth..real_max_depth] were set when initializing a map and copied when resizing resulting in pre-occupied tags. As the gain of the optimization is very small, well, almost nill, remove the whole thing. Signed-off-by: Tejun Heo Acked-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/blkdev.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 235c3414d268..8d7e2f4151d0 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -294,7 +294,6 @@ struct blk_queue_tag { struct list_head busy_list; /* fifo list of busy tags */ int busy; /* current depth */ int max_depth; /* what we will send to device */ - int real_max_depth; /* what the array can hold */ atomic_t refcnt; /* map can be shared */ }; -- cgit v1.2.1 From f7d37d028dfba90b1b747f8ac685bf0959aeda8b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 23 Jun 2005 00:08:50 -0700 Subject: [PATCH] blk: remove BLK_TAGS_{PER_LONG|MASK} Replace BLK_TAGS_PER_LONG with BITS_PER_LONG and remove unused BLK_TAGS_MASK. Signed-off-by: Tejun Heo Acked-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/blkdev.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 8d7e2f4151d0..60272141ff19 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -285,9 +285,6 @@ enum blk_queue_state { Queue_up, }; -#define BLK_TAGS_PER_LONG (sizeof(unsigned long) * 8) -#define BLK_TAGS_MASK (BLK_TAGS_PER_LONG - 1) - struct blk_queue_tag { struct request **tag_index; /* map of busy tags */ unsigned long *tag_map; /* bit map of free/busy tags */ -- cgit v1.2.1 From 55c888d6d09a0df236adfaf8ccf06ff5d0646775 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 23 Jun 2005 00:08:56 -0700 Subject: [PATCH] timers fixes/improvements This patch tries to solve following problems: 1. del_timer_sync() is racy. The timer can be fired again after del_timer_sync have checked all cpus and before it will recheck timer_pending(). 2. It has scalability problems. All cpus are scanned to determine if the timer is running on that cpu. With this patch del_timer_sync is O(1) and no slower than plain del_timer(pending_timer), unless it has to actually wait for completion of the currently running timer. The only restriction is that the recurring timer should not use add_timer_on(). 3. The timers are not serialized wrt to itself. If CPU_0 does mod_timer(jiffies+1) while the timer is currently running on CPU 1, it is quite possible that local interrupt on CPU_0 will start that timer before it finished on CPU_1. 4. The timers locking is suboptimal. __mod_timer() takes 3 locks at once and still requires wmb() in del_timer/run_timers. The new implementation takes 2 locks sequentially and does not need memory barriers. Currently ->base != NULL means that the timer is pending. In that case ->base.lock is used to lock the timer. __mod_timer also takes timer->lock because ->base can be == NULL. This patch uses timer->entry.next != NULL as indication that the timer is pending. So it does __list_del(), entry->next = NULL instead of list_del() when the timer is deleted. The ->base field is used for hashed locking only, it is initialized in init_timer() which sets ->base = per_cpu(tvec_bases). When the tvec_bases.lock is locked, it means that all timers which are tied to this base via timer->base are locked, and the base itself is locked too. So __run_timers/migrate_timers can safely modify all timers which could be found on ->tvX lists (pending timers). When the timer's base is locked, and the timer removed from ->entry list (which means that _run_timers/migrate_timers can't see this timer), it is possible to set timer->base = NULL and drop the lock: the timer remains locked. This patch adds lock_timer_base() helper, which waits for ->base != NULL, locks the ->base, and checks it is still the same. __mod_timer() schedules the timer on the local CPU and changes it's base. However, it does not lock both old and new bases at once. It locks the timer via lock_timer_base(), deletes the timer, sets ->base = NULL, and unlocks old base. Then __mod_timer() locks new_base, sets ->base = new_base, and adds this timer. This simplifies the code, because AB-BA deadlock is not possible. __mod_timer() also ensures that the timer's base is not changed while the timer's handler is running on the old base. __run_timers(), del_timer() do not change ->base anymore, they only clear pending flag. So del_timer_sync() can test timer->base->running_timer == timer to detect whether it is running or not. We don't need timer_list->lock anymore, this patch kills it. We also don't need barriers. del_timer() and __run_timers() used smp_wmb() before clearing timer's pending flag. It was needed because __mod_timer() did not lock old_base if the timer is not pending, so __mod_timer()->list_add() could race with del_timer()->list_del(). With this patch these functions are serialized through base->lock. One problem. TIMER_INITIALIZER can't use per_cpu(tvec_bases). So this patch adds global struct timer_base_s { spinlock_t lock; struct timer_list *running_timer; } __init_timer_base; which is used by TIMER_INITIALIZER. The corresponding fields in tvec_t_base_s struct are replaced by struct timer_base_s t_base. It is indeed ugly. But this can't have scalability problems. The global __init_timer_base.lock is used only when __mod_timer() is called for the first time AND the timer was compile time initialized. After that the timer migrates to the local CPU. Signed-off-by: Oleg Nesterov Acked-by: Ingo Molnar Signed-off-by: Renaud Lienhart Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/timer.h | 30 +++++++++--------------------- 1 file changed, 9 insertions(+), 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/timer.h b/include/linux/timer.h index 90db1cc62ddd..2e78fedfc069 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -6,45 +6,33 @@ #include #include -struct tvec_t_base_s; +struct timer_base_s; struct timer_list { struct list_head entry; unsigned long expires; - spinlock_t lock; unsigned long magic; void (*function)(unsigned long); unsigned long data; - struct tvec_t_base_s *base; + struct timer_base_s *base; }; #define TIMER_MAGIC 0x4b87ad6e +extern struct timer_base_s __init_timer_base; + #define TIMER_INITIALIZER(_function, _expires, _data) { \ .function = (_function), \ .expires = (_expires), \ .data = (_data), \ - .base = NULL, \ + .base = &__init_timer_base, \ .magic = TIMER_MAGIC, \ - .lock = SPIN_LOCK_UNLOCKED, \ } -/*** - * init_timer - initialize a timer. - * @timer: the timer to be initialized - * - * init_timer() must be done to a timer prior calling *any* of the - * other timer functions. - */ -static inline void init_timer(struct timer_list * timer) -{ - timer->base = NULL; - timer->magic = TIMER_MAGIC; - spin_lock_init(&timer->lock); -} +void fastcall init_timer(struct timer_list * timer); /*** * timer_pending - is a timer pending? @@ -58,7 +46,7 @@ static inline void init_timer(struct timer_list * timer) */ static inline int timer_pending(const struct timer_list * timer) { - return timer->base != NULL; + return timer->entry.next != NULL; } extern void add_timer_on(struct timer_list *timer, int cpu); @@ -89,12 +77,12 @@ static inline void add_timer(struct timer_list * timer) #ifdef CONFIG_SMP extern int del_timer_sync(struct timer_list *timer); - extern int del_singleshot_timer_sync(struct timer_list *timer); #else # define del_timer_sync(t) del_timer(t) -# define del_singleshot_timer_sync(t) del_timer(t) #endif +#define del_singleshot_timer_sync(t) del_timer_sync(t) + extern void init_timers(void); extern void run_local_timers(void); extern void it_real_fn(unsigned long); -- cgit v1.2.1 From fd450b7318b75343fd76b3d95416853e34e72c95 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Thu, 23 Jun 2005 00:08:59 -0700 Subject: [PATCH] timers: introduce try_to_del_timer_sync() This patch splits del_timer_sync() into 2 functions. The new one, try_to_del_timer_sync(), returns -1 when it hits executing timer. It can be used in interrupt context, or when the caller hold locks which can prevent completion of the timer's handler. NOTE. Currently it can't be used in interrupt context in UP case, because ->running_timer is used only with CONFIG_SMP. Should the need arise, it is possible to kill #ifdef CONFIG_SMP in set_running_timer(), it is cheap. Signed-off-by: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/timer.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/timer.h b/include/linux/timer.h index 2e78fedfc069..221f81ac2002 100644 --- a/include/linux/timer.h +++ b/include/linux/timer.h @@ -76,9 +76,11 @@ static inline void add_timer(struct timer_list * timer) } #ifdef CONFIG_SMP + extern int try_to_del_timer_sync(struct timer_list *timer); extern int del_timer_sync(struct timer_list *timer); #else -# define del_timer_sync(t) del_timer(t) +# define try_to_del_timer_sync(t) del_timer(t) +# define del_timer_sync(t) del_timer(t) #endif #define del_singleshot_timer_sync(t) del_timer_sync(t) -- cgit v1.2.1 From 991114c6fa6a21d1fa4d544abe78592352860c82 Mon Sep 17 00:00:00 2001 From: Alexander Viro Date: Thu, 23 Jun 2005 00:09:01 -0700 Subject: [PATCH] fix for prune_icache()/forced final iput() races Based on analysis and a patch from Russ Weight There is a race condition that can occur if an inode is allocated and then released (using iput) during the ->fill_super functions. The race condition is between kswapd and mount. For most filesystems this can only happen in an error path when kswapd is running concurrently. For isofs, however, the error can occur in a more common code path (which is how the bug was found). The logic here is "we want final iput() to free inode *now* instead of letting it sit in cache if fs is going down or had not quite come up". The problem is with kswapd seeing such inodes in the middle of being killed and happily taking over. The clean solution would be to tell kswapd to leave those inodes alone and let our final iput deal with them. I.e. add a new flag (I_FORCED_FREEING), set it before write_inode_now() there and make prune_icache() leave those alone. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index e5a8db00df29..3622e952e98c 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1025,6 +1025,7 @@ struct super_operations { #define I_FREEING 16 #define I_CLEAR 32 #define I_NEW 64 +#define I_WILL_FREE 128 #define I_DIRTY (I_DIRTY_SYNC | I_DIRTY_DATASYNC | I_DIRTY_PAGES) -- cgit v1.2.1 From 543537bd922692bc978e2e356fcd8bfc9c2ee7d5 Mon Sep 17 00:00:00 2001 From: Paulo Marques Date: Thu, 23 Jun 2005 00:09:02 -0700 Subject: [PATCH] create a kstrdup library function This patch creates a new kstrdup library function and changes the "local" implementations in several places to use this function. Most of the changes come from the sound and net subsystems. The sound part had already been acknowledged by Takashi Iwai and the net part by David S. Miller. I left UML alone for now because I would need more time to read the code carefully before making changes there. Signed-off-by: Paulo Marques Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/netdevice.h | 4 ---- include/linux/string.h | 2 ++ 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d6afd440cf7b..d89816ad642f 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -925,10 +925,6 @@ extern int skb_checksum_help(struct sk_buff *skb, int inward); extern void net_enable_timestamp(void); extern void net_disable_timestamp(void); -#ifdef CONFIG_SYSCTL -extern char *net_sysctl_strdup(const char *s); -#endif - #endif /* __KERNEL__ */ #endif /* _LINUX_DEV_H */ diff --git a/include/linux/string.h b/include/linux/string.h index b9fc59469956..93994c613095 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -88,6 +88,8 @@ extern int memcmp(const void *,const void *,__kernel_size_t); extern void * memchr(const void *,int,__kernel_size_t); #endif +extern char *kstrdup(const char *s, int gfp); + #ifdef __cplusplus } #endif -- cgit v1.2.1 From 35a82d1a53e1a9ad54efafcc940f9335beaed5c3 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Thu, 23 Jun 2005 00:09:06 -0700 Subject: [PATCH] optimise loop driver a bit Looks like locking can be optimised quite a lot. Increase lock widths slightly so lo_lock is taken fewer times per request. Also it was quite trivial to cover lo_pending with that lock, and remove the atomic requirement. This also makes memory ordering explicitly correct, which is nice (not that I particularly saw any mem ordering bugs). Test was reading 4 250MB files in parallel on ext2-on-tmpfs filesystem (1K block size, 4K page size). System is 2 socket Xeon with HT (4 thread). intel:/home/npiggin# umount /dev/loop0 ; mount /dev/loop0 /mnt/loop ; /usr/bin/time ./mtloop.sh Before: 0.24user 5.51system 0:02.84elapsed 202%CPU (0avgtext+0avgdata 0maxresident)k 0.19user 5.52system 0:02.88elapsed 198%CPU (0avgtext+0avgdata 0maxresident)k 0.19user 5.57system 0:02.89elapsed 198%CPU (0avgtext+0avgdata 0maxresident)k 0.22user 5.51system 0:02.90elapsed 197%CPU (0avgtext+0avgdata 0maxresident)k 0.19user 5.44system 0:02.91elapsed 193%CPU (0avgtext+0avgdata 0maxresident)k After: 0.07user 2.34system 0:01.68elapsed 143%CPU (0avgtext+0avgdata 0maxresident)k 0.06user 2.37system 0:01.68elapsed 144%CPU (0avgtext+0avgdata 0maxresident)k 0.06user 2.39system 0:01.68elapsed 145%CPU (0avgtext+0avgdata 0maxresident)k 0.06user 2.36system 0:01.68elapsed 144%CPU (0avgtext+0avgdata 0maxresident)k 0.06user 2.42system 0:01.68elapsed 147%CPU (0avgtext+0avgdata 0maxresident)k Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/loop.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/loop.h b/include/linux/loop.h index 8220d9c9da00..53fa51595443 100644 --- a/include/linux/loop.h +++ b/include/linux/loop.h @@ -61,7 +61,7 @@ struct loop_device { struct semaphore lo_sem; struct semaphore lo_ctl_mutex; struct semaphore lo_bh_mutex; - atomic_t lo_pending; + int lo_pending; request_queue_t *lo_queue; }; -- cgit v1.2.1 From ac20427ef6aa63da663bdc88b71d16f7394f5e23 Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Thu, 23 Jun 2005 00:09:11 -0700 Subject: [PATCH] add check to /proc/devices read routines Patch to add check to get_chrdev_list and get_blkdev_list to prevent reads of /proc/devices from spilling over the provided page if more than 4096 bytes of string data are generated from all the registered character and block devices in a system Signed-off-by: Neil Horman Cc: Christoph Hellwig Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/genhd.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/genhd.h b/include/linux/genhd.h index af26dc718ef6..01796c41c951 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -224,7 +224,7 @@ static inline void free_disk_stats(struct gendisk *disk) extern void disk_round_stats(struct gendisk *disk); /* drivers/block/genhd.c */ -extern int get_blkdev_list(char *); +extern int get_blkdev_list(char *, int); extern void add_disk(struct gendisk *disk); extern void del_gendisk(struct gendisk *gp); extern void unlink_gendisk(struct gendisk *gp); -- cgit v1.2.1 From 84de856ed30c568c2bb7b9ac0679772bd2737d9b Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Jun 2005 00:09:16 -0700 Subject: [PATCH] quota: consolidate code surrounding vfs_quota_on_mount Move some code duplicated in both callers into vfs_quota_on_mount Signed-off-by: Christoph Hellwig Acked-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/quotaops.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h index e57baa85e744..d211507ab246 100644 --- a/include/linux/quotaops.h +++ b/include/linux/quotaops.h @@ -39,7 +39,8 @@ extern int dquot_commit_info(struct super_block *sb, int type); extern int dquot_mark_dquot_dirty(struct dquot *dquot); extern int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path); -extern int vfs_quota_on_mount(int type, int format_id, struct dentry *dentry); +extern int vfs_quota_on_mount(struct super_block *sb, char *qf_name, + int format_id, int type); extern int vfs_quota_off(struct super_block *sb, int type); #define vfs_quota_off_mount(sb, type) vfs_quota_off(sb, type) extern int vfs_quota_sync(struct super_block *sb, int type); -- cgit v1.2.1 From b94cce926b2b902b79380ccba370d6f9f2980de0 Mon Sep 17 00:00:00 2001 From: Hien Nguyen Date: Thu, 23 Jun 2005 00:09:19 -0700 Subject: [PATCH] kprobes: function-return probes This patch adds function-return probes to kprobes for the i386 architecture. This enables you to establish a handler to be run when a function returns. 1. API Two new functions are added to kprobes: int register_kretprobe(struct kretprobe *rp); void unregister_kretprobe(struct kretprobe *rp); 2. Registration and unregistration 2.1 Register To register a function-return probe, the user populates the following fields in a kretprobe object and calls register_kretprobe() with the kretprobe address as an argument: kp.addr - the function's address handler - this function is run after the ret instruction executes, but before control returns to the return address in the caller. maxactive - The maximum number of instances of the probed function that can be active concurrently. For example, if the function is non- recursive and is called with a spinlock or mutex held, maxactive = 1 should be enough. If the function is non-recursive and can never relinquish the CPU (e.g., via a semaphore or preemption), NR_CPUS should be enough. maxactive is used to determine how many kretprobe_instance objects to allocate for this particular probed function. If maxactive <= 0, it is set to a default value (if CONFIG_PREEMPT maxactive=max(10, 2 * NR_CPUS) else maxactive=NR_CPUS) For example: struct kretprobe rp; rp.kp.addr = /* entrypoint address */ rp.handler = /*return probe handler */ rp.maxactive = /* e.g., 1 or NR_CPUS or 0, see the above explanation */ register_kretprobe(&rp); The following field may also be of interest: nmissed - Initialized to zero when the function-return probe is registered, and incremented every time the probed function is entered but there is no kretprobe_instance object available for establishing the function-return probe (i.e., because maxactive was set too low). 2.2 Unregister To unregiter a function-return probe, the user calls unregister_kretprobe() with the same kretprobe object as registered previously. If a probed function is running when the return probe is unregistered, the function will return as expected, but the handler won't be run. 3. Limitations 3.1 This patch supports only the i386 architecture, but patches for x86_64 and ppc64 are anticipated soon. 3.2 Return probes operates by replacing the return address in the stack (or in a known register, such as the lr register for ppc). This may cause __builtin_return_address(0), when invoked from the return-probed function, to return the address of the return-probes trampoline. 3.3 This implementation uses the "Multiprobes at an address" feature in 2.6.12-rc3-mm3. 3.4 Due to a limitation in multi-probes, you cannot currently establish a return probe and a jprobe on the same function. A patch to remove this limitation is being tested. This feature is required by SystemTap (http://sourceware.org/systemtap), and reflects ideas contributed by several SystemTap developers, including Will Cohen and Ananth Mavinakayanahalli. Signed-off-by: Hien Nguyen Signed-off-by: Prasanna S Panchamukhi Signed-off-by: Frederik Deweerdt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kprobes.h | 90 +++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 88 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 99ddba5a4e00..fba39f87efec 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -25,21 +25,31 @@ * Rusty Russell). * 2004-July Suparna Bhattacharya added jumper probes * interface to access function arguments. + * 2005-May Hien Nguyen and Jim Keniston + * and Prasanna S Panchamukhi + * added function-return probes. */ #include #include #include #include +#include + #include struct kprobe; struct pt_regs; +struct kretprobe; +struct kretprobe_instance; typedef int (*kprobe_pre_handler_t) (struct kprobe *, struct pt_regs *); typedef int (*kprobe_break_handler_t) (struct kprobe *, struct pt_regs *); typedef void (*kprobe_post_handler_t) (struct kprobe *, struct pt_regs *, unsigned long flags); typedef int (*kprobe_fault_handler_t) (struct kprobe *, struct pt_regs *, int trapnr); +typedef int (*kretprobe_handler_t) (struct kretprobe_instance *, + struct pt_regs *); + struct kprobe { struct hlist_node hlist; @@ -85,6 +95,62 @@ struct jprobe { kprobe_opcode_t *entry; /* probe handling code to jump to */ }; +#ifdef ARCH_SUPPORTS_KRETPROBES +extern int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs); +extern void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs, + unsigned long flags); +extern struct task_struct *arch_get_kprobe_task(void *ptr); +extern void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs); +extern void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock); +#else /* ARCH_SUPPORTS_KRETPROBES */ +static inline void kretprobe_trampoline(void) +{ +} +static inline int trampoline_probe_handler(struct kprobe *p, + struct pt_regs *regs) +{ + return 0; +} +static inline void trampoline_post_handler(struct kprobe *p, + struct pt_regs *regs, unsigned long flags) +{ +} +static inline void arch_prepare_kretprobe(struct kretprobe *rp, + struct pt_regs *regs) +{ +} +static inline void arch_kprobe_flush_task(struct task_struct *tk) +{ +} +#define arch_get_kprobe_task(ptr) ((struct task_struct *)NULL) +#endif /* ARCH_SUPPORTS_KRETPROBES */ +/* + * Function-return probe - + * Note: + * User needs to provide a handler function, and initialize maxactive. + * maxactive - The maximum number of instances of the probed function that + * can be active concurrently. + * nmissed - tracks the number of times the probed function's return was + * ignored, due to maxactive being too low. + * + */ +struct kretprobe { + struct kprobe kp; + kretprobe_handler_t handler; + int maxactive; + int nmissed; + struct hlist_head free_instances; + struct hlist_head used_instances; +}; + +struct kretprobe_instance { + struct hlist_node uflist; /* either on free list or used list */ + struct hlist_node hlist; + struct kretprobe *rp; + void *ret_addr; + void *stack_addr; +}; + #ifdef CONFIG_KPROBES /* Locks kprobe: irq must be disabled */ void lock_kprobes(void); @@ -104,6 +170,7 @@ extern void show_registers(struct pt_regs *regs); /* Get the kprobe at this addr (if any). Must have called lock_kprobes */ struct kprobe *get_kprobe(void *addr); +struct hlist_head * kretprobe_inst_table_head(struct task_struct *tsk); int register_kprobe(struct kprobe *p); void unregister_kprobe(struct kprobe *p); @@ -113,7 +180,16 @@ int register_jprobe(struct jprobe *p); void unregister_jprobe(struct jprobe *p); void jprobe_return(void); -#else +int register_kretprobe(struct kretprobe *rp); +void unregister_kretprobe(struct kretprobe *rp); + +struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp); +struct kretprobe_instance *get_rp_inst(void *sara); +struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk); +void add_rp_inst(struct kretprobe_instance *ri); +void kprobe_flush_task(struct task_struct *tk); +void recycle_rp_inst(struct kretprobe_instance *ri); +#else /* CONFIG_KPROBES */ static inline int kprobe_running(void) { return 0; @@ -135,5 +211,15 @@ static inline void unregister_jprobe(struct jprobe *p) static inline void jprobe_return(void) { } -#endif +static inline int register_kretprobe(struct kretprobe *rp) +{ + return -ENOSYS; +} +static inline void unregister_kretprobe(struct kretprobe *rp) +{ +} +static inline void kprobe_flush_task(struct task_struct *tk) +{ +} +#endif /* CONFIG_KPROBES */ #endif /* _LINUX_KPROBES_H */ -- cgit v1.2.1 From 7e1048b11c5afe79aac46a42e3ccec86b8365c6d Mon Sep 17 00:00:00 2001 From: Rusty Lynch Date: Thu, 23 Jun 2005 00:09:25 -0700 Subject: [PATCH] Move kprobe [dis]arming into arch specific code The architecture independent code of the current kprobes implementation is arming and disarming kprobes at registration time. The problem is that the code is assuming that arming and disarming is a just done by a simple write of some magic value to an address. This is problematic for ia64 where our instructions look more like structures, and we can not insert break points by just doing something like: *p->addr = BREAKPOINT_INSTRUCTION; The following patch to 2.6.12-rc4-mm2 adds two new architecture dependent functions: * void arch_arm_kprobe(struct kprobe *p) * void arch_disarm_kprobe(struct kprobe *p) and then adds the new functions for each of the architectures that already implement kprobes (spar64/ppc64/i386/x86_64). I thought arch_[dis]arm_kprobe was the most descriptive of what was really happening, but each of the architectures already had a disarm_kprobe() function that was really a "disarm and do some other clean-up items as needed when you stumble across a recursive kprobe." So... I took the liberty of changing the code that was calling disarm_kprobe() to call arch_disarm_kprobe(), and then do the cleanup in the block of code dealing with the recursive kprobe case. So far this patch as been tested on i386, x86_64, and ppc64, but still needs to be tested in sparc64. Signed-off-by: Rusty Lynch Signed-off-by: Anil S Keshavamurthy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kprobes.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index fba39f87efec..0f90466fb8b0 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -165,6 +165,8 @@ static inline int kprobe_running(void) extern int arch_prepare_kprobe(struct kprobe *p); extern void arch_copy_kprobe(struct kprobe *p); +extern void arch_arm_kprobe(struct kprobe *p); +extern void arch_disarm_kprobe(struct kprobe *p); extern void arch_remove_kprobe(struct kprobe *p); extern void show_registers(struct pt_regs *regs); -- cgit v1.2.1 From 0aa55e4d7db822059fe8132fe9f2b7773c48216c Mon Sep 17 00:00:00 2001 From: Hien Nguyen Date: Thu, 23 Jun 2005 00:09:26 -0700 Subject: [PATCH] kprobes: moves lock-unlock to non-arch kprobe_flush_task This patch moves the lock/unlock of the arch specific kprobe_flush_task() to the non-arch specific kprobe_flusk_task(). Signed-off-by: Hien Nguyen Acked-by: Prasanna S Panchamukhi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kprobes.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 0f90466fb8b0..461391decc46 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -33,7 +33,6 @@ #include #include #include -#include #include @@ -101,7 +100,7 @@ extern void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs, unsigned long flags); extern struct task_struct *arch_get_kprobe_task(void *ptr); extern void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs); -extern void arch_kprobe_flush_task(struct task_struct *tk, spinlock_t *kp_lock); +extern void arch_kprobe_flush_task(struct task_struct *tk); #else /* ARCH_SUPPORTS_KRETPROBES */ static inline void kretprobe_trampoline(void) { -- cgit v1.2.1 From ea32c65cc2d2294c04e9f81d0578a6f51febfdbf Mon Sep 17 00:00:00 2001 From: Prasanna S Panchamukhi Date: Thu, 23 Jun 2005 00:09:36 -0700 Subject: [PATCH] kprobes: Temporary disarming of reentrant probe In situations where a kprobes handler calls a routine which has a probe on it, then kprobes_handler() disarms the new probe forever. This patch removes the above limitation by temporarily disarming the new probe. When the another probe hits while handling the old probe, the kprobes_handler() saves previous kprobes state and handles the new probe without calling the new kprobes registered handlers. kprobe_post_handler() restores back the previous kprobes state and the normal execution continues. However on x86_64 architecture, re-rentrancy is provided only through pre_handler(). If a routine having probe is referenced through post_handler(), then the probes on that routine are disarmed forever, since the exception stack is gets changed after the processor single steps the instruction of the new probe. This patch includes generic changes to support temporary disarming on reentrancy of probes. Signed-of-by: Prasanna S Panchamukhi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kprobes.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 461391decc46..5e1a7b0d7b3f 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -36,6 +36,12 @@ #include +/* kprobe_status settings */ +#define KPROBE_HIT_ACTIVE 0x00000001 +#define KPROBE_HIT_SS 0x00000002 +#define KPROBE_REENTER 0x00000004 +#define KPROBE_HIT_SSDONE 0x00000008 + struct kprobe; struct pt_regs; struct kretprobe; @@ -55,6 +61,9 @@ struct kprobe { /* list of kprobes for multi-handler support */ struct list_head list; + /*count the number of times this probe was temporarily disarmed */ + unsigned long nmissed; + /* location of the probe point */ kprobe_opcode_t *addr; -- cgit v1.2.1 From d6e711448137ca3301512cec41a2c2ce852b3d0a Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Thu, 23 Jun 2005 00:09:43 -0700 Subject: [PATCH] setuid core dump Add a new `suid_dumpable' sysctl: This value can be used to query and set the core dump mode for setuid or otherwise protected/tainted binaries. The modes are 0 - (default) - traditional behaviour. Any process which has changed privilege levels or is execute only will not be dumped 1 - (debug) - all processes dump core when possible. The core dump is owned by the current user and no security is applied. This is intended for system debugging situations only. Ptrace is unchecked. 2 - (suidsafe) - any binary which normally would not be dumped is dumped readable by root only. This allows the end user to remove such a dump but not access it directly. For security reasons core dumps in this mode will not overwrite one another or other files. This mode is appropriate when adminstrators are attempting to debug problems in a normal environment. (akpm: > > +EXPORT_SYMBOL(suid_dumpable); > > EXPORT_SYMBOL_GPL? No problem to me. > > if (current->euid == current->uid && current->egid == current->gid) > > current->mm->dumpable = 1; > > Should this be SUID_DUMP_USER? Actually the feedback I had from last time was that the SUID_ defines should go because its clearer to follow the numbers. They can go everywhere (and there are lots of places where dumpable is tested/used as a bool in untouched code) > Maybe this should be renamed to `dump_policy' or something. Doing that > would help us catch any code which isn't using the #defines, too. Fair comment. The patch was designed to be easy to maintain for Red Hat rather than for merging. Changing that field would create a gigantic diff because it is used all over the place. ) Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/binfmts.h | 5 +++++ include/linux/sched.h | 2 +- include/linux/sysctl.h | 1 + 3 files changed, 7 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h index 7e736e201c46..c1e82c514443 100644 --- a/include/linux/binfmts.h +++ b/include/linux/binfmts.h @@ -69,6 +69,11 @@ extern void remove_arg_zero(struct linux_binprm *); extern int search_binary_handler(struct linux_binprm *,struct pt_regs *); extern int flush_old_exec(struct linux_binprm * bprm); +extern int suid_dumpable; +#define SUID_DUMP_DISABLE 0 /* No setuid dumping */ +#define SUID_DUMP_USER 1 /* Dump as user of process */ +#define SUID_DUMP_ROOT 2 /* Dump as root */ + /* Stack area protections */ #define EXSTACK_DEFAULT 0 /* Whatever the arch defaults to */ #define EXSTACK_DISABLE_X 1 /* Disable executable stacks */ diff --git a/include/linux/sched.h b/include/linux/sched.h index b58afd97a180..901742f92389 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -246,7 +246,7 @@ struct mm_struct { unsigned long saved_auxv[42]; /* for /proc/PID/auxv */ - unsigned dumpable:1; + unsigned dumpable:2; cpumask_t cpu_vm_mask; /* Architecture-specific MM context */ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index a17745c80a91..614e939c78a4 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -136,6 +136,7 @@ enum KERN_UNKNOWN_NMI_PANIC=66, /* int: unknown nmi panic flag */ KERN_BOOTLOADER_TYPE=67, /* int: boot loader type */ KERN_RANDOMIZE=68, /* int: randomize virtual address space */ + KERN_SETUID_DUMPABLE=69, /* int: behaviour of dumps for setuid core */ }; -- cgit v1.2.1 From ef3daeda7b58f046f94b26637d500354038d39f4 Mon Sep 17 00:00:00 2001 From: Yoav Zach Date: Thu, 23 Jun 2005 00:09:58 -0700 Subject: [PATCH] Don't force O_LARGEFILE for 32 bit processes on ia64 In ia64 kernel, the O_LARGEFILE flag is forced when opening a file. This is problematic for execution of 32 bit processes, which are not largefile aware, either by SW emulation or by HW execution. For such processes, the problem is two-fold: 1) When trying to open a file that is larger than 4G the operation should fail, but it's not 2) Writing to offset larger than 4G should fail, but it's not The proposed patch takes advantage of the way 32 bit processes are identified in ia64 systems. Such processes have PER_LINUX32 for their personality. With the patch, the ia64 kernel will not enforce the O_LARGEFILE flag if the current process has PER_LINUX32 set. The behavior for all other architectures remains unchanged. Signed-off-by: Yoav Zach Acked-by: Tony Luck Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fcntl.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fcntl.h b/include/linux/fcntl.h index 704fb76b6334..8a7c82151de9 100644 --- a/include/linux/fcntl.h +++ b/include/linux/fcntl.h @@ -25,6 +25,10 @@ #ifdef __KERNEL__ +#ifndef force_o_largefile +#define force_o_largefile() (BITS_PER_LONG != 32) +#endif + #if BITS_PER_LONG == 32 #define IS_GETLK32(cmd) ((cmd) == F_GETLK) #define IS_SETLK32(cmd) ((cmd) == F_SETLK) -- cgit v1.2.1 From 46c271bedd2c8444b1d05bc44928beec0c07debc Mon Sep 17 00:00:00 2001 From: Peter Osterlund Date: Thu, 23 Jun 2005 00:10:02 -0700 Subject: [PATCH] Improve CD/DVD packet driver write performance This patch improves write performance for the CD/DVD packet writing driver. The logic for switching between reading and writing has been changed so that streaming writes are no longer interrupted by read requests. Signed-off-by: Peter Osterlund Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pktcdvd.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/pktcdvd.h b/include/linux/pktcdvd.h index 4e2d2a942ecb..4b32bce9a289 100644 --- a/include/linux/pktcdvd.h +++ b/include/linux/pktcdvd.h @@ -159,7 +159,7 @@ struct packet_iosched struct bio *read_queue_tail; struct bio *write_queue; struct bio *write_queue_tail; - int high_prio_read; /* An important read request has been queued */ + sector_t last_write; /* The sector where the last write ended */ int successive_reads; }; -- cgit v1.2.1 From bb93e3a52f8db7210258a1a2134cced0b78a46e1 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 23 Jun 2005 00:10:15 -0700 Subject: [PATCH] block: add unlocked_ioctl support for block devices This patch allows block device drivers to convert their ioctl functions to unlocked_ioctl() like character devices and other subsystems. All functions that were called with the BKL held before are still used that way, but I would not be surprised if it could be removed from the ioctl functions in drivers/block/ioctl.c themselves. As a side note, I found that compat_blkdev_ioctl() acquires the BKL as well, which looks like a bug. I have checked that every user of disk->fops->compat_ioctl() in the current git tree gets the BKL itself, so it could easily be removed from compat_blkdev_ioctl(). Signed-off-by: Arnd Bergmann Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 3622e952e98c..9b1278e21279 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -884,6 +884,7 @@ struct block_device_operations { int (*open) (struct inode *, struct file *); int (*release) (struct inode *, struct file *); int (*ioctl) (struct inode *, struct file *, unsigned, unsigned long); + long (*unlocked_ioctl) (struct file *, unsigned, unsigned long); long (*compat_ioctl) (struct file *, unsigned, unsigned long); int (*media_changed) (struct gendisk *); int (*revalidate_disk) (struct gendisk *); -- cgit v1.2.1 From 45778ca819accab1a4a3378b3566cab0f189164f Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 23 Jun 2005 00:10:17 -0700 Subject: [PATCH] Remove f_error field from struct file The following patch removes the f_error field and all checks of f_error. Trond said: f_error was introduced for NFS, and made sense when we were guaranteed always to have a file pointer around when write errors occurred. Since then, we have (for various reasons) had to introduce the nfs_open_context in order to track the file read/write state, and it made sense to move our f_error tracking there too. Signed-off-by: Christoph Lameter Acked-by: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 9b1278e21279..517bf4966bf5 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -581,7 +581,6 @@ struct file { atomic_t f_count; unsigned int f_flags; mode_t f_mode; - int f_error; loff_t f_pos; struct fown_struct f_owner; unsigned int f_uid, f_gid; -- cgit v1.2.1 From f9fd27a253d5e0b23531d12ce7ad15b6535d4486 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Jun 2005 00:10:19 -0700 Subject: [PATCH] acl endianess annotations Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/posix_acl_xattr.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/posix_acl_xattr.h b/include/linux/posix_acl_xattr.h index 5efd0a6dad94..fe271c1947b2 100644 --- a/include/linux/posix_acl_xattr.h +++ b/include/linux/posix_acl_xattr.h @@ -23,13 +23,13 @@ #define ACL_UNDEFINED_ID (-1) typedef struct { - __u16 e_tag; - __u16 e_perm; - __u32 e_id; + __le16 e_tag; + __le16 e_perm; + __le32 e_id; } posix_acl_xattr_entry; typedef struct { - __u32 a_version; + __le32 a_version; posix_acl_xattr_entry a_entries[0]; } posix_acl_xattr_header; -- cgit v1.2.1 From 9a59f452abe11f569e13ec16c51e6d61c54b9838 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Jun 2005 00:10:19 -0700 Subject: [PATCH] remove This file duplicates , using slightly different names. Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/posix_acl_xattr.h | 3 +++ include/linux/reiserfs_acl.h | 1 - 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/posix_acl_xattr.h b/include/linux/posix_acl_xattr.h index fe271c1947b2..6e53c34035cd 100644 --- a/include/linux/posix_acl_xattr.h +++ b/include/linux/posix_acl_xattr.h @@ -52,4 +52,7 @@ posix_acl_xattr_count(size_t size) return size / sizeof(posix_acl_xattr_entry); } +struct posix_acl *posix_acl_from_xattr(const void *value, size_t size); +int posix_acl_to_xattr(const struct posix_acl *acl, void *buffer, size_t size); + #endif /* _POSIX_ACL_XATTR_H */ diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h index 2aef9c3f5ce8..0760507a545b 100644 --- a/include/linux/reiserfs_acl.h +++ b/include/linux/reiserfs_acl.h @@ -1,6 +1,5 @@ #include #include -#include #define REISERFS_ACL_VERSION 0x0001 -- cgit v1.2.1 From c43dc2fd885b5658cfd7cedb7bcca20910c517a4 Mon Sep 17 00:00:00 2001 From: Benjamin LaHaise Date: Thu, 23 Jun 2005 00:10:27 -0700 Subject: [PATCH] aio: make wait_queue ->task ->private In the upcoming aio_down patch, it is useful to store a private data pointer in the kiocb's wait_queue. Since we provide our own wake up function and do not require the task_struct pointer, it makes sense to convert the task pointer into a generic private pointer. Signed-off-by: Benjamin LaHaise Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/wait.h | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/wait.h b/include/linux/wait.h index c9486c3efb4a..d38c9fecdc36 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -33,7 +33,7 @@ int default_wake_function(wait_queue_t *wait, unsigned mode, int sync, void *key struct __wait_queue { unsigned int flags; #define WQ_FLAG_EXCLUSIVE 0x01 - struct task_struct * task; + void *private; wait_queue_func_t func; struct list_head task_list; }; @@ -60,7 +60,7 @@ typedef struct __wait_queue_head wait_queue_head_t; */ #define __WAITQUEUE_INITIALIZER(name, tsk) { \ - .task = tsk, \ + .private = tsk, \ .func = default_wake_function, \ .task_list = { NULL, NULL } } @@ -86,7 +86,7 @@ static inline void init_waitqueue_head(wait_queue_head_t *q) static inline void init_waitqueue_entry(wait_queue_t *q, struct task_struct *p) { q->flags = 0; - q->task = p; + q->private = p; q->func = default_wake_function; } @@ -94,7 +94,7 @@ static inline void init_waitqueue_func_entry(wait_queue_t *q, wait_queue_func_t func) { q->flags = 0; - q->task = NULL; + q->private = NULL; q->func = func; } @@ -110,7 +110,7 @@ static inline int waitqueue_active(wait_queue_head_t *q) * aio specifies a wait queue entry with an async notification * callback routine, not associated with any task. */ -#define is_sync_wait(wait) (!(wait) || ((wait)->task)) +#define is_sync_wait(wait) (!(wait) || ((wait)->private)) extern void FASTCALL(add_wait_queue(wait_queue_head_t *q, wait_queue_t * wait)); extern void FASTCALL(add_wait_queue_exclusive(wait_queue_head_t *q, wait_queue_t * wait)); @@ -384,7 +384,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); #define DEFINE_WAIT(name) \ wait_queue_t name = { \ - .task = current, \ + .private = current, \ .func = autoremove_wake_function, \ .task_list = LIST_HEAD_INIT((name).task_list), \ } @@ -393,7 +393,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); struct wait_bit_queue name = { \ .key = __WAIT_BIT_KEY_INITIALIZER(word, bit), \ .wait = { \ - .task = current, \ + .private = current, \ .func = wake_bit_function, \ .task_list = \ LIST_HEAD_INIT((name).wait.task_list), \ @@ -402,7 +402,7 @@ int wake_bit_function(wait_queue_t *wait, unsigned mode, int sync, void *key); #define init_wait(wait) \ do { \ - (wait)->task = current; \ + (wait)->private = current; \ (wait)->func = autoremove_wake_function; \ INIT_LIST_HEAD(&(wait)->task_list); \ } while (0) -- cgit v1.2.1 From bfb07599da289881d3bcbb601a110e997fc7444b Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 23 Jun 2005 00:10:32 -0700 Subject: [PATCH] Introduce tty_unregister_ldisc() It's a bit strange to see tty_register_ldisc call in modules' exit functions. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/tty.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/tty.h b/include/linux/tty.h index 1b76106272d3..59ff42c629ec 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -345,6 +345,7 @@ extern int tty_check_change(struct tty_struct * tty); extern void stop_tty(struct tty_struct * tty); extern void start_tty(struct tty_struct * tty); extern int tty_register_ldisc(int disc, struct tty_ldisc *new_ldisc); +extern int tty_unregister_ldisc(int disc); extern int tty_register_driver(struct tty_driver *driver); extern int tty_unregister_driver(struct tty_driver *driver); extern void tty_register_device(struct tty_driver *driver, unsigned index, struct device *dev); -- cgit v1.2.1 From 4749f32da939d4e4160541b2cadc22492bb507ec Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 23 Jun 2005 11:36:56 +0200 Subject: [PATCH] better USB_MON dependencies This makes the USB_MON less confusing. Signed-off-by: Adrian Bunk Signed-off-by: Linus Torvalds --- include/linux/usb.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/usb.h b/include/linux/usb.h index 3d508bf08402..eb282b581546 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -290,7 +290,7 @@ struct usb_bus { struct class_device *class_dev; /* class device for this bus */ struct kref kref; /* handles reference counting this bus */ void (*release)(struct usb_bus *bus); /* function to destroy this bus's memory */ -#if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) +#if defined(CONFIG_USB_MON) struct mon_bus *mon_bus; /* non-null when associated */ int monitored; /* non-zero when monitored */ #endif -- cgit v1.2.1 From 317a76f9a44b437d6301718f4e5d08bd93f98da7 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 12:19:55 -0700 Subject: [TCP]: Add pluggable congestion control algorithm infrastructure. Allow TCP to have multiple pluggable congestion control algorithms. Algorithms are defined by a set of operations and can be built in or modules. The legacy "new RENO" algorithm is used as a starting point and fallback. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/sysctl.h | 9 +-------- include/linux/tcp.h | 49 ++++++++++--------------------------------------- 2 files changed, 11 insertions(+), 47 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 614e939c78a4..72965bfe6cfb 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -333,21 +333,14 @@ enum NET_TCP_FRTO=92, NET_TCP_LOW_LATENCY=93, NET_IPV4_IPFRAG_SECRET_INTERVAL=94, - NET_TCP_WESTWOOD=95, NET_IPV4_IGMP_MAX_MSF=96, NET_TCP_NO_METRICS_SAVE=97, - NET_TCP_VEGAS=98, - NET_TCP_VEGAS_ALPHA=99, - NET_TCP_VEGAS_BETA=100, - NET_TCP_VEGAS_GAMMA=101, - NET_TCP_BIC=102, - NET_TCP_BIC_FAST_CONVERGENCE=103, - NET_TCP_BIC_LOW_WINDOW=104, NET_TCP_DEFAULT_WIN_SCALE=105, NET_TCP_MODERATE_RCVBUF=106, NET_TCP_TSO_WIN_DIVISOR=107, NET_TCP_BIC_BETA=108, NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR=109, + NET_TCP_CONG_CONTROL=110, }; enum { diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 97a7c9e03df5..3ea75dd6640a 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -203,13 +203,6 @@ struct tcp_sack_block { __u32 end_seq; }; -enum tcp_congestion_algo { - TCP_RENO=0, - TCP_VEGAS, - TCP_WESTWOOD, - TCP_BIC, -}; - struct tcp_options_received { /* PAWS/RTTM data */ long ts_recent_stamp;/* Time we stored ts_recent (for aging) */ @@ -305,7 +298,7 @@ struct tcp_sock { __u8 reordering; /* Packet reordering metric. */ __u8 frto_counter; /* Number of new acks after RTO */ - __u8 adv_cong; /* Using Vegas, Westwood, or BIC */ + __u8 unused; __u8 defer_accept; /* User waits for some data after accept() */ /* RTT measurement */ @@ -401,37 +394,10 @@ struct tcp_sock { __u32 time; } rcvq_space; -/* TCP Westwood structure */ - struct { - __u32 bw_ns_est; /* first bandwidth estimation..not too smoothed 8) */ - __u32 bw_est; /* bandwidth estimate */ - __u32 rtt_win_sx; /* here starts a new evaluation... */ - __u32 bk; - __u32 snd_una; /* used for evaluating the number of acked bytes */ - __u32 cumul_ack; - __u32 accounted; - __u32 rtt; - __u32 rtt_min; /* minimum observed RTT */ - } westwood; - -/* Vegas variables */ - struct { - __u32 beg_snd_nxt; /* right edge during last RTT */ - __u32 beg_snd_una; /* left edge during last RTT */ - __u32 beg_snd_cwnd; /* saves the size of the cwnd */ - __u8 doing_vegas_now;/* if true, do vegas for this RTT */ - __u16 cntRTT; /* # of RTTs measured within last RTT */ - __u32 minRTT; /* min of RTTs measured within last RTT (in usec) */ - __u32 baseRTT; /* the min of all Vegas RTT measurements seen (in usec) */ - } vegas; - - /* BI TCP Parameters */ - struct { - __u32 cnt; /* increase cwnd by 1 after this number of ACKs */ - __u32 last_max_cwnd; /* last maximium snd_cwnd */ - __u32 last_cwnd; /* the last snd_cwnd */ - __u32 last_stamp; /* time when updated last_cwnd */ - } bictcp; + /* Pluggable TCP congestion control hook */ + struct tcp_congestion_ops *ca_ops; + u32 ca_priv[16]; +#define TCP_CA_PRIV_SIZE (16*sizeof(u32)) }; static inline struct tcp_sock *tcp_sk(const struct sock *sk) @@ -439,6 +405,11 @@ static inline struct tcp_sock *tcp_sk(const struct sock *sk) return (struct tcp_sock *)sk; } +static inline void *tcp_ca(const struct tcp_sock *tp) +{ + return (void *) tp->ca_priv; +} + #endif #endif /* _LINUX_TCP_H */ -- cgit v1.2.1 From 056ede6cface66b400cd3b8e60ed077cc5b85c18 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 12:21:28 -0700 Subject: [TCP]: Report congestion control algorithm in tcp_diag. Enhancement to the tcp_diag interface used by the iproute2 ss command to report the tcp congestion control being used by a socket. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/tcp_diag.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tcp_diag.h b/include/linux/tcp_diag.h index ceee962e1d15..7a5996743946 100644 --- a/include/linux/tcp_diag.h +++ b/include/linux/tcp_diag.h @@ -99,9 +99,10 @@ enum TCPDIAG_MEMINFO, TCPDIAG_INFO, TCPDIAG_VEGASINFO, + TCPDIAG_CONG, }; -#define TCPDIAG_MAX TCPDIAG_VEGASINFO +#define TCPDIAG_MAX TCPDIAG_CONG /* TCPDIAG_MEM */ @@ -123,5 +124,4 @@ struct tcpvegas_info { __u32 tcpv_minrtt; }; - #endif /* _TCP_DIAG_H_ */ -- cgit v1.2.1 From c1ebcdb8c422cd73f54bcd2b9953e443a47667e5 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:08:59 -0700 Subject: [NET]: Remove obsolete fastroute stats. Remove last vestiages of fastroute code that is no longer used. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d89816ad642f..c2e15e381a58 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -165,11 +165,6 @@ struct netif_rx_stats unsigned dropped; unsigned time_squeeze; unsigned throttled; - unsigned fastroute_hit; - unsigned fastroute_success; - unsigned fastroute_defer; - unsigned fastroute_deferred_out; - unsigned fastroute_latency_reduction; unsigned cpu_collision; }; -- cgit v1.2.1 From 34008d8c631d067caffa136313260525f3ae48a2 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:10:00 -0700 Subject: [NET]: Remove obsolete netif_rx congestion sensing mechanism. Remove the congestion sensing mechanism from netif_rx, and always return either full or empty. Almost no driver checks the return value from netif_rx, and those that do only use it for debug messages. The original design of netif_rx was to do flow control based on the receive queue, but NAPI has supplanted this and no driver uses the feedback. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c2e15e381a58..718ad579c65c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -558,8 +558,6 @@ static inline int unregister_gifconf(unsigned int family) struct softnet_data { int throttle; - int cng_level; - int avg_blog; struct sk_buff_head input_pkt_queue; struct list_head poll_list; struct net_device *output_queue; -- cgit v1.2.1 From 31aa02c53c84658f6694f319f09e232ede27be5a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:12:48 -0700 Subject: [NET]: Eliminate netif_rx massive packet drops. Eliminate the throttling behaviour when the netif receive queue fills because it behaves badly when using high speed networks under load. The throttling cause multiple packet drops that cause TCP to go into slow start mode. The same effective patch has been part of BIC TCP and H-TCP as well as part of Web100. The existing code drops 100's of packets when the queue fills; this changes it to individual packet drop-tail. Signed-off-by: Stephen Hemmminger Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 718ad579c65c..3a0ed7f9e801 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -164,7 +164,6 @@ struct netif_rx_stats unsigned total; unsigned dropped; unsigned time_squeeze; - unsigned throttled; unsigned cpu_collision; }; @@ -557,10 +556,9 @@ static inline int unregister_gifconf(unsigned int family) struct softnet_data { - int throttle; + struct net_device *output_queue; struct sk_buff_head input_pkt_queue; struct list_head poll_list; - struct net_device *output_queue; struct sk_buff *completion_queue; struct net_device backlog_dev; /* Sorry. 8) */ -- cgit v1.2.1 From 51b0bdedb8e784d0d969a6b77151911130812400 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:14:40 -0700 Subject: [NET]: Separate two usages of netdev_max_backlog. Separate out the two uses of netdev_max_backlog. One controls the upper bound on packets processed per softirq, the new name for this is netdev_budget; the other controls the limit on packets queued via netif_rx. Increase the max_backlog default to account for faster processors. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/sysctl.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 72965bfe6cfb..ebfe1250f0a4 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -243,6 +243,7 @@ enum NET_CORE_MOD_CONG=16, NET_CORE_DEV_WEIGHT=17, NET_CORE_SOMAXCONN=18, + NET_CORE_BUDGET=19, }; /* /proc/sys/net/ethernet */ -- cgit v1.2.1 From 5f8ef48d240963093451bcf83df89f1a1364f51d Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 23 Jun 2005 20:37:36 -0700 Subject: [TCP]: Allow choosing TCP congestion control via sockopt. Allow using setsockopt to set TCP congestion control to use on a per socket basis. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/tcp.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 3ea75dd6640a..dfd93d03f5d2 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -127,6 +127,7 @@ enum { #define TCP_WINDOW_CLAMP 10 /* Bound advertised window */ #define TCP_INFO 11 /* Information about this connection. */ #define TCP_QUICKACK 12 /* Block/reenable quick acks */ +#define TCP_CONGESTION 13 /* Congestion control algorithm */ #define TCPI_OPT_TIMESTAMPS 1 #define TCPI_OPT_SACK 2 -- cgit v1.2.1 From 2de4ff7bd658c97fb357efa3095a509674dacb5a Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 20:49:30 -0700 Subject: [LIB]: Textsearch infrastructure. The textsearch infrastructure provides text searching facitilies for both linear and non-linear data. Individual search algorithms are implemented in modules and chosen by the user. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/textsearch.h | 180 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 180 insertions(+) create mode 100644 include/linux/textsearch.h (limited to 'include/linux') diff --git a/include/linux/textsearch.h b/include/linux/textsearch.h new file mode 100644 index 000000000000..941f45ac117a --- /dev/null +++ b/include/linux/textsearch.h @@ -0,0 +1,180 @@ +#ifndef __LINUX_TEXTSEARCH_H +#define __LINUX_TEXTSEARCH_H + +#ifdef __KERNEL__ + +#include +#include +#include +#include +#include + +struct ts_config; + +/** + * TS_AUTOLOAD - Automatically load textsearch modules when needed + */ +#define TS_AUTOLOAD 1 + +/** + * struct ts_state - search state + * @offset: offset for next match + * @cb: control buffer, for persistant variables of get_next_block() + */ +struct ts_state +{ + unsigned int offset; + char cb[40]; +}; + +/** + * struct ts_ops - search module operations + * @name: name of search algorithm + * @init: initialization function to prepare a search + * @find: find the next occurrence of the pattern + * @destroy: destroy algorithm specific parts of a search configuration + * @get_pattern: return head of pattern + * @get_pattern_len: return length of pattern + * @owner: module reference to algorithm + */ +struct ts_ops +{ + const char *name; + struct ts_config * (*init)(const void *, unsigned int, int); + unsigned int (*find)(struct ts_config *, + struct ts_state *); + void (*destroy)(struct ts_config *); + void * (*get_pattern)(struct ts_config *); + unsigned int (*get_pattern_len)(struct ts_config *); + struct module *owner; + struct list_head list; +}; + +/** + * struct ts_config - search configuration + * @ops: operations of chosen algorithm + * @get_next_block: callback to fetch the next block to search in + * @finish: callback to finalize a search + */ +struct ts_config +{ + struct ts_ops *ops; + + /** + * get_next_block - fetch next block of data + * @consumed: number of bytes consumed by the caller + * @dst: destination buffer + * @conf: search configuration + * @state: search state + * + * Called repeatedly until 0 is returned. Must assign the + * head of the next block of data to &*dst and return the length + * of the block or 0 if at the end. consumed == 0 indicates + * a new search. May store/read persistant values in state->cb. + */ + unsigned int (*get_next_block)(unsigned int consumed, + const u8 **dst, + struct ts_config *conf, + struct ts_state *state); + + /** + * finish - finalize/clean a series of get_next_block() calls + * @conf: search configuration + * @state: search state + * + * Called after the last use of get_next_block(), may be used + * to cleanup any leftovers. + */ + void (*finish)(struct ts_config *conf, + struct ts_state *state); +}; + +/** + * textsearch_next - continue searching for a pattern + * @conf: search configuration + * @state: search state + * + * Continues a search looking for more occurrences of the pattern. + * textsearch_find() must be called to find the first occurrence + * in order to reset the state. + * + * Returns the position of the next occurrence of the pattern or + * UINT_MAX if not match was found. + */ +static inline unsigned int textsearch_next(struct ts_config *conf, + struct ts_state *state) +{ + unsigned int ret = conf->ops->find(conf, state); + + if (conf->finish) + conf->finish(conf, state); + + return ret; +} + +/** + * textsearch_find - start searching for a pattern + * @conf: search configuration + * @state: search state + * + * Returns the position of first occurrence of the pattern or + * UINT_MAX if no match was found. + */ +static inline unsigned int textsearch_find(struct ts_config *conf, + struct ts_state *state) +{ + state->offset = 0; + return textsearch_next(conf, state); +} + +/** + * textsearch_get_pattern - return head of the pattern + * @conf: search configuration + */ +static inline void *textsearch_get_pattern(struct ts_config *conf) +{ + return conf->ops->get_pattern(conf); +} + +/** + * textsearch_get_pattern_len - return length of the pattern + * @conf: search configuration + */ +static inline unsigned int textsearch_get_pattern_len(struct ts_config *conf) +{ + return conf->ops->get_pattern_len(conf); +} + +extern int textsearch_register(struct ts_ops *); +extern int textsearch_unregister(struct ts_ops *); +extern struct ts_config *textsearch_prepare(const char *, const void *, + unsigned int, int, int); +extern void textsearch_destroy(struct ts_config *conf); +extern unsigned int textsearch_find_continuous(struct ts_config *, + struct ts_state *, + const void *, unsigned int); + + +#define TS_PRIV_ALIGNTO 8 +#define TS_PRIV_ALIGN(len) (((len) + TS_PRIV_ALIGNTO-1) & ~(TS_PRIV_ALIGNTO-1)) + +static inline struct ts_config *alloc_ts_config(size_t payload, int gfp_mask) +{ + struct ts_config *conf; + + conf = kmalloc(TS_PRIV_ALIGN(sizeof(*conf)) + payload, gfp_mask); + if (conf == NULL) + return ERR_PTR(-ENOMEM); + + memset(conf, 0, TS_PRIV_ALIGN(sizeof(*conf)) + payload); + return conf; +} + +static inline void *ts_config_priv(struct ts_config *conf) +{ + return ((u8 *) conf + TS_PRIV_ALIGN(sizeof(struct ts_config))); +} + +#endif /* __KERNEL__ */ + +#endif -- cgit v1.2.1 From 6408f79cce401e1bfecf923e7156f84f96e021e3 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 20:59:16 -0700 Subject: [LIB]: Naive finite state machine based textsearch A finite state machine consists of n states (struct ts_fsm_token) representing the pattern as a finite automation. The data is read sequentially on a octet basis. Every state token specifies the number of recurrences and the type of value accepted which can be either a specific character or ctype based set of characters. The available type of recurrences include 1, (0|1), [0 n], and [1 n]. The algorithm differs between strict/non-strict mode specyfing whether the pattern has to start at the first octect. Strict mode is enabled by default and can be disabled by inserting TS_FSM_HEAD_IGNORE as the first token in the chain. The runtime performance of the algorithm should be around O(n), however while in strict mode the average runtime can be better. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/textsearch_fsm.h | 48 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 48 insertions(+) create mode 100644 include/linux/textsearch_fsm.h (limited to 'include/linux') diff --git a/include/linux/textsearch_fsm.h b/include/linux/textsearch_fsm.h new file mode 100644 index 000000000000..fdfa078c66e5 --- /dev/null +++ b/include/linux/textsearch_fsm.h @@ -0,0 +1,48 @@ +#ifndef __LINUX_TEXTSEARCH_FSM_H +#define __LINUX_TEXTSEARCH_FSM_H + +#include + +enum { + TS_FSM_SPECIFIC, /* specific character */ + TS_FSM_WILDCARD, /* any character */ + TS_FSM_DIGIT, /* isdigit() */ + TS_FSM_XDIGIT, /* isxdigit() */ + TS_FSM_PRINT, /* isprint() */ + TS_FSM_ALPHA, /* isalpha() */ + TS_FSM_ALNUM, /* isalnum() */ + TS_FSM_ASCII, /* isascii() */ + TS_FSM_CNTRL, /* iscntrl() */ + TS_FSM_GRAPH, /* isgraph() */ + TS_FSM_LOWER, /* islower() */ + TS_FSM_UPPER, /* isupper() */ + TS_FSM_PUNCT, /* ispunct() */ + TS_FSM_SPACE, /* isspace() */ + __TS_FSM_TYPE_MAX, +}; +#define TS_FSM_TYPE_MAX (__TS_FSM_TYPE_MAX - 1) + +enum { + TS_FSM_SINGLE, /* 1 occurrence */ + TS_FSM_PERHAPS, /* 1 or 0 occurrence */ + TS_FSM_ANY, /* 0..n occurrences */ + TS_FSM_MULTI, /* 1..n occurrences */ + TS_FSM_HEAD_IGNORE, /* 0..n ignored occurrences at head */ + __TS_FSM_RECUR_MAX, +}; +#define TS_FSM_RECUR_MAX (__TS_FSM_RECUR_MAX - 1) + +/** + * struct ts_fsm_token - state machine token (state) + * @type: type of token + * @recur: number of recurrences + * @value: character value for TS_FSM_SPECIFIC + */ +struct ts_fsm_token +{ + __u16 type; + __u8 recur; + __u8 value; +}; + +#endif -- cgit v1.2.1 From 677e90eda3bd8cfde0b748daaa46476162a03950 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 20:59:51 -0700 Subject: [NET]: Zerocopy sequential reading of skb data Implements sequential reading for both linear and non-linear skb data at zerocopy cost. The data is returned in chunks of arbitary length, therefore random access is not possible. Usage: from := 0 to := 128 state := undef data := undef len := undef consumed := 0 skb_prepare_seq_read(skb, from, to, &state) while (len = skb_seq_read(consumed, &data, &state)) != 0 do /* do something with 'data' of length 'len' */ if abort then /* abort read if we don't wait for * skb_seq_read() to return 0 */ skb_abort_seq_read(&state) return endif /* not necessary to consume all of 'len' */ consumed += len done Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/skbuff.h | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index d7c839a21842..171a37dff83a 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -321,6 +321,24 @@ extern void skb_over_panic(struct sk_buff *skb, int len, extern void skb_under_panic(struct sk_buff *skb, int len, void *here); +struct skb_seq_state +{ + __u32 lower_offset; + __u32 upper_offset; + __u32 frag_idx; + __u32 stepped_offset; + struct sk_buff *root_skb; + struct sk_buff *cur_skb; + __u8 *frag_data; +}; + +extern void skb_prepare_seq_read(struct sk_buff *skb, + unsigned int from, unsigned int to, + struct skb_seq_state *st); +extern unsigned int skb_seq_read(unsigned int consumed, const u8 **data, + struct skb_seq_state *st); +extern void skb_abort_seq_read(struct skb_seq_state *st); + /* Internal */ #define skb_shinfo(SKB) ((struct skb_shared_info *)((SKB)->end)) -- cgit v1.2.1 From 3fc7e8a6d842f72d16d2623b1022814a635ab961 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 21:00:17 -0700 Subject: [NET]: skb_find_text() - Find a text pattern in skb data Finds a pattern in the skb data according to the specified textsearch configuration. Use textsearch_next() to retrieve subsequent occurrences of the pattern. Returns the offset to the first occurrence or UINT_MAX if no match was found. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/skbuff.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 171a37dff83a..416a2e4024b2 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -27,6 +27,7 @@ #include #include #include +#include #include #define HAVE_ALLOC_SKB /* For the drivers to know */ @@ -339,6 +340,10 @@ extern unsigned int skb_seq_read(unsigned int consumed, const u8 **data, struct skb_seq_state *st); extern void skb_abort_seq_read(struct skb_seq_state *st); +extern unsigned int skb_find_text(struct sk_buff *skb, unsigned int from, + unsigned int to, struct ts_config *config, + struct ts_state *state); + /* Internal */ #define skb_shinfo(SKB) ((struct skb_shared_info *)((SKB)->end)) -- cgit v1.2.1 From d675c989ed2d4ba23dff615330b04371aea83534 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Thu, 23 Jun 2005 21:00:58 -0700 Subject: [PKT_SCHED]: Packet classification based on textsearch (ematch) Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/pkt_cls.h | 1 + include/linux/rtnetlink.h | 7 +++++-- include/linux/tc_ematch/tc_em_text.h | 19 +++++++++++++++++++ 3 files changed, 25 insertions(+), 2 deletions(-) create mode 100644 include/linux/tc_ematch/tc_em_text.h (limited to 'include/linux') diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h index d2aa214d6803..25d2d67c1faf 100644 --- a/include/linux/pkt_cls.h +++ b/include/linux/pkt_cls.h @@ -408,6 +408,7 @@ enum TCF_EM_NBYTE, TCF_EM_U32, TCF_EM_META, + TCF_EM_TEXT, __TCF_EM_MAX }; diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index e68dbf0bf579..d021888b58f1 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -892,10 +892,13 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi goto rtattr_failure; \ __rta_fill(skb, attrtype, attrlen, data); }) -#define RTA_PUT_NOHDR(skb, attrlen, data) \ +#define RTA_APPEND(skb, attrlen, data) \ ({ if (unlikely(skb_tailroom(skb) < (int)(attrlen))) \ goto rtattr_failure; \ - memcpy(skb_put(skb, RTA_ALIGN(attrlen)), data, attrlen); }) + memcpy(skb_put(skb, attrlen), data, attrlen); }) + +#define RTA_PUT_NOHDR(skb, attrlen, data) \ + RTA_APPEND(skb, RTA_ALIGN(attrlen), data) #define RTA_PUT_U8(skb, attrtype, value) \ ({ u8 _tmp = (value); \ diff --git a/include/linux/tc_ematch/tc_em_text.h b/include/linux/tc_ematch/tc_em_text.h new file mode 100644 index 000000000000..7cd43e99c7f5 --- /dev/null +++ b/include/linux/tc_ematch/tc_em_text.h @@ -0,0 +1,19 @@ +#ifndef __LINUX_TC_EM_TEXT_H +#define __LINUX_TC_EM_TEXT_H + +#include + +#define TC_EM_TEXT_ALGOSIZ 16 + +struct tcf_em_text +{ + char algo[TC_EM_TEXT_ALGOSIZ]; + __u16 from_offset; + __u16 to_offset; + __u16 pattern_len; + __u8 from_layer:4; + __u8 to_layer:4; + __u8 pad; +}; + +#endif -- cgit v1.2.1 From 76d8aeabfeb1c42641a81c44280177b9a08670d8 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 23 Jun 2005 22:00:49 -0700 Subject: [PATCH] keys: Discard key spinlock and use RCU for key payload The attached patch changes the key implementation in a number of ways: (1) It removes the spinlock from the key structure. (2) The key flags are now accessed using atomic bitops instead of write-locking the key spinlock and using C bitwise operators. The three instantiation flags are dealt with with the construction semaphore held during the request_key/instantiate/negate sequence, thus rendering the spinlock superfluous. The key flags are also now bit numbers not bit masks. (3) The key payload is now accessed using RCU. This permits the recursive keyring search algorithm to be simplified greatly since no locks need be taken other than the usual RCU preemption disablement. Searching now does not require any locks or semaphores to be held; merely that the starting keyring be pinned. (4) The keyring payload now includes an RCU head so that it can be disposed of by call_rcu(). This requires that the payload be copied on unlink to prevent introducing races in copy-down vs search-up. (5) The user key payload is now a structure with the data following it. It includes an RCU head like the keyring payload and for the same reason. It also contains a data length because the data length in the key may be changed on another CPU whilst an RCU protected read is in progress on the payload. This would then see the supposed RCU payload and the on-key data length getting out of sync. I'm tempted to drop the key's datalen entirely, except that it's used in conjunction with quota management and so is a little tricky to get rid of. (6) Update the keys documentation. Signed-Off-By: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/key-ui.h | 6 ++++-- include/linux/key.h | 25 +++++++++++++++---------- 2 files changed, 19 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/key-ui.h b/include/linux/key-ui.h index 60cc7b762e78..159ca8d54e9a 100644 --- a/include/linux/key-ui.h +++ b/include/linux/key-ui.h @@ -31,8 +31,10 @@ extern spinlock_t key_serial_lock; * subscribed */ struct keyring_list { - unsigned maxkeys; /* max keys this list can hold */ - unsigned nkeys; /* number of keys currently held */ + struct rcu_head rcu; /* RCU deletion hook */ + unsigned short maxkeys; /* max keys this list can hold */ + unsigned short nkeys; /* number of keys currently held */ + unsigned short delkey; /* key to be unlinked by RCU */ struct key *keys[0]; }; diff --git a/include/linux/key.h b/include/linux/key.h index 6aa46d0e812f..2c24ffaca86f 100644 --- a/include/linux/key.h +++ b/include/linux/key.h @@ -18,7 +18,7 @@ #include #include #include -#include +#include #include #ifdef __KERNEL__ @@ -78,7 +78,6 @@ struct key { key_serial_t serial; /* key serial number */ struct rb_node serial_node; struct key_type *type; /* type of key */ - rwlock_t lock; /* examination vs change lock */ struct rw_semaphore sem; /* change vs change sem */ struct key_user *user; /* owner of this key */ time_t expiry; /* time at which key expires (or 0) */ @@ -86,14 +85,10 @@ struct key { gid_t gid; key_perm_t perm; /* access permissions */ unsigned short quotalen; /* length added to quota */ - unsigned short datalen; /* payload data length */ - unsigned short flags; /* status flags (change with lock writelocked) */ -#define KEY_FLAG_INSTANTIATED 0x00000001 /* set if key has been instantiated */ -#define KEY_FLAG_DEAD 0x00000002 /* set if key type has been deleted */ -#define KEY_FLAG_REVOKED 0x00000004 /* set if key had been revoked */ -#define KEY_FLAG_IN_QUOTA 0x00000008 /* set if key consumes quota */ -#define KEY_FLAG_USER_CONSTRUCT 0x00000010 /* set if key is being constructed in userspace */ -#define KEY_FLAG_NEGATIVE 0x00000020 /* set if key is negative */ + unsigned short datalen; /* payload data length + * - may not match RCU dereferenced payload + * - payload should contain own length + */ #ifdef KEY_DEBUGGING unsigned magic; @@ -101,6 +96,14 @@ struct key { #define KEY_DEBUG_MAGIC_X 0xf8e9dacbu #endif + unsigned long flags; /* status flags (change with bitops) */ +#define KEY_FLAG_INSTANTIATED 0 /* set if key has been instantiated */ +#define KEY_FLAG_DEAD 1 /* set if key type has been deleted */ +#define KEY_FLAG_REVOKED 2 /* set if key had been revoked */ +#define KEY_FLAG_IN_QUOTA 3 /* set if key consumes quota */ +#define KEY_FLAG_USER_CONSTRUCT 4 /* set if key is being constructed in userspace */ +#define KEY_FLAG_NEGATIVE 5 /* set if key is negative */ + /* the description string * - this is used to match a key against search criteria * - this should be a printable string @@ -250,6 +253,8 @@ extern int keyring_add_key(struct key *keyring, extern struct key *key_lookup(key_serial_t id); +extern void keyring_replace_payload(struct key *key, void *replacement); + #define key_serial(key) ((key) ? (key)->serial : 0) /* -- cgit v1.2.1 From 7888e7ff4ee579442128d7d12a9c9dbf2cf7de6a Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 23 Jun 2005 22:00:51 -0700 Subject: [PATCH] Keys: Pass session keyring to call_usermodehelper() The attached patch makes it possible to pass a session keyring through to the process spawned by call_usermodehelper(). This allows patch 3/3 to pass an authorisation key through to /sbin/request-key, thus permitting better access controls when doing just-in-time key creation. Signed-Off-By: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/key.h | 10 +++++++++- include/linux/kmod.h | 13 ++++++++++++- 2 files changed, 21 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/key.h b/include/linux/key.h index 2c24ffaca86f..2bfbf88d2740 100644 --- a/include/linux/key.h +++ b/include/linux/key.h @@ -273,14 +273,22 @@ extern void key_fsuid_changed(struct task_struct *tsk); extern void key_fsgid_changed(struct task_struct *tsk); extern void key_init(void); +#define __install_session_keyring(tsk, keyring) \ +({ \ + struct key *old_session = tsk->signal->session_keyring; \ + tsk->signal->session_keyring = keyring; \ + old_session; \ +}) + #else /* CONFIG_KEYS */ #define key_validate(k) 0 #define key_serial(k) 0 -#define key_get(k) NULL +#define key_get(k) ({ NULL; }) #define key_put(k) do { } while(0) #define alloc_uid_keyring(u) 0 #define switch_uid_keyring(u) do { } while(0) +#define __install_session_keyring(t, k) ({ NULL; }) #define copy_keys(f,t) 0 #define copy_thread_group_keys(t) 0 #define exit_keys(t) do { } while(0) diff --git a/include/linux/kmod.h b/include/linux/kmod.h index 95d0e4b0814d..e4a231549407 100644 --- a/include/linux/kmod.h +++ b/include/linux/kmod.h @@ -19,6 +19,7 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ +#include #include #include #include @@ -34,7 +35,17 @@ static inline int request_module(const char * name, ...) { return -ENOSYS; } #endif #define try_then_request_module(x, mod...) ((x) ?: (request_module(mod), (x))) -extern int call_usermodehelper(char *path, char *argv[], char *envp[], int wait); + +struct key; +extern int call_usermodehelper_keys(char *path, char *argv[], char *envp[], + struct key *session_keyring, int wait); + +static inline int +call_usermodehelper(char *path, char **argv, char **envp, int wait) +{ + return call_usermodehelper_keys(path, argv, envp, NULL, wait); +} + extern void usermodehelper_init(void); #endif /* __LINUX_KMOD_H__ */ -- cgit v1.2.1 From 3e30148c3d524a9c1c63ca28261bc24c457eb07a Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 23 Jun 2005 22:00:56 -0700 Subject: [PATCH] Keys: Make request-key create an authorisation key The attached patch makes the following changes: (1) There's a new special key type called ".request_key_auth". This is an authorisation key for when one process requests a key and another process is started to construct it. This type of key cannot be created by the user; nor can it be requested by kernel services. Authorisation keys hold two references: (a) Each refers to a key being constructed. When the key being constructed is instantiated the authorisation key is revoked, rendering it of no further use. (b) The "authorising process". This is either: (i) the process that called request_key(), or: (ii) if the process that called request_key() itself had an authorisation key in its session keyring, then the authorising process referred to by that authorisation key will also be referred to by the new authorisation key. This means that the process that initiated a chain of key requests will authorise the lot of them, and will, by default, wind up with the keys obtained from them in its keyrings. (2) request_key() creates an authorisation key which is then passed to /sbin/request-key in as part of a new session keyring. (3) When request_key() is searching for a key to hand back to the caller, if it comes across an authorisation key in the session keyring of the calling process, it will also search the keyrings of the process specified therein and it will use the specified process's credentials (fsuid, fsgid, groups) to do that rather than the calling process's credentials. This allows a process started by /sbin/request-key to find keys belonging to the authorising process. (4) A key can be read, even if the process executing KEYCTL_READ doesn't have direct read or search permission if that key is contained within the keyrings of a process specified by an authorisation key found within the calling process's session keyring, and is searchable using the credentials of the authorising process. This allows a process started by /sbin/request-key to read keys belonging to the authorising process. (5) The magic KEY_SPEC_*_KEYRING key IDs when passed to KEYCTL_INSTANTIATE or KEYCTL_NEGATE will specify a keyring of the authorising process, rather than the process doing the instantiation. (6) One of the process keyrings can be nominated as the default to which request_key() should attach new keys if not otherwise specified. This is done with KEYCTL_SET_REQKEY_KEYRING and one of the KEY_REQKEY_DEFL_* constants. The current setting can also be read using this call. (7) request_key() is partially interruptible. If it is waiting for another process to finish constructing a key, it can be interrupted. This permits a request-key cycle to be broken without recourse to rebooting. Signed-Off-By: David Howells Signed-Off-By: Benoit Boissinot Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/key-ui.h | 41 +++++++++++++++++++++++++++++++++++++++-- include/linux/key.h | 9 ++++----- include/linux/keyctl.h | 11 +++++++++++ include/linux/sched.h | 8 +++++--- 4 files changed, 59 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/key-ui.h b/include/linux/key-ui.h index 159ca8d54e9a..cc326174a808 100644 --- a/include/linux/key-ui.h +++ b/include/linux/key-ui.h @@ -1,4 +1,4 @@ -/* key-ui.h: key userspace interface stuff for use by keyfs +/* key-ui.h: key userspace interface stuff * * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) @@ -84,8 +84,45 @@ static inline int key_any_permission(const struct key *key, key_perm_t perm) return kperm != 0; } +static inline int key_task_groups_search(struct task_struct *tsk, gid_t gid) +{ + int ret; + + task_lock(tsk); + ret = groups_search(tsk->group_info, gid); + task_unlock(tsk); + return ret; +} + +static inline int key_task_permission(const struct key *key, + struct task_struct *context, + key_perm_t perm) +{ + key_perm_t kperm; + + if (key->uid == context->fsuid) { + kperm = key->perm >> 16; + } + else if (key->gid != -1 && + key->perm & KEY_GRP_ALL && ( + key->gid == context->fsgid || + key_task_groups_search(context, key->gid) + ) + ) { + kperm = key->perm >> 8; + } + else { + kperm = key->perm; + } + + kperm = kperm & perm & KEY_ALL; + + return kperm == perm; + +} -extern struct key *lookup_user_key(key_serial_t id, int create, int part, +extern struct key *lookup_user_key(struct task_struct *context, + key_serial_t id, int create, int partial, key_perm_t perm); extern long join_session_keyring(const char *name); diff --git a/include/linux/key.h b/include/linux/key.h index 2bfbf88d2740..970bbd916cf4 100644 --- a/include/linux/key.h +++ b/include/linux/key.h @@ -199,10 +199,12 @@ extern int key_payload_reserve(struct key *key, size_t datalen); extern int key_instantiate_and_link(struct key *key, const void *data, size_t datalen, - struct key *keyring); + struct key *keyring, + struct key *instkey); extern int key_negate_and_link(struct key *key, unsigned timeout, - struct key *keyring); + struct key *keyring, + struct key *instkey); extern void key_revoke(struct key *key); extern void key_put(struct key *key); @@ -245,9 +247,6 @@ extern struct key *keyring_search(struct key *keyring, struct key_type *type, const char *description); -extern struct key *search_process_keyrings(struct key_type *type, - const char *description); - extern int keyring_add_key(struct key *keyring, struct key *key); diff --git a/include/linux/keyctl.h b/include/linux/keyctl.h index 381dedc370a3..8d7c59a29e09 100644 --- a/include/linux/keyctl.h +++ b/include/linux/keyctl.h @@ -20,6 +20,16 @@ #define KEY_SPEC_USER_SESSION_KEYRING -5 /* - key ID for UID-session keyring */ #define KEY_SPEC_GROUP_KEYRING -6 /* - key ID for GID-specific keyring */ +/* request-key default keyrings */ +#define KEY_REQKEY_DEFL_NO_CHANGE -1 +#define KEY_REQKEY_DEFL_DEFAULT 0 +#define KEY_REQKEY_DEFL_THREAD_KEYRING 1 +#define KEY_REQKEY_DEFL_PROCESS_KEYRING 2 +#define KEY_REQKEY_DEFL_SESSION_KEYRING 3 +#define KEY_REQKEY_DEFL_USER_KEYRING 4 +#define KEY_REQKEY_DEFL_USER_SESSION_KEYRING 5 +#define KEY_REQKEY_DEFL_GROUP_KEYRING 6 + /* keyctl commands */ #define KEYCTL_GET_KEYRING_ID 0 /* ask for a keyring's ID */ #define KEYCTL_JOIN_SESSION_KEYRING 1 /* join or start named session keyring */ @@ -35,5 +45,6 @@ #define KEYCTL_READ 11 /* read a key or keyring's contents */ #define KEYCTL_INSTANTIATE 12 /* instantiate a partially constructed key */ #define KEYCTL_NEGATE 13 /* negate a partially constructed key */ +#define KEYCTL_SET_REQKEY_KEYRING 14 /* set default request-key keyring */ #endif /* _LINUX_KEYCTL_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 901742f92389..2c69682b0444 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -561,9 +561,10 @@ struct group_info { groups_free(group_info); \ } while (0) -struct group_info *groups_alloc(int gidsetsize); -void groups_free(struct group_info *group_info); -int set_current_groups(struct group_info *group_info); +extern struct group_info *groups_alloc(int gidsetsize); +extern void groups_free(struct group_info *group_info); +extern int set_current_groups(struct group_info *group_info); +extern int groups_search(struct group_info *group_info, gid_t grp); /* access the groups "array" with this macro */ #define GROUP_AT(gi, i) \ ((gi)->blocks[(i)/NGROUPS_PER_BLOCK][(i)%NGROUPS_PER_BLOCK]) @@ -660,6 +661,7 @@ struct task_struct { struct user_struct *user; #ifdef CONFIG_KEYS struct key *thread_keyring; /* keyring private to this thread */ + unsigned char jit_keyring; /* default keyring to attach requested keys to */ #endif int oomkilladj; /* OOM kill score adjustment (bit shift). */ char comm[TASK_COMM_LEN]; /* executable name excluding path -- cgit v1.2.1 From 92198f7eaa5df3479341dd8fa20c2c81aa3b1e25 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 23 Jun 2005 22:00:59 -0700 Subject: [PATCH] pass iocb to dio_iodone_t XFS will have to look at iocb->private to fix aio+dio. No other filesystem is using the blockdev_direct_IO* end_io callback. Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 517bf4966bf5..83857d8070d3 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -220,6 +220,7 @@ extern int dir_notify_enable; struct iovec; struct nameidata; +struct kiocb; struct pipe_inode_info; struct poll_table_struct; struct kstatfs; @@ -240,7 +241,7 @@ typedef int (get_block_t)(struct inode *inode, sector_t iblock, typedef int (get_blocks_t)(struct inode *inode, sector_t iblock, unsigned long max_blocks, struct buffer_head *bh_result, int create); -typedef void (dio_iodone_t)(struct inode *inode, loff_t offset, +typedef void (dio_iodone_t)(struct kiocb *iocb, loff_t offset, ssize_t bytes, void *private); /* @@ -302,7 +303,6 @@ struct iattr { struct page; struct address_space; struct writeback_control; -struct kiocb; struct address_space_operations { int (*writepage)(struct page *page, struct writeback_control *wbc); -- cgit v1.2.1 From 4e5117ba0af4582b6ec9164874f719d7f3f1eb2b Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 23 Jun 2005 22:01:03 -0700 Subject: [PATCH] quota: improve credits estimates Improve estimates on the number of needed credits for quota transaction. Now we distinguish blocks that might need to be allocated and blocks that only need to be rewritten. Also we distinguish deleting of a quota structure and creating of a new one. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/dqblk_v1.h | 6 ++++++ include/linux/dqblk_v2.h | 6 ++++++ include/linux/quota.h | 7 +++++-- 3 files changed, 17 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dqblk_v1.h b/include/linux/dqblk_v1.h index 42fbf4797156..57f1250d5a52 100644 --- a/include/linux/dqblk_v1.h +++ b/include/linux/dqblk_v1.h @@ -11,6 +11,12 @@ /* Root squash turned on */ #define V1_DQF_RSQUASH 1 +/* Numbers of blocks needed for updates */ +#define V1_INIT_ALLOC 1 +#define V1_INIT_REWRITE 1 +#define V1_DEL_ALLOC 0 +#define V1_DEL_REWRITE 2 + /* Special information about quotafile */ struct v1_mem_dqinfo { }; diff --git a/include/linux/dqblk_v2.h b/include/linux/dqblk_v2.h index 4a6c5f6867bb..4f853322cb7f 100644 --- a/include/linux/dqblk_v2.h +++ b/include/linux/dqblk_v2.h @@ -10,6 +10,12 @@ /* id numbers of quota format */ #define QFMT_VFS_V0 2 +/* Numbers of blocks needed for updates */ +#define V2_INIT_ALLOC 4 +#define V2_INIT_REWRITE 2 +#define V2_DEL_ALLOC 0 +#define V2_DEL_REWRITE 6 + /* Inmemory copy of version specific information */ struct v2_mem_dqinfo { unsigned int dqi_blocks; diff --git a/include/linux/quota.h b/include/linux/quota.h index ac5b90f4f256..700ead45084f 100644 --- a/include/linux/quota.h +++ b/include/linux/quota.h @@ -138,8 +138,11 @@ struct if_dqinfo { #include /* Maximal numbers of writes for quota operation (insert/delete/update) - * (over all formats) - info block, 4 pointer blocks, data block */ -#define DQUOT_MAX_WRITES 6 + * (over VFS all formats) */ +#define DQUOT_INIT_ALLOC max(V1_INIT_ALLOC, V2_INIT_ALLOC) +#define DQUOT_INIT_REWRITE max(V1_INIT_REWRITE, V2_INIT_REWRITE) +#define DQUOT_DEL_ALLOC max(V1_DEL_ALLOC, V2_DEL_ALLOC) +#define DQUOT_DEL_REWRITE max(V1_DEL_REWRITE, V2_DEL_REWRITE) /* * Data for one user/group kept in memory -- cgit v1.2.1 From 1f54587bea84a35125c95e19b98c2f464c50871b Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 23 Jun 2005 22:01:04 -0700 Subject: [PATCH] quota: ext3: Improve quota credit estimates Use improved credits estimates for quota operations. Also reserve a space for a quota operation in a transaction only if filesystem was mounted with some quota options. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ext3_fs.h | 1 + include/linux/ext3_jbd.h | 19 +++++++++++-------- 2 files changed, 12 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h index 74ad31781e3e..4b6e1ab216a5 100644 --- a/include/linux/ext3_fs.h +++ b/include/linux/ext3_fs.h @@ -358,6 +358,7 @@ struct ext3_inode { #define EXT3_MOUNT_RESERVATION 0x10000 /* Preallocation */ #define EXT3_MOUNT_BARRIER 0x20000 /* Use block barriers */ #define EXT3_MOUNT_NOBH 0x40000 /* No bufferheads */ +#define EXT3_MOUNT_QUOTA 0x80000 /* Some quota option set */ /* Compatibility, for having both ext2_fs.h and ext3_fs.h included at once */ #ifndef _LINUX_EXT2_FS_H diff --git a/include/linux/ext3_jbd.h b/include/linux/ext3_jbd.h index e8292af9033b..c8307c02dd07 100644 --- a/include/linux/ext3_jbd.h +++ b/include/linux/ext3_jbd.h @@ -42,15 +42,15 @@ * superblock only gets updated once, of course, so don't bother * counting that again for the quota updates. */ -#define EXT3_DATA_TRANS_BLOCKS (EXT3_SINGLEDATA_TRANS_BLOCKS + \ +#define EXT3_DATA_TRANS_BLOCKS(sb) (EXT3_SINGLEDATA_TRANS_BLOCKS + \ EXT3_XATTR_TRANS_BLOCKS - 2 + \ - 2*EXT3_QUOTA_TRANS_BLOCKS) + 2*EXT3_QUOTA_TRANS_BLOCKS(sb)) /* Delete operations potentially hit one directory's namespace plus an * entire inode, plus arbitrary amounts of bitmap/indirection data. Be * generous. We can grow the delete transaction later if necessary. */ -#define EXT3_DELETE_TRANS_BLOCKS (2 * EXT3_DATA_TRANS_BLOCKS + 64) +#define EXT3_DELETE_TRANS_BLOCKS(sb) (2 * EXT3_DATA_TRANS_BLOCKS(sb) + 64) /* Define an arbitrary limit for the amount of data we will anticipate * writing to any given transaction. For unbounded transactions such as @@ -74,14 +74,17 @@ #ifdef CONFIG_QUOTA /* Amount of blocks needed for quota update - we know that the structure was * allocated so we need to update only inode+data */ -#define EXT3_QUOTA_TRANS_BLOCKS 2 +#define EXT3_QUOTA_TRANS_BLOCKS(sb) (test_opt(sb, QUOTA) ? 2 : 0) /* Amount of blocks needed for quota insert/delete - we do some block writes * but inode, sb and group updates are done only once */ -#define EXT3_QUOTA_INIT_BLOCKS (DQUOT_MAX_WRITES*\ - (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3) +#define EXT3_QUOTA_INIT_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_INIT_ALLOC*\ + (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_INIT_REWRITE) : 0) +#define EXT3_QUOTA_DEL_BLOCKS(sb) (test_opt(sb, QUOTA) ? (DQUOT_DEL_ALLOC*\ + (EXT3_SINGLEDATA_TRANS_BLOCKS-3)+3+DQUOT_DEL_REWRITE) : 0) #else -#define EXT3_QUOTA_TRANS_BLOCKS 0 -#define EXT3_QUOTA_INIT_BLOCKS 0 +#define EXT3_QUOTA_TRANS_BLOCKS(sb) 0 +#define EXT3_QUOTA_INIT_BLOCKS(sb) 0 +#define EXT3_QUOTA_DEL_BLOCKS(sb) 0 #endif int -- cgit v1.2.1 From 556a2a45bce1740f035befaa7201e4ad836c7257 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 23 Jun 2005 22:01:06 -0700 Subject: [PATCH] quota: reiserfs: improve quota credit estimates Use improved credits estimates for quota operations. Also reserve space for a quota operation in a transaction only if filesystem was mounted with some quota option. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/reiserfs_fs.h | 15 +++++++++++---- include/linux/reiserfs_fs_sb.h | 2 ++ 2 files changed, 13 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index 32148625fc2f..4c7c5689ad93 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -1644,11 +1644,18 @@ struct reiserfs_journal_header { #define JOURNAL_MAX_TRANS_AGE 30 #define JOURNAL_PER_BALANCE_CNT (3 * (MAX_HEIGHT-2) + 9) #ifdef CONFIG_QUOTA -#define REISERFS_QUOTA_TRANS_BLOCKS 2 /* We need to update data and inode (atime) */ -#define REISERFS_QUOTA_INIT_BLOCKS (DQUOT_MAX_WRITES*(JOURNAL_PER_BALANCE_CNT+2)+1) /* 1 balancing, 1 bitmap, 1 data per write + stat data update */ +/* We need to update data and inode (atime) */ +#define REISERFS_QUOTA_TRANS_BLOCKS(s) (REISERFS_SB(s)->s_mount_opt & (1<s_mount_opt & (1<s_mount_opt & (1< Date: Thu, 23 Jun 2005 22:01:42 -0700 Subject: [PATCH] ipmi: add power cycle capability This patch to adds "power cycle" functionality to the IPMI power off module ipmi_poweroff. It also contains changes to support procfs control of the feature. The power cycle action is considered an optional chassis control in the IPMI specification. However, it is definitely useful when the hardware supports it. A power cycle is usually required in order to reset a firmware in a bad state. This action is critical to allow remote management of servers. The implementation adds power cycle as optional to the ipmi_poweroff module. It can be modified dynamically through the proc entry mentioned above. During a power down and enabled, the power cycle command is sent to the BMC firmware. If it fails either due to non-support or some error, it will retry to send the command as power off. Signed-off-by: Christopher A. Poblete Signed-off-by: Corey Minyard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ipmi.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/ipmi.h b/include/linux/ipmi.h index 2ec265e1045f..596ca6130159 100644 --- a/include/linux/ipmi.h +++ b/include/linux/ipmi.h @@ -209,6 +209,11 @@ struct kernel_ipmi_msg #include #include +#ifdef CONFIG_PROC_FS +#include +extern struct proc_dir_entry *proc_ipmi_root; +#endif /* CONFIG_PROC_FS */ + /* Opaque type for a IPMI message user. One of these is needed to send and receive messages. */ typedef struct ipmi_user *ipmi_user_t; -- cgit v1.2.1 From a6df7da8f7ee99e6fd1995fad852bacb978a6447 Mon Sep 17 00:00:00 2001 From: Kylene Hall Date: Thu, 23 Jun 2005 22:02:04 -0700 Subject: [PATCH] tpm: TPMs on additional LPC bus Add support for TPMs on additional LPC buses. Signed-off-by: Kylene Hall Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pci_ids.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 63e89e47b8e9..bf608808a60c 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1568,6 +1568,7 @@ #define PCI_DEVICE_ID_SERVERWORKS_OSB4USB 0x0220 #define PCI_DEVICE_ID_SERVERWORKS_CSB5USB PCI_DEVICE_ID_SERVERWORKS_OSB4USB #define PCI_DEVICE_ID_SERVERWORKS_CSB6USB 0x0221 +#define PCI_DEVICE_ID_SERVERWORKS_CSB6LPC 0x0227 #define PCI_DEVICE_ID_SERVERWORKS_GCLE 0x0225 #define PCI_DEVICE_ID_SERVERWORKS_GCLE2 0x0227 #define PCI_DEVICE_ID_SERVERWORKS_CSB5ISA 0x0230 -- cgit v1.2.1 From 61fbfa8129c1771061a0e9f47747854293081c5b Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:11 -0700 Subject: [PATCH] I2O: bugfixes and compability enhancements Changes: - Fixed sysfs bug where user and parent links where added to the I2O device itself - Fixed bug when calculating TID for the event handler and cleaned up the workflow of i2o_driver_dispatch() - Fixed oops when no I2O device could be found for an event delivered to Exec-OSM - Fixed initialization of spinlock in Exec-OSM - Fixed memory leak in i2o_cfg_passthru() and i2o_cfg_passthru() - Removed MTRR support - Added PCI ID of Promise SX6000 with firmware >= 1.20.x.x - Turn of caching for ioremapped memory of in_queue - Added initialization sequence for Promise controllers - Moved definition of u8 / u16 / u32 for raidutils before first use Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2o-dev.h | 16 ++++++++-------- include/linux/i2o.h | 5 ----- 2 files changed, 8 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index ef7f644dd873..3414325bdcfd 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -24,6 +24,14 @@ #define MAX_I2O_CONTROLLERS 32 //#include +#ifndef __KERNEL__ + +typedef unsigned char u8; +typedef unsigned short u16; +typedef unsigned int u32; + +#endif /* __KERNEL__ */ + /* * I2O Control IOCTLs and structures @@ -126,14 +134,6 @@ struct i2o_evt_get { #define I2O_BUS_CARDBUS 7 #define I2O_BUS_UNKNOWN 0x80 -#ifndef __KERNEL__ - -typedef unsigned char u8; -typedef unsigned short u16; -typedef unsigned int u32; - -#endif /* __KERNEL__ */ - typedef struct _i2o_pci_bus { u8 PciFunctionNumber; u8 PciDeviceNumber; diff --git a/include/linux/i2o.h b/include/linux/i2o.h index ea9a3ad4b67f..40e45a83d3fb 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -152,11 +152,6 @@ struct i2o_controller { unsigned int raptor:1; /* split bar */ unsigned int promise:1; /* Promise controller */ -#ifdef CONFIG_MTRR - int mtrr_reg0; - int mtrr_reg1; -#endif - struct list_head devices; /* list of I2O devices */ struct notifier_block *event_notifer; /* Events */ -- cgit v1.2.1 From f88e119c4b824a5017456fa094950d0f4092d96c Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:14 -0700 Subject: [PATCH] I2O: first code cleanup of spare warnings and unused functions Changes: - Removed unnecessary checking of NULL before calling kfree() - Make some functions static - Changed pr_debug() into osm_debug() - Use i2o_msg_in_to_virt() for getting a pointer to the message frame - Cleaned up some comments - Changed some le32_to_cpu() into readl() where necessary - Make error messages of OSM's look the same - Cleaned up error handling in i2o_block_end_request() - Removed unused error handling of failed messages in Block-OSM, which are not allowed by the I2O spec - Corrected the blocksize detection in i2o_block - Added hrt and lct sysfs-attribute to controller - Call done() function in SCSI-OSM after freeing DMA buffers - Removed unneeded variable for message size calculation in i2o_scsi_queuecommand() - Make some changes to remove sparse warnings - Reordered some functions - Cleaned up controller initialization - Replaced some magic numbers by defines - Removed unnecessary dma_sync_single_for_cpu() call on coherent DMA - Removed some unused fields in i2o_controller and removed some unused functions Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2o.h | 74 ++++++++++++----------------------------------------- 1 file changed, 16 insertions(+), 58 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 40e45a83d3fb..e8cd11290010 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -153,12 +153,10 @@ struct i2o_controller { unsigned int promise:1; /* Promise controller */ struct list_head devices; /* list of I2O devices */ - - struct notifier_block *event_notifer; /* Events */ - atomic_t users; struct list_head list; /* Controller list */ - void __iomem *post_port; /* Inbout port address */ - void __iomem *reply_port; /* Outbound port address */ + + void __iomem *in_port; /* Inbout port address */ + void __iomem *out_port; /* Outbound port address */ void __iomem *irq_mask; /* Interrupt register address */ /* Dynamic LCT related data */ @@ -182,9 +180,6 @@ struct i2o_controller { struct resource io_resource; /* I/O resource allocated to the IOP */ struct resource mem_resource; /* Mem resource allocated to the IOP */ - struct proc_dir_entry *proc_entry; /* /proc dir */ - - struct list_head bus_list; /* list of busses on IOP */ struct device device; struct i2o_device *exec; /* Executive */ #if BITS_PER_LONG == 64 @@ -380,49 +375,10 @@ extern int i2o_device_claim_release(struct i2o_device *); /* Exec OSM functions */ extern int i2o_exec_lct_get(struct i2o_controller *); -/* device to i2o_device and driver to i2o_driver convertion functions */ +/* device / driver conversion functions */ #define to_i2o_driver(drv) container_of(drv,struct i2o_driver, driver) #define to_i2o_device(dev) container_of(dev, struct i2o_device, device) - -/* - * Messenger inlines - */ -static inline u32 I2O_POST_READ32(struct i2o_controller *c) -{ - rmb(); - return readl(c->post_port); -}; - -static inline void I2O_POST_WRITE32(struct i2o_controller *c, u32 val) -{ - wmb(); - writel(val, c->post_port); -}; - -static inline u32 I2O_REPLY_READ32(struct i2o_controller *c) -{ - rmb(); - return readl(c->reply_port); -}; - -static inline void I2O_REPLY_WRITE32(struct i2o_controller *c, u32 val) -{ - wmb(); - writel(val, c->reply_port); -}; - -static inline u32 I2O_IRQ_READ32(struct i2o_controller *c) -{ - rmb(); - return readl(c->irq_mask); -}; - -static inline void I2O_IRQ_WRITE32(struct i2o_controller *c, u32 val) -{ - wmb(); - writel(val, c->irq_mask); - wmb(); -}; +#define to_i2o_controller(dev) container_of(dev, struct i2o_controller, device) /** * i2o_msg_get - obtain an I2O message from the IOP @@ -440,10 +396,12 @@ static inline void I2O_IRQ_WRITE32(struct i2o_controller *c, u32 val) static inline u32 i2o_msg_get(struct i2o_controller *c, struct i2o_message __iomem **msg) { - u32 m; + u32 m = readl(c->in_port); - if ((m = I2O_POST_READ32(c)) != I2O_QUEUE_EMPTY) + if (m != I2O_QUEUE_EMPTY) { *msg = c->in_queue.virt + m; + rmb(); + } return m; }; @@ -457,7 +415,8 @@ static inline u32 i2o_msg_get(struct i2o_controller *c, */ static inline void i2o_msg_post(struct i2o_controller *c, u32 m) { - I2O_POST_WRITE32(c, m); + wmb(); + writel(m, c->in_port); }; /** @@ -486,12 +445,10 @@ static inline int i2o_msg_post_wait(struct i2o_controller *c, u32 m, * The I2O controller must be informed that the reply message is not needed * anymore. If you forget to flush the reply, the message frame can't be * used by the controller anymore and is therefore lost. - * - * FIXME: is there a timeout after which the controller reuse the message? */ static inline void i2o_flush_reply(struct i2o_controller *c, u32 m) { - I2O_REPLY_WRITE32(c, m); + writel(m, c->out_port); }; /** @@ -505,8 +462,9 @@ static inline void i2o_flush_reply(struct i2o_controller *c, u32 m) * work for sender side messages as they are ioremap objects * provided by the I2O controller. */ -static inline struct i2o_message *i2o_msg_out_to_virt(struct i2o_controller *c, - u32 m) +static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct + i2o_controller *c, + u32 m) { BUG_ON(m < c->out_queue.phys || m >= c->out_queue.phys + c->out_queue.len); @@ -917,7 +875,7 @@ extern void i2o_debug_state(struct i2o_controller *c); #define I2OVER15 0x0001 #define I2OVER20 0x0002 -/* Default is 1.5, FIXME: Need support for both 1.5 and 2.0 */ +/* Default is 1.5 */ #define I2OVERSION I2OVER15 #define SGL_OFFSET_0 I2OVERSION -- cgit v1.2.1 From f10378fff658f61307496e0ae00095041725cf07 Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:16 -0700 Subject: [PATCH] I2O: new sysfs attributes and Adaptec specific block device access and 64-bit DMA support Changes: - Added Bus-OSM which could be used by user space programs to reset a channel on the controller - Make ioctl's in Config-OSM obsolete in prefer for sysfs attributes and move those to its own file - Added sysfs attribute for firmware read and write access for I2O controllers - Added special handling of firmware read and write access for Adaptec controllers - Added vendor id and product id as sysfs-attribute to Executive classes - Added automatic notification of LCT change handling to Exec-OSM - Added flushing function to Block-OSM for later barrier implementation - Use PRIVATE messages for Block access on Adaptec controllers, which are faster then BLOCK class access - Cleaned up support for Promise controller - New messages are now detected using the IRQ status register as suggested by the I2O spec - Added i2o_dma_high() and i2o_dma_low() functions - Added facility for SG tablesize calculation when using 32-bit and 64-bit DMA addresses - Added i2o_dma_map_single() and i2o_dma_map_sg() which could build the SG list for 32-bit as well as 64-bit DMA addresses Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2o-dev.h | 6 +- include/linux/i2o.h | 321 +++++++++++++++++++++++++++++++++++++++--------- 2 files changed, 270 insertions(+), 57 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index 3414325bdcfd..90c984ecd521 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -32,6 +32,10 @@ typedef unsigned int u32; #endif /* __KERNEL__ */ +/* + * Vendors + */ +#define I2O_VENDOR_DPT 0x001b /* * I2O Control IOCTLs and structures @@ -333,7 +337,7 @@ typedef struct _i2o_status_block { #define I2O_CLASS_ATE_PERIPHERAL 0x061 #define I2O_CLASS_FLOPPY_CONTROLLER 0x070 #define I2O_CLASS_FLOPPY_DEVICE 0x071 -#define I2O_CLASS_BUS_ADAPTER_PORT 0x080 +#define I2O_CLASS_BUS_ADAPTER 0x080 #define I2O_CLASS_PEER_TRANSPORT_AGENT 0x090 #define I2O_CLASS_PEER_TRANSPORT 0x091 #define I2O_CLASS_END 0xfff diff --git a/include/linux/i2o.h b/include/linux/i2o.h index e8cd11290010..497ea574f96b 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -157,7 +157,8 @@ struct i2o_controller { void __iomem *in_port; /* Inbout port address */ void __iomem *out_port; /* Outbound port address */ - void __iomem *irq_mask; /* Interrupt register address */ + void __iomem *irq_status; /* Interrupt status register address */ + void __iomem *irq_mask; /* Interrupt mask register address */ /* Dynamic LCT related data */ @@ -242,15 +243,6 @@ extern int i2o_msg_post_wait_mem(struct i2o_controller *, u32, unsigned long, extern void i2o_msg_nop(struct i2o_controller *, u32); static inline void i2o_flush_reply(struct i2o_controller *, u32); -/* DMA handling functions */ -static inline int i2o_dma_alloc(struct device *, struct i2o_dma *, size_t, - unsigned int); -static inline void i2o_dma_free(struct device *, struct i2o_dma *); -int i2o_dma_realloc(struct device *, struct i2o_dma *, size_t, unsigned int); - -static inline int i2o_dma_map(struct device *, struct i2o_dma *); -static inline void i2o_dma_unmap(struct device *, struct i2o_dma *); - /* IOP functions */ extern int i2o_status_get(struct i2o_controller *); @@ -275,6 +267,16 @@ static inline u32 i2o_ptr_high(void *ptr) { return (u32) ((u64) ptr >> 32); }; + +static inline u32 i2o_dma_low(dma_addr_t dma_addr) +{ + return (u32) (u64) dma_addr; +}; + +static inline u32 i2o_dma_high(dma_addr_t dma_addr) +{ + return (u32) ((u64) dma_addr >> 32); +}; #else static inline u32 i2o_cntxt_list_add(struct i2o_controller *c, void *ptr) { @@ -305,8 +307,246 @@ static inline u32 i2o_ptr_high(void *ptr) { return 0; }; + +static inline u32 i2o_dma_low(dma_addr_t dma_addr) +{ + return (u32) dma_addr; +}; + +static inline u32 i2o_dma_high(dma_addr_t dma_addr) +{ + return 0; +}; +#endif + +/** + * i2o_sg_tablesize - Calculate the maximum number of elements in a SGL + * @c: I2O controller for which the calculation should be done + * @body_size: maximum body size used for message in 32-bit words. + * + * Return the maximum number of SG elements in a SG list. + */ +static inline u16 i2o_sg_tablesize(struct i2o_controller *c, u16 body_size) +{ + i2o_status_block *sb = c->status_block.virt; + u16 sg_count = + (sb->inbound_frame_size - sizeof(struct i2o_message) / 4) - + body_size; + + if (c->pae_support) { + /* + * for 64-bit a SG attribute element must be added and each + * SG element needs 12 bytes instead of 8. + */ + sg_count -= 2; + sg_count /= 3; + } else + sg_count /= 2; + + if (c->short_req && (sg_count > 8)) + sg_count = 8; + + return sg_count; +}; + +/** + * i2o_dma_map_single - Map pointer to controller and fill in I2O message. + * @c: I2O controller + * @ptr: pointer to the data which should be mapped + * @size: size of data in bytes + * @direction: DMA_TO_DEVICE / DMA_FROM_DEVICE + * @sg_ptr: pointer to the SG list inside the I2O message + * + * This function does all necessary DMA handling and also writes the I2O + * SGL elements into the I2O message. For details on DMA handling see also + * dma_map_single(). The pointer sg_ptr will only be set to the end of the + * SG list if the allocation was successful. + * + * Returns DMA address which must be checked for failures using + * dma_mapping_error(). + */ +static inline dma_addr_t i2o_dma_map_single(struct i2o_controller *c, void *ptr, + size_t size, + enum dma_data_direction direction, + u32 __iomem ** sg_ptr) +{ + u32 sg_flags; + u32 __iomem *mptr = *sg_ptr; + dma_addr_t dma_addr; + + switch (direction) { + case DMA_TO_DEVICE: + sg_flags = 0xd4000000; + break; + case DMA_FROM_DEVICE: + sg_flags = 0xd0000000; + break; + default: + return 0; + } + + dma_addr = dma_map_single(&c->pdev->dev, ptr, size, direction); + if (!dma_mapping_error(dma_addr)) { +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if ((sizeof(dma_addr_t) > 4) && c->pae_support) { + writel(0x7C020002, mptr++); + writel(PAGE_SIZE, mptr++); + } +#endif + + writel(sg_flags | size, mptr++); + writel(i2o_dma_low(dma_addr), mptr++); +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if ((sizeof(dma_addr_t) > 4) && c->pae_support) + writel(i2o_dma_high(dma_addr), mptr++); +#endif + *sg_ptr = mptr; + } + return dma_addr; +}; + +/** + * i2o_dma_map_sg - Map a SG List to controller and fill in I2O message. + * @c: I2O controller + * @sg: SG list to be mapped + * @sg_count: number of elements in the SG list + * @direction: DMA_TO_DEVICE / DMA_FROM_DEVICE + * @sg_ptr: pointer to the SG list inside the I2O message + * + * This function does all necessary DMA handling and also writes the I2O + * SGL elements into the I2O message. For details on DMA handling see also + * dma_map_sg(). The pointer sg_ptr will only be set to the end of the SG + * list if the allocation was successful. + * + * Returns 0 on failure or 1 on success. + */ +static inline int i2o_dma_map_sg(struct i2o_controller *c, + struct scatterlist *sg, int sg_count, + enum dma_data_direction direction, + u32 __iomem ** sg_ptr) +{ + u32 sg_flags; + u32 __iomem *mptr = *sg_ptr; + + switch (direction) { + case DMA_TO_DEVICE: + sg_flags = 0x14000000; + break; + case DMA_FROM_DEVICE: + sg_flags = 0x10000000; + break; + default: + return 0; + } + + sg_count = dma_map_sg(&c->pdev->dev, sg, sg_count, direction); + if (!sg_count) + return 0; + +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if ((sizeof(dma_addr_t) > 4) && c->pae_support) { + writel(0x7C020002, mptr++); + writel(PAGE_SIZE, mptr++); + } #endif + while (sg_count-- > 0) { + if (!sg_count) + sg_flags |= 0xC0000000; + writel(sg_flags | sg_dma_len(sg), mptr++); + writel(i2o_dma_low(sg_dma_address(sg)), mptr++); +#ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64 + if ((sizeof(dma_addr_t) > 4) && c->pae_support) + writel(i2o_dma_high(sg_dma_address(sg)), mptr++); +#endif + sg++; + } + *sg_ptr = mptr; + + return 1; +}; + +/** + * i2o_dma_alloc - Allocate DMA memory + * @dev: struct device pointer to the PCI device of the I2O controller + * @addr: i2o_dma struct which should get the DMA buffer + * @len: length of the new DMA memory + * @gfp_mask: GFP mask + * + * Allocate a coherent DMA memory and write the pointers into addr. + * + * Returns 0 on success or -ENOMEM on failure. + */ +static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr, + size_t len, unsigned int gfp_mask) +{ + struct pci_dev *pdev = to_pci_dev(dev); + int dma_64 = 0; + + if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) { + dma_64 = 1; + if (pci_set_dma_mask(pdev, DMA_32BIT_MASK)) + return -ENOMEM; + } + + addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask); + + if ((sizeof(dma_addr_t) > 4) && dma_64) + if (pci_set_dma_mask(pdev, DMA_64BIT_MASK)) + printk(KERN_WARNING "i2o: unable to set 64-bit DMA"); + + if (!addr->virt) + return -ENOMEM; + + memset(addr->virt, 0, len); + addr->len = len; + + return 0; +}; + +/** + * i2o_dma_free - Free DMA memory + * @dev: struct device pointer to the PCI device of the I2O controller + * @addr: i2o_dma struct which contains the DMA buffer + * + * Free a coherent DMA memory and set virtual address of addr to NULL. + */ +static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr) +{ + if (addr->virt) { + if (addr->phys) + dma_free_coherent(dev, addr->len, addr->virt, + addr->phys); + else + kfree(addr->virt); + addr->virt = NULL; + } +}; + +/** + * i2o_dma_realloc - Realloc DMA memory + * @dev: struct device pointer to the PCI device of the I2O controller + * @addr: pointer to a i2o_dma struct DMA buffer + * @len: new length of memory + * @gfp_mask: GFP mask + * + * If there was something allocated in the addr, free it first. If len > 0 + * than try to allocate it and write the addresses back to the addr + * structure. If len == 0 set the virtual address to NULL. + * + * Returns the 0 on success or negative error code on failure. + */ +static inline int i2o_dma_realloc(struct device *dev, struct i2o_dma *addr, + size_t len, unsigned int gfp_mask) +{ + i2o_dma_free(dev, addr); + + if (len) + return i2o_dma_alloc(dev, addr, len, gfp_mask); + + return 0; +}; + /* I2O driver (OSM) functions */ extern int i2o_driver_register(struct i2o_driver *); extern void i2o_driver_unregister(struct i2o_driver *); @@ -375,10 +615,11 @@ extern int i2o_device_claim_release(struct i2o_device *); /* Exec OSM functions */ extern int i2o_exec_lct_get(struct i2o_controller *); -/* device / driver conversion functions */ +/* device / driver / kobject conversion functions */ #define to_i2o_driver(drv) container_of(drv,struct i2o_driver, driver) #define to_i2o_device(dev) container_of(dev, struct i2o_device, device) #define to_i2o_controller(dev) container_of(dev, struct i2o_controller, device) +#define kobj_to_i2o_device(kobj) to_i2o_device(container_of(kobj, struct device, kobj)) /** * i2o_msg_get - obtain an I2O message from the IOP @@ -466,8 +707,10 @@ static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct i2o_controller *c, u32 m) { - BUG_ON(m < c->out_queue.phys - || m >= c->out_queue.phys + c->out_queue.len); + if (unlikely + (m < c->out_queue.phys + || m >= c->out_queue.phys + c->out_queue.len)) + return NULL; return c->out_queue.virt + (m - c->out_queue.phys); }; @@ -532,48 +775,6 @@ static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr) } }; -/** - * i2o_dma_map - Map the memory to DMA - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: i2o_dma struct which should be mapped - * - * Map the memory in addr->virt to coherent DMA memory and write the - * physical address into addr->phys. - * - * Returns 0 on success or -ENOMEM on failure. - */ -static inline int i2o_dma_map(struct device *dev, struct i2o_dma *addr) -{ - if (!addr->virt) - return -EFAULT; - - if (!addr->phys) - addr->phys = dma_map_single(dev, addr->virt, addr->len, - DMA_BIDIRECTIONAL); - if (!addr->phys) - return -ENOMEM; - - return 0; -}; - -/** - * i2o_dma_unmap - Unmap the DMA memory - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: i2o_dma struct which should be unmapped - * - * Unmap the memory in addr->virt from DMA memory. - */ -static inline void i2o_dma_unmap(struct device *dev, struct i2o_dma *addr) -{ - if (!addr->virt) - return; - - if (addr->phys) { - dma_unmap_single(dev, addr->phys, addr->len, DMA_BIDIRECTIONAL); - addr->phys = 0; - } -}; - /* * Endian handling wrapped into the macro - keeps the core code * cleaner. @@ -725,6 +926,14 @@ extern void i2o_debug_state(struct i2o_controller *c); #define I2O_CMD_SCSI_ABORT 0x83 #define I2O_CMD_SCSI_BUSRESET 0x27 +/* + * Bus Adapter Class + */ +#define I2O_CMD_BUS_ADAPTER_RESET 0x85 +#define I2O_CMD_BUS_RESET 0x87 +#define I2O_CMD_BUS_SCAN 0x89 +#define I2O_CMD_BUS_QUIESCE 0x8b + /* * Random Block Storage Class */ @@ -948,7 +1157,7 @@ extern void i2o_debug_state(struct i2o_controller *c); /* request queue sizes */ #define I2O_MAX_SECTORS 1024 -#define I2O_MAX_SEGMENTS 128 +#define I2O_MAX_PHYS_SEGMENTS MAX_PHYS_SEGMENTS #define I2O_REQ_MEMPOOL_SIZE 32 -- cgit v1.2.1 From b2aaee33fbb354a2f08121aa1c1be55841102761 Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:19 -0700 Subject: [PATCH] I2O: Adaptec specific SG_IO access, firmware access through sysfs and 2400A workaround Changes: - Provide SG_IO access to BLOCK and EXECUTIVE class on Adaptec controllers - Use PRIVATE messages in SCSI-OSM because on some controllers normal SCSI class commands like READ or READ CAPACITY cause errors - Use new DMA and SG list creation function - Added workaround to limit sectors per request for Adaptec 2400A controllers Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2o-dev.h | 22 ++++++++++++++++++++++ include/linux/i2o.h | 37 ++++++++++++++++++++++++++----------- 2 files changed, 48 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index 90c984ecd521..d4a08d29e36d 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -32,6 +32,13 @@ typedef unsigned int u32; #endif /* __KERNEL__ */ +/* + * Software module types + */ +#define I2O_SOFTWARE_MODULE_IRTOS 0x11 +#define I2O_SOFTWARE_MODULE_IOP_PRIVATE 0x22 +#define I2O_SOFTWARE_MODULE_IOP_CONFIG 0x23 + /* * Vendors */ @@ -125,6 +132,10 @@ struct i2o_evt_get { int lost; }; +typedef struct i2o_sg_io_hdr { + unsigned int flags; /* see I2O_DPT_SG_IO_FLAGS */ +} i2o_sg_io_hdr_t; + /************************************************************************** * HRT related constants and structures **************************************************************************/ @@ -403,4 +414,15 @@ typedef struct _i2o_status_block { #define ADAPTER_STATE_FAILED 0x10 #define ADAPTER_STATE_FAULTED 0x11 + +/* + * DPT / Adaptec specific values for i2o_sg_io_hdr flags. + */ +#define I2O_DPT_SG_FLAG_INTERPRET 0x00010000 +#define I2O_DPT_SG_FLAG_PHYSICAL 0x00020000 + +#define I2O_DPT_FLASH_FRAG_SIZE 0x10000 +#define I2O_DPT_FLASH_READ 0x0101 +#define I2O_DPT_FLASH_WRITE 0x0102 + #endif /* _I2O_DEV_H */ diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 497ea574f96b..2039a87c2b91 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -147,10 +147,13 @@ struct i2o_controller { struct pci_dev *pdev; /* PCI device */ - unsigned int short_req:1; /* use small block sizes */ - unsigned int no_quiesce:1; /* dont quiesce before reset */ - unsigned int raptor:1; /* split bar */ unsigned int promise:1; /* Promise controller */ + unsigned int adaptec:1; /* DPT / Adaptec controller */ + unsigned int raptor:1; /* split bar */ + unsigned int no_quiesce:1; /* dont quiesce before reset */ + unsigned int short_req:1; /* use small block sizes */ + unsigned int limit_sectors:1; /* limit number of sectors / request */ + unsigned int pae_support:1; /* controller has 64-bit SGL support */ struct list_head devices; /* list of I2O devices */ struct list_head list; /* Controller list */ @@ -746,7 +749,21 @@ static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct i2o_controll static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr, size_t len, unsigned int gfp_mask) { + struct pci_dev *pdev = to_pci_dev(dev); + int dma_64 = 0; + + if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) { + dma_64 = 1; + if(pci_set_dma_mask(pdev, DMA_32BIT_MASK)) + return -ENOMEM; + } + addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask); + + if ((sizeof(dma_addr_t) > 4) && dma_64) + if(pci_set_dma_mask(pdev, DMA_64BIT_MASK)) + printk(KERN_WARNING "i2o: unable to set 64-bit DMA"); + if (!addr->virt) return -ENOMEM; @@ -946,7 +963,7 @@ extern void i2o_debug_state(struct i2o_controller *c); #define I2O_CMD_BLOCK_MEJECT 0x43 #define I2O_CMD_BLOCK_POWER 0x70 -#define I2O_PRIVATE_MSG 0xFF +#define I2O_CMD_PRIVATE 0xFF /* Command status values */ @@ -1095,9 +1112,9 @@ extern void i2o_debug_state(struct i2o_controller *c); #define SGL_OFFSET_8 (0x0080 | I2OVERSION) #define SGL_OFFSET_9 (0x0090 | I2OVERSION) #define SGL_OFFSET_10 (0x00A0 | I2OVERSION) - -#define TRL_OFFSET_5 (0x0050 | I2OVERSION) -#define TRL_OFFSET_6 (0x0060 | I2OVERSION) +#define SGL_OFFSET_11 (0x00B0 | I2OVERSION) +#define SGL_OFFSET_12 (0x00C0 | I2OVERSION) +#define SGL_OFFSET(x) (((x)<<4) | I2OVERSION) /* Transaction Reply Lists (TRL) Control Word structure */ #define TRL_SINGLE_FIXED_LENGTH 0x00 @@ -1130,7 +1147,6 @@ extern void i2o_debug_state(struct i2o_controller *c); #define HOST_TID 1 #define MSG_FRAME_SIZE 128 /* i2o_scsi assumes >= 32 */ -#define REPLY_FRAME_SIZE 17 #define SG_TABLESIZE 30 #define NMBR_MSG_FRAMES 128 @@ -1155,11 +1171,10 @@ extern void i2o_debug_state(struct i2o_controller *c); #define I2O_HRT_GET_TRIES 3 #define I2O_LCT_GET_TRIES 3 -/* request queue sizes */ +/* defines for max_sectors and max_phys_segments */ #define I2O_MAX_SECTORS 1024 +#define I2O_MAX_SECTORS_LIMITED 256 #define I2O_MAX_PHYS_SEGMENTS MAX_PHYS_SEGMENTS -#define I2O_REQ_MEMPOOL_SIZE 32 - #endif /* __KERNEL__ */ #endif /* _I2O_H */ -- cgit v1.2.1 From 9e87545f06930c1d294423a8091d1077e7444a47 Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:21 -0700 Subject: [PATCH] I2O: second code cleanup of sparse warnings and unneeded syncronization Changes: - Added header "core.h" for i2o_core.ko internal definitions - More sparse fixes - Changed display of TID's in sysfs attributes from XXX to 0xXXX - Use the right functions for accessing I/O and normal memory - Removed error handling of SCSI device errors and let the SCSI layer take care of it - Added new device / removed device handling to SCSI-OSM - Make status access volatile - Cleaned up activation of I2O controller - Removed unnecessary wmb() and rmb() calls - Use own struct i2o_io for I/O memory instead of struct i2o_dma Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2o-dev.h | 23 +++++----- include/linux/i2o.h | 116 +++++++++++++----------------------------------- 2 files changed, 42 insertions(+), 97 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2o-dev.h b/include/linux/i2o-dev.h index d4a08d29e36d..36fd18cdad28 100644 --- a/include/linux/i2o-dev.h +++ b/include/linux/i2o-dev.h @@ -32,18 +32,6 @@ typedef unsigned int u32; #endif /* __KERNEL__ */ -/* - * Software module types - */ -#define I2O_SOFTWARE_MODULE_IRTOS 0x11 -#define I2O_SOFTWARE_MODULE_IOP_PRIVATE 0x22 -#define I2O_SOFTWARE_MODULE_IOP_CONFIG 0x23 - -/* - * Vendors - */ -#define I2O_VENDOR_DPT 0x001b - /* * I2O Control IOCTLs and structures */ @@ -414,6 +402,17 @@ typedef struct _i2o_status_block { #define ADAPTER_STATE_FAILED 0x10 #define ADAPTER_STATE_FAULTED 0x11 +/* + * Software module types + */ +#define I2O_SOFTWARE_MODULE_IRTOS 0x11 +#define I2O_SOFTWARE_MODULE_IOP_PRIVATE 0x22 +#define I2O_SOFTWARE_MODULE_IOP_CONFIG 0x23 + +/* + * Vendors + */ +#define I2O_VENDOR_DPT 0x001b /* * DPT / Adaptec specific values for i2o_sg_io_hdr flags. diff --git a/include/linux/i2o.h b/include/linux/i2o.h index 2039a87c2b91..be937d0372a7 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -119,12 +119,21 @@ struct i2o_driver { }; /* - * Contains all information which are necessary for DMA operations + * Contains DMA mapped address information */ struct i2o_dma { void *virt; dma_addr_t phys; - u32 len; + size_t len; +}; + +/* + * Contains IO mapped address information + */ +struct i2o_io { + void __iomem *virt; + unsigned long phys; + unsigned long len; }; /* @@ -173,8 +182,8 @@ struct i2o_controller { struct semaphore lct_lock; /* Lock for LCT updates */ struct i2o_dma status_block; /* IOP status block */ - struct i2o_dma base; /* controller messaging unit */ - struct i2o_dma in_queue; /* inbound message queue Host->IOP */ + struct i2o_io base; /* controller messaging unit */ + struct i2o_io in_queue; /* inbound message queue Host->IOP */ struct i2o_dma out_queue; /* outbound message queue IOP->Host */ unsigned int battery:1; /* Has a battery backup */ @@ -185,6 +194,7 @@ struct i2o_controller { struct resource mem_resource; /* Mem resource allocated to the IOP */ struct device device; + struct class_device classdev; /* I2O controller class */ struct i2o_device *exec; /* Executive */ #if BITS_PER_LONG == 64 spinlock_t context_list_lock; /* lock for context_list */ @@ -235,9 +245,10 @@ struct i2o_sys_tbl { extern struct list_head i2o_controllers; /* Message functions */ -static inline u32 i2o_msg_get(struct i2o_controller *, struct i2o_message __iomem **); -extern u32 i2o_msg_get_wait(struct i2o_controller *, struct i2o_message __iomem **, - int); +static inline u32 i2o_msg_get(struct i2o_controller *, + struct i2o_message __iomem **); +extern u32 i2o_msg_get_wait(struct i2o_controller *, + struct i2o_message __iomem **, int); static inline void i2o_msg_post(struct i2o_controller *, u32); static inline int i2o_msg_post_wait(struct i2o_controller *, u32, unsigned long); @@ -638,14 +649,12 @@ extern int i2o_exec_lct_get(struct i2o_controller *); * available returns I2O_QUEUE_EMPTY and msg is leaved untouched. */ static inline u32 i2o_msg_get(struct i2o_controller *c, - struct i2o_message __iomem **msg) + struct i2o_message __iomem ** msg) { u32 m = readl(c->in_port); - if (m != I2O_QUEUE_EMPTY) { + if (m != I2O_QUEUE_EMPTY) *msg = c->in_queue.virt + m; - rmb(); - } return m; }; @@ -659,7 +668,6 @@ static inline u32 i2o_msg_get(struct i2o_controller *c, */ static inline void i2o_msg_post(struct i2o_controller *c, u32 m) { - wmb(); writel(m, c->in_port); }; @@ -706,14 +714,11 @@ static inline void i2o_flush_reply(struct i2o_controller *c, u32 m) * work for sender side messages as they are ioremap objects * provided by the I2O controller. */ -static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct - i2o_controller *c, - u32 m) +static inline struct i2o_message *i2o_msg_out_to_virt(struct i2o_controller *c, + u32 m) { - if (unlikely - (m < c->out_queue.phys - || m >= c->out_queue.phys + c->out_queue.len)) - return NULL; + BUG_ON(m < c->out_queue.phys + || m >= c->out_queue.phys + c->out_queue.len); return c->out_queue.virt + (m - c->out_queue.phys); }; @@ -729,69 +734,13 @@ static inline struct i2o_message __iomem *i2o_msg_out_to_virt(struct * work for receive side messages as they are kmalloc objects * in a different pool. */ -static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct i2o_controller *c, - u32 m) +static inline struct i2o_message __iomem *i2o_msg_in_to_virt(struct + i2o_controller *c, + u32 m) { return c->in_queue.virt + m; }; -/** - * i2o_dma_alloc - Allocate DMA memory - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: i2o_dma struct which should get the DMA buffer - * @len: length of the new DMA memory - * @gfp_mask: GFP mask - * - * Allocate a coherent DMA memory and write the pointers into addr. - * - * Returns 0 on success or -ENOMEM on failure. - */ -static inline int i2o_dma_alloc(struct device *dev, struct i2o_dma *addr, - size_t len, unsigned int gfp_mask) -{ - struct pci_dev *pdev = to_pci_dev(dev); - int dma_64 = 0; - - if ((sizeof(dma_addr_t) > 4) && (pdev->dma_mask == DMA_64BIT_MASK)) { - dma_64 = 1; - if(pci_set_dma_mask(pdev, DMA_32BIT_MASK)) - return -ENOMEM; - } - - addr->virt = dma_alloc_coherent(dev, len, &addr->phys, gfp_mask); - - if ((sizeof(dma_addr_t) > 4) && dma_64) - if(pci_set_dma_mask(pdev, DMA_64BIT_MASK)) - printk(KERN_WARNING "i2o: unable to set 64-bit DMA"); - - if (!addr->virt) - return -ENOMEM; - - memset(addr->virt, 0, len); - addr->len = len; - - return 0; -}; - -/** - * i2o_dma_free - Free DMA memory - * @dev: struct device pointer to the PCI device of the I2O controller - * @addr: i2o_dma struct which contains the DMA buffer - * - * Free a coherent DMA memory and set virtual address of addr to NULL. - */ -static inline void i2o_dma_free(struct device *dev, struct i2o_dma *addr) -{ - if (addr->virt) { - if (addr->phys) - dma_free_coherent(dev, addr->len, addr->virt, - addr->phys); - else - kfree(addr->virt); - addr->virt = NULL; - } -}; - /* * Endian handling wrapped into the macro - keeps the core code * cleaner. @@ -1141,16 +1090,13 @@ extern void i2o_debug_state(struct i2o_controller *c); #define ELEVEN_WORD_MSG_SIZE 0x000B0000 #define I2O_MESSAGE_SIZE(x) ((x)<<16) -/* Special TID Assignments */ - +/* special TID assignments */ #define ADAPTER_TID 0 #define HOST_TID 1 -#define MSG_FRAME_SIZE 128 /* i2o_scsi assumes >= 32 */ -#define SG_TABLESIZE 30 -#define NMBR_MSG_FRAMES 128 - -#define MSG_POOL_SIZE (MSG_FRAME_SIZE*NMBR_MSG_FRAMES*sizeof(u32)) +/* outbound queue defines */ +#define I2O_MAX_OUTBOUND_MSG_FRAMES 128 +#define I2O_OUTBOUND_MSG_FRAME_SIZE 128 /* in 32-bit words */ #define I2O_POST_WAIT_OK 0 #define I2O_POST_WAIT_TIMEOUT -ETIMEDOUT -- cgit v1.2.1 From f33213ecf49c98da4e85121b592c3bea8057c2e6 Mon Sep 17 00:00:00 2001 From: Markus Lidel Date: Thu, 23 Jun 2005 22:02:23 -0700 Subject: [PATCH] I2O: Lindent run and replacement of printk through osm printing functions Lindent run and replaced printk() through the corresponding osm_*() function Signed-off-by: Markus Lidel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2o.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/i2o.h b/include/linux/i2o.h index be937d0372a7..bdc286ec947c 100644 --- a/include/linux/i2o.h +++ b/include/linux/i2o.h @@ -156,8 +156,8 @@ struct i2o_controller { struct pci_dev *pdev; /* PCI device */ - unsigned int promise:1; /* Promise controller */ - unsigned int adaptec:1; /* DPT / Adaptec controller */ + unsigned int promise:1; /* Promise controller */ + unsigned int adaptec:1; /* DPT / Adaptec controller */ unsigned int raptor:1; /* split bar */ unsigned int no_quiesce:1; /* dont quiesce before reset */ unsigned int short_req:1; /* use small block sizes */ @@ -174,7 +174,7 @@ struct i2o_controller { /* Dynamic LCT related data */ - struct i2o_dma status; /* status of IOP */ + struct i2o_dma status; /* IOP status block */ struct i2o_dma hrt; /* HW Resource Table */ i2o_lct *lct; /* Logical Config Table */ @@ -186,7 +186,7 @@ struct i2o_controller { struct i2o_io in_queue; /* inbound message queue Host->IOP */ struct i2o_dma out_queue; /* outbound message queue IOP->Host */ - unsigned int battery:1; /* Has a battery backup */ + unsigned int battery:1; /* Has a battery backup */ unsigned int io_alloc:1; /* An I/O resource was allocated */ unsigned int mem_alloc:1; /* A memory resource was allocated */ -- cgit v1.2.1 From 496400014f22c4dbdbc1e89249a2feba46939708 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:02:58 -0700 Subject: [PATCH] nfsd4: fix fh_expire_type We're returning NFS4_FH_NOEXPIRE_WITH_OPEN | NFS4_FH_VOL_RENAME for the fh_expire_type attribute. This is incorrect: 1. The spec actually only allows NOEXPIRE_WITH_OPEN when VOLATILE_ANY is also set. 2. Filehandles for open files can expire, if the file is removed and there is a reboot. 3. Filehandles are only volatile on rename in the nosubtree check case. Unfortunately, there's no way to indicate that we only expire on remove. So our only choice is FH4_VOLATILE_ANY. Although it's redundant, we also set FH4_VOL_RENAME in the subtree check case, since subtreecheck does actually cause problems in practice and it seems possibly useful to give clients some way to distinguish that case. Fix a mispelled #define while we're at it. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfs4.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs4.h b/include/linux/nfs4.h index 5bb5b2fd7ba2..0c1c306cdaec 100644 --- a/include/linux/nfs4.h +++ b/include/linux/nfs4.h @@ -28,7 +28,7 @@ #define NFS4_ACCESS_DELETE 0x0010 #define NFS4_ACCESS_EXECUTE 0x0020 -#define NFS4_FH_PERISTENT 0x0000 +#define NFS4_FH_PERSISTENT 0x0000 #define NFS4_FH_NOEXPIRE_WITH_OPEN 0x0001 #define NFS4_FH_VOLATILE_ANY 0x0002 #define NFS4_FH_VOL_MIGRATION 0x0004 -- cgit v1.2.1 From 8beefa249371f55432394ac96864c83b0b309c28 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:08 -0700 Subject: [PATCH] nfsd4: rename nfs4_file fields Trivial renaming patch: I can never remember, while looking at various lists relating the nfsd4 state structures, which are the "heads" and which are items on other lists, or which structures are actually on the various lists. The following convention helps me: given structures foo and bar, with foo containing the head of a list of bars, use "bars" for the name of the head of the list contained in the struct foo, and use "per_foo" for the entries in the struct bars. Go ahead and do this for struct nfs4_file. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index b6b2fe1e7c63..2c3b42674a4c 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -218,8 +218,8 @@ struct nfs4_stateowner { */ struct nfs4_file { struct list_head fi_hash; /* hash by "struct inode *" */ - struct list_head fi_perfile; /* list: nfs4_stateid */ - struct list_head fi_del_perfile; /* list: nfs4_delegation */ + struct list_head fi_stateids; + struct list_head fi_delegations; struct inode *fi_inode; u32 fi_id; /* used with stateowner->so_id * for stateid_hashtbl hash */ -- cgit v1.2.1 From 13cd21845d6a9729ca95e36ae6e8c669623fbfd4 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:10 -0700 Subject: [PATCH] nfsd4: reference count struct nfs4_file Add a struct kref to each nfs4_file and take a reference to it from each stateid and delegation that refers to it. The atomicity guarantees are overkill given that all this stuff is done under the single nfsd4 state lock, but a) we'd like finer-grained locking some day, and b) this simplifies the cleanup of the structures a bit, something that has previously been a bit complicated and bug-prone. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 2c3b42674a4c..296e6429fc3b 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -217,6 +217,7 @@ struct nfs4_stateowner { * share_acces, share_deny on the file. */ struct nfs4_file { + struct kref fi_ref; struct list_head fi_hash; /* hash by "struct inode *" */ struct list_head fi_stateids; struct list_head fi_delegations; -- cgit v1.2.1 From 7b190fecfa33d72bcf74c9473134c2ad14ae9545 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:23 -0700 Subject: [PATCH] knfsd: nfsd4: delegation recovery Allow recovery of delegations after reboot. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/xdr4.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/xdr4.h b/include/linux/nfsd/xdr4.h index a1f5ad0be1bf..4d24d65c0e88 100644 --- a/include/linux/nfsd/xdr4.h +++ b/include/linux/nfsd/xdr4.h @@ -210,6 +210,7 @@ struct nfsd4_open { u32 op_share_access; /* request */ u32 op_share_deny; /* request */ stateid_t op_stateid; /* response */ + u32 op_recall; /* recall */ struct nfsd4_change_info op_cinfo; /* response */ u32 op_rflags; /* response */ int op_truncate; /* used during processing */ -- cgit v1.2.1 From 76a3550ec50ed86885a10a767ebaebb7c9104721 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:26 -0700 Subject: [PATCH] knfsd: nfsd4: rename nfs4_state_init Somewhat gratuitous rename to simplify following patch. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/nfsd.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 4bf931d5ff56..3855fdc5af77 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -145,12 +145,12 @@ int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *); * NFSv4 State */ #ifdef CONFIG_NFSD_V4 -int nfs4_state_init(void); +int nfs4_state_start(void); void nfs4_state_shutdown(void); time_t nfs4_lease_time(void); void nfs4_reset_lease(time_t leasetime); #else -static inline int nfs4_state_init(void){return 0;} +static inline int nfs4_state_start(void){return 0;} static inline void nfs4_state_shutdown(void){} static inline time_t nfs4_lease_time(void){return 0;} static inline void nfs4_reset_lease(time_t leasetime){} -- cgit v1.2.1 From ac4d8ff2a57179de3ef7834c6ab3fac430b0a05d Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:30 -0700 Subject: [PATCH] knfsd: nfsd4: clean up state initialization Separate out stuff that needs initialization on startup from stuff that only needs initialization on module init from static data. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/nfsd.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 3855fdc5af77..21c6e9d86e4f 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -145,11 +145,13 @@ int nfsd_set_posix_acl(struct svc_fh *, int, struct posix_acl *); * NFSv4 State */ #ifdef CONFIG_NFSD_V4 +void nfs4_state_init(void); int nfs4_state_start(void); void nfs4_state_shutdown(void); time_t nfs4_lease_time(void); void nfs4_reset_lease(time_t leasetime); #else +static inline void nfs4_state_init(void){}; static inline int nfs4_state_start(void){return 0;} static inline void nfs4_state_shutdown(void){} static inline time_t nfs4_lease_time(void){return 0;} -- cgit v1.2.1 From bd0b1e954e3ba3e5d2cab941458cf98206471bd2 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:35 -0700 Subject: [PATCH] knfsd: nfsd4: idmap initialization Adopt standard kernel style by defining a no-op function instead of putting ifdef's in the code where the function is called. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd_idmap.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfsd_idmap.h b/include/linux/nfsd_idmap.h index 9bb7f30e923b..e82746fcad14 100644 --- a/include/linux/nfsd_idmap.h +++ b/include/linux/nfsd_idmap.h @@ -43,8 +43,13 @@ /* XXX from linux/nfs_idmap.h */ #define IDMAP_NAMESZ 128 +#ifdef CONFIG_NFSD_V4 void nfsd_idmap_init(void); void nfsd_idmap_shutdown(void); +#else +static inline void nfsd_idmap_init(void) {}; +static inline void nfsd_idmap_shutdown(void) {}; +#endif int nfsd_map_name_to_uid(struct svc_rqst *, const char *, size_t, __u32 *); int nfsd_map_name_to_gid(struct svc_rqst *, const char *, size_t, __u32 *); -- cgit v1.2.1 From a55370a3c0106106a975c5a09cee800611d0cf50 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:03:52 -0700 Subject: [PATCH] knfsd: nfsd4: reboot hash For the purposes of reboot recovery we keep a directory with subdirectories each having a name that is the ascii hex representation of the md5 sum of a client identifier for an active client. This adds the code to calculate that name. We also use it for the purposes of comparing clients, so if someone ever manages to find two client names that are md5 collisions, then we'll return clid_inuse to the second. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 296e6429fc3b..fdaa84addadb 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -109,6 +109,8 @@ struct nfs4_callback { struct rpc_clnt * cb_client; }; +#define HEXDIR_LEN 33 /* hex version of 16 byte md5 of cl_name plus '\0' */ + /* * struct nfs4_client - one per client. Clientids live here. * o Each nfs4_client is hashed by clientid. @@ -126,6 +128,7 @@ struct nfs4_client { struct list_head cl_del_perclnt; /* list: delegations */ struct list_head cl_lru; /* tail queue */ struct xdr_netobj cl_name; /* id generated by client */ + char cl_recdir[HEXDIR_LEN]; /* recovery dir */ nfs4_verifier cl_verifier; /* generated by client */ time_t cl_time; /* time of last lease renewal */ u32 cl_addr; /* client ipaddress */ @@ -143,7 +146,7 @@ struct nfs4_client { */ struct nfs4_client_reclaim { struct list_head cr_strhash; /* hash by cr_name */ - struct xdr_netobj cr_name; /* id generated by client */ + char cr_recdir[HEXDIR_LEN]; /* recover dir */ }; static inline void @@ -283,6 +286,7 @@ extern void nfs4_free_stateowner(struct kref *kref); extern void nfsd4_probe_callback(struct nfs4_client *clp); extern void nfsd4_cb_recall(struct nfs4_delegation *dp); extern void nfs4_put_delegation(struct nfs4_delegation *dp); +extern int nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname); static inline void nfs4_put_stateowner(struct nfs4_stateowner *so) -- cgit v1.2.1 From fd39ca9a808c6026989bc2188868a0574eb37108 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:03 -0700 Subject: [PATCH] knfsd: nfsd4: make needlessly global code static This patch contains the following possible cleanups: - make needlessly global code static Signed-off-by: Adrian Bunk Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 8 -------- 1 file changed, 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index fdaa84addadb..0e18ae22127d 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -61,11 +61,6 @@ typedef struct { #define si_stateownerid si_opaque.so_stateownerid #define si_fileid si_opaque.so_fileid -extern stateid_t zerostateid; -extern stateid_t onestateid; - -#define ZERO_STATEID(stateid) (!memcmp((stateid), &zerostateid, sizeof(stateid_t))) -#define ONE_STATEID(stateid) (!memcmp((stateid), &onestateid, sizeof(stateid_t))) struct nfs4_cb_recall { u32 cbr_ident; @@ -271,12 +266,9 @@ struct nfs4_stateid { ((err) != nfserr_stale_stateid) && \ ((err) != nfserr_bad_stateid)) -extern time_t nfs4_laundromat(void); extern int nfsd4_renew(clientid_t *clid); extern int nfs4_preprocess_stateid_op(struct svc_fh *current_fh, stateid_t *stateid, int flags, struct file **filp); -extern int nfs4_share_conflict(struct svc_fh *current_fh, - unsigned int deny_type); extern void nfs4_lock_state(void); extern void nfs4_unlock_state(void); extern int nfs4_in_grace(void); -- cgit v1.2.1 From ea1da636e956ad1591a74904f23d98bbc26a644b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:17 -0700 Subject: [PATCH] knfsd: nfsd4: rename state list fields Trivial renaming patch: I can never remember, while looking at various lists relating the nfsd4 state structures, which are the "heads" and which are items on other lists, or which structures are actually on the various lists. The following convention helps me: given structures foo and bar, with foo containing the head of a list of bars, use "bars" for the name of the head of the list contained in the struct foo, and use "per_foo" for the entries in the struct bars. Already done for struct nfs4_file; go ahead and do it for the other nfsd4 state structures. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 0e18ae22127d..f4f27b76ee64 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -72,8 +72,8 @@ struct nfs4_cb_recall { }; struct nfs4_delegation { - struct list_head dl_del_perfile; /* nfs4_file->fi_del_perfile */ - struct list_head dl_del_perclnt; /* nfs4_client->cl_del_perclnt*/ + struct list_head dl_perfile; + struct list_head dl_perclnt; struct list_head dl_recall_lru; /* delegation recalled */ atomic_t dl_count; /* ref count */ struct nfs4_client *dl_client; @@ -119,8 +119,8 @@ struct nfs4_callback { struct nfs4_client { struct list_head cl_idhash; /* hash by cl_clientid.id */ struct list_head cl_strhash; /* hash by cl_name */ - struct list_head cl_perclient; /* list: stateowners */ - struct list_head cl_del_perclnt; /* list: delegations */ + struct list_head cl_openowners; + struct list_head cl_delegations; struct list_head cl_lru; /* tail queue */ struct xdr_netobj cl_name; /* id generated by client */ char cl_recdir[HEXDIR_LEN]; /* recovery dir */ @@ -195,9 +195,9 @@ struct nfs4_stateowner { struct kref so_ref; struct list_head so_idhash; /* hash by so_id */ struct list_head so_strhash; /* hash by op_name */ - struct list_head so_perclient; /* nfs4_client->cl_perclient */ - struct list_head so_perfilestate; /* list: nfs4_stateid */ - struct list_head so_perlockowner; /* nfs4_stateid->st_perlockowner */ + struct list_head so_perclient; + struct list_head so_stateids; + struct list_head so_perstateid; /* for lockowners only */ struct list_head so_close_lru; /* tail queue */ time_t so_time; /* time of placement on so_close_lru */ int so_is_open_owner; /* 1=openowner,0=lockowner */ @@ -240,8 +240,8 @@ struct nfs4_file { struct nfs4_stateid { struct list_head st_hash; struct list_head st_perfile; - struct list_head st_perfilestate; - struct list_head st_perlockowner; + struct list_head st_perstateowner; + struct list_head st_lockowners; struct nfs4_stateowner * st_stateowner; struct nfs4_file * st_file; stateid_t st_stateid; -- cgit v1.2.1 From cb36d6345752fa24827044c68e15f6708a40d9f6 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:23 -0700 Subject: [PATCH] knfsd: nfsd4: remove cb_parsed The cb_parsed field is only used by probe_callback, to determine whether the callback information has been filled in by setclientid. But there is no way that probe_callback() can be called without that having already happened, so that check is superfluous, as is cb_parsed. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index f4f27b76ee64..83d29ec03a58 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -92,7 +92,6 @@ struct nfs4_delegation { /* client delegation callback info */ struct nfs4_callback { /* SETCLIENTID info */ - u32 cb_parsed; /* addr parsed */ u32 cb_addr; unsigned short cb_port; u32 cb_prog; -- cgit v1.2.1 From 190e4fbf96037e5e526ba3210f2bcc2a3b6fe964 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:25 -0700 Subject: [PATCH] knfsd: nfsd4: initialize recovery directory NFSv4 clients are required to know what state they have on the server so that they can reclaim it on server reboot. However, it is possible for pathalogical combinations of server reboots and network partitions to leave a client in a state where it cannot know whether it has lost its state on the server. For this reason, rfc3530 requires that we store some information about clients to stable storage. So we maintain a directory /var/lib/nfs/v4recovery with a subdirectory for each client with active state. We leave open the possibility of including files underneath each such subdirectory with information about the client, but for now the subdirectories are empty. We create a client subdirectory whenever a client makes its first non-reclaim open_confirm. We remove a client subdirectory whenever either a) its lease expires, or b) the grace period ends without it reclaiming anything. When handling reclaims, we allow the reclaim if and only if the client doing the reclaim has a subdirectory. This patch adds just the code to scan the recovery directory on nfsd startup. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 83d29ec03a58..19481ab122df 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -278,6 +278,10 @@ extern void nfsd4_probe_callback(struct nfs4_client *clp); extern void nfsd4_cb_recall(struct nfs4_delegation *dp); extern void nfs4_put_delegation(struct nfs4_delegation *dp); extern int nfs4_make_rec_clidname(char *clidname, struct xdr_netobj *clname); +extern void nfsd4_init_recdir(char *recdir_name); +extern int nfsd4_recdir_load(void); +extern void nfsd4_shutdown_recdir(void); +extern int nfs4_client_to_reclaim(const char *name); static inline void nfs4_put_stateowner(struct nfs4_stateowner *so) -- cgit v1.2.1 From c7b9a45927e74c81d6562153f7fde9d32da00159 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:30 -0700 Subject: [PATCH] knfsd: nfsd4: reboot recovery This patch adds the code to create and remove client subdirectories from the recovery directory, as described in the previous patch comment. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 19481ab122df..a84a3fa99be1 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -131,6 +131,7 @@ struct nfs4_client { nfs4_verifier cl_confirm; /* generated by server */ struct nfs4_callback cl_callback; /* callback info */ atomic_t cl_count; /* ref count */ + u32 cl_firststate; /* recovery dir creation */ }; /* struct nfs4_client_reset @@ -282,6 +283,10 @@ extern void nfsd4_init_recdir(char *recdir_name); extern int nfsd4_recdir_load(void); extern void nfsd4_shutdown_recdir(void); extern int nfs4_client_to_reclaim(const char *name); +extern int nfs4_has_reclaimed_state(const char *name); +extern void nfsd4_recdir_purge_old(void); +extern int nfsd4_create_clid_dir(struct nfs4_client *clp); +extern void nfsd4_remove_clid_dir(struct nfs4_client *clp); static inline void nfs4_put_stateowner(struct nfs4_stateowner *so) -- cgit v1.2.1 From 0964a3d3f1aa96468091924f6b0c391a46dc6d0b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 23 Jun 2005 22:04:32 -0700 Subject: [PATCH] knfsd: nfsd4 reboot dirname fix Set the recovery directory via /proc/fs/nfsd/nfs4recoverydir. It may be changed any time, but is used only on startup. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/nfsd.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 21c6e9d86e4f..5791dfd30dd0 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -150,12 +150,14 @@ int nfs4_state_start(void); void nfs4_state_shutdown(void); time_t nfs4_lease_time(void); void nfs4_reset_lease(time_t leasetime); +int nfs4_reset_recoverydir(char *recdir); #else static inline void nfs4_state_init(void){}; static inline int nfs4_state_start(void){return 0;} static inline void nfs4_state_shutdown(void){} static inline time_t nfs4_lease_time(void){return 0;} static inline void nfs4_reset_lease(time_t leasetime){} +static inline int nfs4_reset_recoverydir(char *recdir) {return 0;} #endif /* -- cgit v1.2.1 From c988d2b2845495373f666a381d354a7f80981d62 Mon Sep 17 00:00:00 2001 From: Matt Domsch Date: Thu, 23 Jun 2005 22:05:15 -0700 Subject: [PATCH] modules: add version and srcversion to sysfs This patch adds version and srcversion files to /sys/module/${modulename} containing the version and srcversion fields of the module's modinfo section (if present). /sys/module/e1000 |-- srcversion `-- version This patch differs slightly from the version posted in January, as it now uses the new kstrdup() call in -mm. Why put this in sysfs? a) Tools like DKMS, which deal with changing out individual kernel modules without replacing the whole kernel, can behave smarter if they can tell the version of a given module. The autoinstaller feature, for example, which determines if your system has a "good" version of a driver (i.e. if the one provided by DKMS has a newer verson than that provided by the kernel package installed), and to automatically compile and install a newer version if DKMS has it but your kernel doesn't yet have that version. b) Because sysadmins manually, or with tools like DKMS, can switch out modules on the file system, you can't count on 'modinfo foo.ko', which looks at /lib/modules/${kernelver}/... actually matching what is loaded into the kernel already. Hence asking sysfs for this. c) as the unbind-driver-from-device work takes shape, it will be possible to rebind a driver that's built-in (no .ko to modinfo for the version) to a newly loaded module. sysfs will have the currently-built-in version info, for comparison. d) tech support scripts can then easily grab the version info for what's running presently - a question I get often. There has been renewed interest in this patch on linux-scsi by driver authors. As the idea originated from GregKH, I leave his Signed-off-by: intact, though the implementation is nearly completely new. Compiled and run on x86 and x86_64. From: Matthew Dobson build fix From: Thierry Vignaud build fix From: Matthew Dobson warning fix Signed-off-by: Greg Kroah-Hartman Signed-off-by: Matt Domsch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/module.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/module.h b/include/linux/module.h index 0e432a0f4aee..f05372b7fe77 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -51,6 +51,9 @@ struct module_attribute { ssize_t (*show)(struct module_attribute *, struct module *, char *); ssize_t (*store)(struct module_attribute *, struct module *, const char *, size_t count); + void (*setup)(struct module *, const char *); + int (*test)(struct module *); + void (*free)(struct module *); }; struct module_kobject @@ -239,6 +242,8 @@ struct module /* Sysfs stuff. */ struct module_kobject mkobj; struct module_param_attrs *param_attrs; + const char *version; + const char *srcversion; /* Exported symbols */ const struct kernel_symbol *syms; -- cgit v1.2.1 From 420edbcc09008342c7b2665453f6b370739aadb0 Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Thu, 23 Jun 2005 22:05:23 -0700 Subject: [PATCH] xip: bdev: execute in place This is the block device related part. The block device operation direct_access now has a struct block_device as first parameter. Signed-off-by: Carsten Otte Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 83857d8070d3..929bf8d20c87 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -885,6 +885,7 @@ struct block_device_operations { int (*ioctl) (struct inode *, struct file *, unsigned, unsigned long); long (*unlocked_ioctl) (struct file *, unsigned, unsigned long); long (*compat_ioctl) (struct file *, unsigned, unsigned long); + int (*direct_access) (struct block_device *, sector_t, unsigned long *); int (*media_changed) (struct gendisk *); int (*revalidate_disk) (struct gendisk *); struct module *owner; -- cgit v1.2.1 From ceffc078528befc008c6f2c2c4decda79eabd534 Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Thu, 23 Jun 2005 22:05:25 -0700 Subject: [PATCH] xip: fs/mm: execute in place - generic_file* file operations do no longer have a xip/non-xip split - filemap_xip.c implements a new set of fops that require get_xip_page aop to work proper. all new fops are exported GPL-only (don't like to see whatever code use those except GPL modules) - __xip_unmap now uses page_check_address, which is no longer static in rmap.c, and defined in linux/rmap.h - mm/filemap.h is now much more clean, plainly having just Linus' inline funcs moved here from filemap.c - fix includes in filemap_xip to make it build cleanly on i386 Signed-off-by: Carsten Otte Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 18 ++++++++++++++++++ include/linux/rmap.h | 6 ++++++ 2 files changed, 24 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 929bf8d20c87..79c0fafc0211 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -330,6 +330,8 @@ struct address_space_operations { int (*releasepage) (struct page *, int); ssize_t (*direct_IO)(int, struct kiocb *, const struct iovec *iov, loff_t offset, unsigned long nr_segs); + struct page* (*get_xip_page)(struct address_space *, sector_t, + int); }; struct backing_dev_info; @@ -1497,6 +1499,22 @@ extern loff_t remote_llseek(struct file *file, loff_t offset, int origin); extern int generic_file_open(struct inode * inode, struct file * filp); extern int nonseekable_open(struct inode * inode, struct file * filp); +#ifdef CONFIG_FS_XIP +extern ssize_t xip_file_aio_read(struct kiocb *iocb, char __user *buf, + size_t count, loff_t pos); +extern ssize_t xip_file_readv(struct file *filp, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos); +extern ssize_t xip_file_sendfile(struct file *in_file, loff_t *ppos, + size_t count, read_actor_t actor, + void *target); +extern int xip_file_mmap(struct file * file, struct vm_area_struct * vma); +extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf, + size_t count, loff_t pos); +extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos); +extern int xip_truncate_page(struct address_space *mapping, loff_t from); +#endif + static inline void do_generic_file_read(struct file * filp, loff_t *ppos, read_descriptor_t * desc, read_actor_t actor) diff --git a/include/linux/rmap.h b/include/linux/rmap.h index 11b484e37ac9..e80fb7ee6efd 100644 --- a/include/linux/rmap.h +++ b/include/linux/rmap.h @@ -92,6 +92,12 @@ static inline void page_dup_rmap(struct page *page) int page_referenced(struct page *, int is_locked, int ignore_token); int try_to_unmap(struct page *); +/* + * Called from mm/filemap_xip.c to unmap empty zero page + */ +pte_t *page_check_address(struct page *, struct mm_struct *, unsigned long); + + /* * Used by swapoff to help locate where page is expected in vma. */ -- cgit v1.2.1 From 6d79125bba55ee82701f1c7d4ebbc1aa20ecbe4e Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Thu, 23 Jun 2005 22:05:26 -0700 Subject: [PATCH] xip: ext2: execute in place These are the ext2 related parts. Ext2 now uses the xip_* file operations along with the get_xip_page aop when mounted with -o xip. Signed-off-by: Carsten Otte Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ext2_fs.h | 25 +++++++++++++------------ include/linux/fs.h | 5 +++++ 2 files changed, 18 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h index fab43527e597..a657130ba03a 100644 --- a/include/linux/ext2_fs.h +++ b/include/linux/ext2_fs.h @@ -300,18 +300,19 @@ struct ext2_inode { /* * Mount flags */ -#define EXT2_MOUNT_CHECK 0x0001 /* Do mount-time checks */ -#define EXT2_MOUNT_OLDALLOC 0x0002 /* Don't use the new Orlov allocator */ -#define EXT2_MOUNT_GRPID 0x0004 /* Create files with directory's group */ -#define EXT2_MOUNT_DEBUG 0x0008 /* Some debugging messages */ -#define EXT2_MOUNT_ERRORS_CONT 0x0010 /* Continue on errors */ -#define EXT2_MOUNT_ERRORS_RO 0x0020 /* Remount fs ro on errors */ -#define EXT2_MOUNT_ERRORS_PANIC 0x0040 /* Panic on errors */ -#define EXT2_MOUNT_MINIX_DF 0x0080 /* Mimics the Minix statfs */ -#define EXT2_MOUNT_NOBH 0x0100 /* No buffer_heads */ -#define EXT2_MOUNT_NO_UID32 0x0200 /* Disable 32-bit UIDs */ -#define EXT2_MOUNT_XATTR_USER 0x4000 /* Extended user attributes */ -#define EXT2_MOUNT_POSIX_ACL 0x8000 /* POSIX Access Control Lists */ +#define EXT2_MOUNT_CHECK 0x000001 /* Do mount-time checks */ +#define EXT2_MOUNT_OLDALLOC 0x000002 /* Don't use the new Orlov allocator */ +#define EXT2_MOUNT_GRPID 0x000004 /* Create files with directory's group */ +#define EXT2_MOUNT_DEBUG 0x000008 /* Some debugging messages */ +#define EXT2_MOUNT_ERRORS_CONT 0x000010 /* Continue on errors */ +#define EXT2_MOUNT_ERRORS_RO 0x000020 /* Remount fs ro on errors */ +#define EXT2_MOUNT_ERRORS_PANIC 0x000040 /* Panic on errors */ +#define EXT2_MOUNT_MINIX_DF 0x000080 /* Mimics the Minix statfs */ +#define EXT2_MOUNT_NOBH 0x000100 /* No buffer_heads */ +#define EXT2_MOUNT_NO_UID32 0x000200 /* Disable 32-bit UIDs */ +#define EXT2_MOUNT_XATTR_USER 0x004000 /* Extended user attributes */ +#define EXT2_MOUNT_POSIX_ACL 0x008000 /* POSIX Access Control Lists */ +#define EXT2_MOUNT_XIP 0x010000 /* Execute in place */ #define clear_opt(o, opt) o &= ~EXT2_MOUNT_##opt #define set_opt(o, opt) o |= EXT2_MOUNT_##opt diff --git a/include/linux/fs.h b/include/linux/fs.h index 79c0fafc0211..7e0501895f35 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1513,6 +1513,11 @@ extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf, extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov, unsigned long nr_segs, loff_t *ppos); extern int xip_truncate_page(struct address_space *mapping, loff_t from); +#else +static inline int xip_truncate_page(struct address_space *mapping, loff_t from) +{ + return 0; +} #endif static inline void do_generic_file_read(struct file * filp, loff_t *ppos, -- cgit v1.2.1 From eb6fe0c388e43b02e261f0fdee60e42f6298d7f7 Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Thu, 23 Jun 2005 22:05:28 -0700 Subject: [PATCH] xip: reduce code duplication This patch reworks filemap_xip.c with the goal to reduce code duplication from mm/filemap.c. It applies agains 2.6.12-rc6-mm1. Instead of implementing the aio functions, this one implements the synchronous read/write functions only. For readv and writev, the generic fallback is used. For aio, we rely on the application doing the fallback. Since our "synchronous" function does memcpy immediately anyway, there is no performance difference between using the fallbacks or implementing each operation. Signed-off-by: Carsten Otte Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 7e0501895f35..3ae8e37bdfc8 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1500,18 +1500,14 @@ extern int generic_file_open(struct inode * inode, struct file * filp); extern int nonseekable_open(struct inode * inode, struct file * filp); #ifdef CONFIG_FS_XIP -extern ssize_t xip_file_aio_read(struct kiocb *iocb, char __user *buf, - size_t count, loff_t pos); -extern ssize_t xip_file_readv(struct file *filp, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos); +extern ssize_t xip_file_read(struct file *filp, char __user *buf, size_t len, + loff_t *ppos); extern ssize_t xip_file_sendfile(struct file *in_file, loff_t *ppos, size_t count, read_actor_t actor, void *target); extern int xip_file_mmap(struct file * file, struct vm_area_struct * vma); -extern ssize_t xip_file_aio_write(struct kiocb *iocb, const char __user *buf, - size_t count, loff_t pos); -extern ssize_t xip_file_writev(struct file *file, const struct iovec *iov, - unsigned long nr_segs, loff_t *ppos); +extern ssize_t xip_file_write(struct file *filp, const char __user *buf, + size_t len, loff_t *ppos); extern int xip_truncate_page(struct address_space *mapping, loff_t from); #else static inline int xip_truncate_page(struct address_space *mapping, loff_t from) -- cgit v1.2.1 From 52c1da39534fb382c061de58b65f678ad74b59f5 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 23 Jun 2005 22:05:33 -0700 Subject: [PATCH] make various thing static Another rollup of patches which give various symbols static scope Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/irq.h | 1 - include/linux/namespace.h | 1 - 2 files changed, 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irq.h b/include/linux/irq.h index 7fc1022be9ee..12277799c007 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -88,7 +88,6 @@ extern fastcall int handle_IRQ_event(unsigned int irq, struct pt_regs *regs, struct irqaction *action); extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs); extern void note_interrupt(unsigned int irq, irq_desc_t *desc, int action_ret); -extern void report_bad_irq(unsigned int irq, irq_desc_t *desc, int action_ret); extern int can_request_irq(unsigned int irq, unsigned long irqflags); extern void init_irq_proc(void); diff --git a/include/linux/namespace.h b/include/linux/namespace.h index 9eca1558d72f..697991b69f9b 100644 --- a/include/linux/namespace.h +++ b/include/linux/namespace.h @@ -12,7 +12,6 @@ struct namespace { struct rw_semaphore sem; }; -extern void umount_tree(struct vfsmount *); extern int copy_namespace(int, struct task_struct *); extern void __put_namespace(struct namespace *namespace); -- cgit v1.2.1 From 75043cb5b386e5a01fd03b88f647dd992de02f97 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 24 Jun 2005 20:52:52 +0000 Subject: [PATCH] fs/qnx4/*: fix sparse warnings This patch fixes sparse warnings in the qnx4fs (and might even make qnx4fs work on big-endian boxes) Signed-off-by: Alexey Dobriyan Signed-off-by: Domen Puncer Signed-off-by: Anders Larsen Signed-off-by: Linus Torvalds --- include/linux/qnx4_fs.h | 18 +++++++++--------- include/linux/qnxtypes.h | 16 ++++++++-------- 2 files changed, 17 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/include/linux/qnx4_fs.h b/include/linux/qnx4_fs.h index 22ba580b0ae8..fc610bb0f733 100644 --- a/include/linux/qnx4_fs.h +++ b/include/linux/qnx4_fs.h @@ -46,11 +46,11 @@ struct qnx4_inode_entry { char di_fname[QNX4_SHORT_NAME_MAX]; qnx4_off_t di_size; qnx4_xtnt_t di_first_xtnt; - __u32 di_xblk; - __s32 di_ftime; - __s32 di_mtime; - __s32 di_atime; - __s32 di_ctime; + __le32 di_xblk; + __le32 di_ftime; + __le32 di_mtime; + __le32 di_atime; + __le32 di_ctime; qnx4_nxtnt_t di_num_xtnts; qnx4_mode_t di_mode; qnx4_muid_t di_uid; @@ -63,18 +63,18 @@ struct qnx4_inode_entry { struct qnx4_link_info { char dl_fname[QNX4_NAME_MAX]; - __u32 dl_inode_blk; + __le32 dl_inode_blk; __u8 dl_inode_ndx; __u8 dl_spare[10]; __u8 dl_status; }; struct qnx4_xblk { - __u32 xblk_next_xblk; - __u32 xblk_prev_xblk; + __le32 xblk_next_xblk; + __le32 xblk_prev_xblk; __u8 xblk_num_xtnts; __u8 xblk_spare[3]; - __s32 xblk_num_blocks; + __le32 xblk_num_blocks; qnx4_xtnt_t xblk_xtnts[QNX4_MAX_XTNTS_PER_XBLK]; char xblk_signature[8]; qnx4_xtnt_t xblk_first_xtnt; diff --git a/include/linux/qnxtypes.h b/include/linux/qnxtypes.h index fb518e318c7c..a3eb1137857b 100644 --- a/include/linux/qnxtypes.h +++ b/include/linux/qnxtypes.h @@ -12,18 +12,18 @@ #ifndef _QNX4TYPES_H #define _QNX4TYPES_H -typedef __u16 qnx4_nxtnt_t; +typedef __le16 qnx4_nxtnt_t; typedef __u8 qnx4_ftype_t; typedef struct { - __u32 xtnt_blk; - __u32 xtnt_size; + __le32 xtnt_blk; + __le32 xtnt_size; } qnx4_xtnt_t; -typedef __u16 qnx4_mode_t; -typedef __u16 qnx4_muid_t; -typedef __u16 qnx4_mgid_t; -typedef __u32 qnx4_off_t; -typedef __u16 qnx4_nlink_t; +typedef __le16 qnx4_mode_t; +typedef __le16 qnx4_muid_t; +typedef __le16 qnx4_mgid_t; +typedef __le32 qnx4_off_t; +typedef __le16 qnx4_nlink_t; #endif -- cgit v1.2.1 From e70c9d5e61c6cb2272c866fc1303e62975006752 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Sat, 25 Jun 2005 14:54:25 -0700 Subject: [PATCH] I8K: use standard DMI interface I8K: Change to use stock dmi infrastructure instead of homegrown parsing code. The driver now requires box's DMI data to match list of supported models so driver can be safely compiled-in by default without fear of it poking into random SMM BIOS code. DMI checks can be ignored with i8k.ignore_dmi option. Signed-off-by: Dmitry Torokhov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/dmi.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/dmi.h b/include/linux/dmi.h index d2bcf556088b..5e93e6dce9a4 100644 --- a/include/linux/dmi.h +++ b/include/linux/dmi.h @@ -9,6 +9,7 @@ enum dmi_field { DMI_SYS_VENDOR, DMI_PRODUCT_NAME, DMI_PRODUCT_VERSION, + DMI_PRODUCT_SERIAL, DMI_BOARD_VENDOR, DMI_BOARD_NAME, DMI_BOARD_VERSION, -- cgit v1.2.1 From 52a119feaad92d44a0e97d01b22afbcbaf3fc079 Mon Sep 17 00:00:00 2001 From: Ashok Raj Date: Sat, 25 Jun 2005 14:54:57 -0700 Subject: [PATCH] make smp_prepare_cpu to a weak function I really wish smp_prepare_cpu() would disappear eventually. In the interim this is ideally a weak function, so we dont end up changing several places to define this dummy in headers. Today since the dummy declaration is done only in drivers/base/cpu.c but the function is called in kernel/power/smp.c i get undefined reference in my cpu hotplug code for x86_64 under development. Signed-off-by: Ashok Raj Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cpu.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/cpu.h b/include/linux/cpu.h index fe0298e5dae1..e8904c0da686 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -69,6 +69,7 @@ extern struct semaphore cpucontrol; register_cpu_notifier(&fn##_nb); \ } int cpu_down(unsigned int cpu); +extern int __attribute__((weak)) smp_prepare_cpu(int cpu); #define cpu_is_offline(cpu) unlikely(!cpu_online(cpu)) #else #define lock_cpu_hotplug() do { } while (0) -- cgit v1.2.1 From e6982c671c560da4a0bc5c908cbcbec12bd5991d Mon Sep 17 00:00:00 2001 From: Ashok Raj Date: Sat, 25 Jun 2005 14:54:58 -0700 Subject: [PATCH] x86_64: Change init sections for CPU hotplug support This patch adds __cpuinit and __cpuinitdata sections that need to exist past boot to support cpu hotplug. Caveat: This is done *only* for EM64T CPU Hotplug support, on request from Andi Kleen. Much of the generic hotplug code in kernel, and none of the other archs that support CPU hotplug today, i386, ia64, ppc64, s390 and parisc dont mark sections with __cpuinit, but only mark them as __devinit, and __devinitdata. If someone is motivated to change generic code, we need to make sure all existing hotplug code does not break, on other arch's that dont use __cpuinit, and __cpudevinit. Signed-off-by: Ashok Raj Acked-by: Andi Kleen Acked-by: Zwane Mwaikambo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/init.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/linux') diff --git a/include/linux/init.h b/include/linux/init.h index 05c83e0521ca..59008c3826cf 100644 --- a/include/linux/init.h +++ b/include/linux/init.h @@ -229,6 +229,18 @@ void __init parse_early_param(void); #define __devexitdata __exitdata #endif +#ifdef CONFIG_HOTPLUG_CPU +#define __cpuinit +#define __cpuinitdata +#define __cpuexit +#define __cpuexitdata +#else +#define __cpuinit __init +#define __cpuinitdata __initdata +#define __cpuexit __exit +#define __cpuexitdata __exitdata +#endif + /* Functions marked as __devexit may be discarded at kernel link time, depending on config options. Newer versions of binutils detect references from retained sections to discarded sections and flag an error. Pointers to -- cgit v1.2.1 From 5a72e04df5470df0ec646029d31e5528167ab1a7 Mon Sep 17 00:00:00 2001 From: Li Shaohua Date: Sat, 25 Jun 2005 14:55:06 -0700 Subject: [PATCH] suspend/resume SMP support Using CPU hotplug to support suspend/resume SMP. Both S3 and S4 use disable/enable_nonboot_cpus API. The S4 part is based on Pavel's original S4 SMP patch. Signed-off-by: Li Shaohua Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/suspend.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 2bf0d5fabcdb..f2e96fdfaae0 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -58,7 +58,7 @@ static inline int software_suspend(void) } #endif -#ifdef CONFIG_SMP +#ifdef CONFIG_SUSPEND_SMP extern void disable_nonboot_cpus(void); extern void enable_nonboot_cpus(void); #else -- cgit v1.2.1 From 620b03276488c3cf103caf1e326bd21f00d3df84 Mon Sep 17 00:00:00 2001 From: Pavel Machek Date: Sat, 25 Jun 2005 14:55:11 -0700 Subject: [PATCH] properly stop devices before poweroff Without this patch, Linux provokes emergency disk shutdowns and similar nastiness. It was in SuSE kernels for some time, IIRC. Signed-off-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pm.h | 33 +++++++++++++++++++++------------ 1 file changed, 21 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pm.h b/include/linux/pm.h index ed2b76e75199..14479325e3f3 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -103,7 +103,8 @@ extern int pm_active; /* * Register a device with power management */ -struct pm_dev __deprecated *pm_register(pm_dev_t type, unsigned long id, pm_callback callback); +struct pm_dev __deprecated * +pm_register(pm_dev_t type, unsigned long id, pm_callback callback); /* * Unregister a device with power management @@ -190,17 +191,18 @@ typedef u32 __bitwise pm_message_t; /* * There are 4 important states driver can be in: * ON -- driver is working - * FREEZE -- stop operations and apply whatever policy is applicable to a suspended driver - * of that class, freeze queues for block like IDE does, drop packets for - * ethernet, etc... stop DMA engine too etc... so a consistent image can be - * saved; but do not power any hardware down. - * SUSPEND - like FREEZE, but hardware is doing as much powersaving as possible. Roughly - * pci D3. + * FREEZE -- stop operations and apply whatever policy is applicable to a + * suspended driver of that class, freeze queues for block like IDE + * does, drop packets for ethernet, etc... stop DMA engine too etc... + * so a consistent image can be saved; but do not power any hardware + * down. + * SUSPEND - like FREEZE, but hardware is doing as much powersaving as + * possible. Roughly pci D3. * - * Unfortunately, current drivers only recognize numeric values 0 (ON) and 3 (SUSPEND). - * We'll need to fix the drivers. So yes, putting 3 to all diferent defines is intentional, - * and will go away as soon as drivers are fixed. Also note that typedef is neccessary, - * we'll probably want to switch to + * Unfortunately, current drivers only recognize numeric values 0 (ON) and 3 + * (SUSPEND). We'll need to fix the drivers. So yes, putting 3 to all different + * defines is intentional, and will go away as soon as drivers are fixed. Also + * note that typedef is neccessary, we'll probably want to switch to * typedef struct pm_message_t { int event; int flags; } pm_message_t * or something similar soon. */ @@ -222,11 +224,18 @@ struct dev_pm_info { extern void device_pm_set_parent(struct device * dev, struct device * parent); -extern int device_suspend(pm_message_t state); extern int device_power_down(pm_message_t state); extern void device_power_up(void); extern void device_resume(void); +#ifdef CONFIG_PM +extern int device_suspend(pm_message_t state); +#else +static inline int device_suspend(pm_message_t state) +{ + return 0; +} +#endif #endif /* __KERNEL__ */ -- cgit v1.2.1 From b2b18660066997420b716c1881a6be8b82700d97 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 25 Jun 2005 14:55:38 -0700 Subject: [PATCH] RCU: clean up a few remaining synchronize_kernel() calls 2.6.12-rc6-mm1 has a few remaining synchronize_kernel()s, some (but not all) in comments. This patch changes these synchronize_kernel() calls (and comments) to synchronize_rcu() or synchronize_sched() as follows: - arch/x86_64/kernel/mce.c mce_read(): change to synchronize_sched() to handle races with machine-check exceptions (synchronize_rcu() would not cut it given RCU implementations intended for hardcore realtime use. - drivers/input/serio/i8042.c i8042_stop(): change to synchronize_sched() to handle races with i8042_interrupt() interrupt handler. Again, synchronize_rcu() would not cut it given RCU implementations intended for hardcore realtime use. - include/*/kdebug.h comments: change to synchronize_sched() to handle races with NMIs. As before, synchronize_rcu() would not cut it... - include/linux/list.h comment: change to synchronize_rcu(), since this comment is for list_del_rcu(). - security/keys/key.c unregister_key_type(): change to synchronize_rcu(), since this is interacting with RCU read side. - security/keys/process_keys.c install_session_keyring(): change to synchronize_rcu(), since this is interacting with RCU read side. Signed-off-by: "Paul E. McKenney" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/list.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/list.h b/include/linux/list.h index 399b51d17218..aab2db21b013 100644 --- a/include/linux/list.h +++ b/include/linux/list.h @@ -185,7 +185,7 @@ static inline void list_del(struct list_head *entry) * list_for_each_entry_rcu(). * * Note that the caller is not permitted to immediately free - * the newly deleted entry. Instead, either synchronize_kernel() + * the newly deleted entry. Instead, either synchronize_rcu() * or call_rcu() must be used to defer freeing until an RCU * grace period has elapsed. */ -- cgit v1.2.1 From 7897986bad8f6cd50d6149345aca7f6480f49464 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:13 -0700 Subject: [PATCH] sched: balance timers Do CPU load averaging over a number of different intervals. Allow each interval to be chosen by sending a parameter to source_load and target_load. 0 is instantaneous, idx > 0 returns a decaying average with the most recent sample weighted at 2^(idx-1). To a maximum of 3 (could be easily increased). So generally a higher number will result in more conservative balancing. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 4 ++++ include/linux/topology.h | 8 ++++++++ 2 files changed, 12 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 2c69682b0444..664981ac1fb6 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -488,6 +488,10 @@ struct sched_domain { unsigned long long cache_hot_time; /* Task considered cache hot (ns) */ unsigned int cache_nice_tries; /* Leave cache hot tasks for # tries */ unsigned int per_cpu_gain; /* CPU % gained by adding domain cpus */ + unsigned int busy_idx; + unsigned int idle_idx; + unsigned int newidle_idx; + unsigned int wake_idx; int flags; /* See SD_* */ /* Runtime fields. */ diff --git a/include/linux/topology.h b/include/linux/topology.h index d70e8972c67f..ae9c2216dfa6 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -89,6 +89,10 @@ .cache_hot_time = 0, \ .cache_nice_tries = 0, \ .per_cpu_gain = 25, \ + .busy_idx = 0, \ + .idle_idx = 0, \ + .newidle_idx = 0, \ + .wake_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ @@ -115,6 +119,10 @@ .cache_hot_time = (5*1000000/2), \ .cache_nice_tries = 1, \ .per_cpu_gain = 100, \ + .busy_idx = 2, \ + .idle_idx = 0, \ + .newidle_idx = 1, \ + .wake_idx = 1, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ -- cgit v1.2.1 From cafb20c1f9976a70d633bb1e1c8c24eab00e4e80 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:17 -0700 Subject: [PATCH] sched: no aggressive idle balancing Remove the very aggressive idle stuff that has recently gone into 2.6 - it is going against the direction we are trying to go. Hopefully we can regain performance through other methods. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/topology.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/topology.h b/include/linux/topology.h index ae9c2216dfa6..b23ec64df7f1 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -127,7 +127,6 @@ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ | SD_WAKE_AFFINE \ - | SD_WAKE_IDLE \ | SD_WAKE_BALANCE, \ .last_balance = jiffies, \ .balance_interval = 1, \ -- cgit v1.2.1 From 147cbb4bbe991452698f0772d8292f22825710ba Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:19 -0700 Subject: [PATCH] sched: balance on fork Reimplement the balance on exec balancing to be sched-domains aware. Use this to also do balance on fork balancing. Make x86_64 do balance on fork over the NUMA domain. The problem that the non sched domains aware blancing became apparent on dual core, multi socket opterons. What we want is for the new tasks to be sent to a different socket, but more often than not, we would first load up our sibling core, or fill two cores of a single remote socket before selecting a new one. This gives large improvements to STREAM on such systems. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 10 ++++++---- include/linux/topology.h | 2 ++ 2 files changed, 8 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 664981ac1fb6..613491d3a875 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -460,10 +460,11 @@ enum idle_type #define SD_LOAD_BALANCE 1 /* Do load balancing on this domain. */ #define SD_BALANCE_NEWIDLE 2 /* Balance when about to become idle */ #define SD_BALANCE_EXEC 4 /* Balance on exec */ -#define SD_WAKE_IDLE 8 /* Wake to idle CPU on task wakeup */ -#define SD_WAKE_AFFINE 16 /* Wake task to waking CPU */ -#define SD_WAKE_BALANCE 32 /* Perform balancing at task wakeup */ -#define SD_SHARE_CPUPOWER 64 /* Domain members share cpu power */ +#define SD_BALANCE_FORK 8 /* Balance on fork, clone */ +#define SD_WAKE_IDLE 16 /* Wake to idle CPU on task wakeup */ +#define SD_WAKE_AFFINE 32 /* Wake task to waking CPU */ +#define SD_WAKE_BALANCE 64 /* Perform balancing at task wakeup */ +#define SD_SHARE_CPUPOWER 128 /* Domain members share cpu power */ struct sched_group { struct sched_group *next; /* Must be a circular list */ @@ -492,6 +493,7 @@ struct sched_domain { unsigned int idle_idx; unsigned int newidle_idx; unsigned int wake_idx; + unsigned int forkexec_idx; int flags; /* See SD_* */ /* Runtime fields. */ diff --git a/include/linux/topology.h b/include/linux/topology.h index b23ec64df7f1..665597207def 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -93,6 +93,7 @@ .idle_idx = 0, \ .newidle_idx = 0, \ .wake_idx = 0, \ + .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ @@ -123,6 +124,7 @@ .idle_idx = 0, \ .newidle_idx = 1, \ .wake_idx = 1, \ + .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ -- cgit v1.2.1 From 68767a0ae428801649d510d9a65bb71feed44dd1 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:20 -0700 Subject: [PATCH] sched: schedstats update for balance on fork Add SCHEDSTAT statistics for sched-balance-fork. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 613491d3a875..36a10781c3f3 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -517,10 +517,16 @@ struct sched_domain { unsigned long alb_failed; unsigned long alb_pushed; - /* sched_balance_exec() stats */ - unsigned long sbe_attempts; + /* SD_BALANCE_EXEC stats */ + unsigned long sbe_cnt; + unsigned long sbe_balanced; unsigned long sbe_pushed; + /* SD_BALANCE_FORK stats */ + unsigned long sbf_cnt; + unsigned long sbf_balanced; + unsigned long sbf_pushed; + /* try_to_wake_up() stats */ unsigned long ttwu_wake_remote; unsigned long ttwu_move_affine; -- cgit v1.2.1 From 687f1661d302bc70ce906594a6d3f615ef075a50 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:21 -0700 Subject: [PATCH] sched: sched tuning Do some basic initial tuning. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/topology.h | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/topology.h b/include/linux/topology.h index 665597207def..0320225e96da 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -91,7 +91,7 @@ .per_cpu_gain = 25, \ .busy_idx = 0, \ .idle_idx = 0, \ - .newidle_idx = 0, \ + .newidle_idx = 1, \ .wake_idx = 0, \ .forkexec_idx = 0, \ .flags = SD_LOAD_BALANCE \ @@ -121,15 +121,14 @@ .cache_nice_tries = 1, \ .per_cpu_gain = 100, \ .busy_idx = 2, \ - .idle_idx = 0, \ - .newidle_idx = 1, \ + .idle_idx = 1, \ + .newidle_idx = 2, \ .wake_idx = 1, \ - .forkexec_idx = 0, \ + .forkexec_idx = 1, \ .flags = SD_LOAD_BALANCE \ | SD_BALANCE_NEWIDLE \ | SD_BALANCE_EXEC \ - | SD_WAKE_AFFINE \ - | SD_WAKE_BALANCE, \ + | SD_WAKE_AFFINE, \ .last_balance = jiffies, \ .balance_interval = 1, \ .nr_balance_failed = 0, \ -- cgit v1.2.1 From 4866cde064afbb6c2a488c265e696879de616daa Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:23 -0700 Subject: [PATCH] sched: cleanup context switch locking Instead of requiring architecture code to interact with the scheduler's locking implementation, provide a couple of defines that can be used by the architecture to request runqueue unlocked context switches, and ask for interrupts to be enabled over the context switch. Also replaces the "switch_lock" used by these architectures with an oncpu flag (note, not a potentially slow bitflag). This eliminates one bus locked memory operation when context switching, and simplifies the task_running function. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/init_task.h | 1 - include/linux/sched.h | 10 ++++++++-- 2 files changed, 8 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index a6a8c1a38d5e..03206a425d7a 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -108,7 +108,6 @@ extern struct group_info init_groups; .blocked = {{0}}, \ .alloc_lock = SPIN_LOCK_UNLOCKED, \ .proc_lock = SPIN_LOCK_UNLOCKED, \ - .switch_lock = SPIN_LOCK_UNLOCKED, \ .journal_info = NULL, \ .cpu_timers = INIT_CPU_TIMERS(tsk.cpu_timers), \ } diff --git a/include/linux/sched.h b/include/linux/sched.h index 36a10781c3f3..d27be9337425 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -368,6 +368,11 @@ struct signal_struct { #endif }; +/* Context switch must be unlocked if interrupts are to be enabled */ +#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW +# define __ARCH_WANT_UNLOCKED_CTXSW +#endif + /* * Bits in flags field of signal_struct. */ @@ -594,6 +599,9 @@ struct task_struct { int lock_depth; /* BKL lock depth */ +#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) + int oncpu; +#endif int prio, static_prio; struct list_head run_list; prio_array_t *array; @@ -716,8 +724,6 @@ struct task_struct { spinlock_t alloc_lock; /* Protection of proc_dentry: nesting proc_lock, dcache_lock, write_lock_irq(&tasklist_lock); */ spinlock_t proc_lock; -/* context-switch lock */ - spinlock_t switch_lock; /* journalling filesystem info */ void *journal_info; -- cgit v1.2.1 From 476d139c218e44e045e4bc6d4cc02b010b343939 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Sat, 25 Jun 2005 14:57:29 -0700 Subject: [PATCH] sched: consolidate sbe sbf Consolidate balance-on-exec with balance-on-fork. This is made easy by the sched-domains RCU patches. As well as the general goodness of code reduction, this allows the runqueues to be unlocked during balance-on-fork. schedstats is a problem. Maybe just have balance-on-event instead of distinguishing fork and exec? Signed-off-by: Nick Piggin Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index d27be9337425..edb2c69a8873 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -930,7 +930,7 @@ extern void FASTCALL(wake_up_new_task(struct task_struct * tsk, #else static inline void kick_process(struct task_struct *tsk) { } #endif -extern void FASTCALL(sched_fork(task_t * p)); +extern void FASTCALL(sched_fork(task_t * p, int clone_flags)); extern void FASTCALL(sched_exit(task_t * p)); extern int in_group_p(gid_t); -- cgit v1.2.1 From 1a20ff27ef75d866730ee796acd811a925af762f Mon Sep 17 00:00:00 2001 From: Dinakar Guniguntala Date: Sat, 25 Jun 2005 14:57:33 -0700 Subject: [PATCH] Dynamic sched domains: sched changes The following patches add dynamic sched domains functionality that was extensively discussed on lkml and lse-tech. I would like to see this added to -mm o The main advantage with this feature is that it ensures that the scheduler load balacing code only balances against the cpus that are in the sched domain as defined by an exclusive cpuset and not all of the cpus in the system. This removes any overhead due to load balancing code trying to pull tasks outside of the cpu exclusive cpuset only to be prevented by the tasks' cpus_allowed mask. o cpu exclusive cpusets are useful for servers running orthogonal workloads such as RT applications requiring low latency and HPC applications that are throughput sensitive o It provides a new API partition_sched_domains in sched.c that makes dynamic sched domains possible. o cpu_exclusive cpusets sets are now associated with a sched domain. Which means that the users can dynamically modify the sched domains through the cpuset file system interface o ia64 sched domain code has been updated to support this feature as well o Currently, this does not support hotplug. (However some of my tests indicate hotplug+preempt is currently broken) o I have tested it extensively on x86. o This should have very minimal impact on performance as none of the fast paths are affected Signed-off-by: Dinakar Guniguntala Acked-by: Paul Jackson Acked-by: Nick Piggin Acked-by: Matthew Dobson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index edb2c69a8873..98c109e4f43d 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -539,6 +539,8 @@ struct sched_domain { #endif }; +extern void partition_sched_domains(cpumask_t *partition1, + cpumask_t *partition2); #ifdef ARCH_HAS_SCHED_DOMAIN /* Useful helpers that arch setup code may use. Defined in kernel/sched.c */ extern cpumask_t cpu_isolated_map; -- cgit v1.2.1 From f8cbd99bd3a023db8d6356d19a5f6f539d367327 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 25 Jun 2005 14:57:39 -0700 Subject: [PATCH] sched: voluntary kernel preemption This patch adds a new preemption model: 'Voluntary Kernel Preemption'. The 3 models can be selected from a new menu: (X) No Forced Preemption (Server) ( ) Voluntary Kernel Preemption (Desktop) ( ) Preemptible Kernel (Low-Latency Desktop) we still default to the stock (Server) preemption model. Voluntary preemption works by adding a cond_resched() (reschedule-if-needed) call to every might_sleep() check. It is lighter than CONFIG_PREEMPT - at the cost of not having as tight latencies. It represents a different latency/complexity/overhead tradeoff. It has no runtime impact at all if disabled. Here are size stats that show how the various preemption models impact the kernel's size: text data bss dec hex filename 3618774 547184 179896 4345854 424ffe vmlinux.stock 3626406 547184 179896 4353486 426dce vmlinux.voluntary +0.2% 3748414 548640 179896 4476950 445016 vmlinux.preempt +3.5% voluntary-preempt is +0.2% of .text, preempt is +3.5%. This feature has been tested for many months by lots of people (and it's also included in the RHEL4 distribution and earlier variants were in Fedora as well), and it's intended for users and distributions who dont want to use full-blown CONFIG_PREEMPT for one reason or another. Signed-off-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kernel.h | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kernel.h b/include/linux/kernel.h index e25b97062ce1..687ba8c9973d 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -58,15 +58,23 @@ struct completion; * be biten later when the calling function happens to sleep when it is not * supposed to. */ +#ifdef CONFIG_PREEMPT_VOLUNTARY +extern int cond_resched(void); +# define might_resched() cond_resched() +#else +# define might_resched() do { } while (0) +#endif + #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP -#define might_sleep() __might_sleep(__FILE__, __LINE__) -#define might_sleep_if(cond) do { if (unlikely(cond)) might_sleep(); } while (0) -void __might_sleep(char *file, int line); + void __might_sleep(char *file, int line); +# define might_sleep() \ + do { __might_sleep(__FILE__, __LINE__); might_resched(); } while (0) #else -#define might_sleep() do {} while(0) -#define might_sleep_if(cond) do {} while (0) +# define might_sleep() do { might_resched(); } while (0) #endif +#define might_sleep_if(cond) do { if (unlikely(cond)) might_sleep(); } while (0) + #define abs(x) ({ \ int __x = (x); \ (__x < 0) ? -__x : __x; \ -- cgit v1.2.1 From dc009d92435f99498cbc579ce76bf28e837e2c14 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Sat, 25 Jun 2005 14:57:52 -0700 Subject: [PATCH] kexec: add kexec syscalls This patch introduces the architecture independent implementation the sys_kexec_load, the compat_sys_kexec_load system calls. Kexec on panic support has been integrated into the core patch and is relatively clean. In addition the hopefully architecture independent option crashkernel=size@location has been docuemented. It's purpose is to reserve space for the panic kernel to live, and where no DMA transfer will ever be setup to access. Signed-off-by: Eric Biederman Signed-off-by: Alexander Nyberg Signed-off-by: Adrian Bunk Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kexec.h | 127 +++++++++++++++++++++++++++++++++++++++++++++++ include/linux/reboot.h | 3 ++ include/linux/syscalls.h | 5 +- 3 files changed, 133 insertions(+), 2 deletions(-) create mode 100644 include/linux/kexec.h (limited to 'include/linux') diff --git a/include/linux/kexec.h b/include/linux/kexec.h new file mode 100644 index 000000000000..e3fc35f4e35f --- /dev/null +++ b/include/linux/kexec.h @@ -0,0 +1,127 @@ +#ifndef LINUX_KEXEC_H +#define LINUX_KEXEC_H + +#ifdef CONFIG_KEXEC +#include +#include +#include +#include +#include + +/* Verify architecture specific macros are defined */ + +#ifndef KEXEC_SOURCE_MEMORY_LIMIT +#error KEXEC_SOURCE_MEMORY_LIMIT not defined +#endif + +#ifndef KEXEC_DESTINATION_MEMORY_LIMIT +#error KEXEC_DESTINATION_MEMORY_LIMIT not defined +#endif + +#ifndef KEXEC_CONTROL_MEMORY_LIMIT +#error KEXEC_CONTROL_MEMORY_LIMIT not defined +#endif + +#ifndef KEXEC_CONTROL_CODE_SIZE +#error KEXEC_CONTROL_CODE_SIZE not defined +#endif + +#ifndef KEXEC_ARCH +#error KEXEC_ARCH not defined +#endif + +/* + * This structure is used to hold the arguments that are used when loading + * kernel binaries. + */ + +typedef unsigned long kimage_entry_t; +#define IND_DESTINATION 0x1 +#define IND_INDIRECTION 0x2 +#define IND_DONE 0x4 +#define IND_SOURCE 0x8 + +#define KEXEC_SEGMENT_MAX 8 +struct kexec_segment { + void __user *buf; + size_t bufsz; + unsigned long mem; /* User space sees this as a (void *) ... */ + size_t memsz; +}; + +#ifdef CONFIG_COMPAT +struct compat_kexec_segment { + compat_uptr_t buf; + compat_size_t bufsz; + compat_ulong_t mem; /* User space sees this as a (void *) ... */ + compat_size_t memsz; +}; +#endif + +struct kimage { + kimage_entry_t head; + kimage_entry_t *entry; + kimage_entry_t *last_entry; + + unsigned long destination; + + unsigned long start; + struct page *control_code_page; + + unsigned long nr_segments; + struct kexec_segment segment[KEXEC_SEGMENT_MAX]; + + struct list_head control_pages; + struct list_head dest_pages; + struct list_head unuseable_pages; + + /* Address of next control page to allocate for crash kernels. */ + unsigned long control_page; + + /* Flags to indicate special processing */ + unsigned int type : 1; +#define KEXEC_TYPE_DEFAULT 0 +#define KEXEC_TYPE_CRASH 1 +}; + + + +/* kexec interface functions */ +extern NORET_TYPE void machine_kexec(struct kimage *image) ATTRIB_NORET; +extern int machine_kexec_prepare(struct kimage *image); +extern void machine_kexec_cleanup(struct kimage *image); +extern asmlinkage long sys_kexec_load(unsigned long entry, + unsigned long nr_segments, struct kexec_segment __user *segments, + unsigned long flags); +#ifdef CONFIG_COMPAT +extern asmlinkage long compat_sys_kexec_load(unsigned long entry, + unsigned long nr_segments, struct compat_kexec_segment __user *segments, + unsigned long flags); +#endif +extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order); +extern void crash_kexec(void); +extern struct kimage *kexec_image; + +#define KEXEC_ON_CRASH 0x00000001 +#define KEXEC_ARCH_MASK 0xffff0000 + +/* These values match the ELF architecture values. + * Unless there is a good reason that should continue to be the case. + */ +#define KEXEC_ARCH_DEFAULT ( 0 << 16) +#define KEXEC_ARCH_386 ( 3 << 16) +#define KEXEC_ARCH_X86_64 (62 << 16) +#define KEXEC_ARCH_PPC (20 << 16) +#define KEXEC_ARCH_PPC64 (21 << 16) +#define KEXEC_ARCH_IA_64 (50 << 16) + +#define KEXEC_FLAGS (KEXEC_ON_CRASH) /* List of defined/legal kexec flags */ + +/* Location of a reserved region to hold the crash kernel. + */ +extern struct resource crashk_res; + +#else /* !CONFIG_KEXEC */ +static inline void crash_kexec(void) { } +#endif /* CONFIG_KEXEC */ +#endif /* LINUX_KEXEC_H */ diff --git a/include/linux/reboot.h b/include/linux/reboot.h index d60fafc8bdc5..c5a05e16edb2 100644 --- a/include/linux/reboot.h +++ b/include/linux/reboot.h @@ -51,6 +51,9 @@ extern void machine_restart(char *cmd); extern void machine_halt(void); extern void machine_power_off(void); +extern void machine_shutdown(void); +extern void machine_crash_shutdown(void); + #endif #endif /* _LINUX_REBOOT_H */ diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index c39f6f72cbbc..7ba8f8f747aa 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -159,8 +159,9 @@ asmlinkage long sys_shutdown(int, int); asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user *arg); asmlinkage long sys_restart_syscall(void); -asmlinkage long sys_kexec_load(void *entry, unsigned long nr_segments, - struct kexec_segment *segments, unsigned long flags); +asmlinkage long sys_kexec_load(unsigned long entry, + unsigned long nr_segments, struct kexec_segment __user *segments, + unsigned long flags); asmlinkage long sys_exit(int error_code); asmlinkage void sys_exit_group(int error_code); -- cgit v1.2.1 From cf13f0eaffa31bf6a145c53c589654b11c72ddc7 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Sat, 25 Jun 2005 14:58:11 -0700 Subject: [PATCH] kexec: s390 support Add kexec support for s390 architecture. From: Milton Miller - Fix passing of first argument to relocate_kernel assembly. - Fix Kconfig description. - Remove wrong comment and comments that describe obvious things. - Allow only KEXEC_TYPE_DEFAULT as image type -> dump not supported. Acked-by: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kexec.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/kexec.h b/include/linux/kexec.h index e3fc35f4e35f..0653a27c3d72 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -114,6 +114,7 @@ extern struct kimage *kexec_image; #define KEXEC_ARCH_PPC (20 << 16) #define KEXEC_ARCH_PPC64 (21 << 16) #define KEXEC_ARCH_IA_64 (50 << 16) +#define KEXEC_ARCH_S390 (22 << 16) #define KEXEC_FLAGS (KEXEC_ON_CRASH) /* List of defined/legal kexec flags */ -- cgit v1.2.1 From 92aa63a5a1bf2e7b0c79e6716d24b76dbbdcf951 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 25 Jun 2005 14:58:18 -0700 Subject: [PATCH] kdump: Retrieve saved max pfn This patch retrieves the max_pfn being used by previous kernel and stores it in a safe location (saved_max_pfn) before it is overwritten due to user defined memory map. This pfn is used to make sure that user does not try to read the physical memory beyond saved_max_pfn. Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/bootmem.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index 500f451ce0c0..82bd8842d11c 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -22,6 +22,10 @@ extern unsigned long min_low_pfn; */ extern unsigned long max_pfn; +#ifdef CONFIG_CRASH_DUMP +extern unsigned long saved_max_pfn; +#endif + /* * node_bootmem_map is a map pointer - the bits represent all physical * memory pages (including holes) on the node. -- cgit v1.2.1 From 60e64d46a58236e3c718074372cab6a5b56a3b15 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 25 Jun 2005 14:58:19 -0700 Subject: [PATCH] kdump: Routines for copying dump pages This patch provides the interfaces necessary to read the dump contents, treating it as a high memory device. Signed off by Hariprasad Nellitheertha Signed-off-by: Eric Biederman Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/crash_dump.h | 13 +++++++++++++ include/linux/highmem.h | 1 + 2 files changed, 14 insertions(+) create mode 100644 include/linux/crash_dump.h (limited to 'include/linux') diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h new file mode 100644 index 000000000000..7d983b817429 --- /dev/null +++ b/include/linux/crash_dump.h @@ -0,0 +1,13 @@ +#ifndef LINUX_CRASH_DUMP_H +#define LINUX_CRASH_DUMP_H + +#ifdef CONFIG_CRASH_DUMP +#include +#include +#include +#include + +extern ssize_t copy_oldmem_page(unsigned long, char *, size_t, + unsigned long, int); +#endif /* CONFIG_CRASH_DUMP */ +#endif /* LINUX_CRASHDUMP_H */ diff --git a/include/linux/highmem.h b/include/linux/highmem.h index 2a7e6c65c882..6bece9280eb7 100644 --- a/include/linux/highmem.h +++ b/include/linux/highmem.h @@ -28,6 +28,7 @@ static inline void *kmap(struct page *page) #define kmap_atomic(page, idx) page_address(page) #define kunmap_atomic(addr, idx) do { } while (0) +#define kmap_atomic_pfn(pfn, idx) page_address(pfn_to_page(pfn)) #define kmap_atomic_to_page(ptr) virt_to_page(ptr) #endif /* CONFIG_HIGHMEM */ -- cgit v1.2.1 From 2030eae52b416a9a9f0ffda74c982b7f1e19496d Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 25 Jun 2005 14:58:20 -0700 Subject: [PATCH] Retrieve elfcorehdr address from command line This patch adds support for retrieving the address of elf core header if one is passed in command line. Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/crash_dump.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h index 7d983b817429..3f25fd1eaa4b 100644 --- a/include/linux/crash_dump.h +++ b/include/linux/crash_dump.h @@ -7,6 +7,7 @@ #include #include +extern unsigned long long elfcorehdr_addr; extern ssize_t copy_oldmem_page(unsigned long, char *, size_t, unsigned long, int); #endif /* CONFIG_CRASH_DUMP */ -- cgit v1.2.1 From 666bfddbe8b8fd4fd44617d6c55193d5ac7edb29 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Sat, 25 Jun 2005 14:58:21 -0700 Subject: [PATCH] kdump: Access dump file in elf format (/proc/vmcore) From: "Vivek Goyal" o Support for /proc/vmcore interface. This interface exports elf core image either in ELF32 or ELF64 format, depending on the format in which elf headers have been stored by crashed kernel. o Added support for CONFIG_VMCORE config option. o Removed the dependency on /proc/kcore. From: "Eric W. Biederman" This patch has been refactored to more closely match the prevailing style in the affected files. And to clearly indicate the dependency between /proc/kcore and proc/vmcore.c From: Hariprasad Nellitheertha This patch contains the code that provides an ELF format interface to the previous kernel's memory post kexec reboot. Signed off by Hariprasad Nellitheertha Signed-off-by: Eric Biederman Signed-off-by: Vivek Goyal Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/crash_dump.h | 4 ++++ include/linux/proc_fs.h | 7 +++++++ 2 files changed, 11 insertions(+) (limited to 'include/linux') diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h index 3f25fd1eaa4b..534d750d922d 100644 --- a/include/linux/crash_dump.h +++ b/include/linux/crash_dump.h @@ -7,8 +7,12 @@ #include #include +#define ELFCORE_ADDR_MAX (-1ULL) extern unsigned long long elfcorehdr_addr; extern ssize_t copy_oldmem_page(unsigned long, char *, size_t, unsigned long, int); +extern struct file_operations proc_vmcore_operations; +extern struct proc_dir_entry *proc_vmcore; + #endif /* CONFIG_CRASH_DUMP */ #endif /* LINUX_CRASHDUMP_H */ diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h index 59e505261fd6..0563581e3a02 100644 --- a/include/linux/proc_fs.h +++ b/include/linux/proc_fs.h @@ -74,6 +74,13 @@ struct kcore_list { size_t size; }; +struct vmcore { + struct list_head list; + unsigned long long paddr; + unsigned long size; + loff_t offset; +}; + #ifdef CONFIG_PROC_FS extern struct proc_dir_entry proc_root; -- cgit v1.2.1 From 6e274d144302068a00794ec22e73520c0615cb6f Mon Sep 17 00:00:00 2001 From: Alexander Nyberg Date: Sat, 25 Jun 2005 14:58:26 -0700 Subject: [PATCH] kdump: Use real pt_regs from exception Makes kexec_crashdump() take a pt_regs * as an argument. This allows to get exact register state at the point of the crash. If we come from direct panic assertion NULL will be passed and the current registers saved before crashdump. This hooks into two places: die(): check the conditions under which we will panic when calling do_exit and go there directly with the pt_regs that caused the fatal fault. die_nmi(): If we receive an NMI lockup while in the kernel use the pt_regs and go directly to crash_kexec(). We're probably nested up badly at this point so this might be the only chance to escape with proper information. Signed-off-by: Alexander Nyberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kexec.h | 8 ++++++-- include/linux/reboot.h | 3 ++- 2 files changed, 8 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 0653a27c3d72..7383173a3a9c 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -99,7 +99,8 @@ extern asmlinkage long compat_sys_kexec_load(unsigned long entry, unsigned long flags); #endif extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order); -extern void crash_kexec(void); +extern void crash_kexec(struct pt_regs *); +int kexec_should_crash(struct task_struct *); extern struct kimage *kexec_image; #define KEXEC_ON_CRASH 0x00000001 @@ -123,6 +124,9 @@ extern struct kimage *kexec_image; extern struct resource crashk_res; #else /* !CONFIG_KEXEC */ -static inline void crash_kexec(void) { } +struct pt_regs; +struct task_struct; +static inline void crash_kexec(struct pt_regs *regs) { } +static inline int kexec_should_crash(struct task_struct *p) { return 0; } #endif /* CONFIG_KEXEC */ #endif /* LINUX_KEXEC_H */ diff --git a/include/linux/reboot.h b/include/linux/reboot.h index c5a05e16edb2..2d4dd23168dd 100644 --- a/include/linux/reboot.h +++ b/include/linux/reboot.h @@ -52,7 +52,8 @@ extern void machine_halt(void); extern void machine_power_off(void); extern void machine_shutdown(void); -extern void machine_crash_shutdown(void); +struct pt_regs; +extern void machine_crash_shutdown(struct pt_regs *); #endif -- cgit v1.2.1 From 72414d3f1d22fc3e311b162fca95c430048d38ce Mon Sep 17 00:00:00 2001 From: Maneesh Soni Date: Sat, 25 Jun 2005 14:58:28 -0700 Subject: [PATCH] kexec code cleanup o Following patch provides purely cosmetic changes and corrects CodingStyle guide lines related certain issues like below in kexec related files o braces for one line "if" statements, "for" loops, o more than 80 column wide lines, o No space after "while", "for" and "switch" key words o Changes: o take-2: Removed the extra tab before "case" key words. o take-3: Put operator at the end of line and space before "*/" Signed-off-by: Maneesh Soni Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kexec.h | 13 ++++++++----- include/linux/syscalls.h | 6 +++--- 2 files changed, 11 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kexec.h b/include/linux/kexec.h index 7383173a3a9c..c8468472aec0 100644 --- a/include/linux/kexec.h +++ b/include/linux/kexec.h @@ -91,14 +91,17 @@ extern NORET_TYPE void machine_kexec(struct kimage *image) ATTRIB_NORET; extern int machine_kexec_prepare(struct kimage *image); extern void machine_kexec_cleanup(struct kimage *image); extern asmlinkage long sys_kexec_load(unsigned long entry, - unsigned long nr_segments, struct kexec_segment __user *segments, - unsigned long flags); + unsigned long nr_segments, + struct kexec_segment __user *segments, + unsigned long flags); #ifdef CONFIG_COMPAT extern asmlinkage long compat_sys_kexec_load(unsigned long entry, - unsigned long nr_segments, struct compat_kexec_segment __user *segments, - unsigned long flags); + unsigned long nr_segments, + struct compat_kexec_segment __user *segments, + unsigned long flags); #endif -extern struct page *kimage_alloc_control_pages(struct kimage *image, unsigned int order); +extern struct page *kimage_alloc_control_pages(struct kimage *image, + unsigned int order); extern void crash_kexec(struct pt_regs *); int kexec_should_crash(struct task_struct *); extern struct kimage *kexec_image; diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 7ba8f8f747aa..52830b6d94e5 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -159,9 +159,9 @@ asmlinkage long sys_shutdown(int, int); asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user *arg); asmlinkage long sys_restart_syscall(void); -asmlinkage long sys_kexec_load(unsigned long entry, - unsigned long nr_segments, struct kexec_segment __user *segments, - unsigned long flags); +asmlinkage long sys_kexec_load(unsigned long entry, unsigned long nr_segments, + struct kexec_segment __user *segments, + unsigned long flags); asmlinkage long sys_exit(int error_code); asmlinkage void sys_exit_group(int error_code); -- cgit v1.2.1 From 8c0e33c133021ee241e9d51255b9fb18eb34ef0e Mon Sep 17 00:00:00 2001 From: Nick Wilson Date: Sat, 25 Jun 2005 14:59:00 -0700 Subject: [PATCH] Use ALIGN to remove duplicate code This patch makes use of ALIGN() to remove duplicate round-up code. Signed-off-by: Nick Wilson Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/a.out.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/a.out.h b/include/linux/a.out.h index af8a1dfa5c32..f913cc3e1b0d 100644 --- a/include/linux/a.out.h +++ b/include/linux/a.out.h @@ -138,7 +138,7 @@ enum machine_type { #endif #endif -#define _N_SEGMENT_ROUND(x) (((x) + SEGMENT_SIZE - 1) & ~(SEGMENT_SIZE - 1)) +#define _N_SEGMENT_ROUND(x) ALIGN(x, SEGMENT_SIZE) #define _N_TXTENDADDR(x) (N_TXTADDR(x)+(x).a_text) -- cgit v1.2.1 From 681ea4b930768444e9d88651c1362b0bf6d2a42b Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sat, 25 Jun 2005 14:59:04 -0700 Subject: [PATCH] drivers/char/nvram.c: possible cleanups This patch contains the following possible cleanups: - make the needlessly global function __nvram_set_checksum static - #if 0 the unused global function nvram_set_checksum - remove the EXPORT_SYMBOL's for both functions Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nvram.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nvram.h b/include/linux/nvram.h index b031e41b5e0d..9189829c131c 100644 --- a/include/linux/nvram.h +++ b/include/linux/nvram.h @@ -20,8 +20,6 @@ extern void __nvram_write_byte(unsigned char c, int i); extern void nvram_write_byte(unsigned char c, int i); extern int __nvram_check_checksum(void); extern int nvram_check_checksum(void); -extern void __nvram_set_checksum(void); -extern void nvram_set_checksum(void); #endif #endif /* _LINUX_NVRAM_H */ -- cgit v1.2.1 From 93d17d3d84b7147e8f07aeeb15ec01aa92c6b564 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Sat, 25 Jun 2005 14:59:10 -0700 Subject: [PATCH] drivers/block/ll_rw_blk.c: cleanups This patch contains the following cleanups: - make needlessly global code static - remove the following unused global functions: - blkdev_scsi_issue_flush_fn - __blk_attempt_remerge - remove the following unused EXPORT_SYMBOL's: - blk_phys_contig_segment - blk_hw_contig_segment - blkdev_scsi_issue_flush_fn - __blk_attempt_remerge Signed-off-by: Adrian Bunk Acked-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/blkdev.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 60272141ff19..b54a0348a890 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -539,15 +539,12 @@ extern void generic_make_request(struct bio *bio); extern void blk_put_request(struct request *); extern void blk_end_sync_rq(struct request *rq); extern void blk_attempt_remerge(request_queue_t *, struct request *); -extern void __blk_attempt_remerge(request_queue_t *, struct request *); extern struct request *blk_get_request(request_queue_t *, int, int); extern void blk_insert_request(request_queue_t *, struct request *, int, void *); extern void blk_requeue_request(request_queue_t *, struct request *); extern void blk_plug_device(request_queue_t *); extern int blk_remove_plug(request_queue_t *); extern void blk_recount_segments(request_queue_t *, struct bio *); -extern int blk_phys_contig_segment(request_queue_t *q, struct bio *, struct bio *); -extern int blk_hw_contig_segment(request_queue_t *q, struct bio *, struct bio *); extern int scsi_cmd_ioctl(struct file *, struct gendisk *, unsigned int, void __user *); extern void blk_start_queue(request_queue_t *q); extern void blk_stop_queue(request_queue_t *q); @@ -631,7 +628,6 @@ extern void blk_queue_dma_alignment(request_queue_t *, int); extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev); extern void blk_queue_ordered(request_queue_t *, int); extern void blk_queue_issue_flush_fn(request_queue_t *, issue_flush_fn *); -extern int blkdev_scsi_issue_flush_fn(request_queue_t *, struct gendisk *, sector_t *); extern struct request *blk_start_pre_flush(request_queue_t *,struct request *); extern int blk_complete_barrier_rq(request_queue_t *, struct request *, int); extern int blk_complete_barrier_rq_locked(request_queue_t *, struct request *, int); @@ -675,8 +671,6 @@ extern int blkdev_issue_flush(struct block_device *, sector_t *); #define blkdev_entry_to_request(entry) list_entry((entry), struct request, queuelist) -extern void drive_stat_acct(struct request *, int, int); - static inline int queue_hardsect_size(request_queue_t *q) { int retval = 512; -- cgit v1.2.1 From 3e1d1d28d99dabe63c64f7f40f1ca1d646de1f73 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Fri, 24 Jun 2005 23:13:50 -0700 Subject: [PATCH] Cleanup patch for process freezing 1. Establish a simple API for process freezing defined in linux/include/sched.h: frozen(process) Check for frozen process freezing(process) Check if a process is being frozen freeze(process) Tell a process to freeze (go to refrigerator) thaw_process(process) Restart process frozen_process(process) Process is frozen now 2. Remove all references to PF_FREEZE and PF_FROZEN from all kernel sources except sched.h 3. Fix numerous locations where try_to_freeze is manually done by a driver 4. Remove the argument that is no longer necessary from two function calls. 5. Some whitespace cleanup 6. Clear potential race in refrigerator (provides an open window of PF_FREEZE cleared before setting PF_FROZEN, recalc_sigpending does not check PF_FROZEN). This patch does not address the problem of freeze_processes() violating the rule that a task may only modify its own flags by setting PF_FREEZE. This is not clean in an SMP environment. freeze(process) is therefore not SMP safe! Signed-off-by: Christoph Lameter Signed-off-by: Linus Torvalds --- include/linux/sched.h | 73 +++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 59 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sched.h b/include/linux/sched.h index 2c69682b0444..e7fd09b0557f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1245,33 +1245,78 @@ extern void normalize_rt_tasks(void); #endif -/* try_to_freeze - * - * Checks whether we need to enter the refrigerator - * and returns 1 if we did so. - */ #ifdef CONFIG_PM -extern void refrigerator(unsigned long); +/* + * Check if a process has been frozen + */ +static inline int frozen(struct task_struct *p) +{ + return p->flags & PF_FROZEN; +} + +/* + * Check if there is a request to freeze a process + */ +static inline int freezing(struct task_struct *p) +{ + return p->flags & PF_FREEZE; +} + +/* + * Request that a process be frozen + * FIXME: SMP problem. We may not modify other process' flags! + */ +static inline void freeze(struct task_struct *p) +{ + p->flags |= PF_FREEZE; +} + +/* + * Wake up a frozen process + */ +static inline int thaw_process(struct task_struct *p) +{ + if (frozen(p)) { + p->flags &= ~PF_FROZEN; + wake_up_process(p); + return 1; + } + return 0; +} + +/* + * freezing is complete, mark process as frozen + */ +static inline void frozen_process(struct task_struct *p) +{ + p->flags = (p->flags & ~PF_FREEZE) | PF_FROZEN; +} + +extern void refrigerator(void); extern int freeze_processes(void); extern void thaw_processes(void); -static inline int try_to_freeze(unsigned long refrigerator_flags) +static inline int try_to_freeze(void) { - if (unlikely(current->flags & PF_FREEZE)) { - refrigerator(refrigerator_flags); + if (freezing(current)) { + refrigerator(); return 1; } else return 0; } #else -static inline void refrigerator(unsigned long flag) {} +static inline int frozen(struct task_struct *p) { return 0; } +static inline int freezing(struct task_struct *p) { return 0; } +static inline void freeze(struct task_struct *p) { BUG(); } +static inline int thaw_process(struct task_struct *p) { return 1; } +static inline void frozen_process(struct task_struct *p) { BUG(); } + +static inline void refrigerator(void) {} static inline int freeze_processes(void) { BUG(); return 0; } static inline void thaw_processes(void) {} -static inline int try_to_freeze(unsigned long refrigerator_flags) -{ - return 0; -} +static inline int try_to_freeze(void) { return 0; } + #endif /* CONFIG_PM */ #endif /* __KERNEL__ */ -- cgit v1.2.1 From 169a3e66637c667b43dab7c319ffd5c99804cad8 Mon Sep 17 00:00:00 2001 From: Jay Vosburgh Date: Sun, 26 Jun 2005 17:54:11 -0400 Subject: bonding: xor/802.3ad improved slave hash Add support for alternate slave selection algorithms to bonding balance-xor and 802.3ad modes. Default mode (what we have now: xor of MAC addresses) is "layer2", new choice is "layer3+4", using IP and port information for hashing to select peer. Originally submitted by Jason Gabler for balance-xor mode; modified by Jay Vosburgh to additionally support 802.3ad mode. Jason's original comment is as follows: The attached patch to the Linux Etherchannel Bonding driver modifies the driver's "balance-xor" mode as follows: - alternate hashing policy support for mode 2 * Added kernel parameter "xmit_policy" to allow the specification of different hashing policies for mode 2. The original mode 2 policy is the default, now found in xmit_hash_policy_layer2(). * Added xmit_hash_policy_layer34() This patch was inspired by hashing policies implemented by Cisco, Foundry and IBM, which are explained in Foundry documentation found at: http://www.foundrynet.com/services/documentation/sribcg/Trunking.html#112750 Signed-off-by: Jason Gabler Signed-off-by: Jay Vosburgh --- include/linux/if_bonding.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/if_bonding.h b/include/linux/if_bonding.h index 57024ce2c74f..84598fa2e9de 100644 --- a/include/linux/if_bonding.h +++ b/include/linux/if_bonding.h @@ -35,6 +35,9 @@ * * 2003/12/01 - Shmulik Hen * - Code cleanup and style changes + * + * 2005/05/05 - Jason Gabler + * - added definitions for various XOR hashing policies */ #ifndef _LINUX_IF_BONDING_H @@ -80,6 +83,10 @@ #define BOND_DEFAULT_MAX_BONDS 1 /* Default maximum number of devices to support */ +/* hashing types */ +#define BOND_XMIT_POLICY_LAYER2 0 /* layer 2 (MAC only), default */ +#define BOND_XMIT_POLICY_LAYER34 1 /* layer 3+4 (IP ^ MAC) */ + typedef struct ifbond { __s32 bond_mode; __s32 num_slaves; -- cgit v1.2.1 From 32e9e25ef20789c24ffa1f41489a13932cf82c77 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 26 Jun 2005 15:28:10 -0700 Subject: [ATALK]: Include asm/byteorder.h in linux/atalk.h We're using __be16 in userland visible types, so we have to include asm/byteorder.h so that works. Signed-off-by: David S. Miller --- include/linux/atalk.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/atalk.h b/include/linux/atalk.h index 09a1451c1159..911c09cb9bf9 100644 --- a/include/linux/atalk.h +++ b/include/linux/atalk.h @@ -1,6 +1,8 @@ #ifndef __LINUX_ATALK_H__ #define __LINUX_ATALK_H__ +#include + /* * AppleTalk networking structures * -- cgit v1.2.1 From f49d16ef2d6f008119d4ee2c895781fb229bad68 Mon Sep 17 00:00:00 2001 From: Manfred Spraul Date: Sun, 26 Jun 2005 11:36:52 +0200 Subject: [PATCH] forcedeth: Add support for new device id This is a multi-part message in MIME format. --- include/linux/pci_ids.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index bf608808a60c..3af7450278b7 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1284,6 +1284,8 @@ #define PCI_DEVICE_ID_NVIDIA_GEFORCE_FX_GO5700_2 0x0348 #define PCI_DEVICE_ID_NVIDIA_QUADRO_FX_GO1000 0x034C #define PCI_DEVICE_ID_NVIDIA_QUADRO_FX_1100 0x034E +#define PCI_DEVICE_ID_NVIDIA_NVENET_14 0x0372 +#define PCI_DEVICE_ID_NVIDIA_NVENET_15 0x0373 #define PCI_VENDOR_ID_IMS 0x10e0 #define PCI_DEVICE_ID_IMS_8849 0x8849 -- cgit v1.2.1 From ec9f47cd6a14ca069bb7552a984c0a338fc7262b Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 27 Jun 2005 11:12:54 +0100 Subject: [PATCH] Serial: Split 8250 port table Add separate files for the different 8250 ISA-based serial boards. Looking across all the various architectures, it seems reasonable that we can key the availability of the configuration options for these beasts to the bus-related symbols (iow, CONFIG_ISA). We also standardise the base baud/uart clock rate for these boards - I'm sure that isn't architecture specific, but is solely dependent on the crystal fitted on the board (which should be the same no matter what type of machine its fitted into.) Signed-off-by: Russell King --- include/linux/serial_8250.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h index 823181af6ddf..3e3c1fa35b06 100644 --- a/include/linux/serial_8250.h +++ b/include/linux/serial_8250.h @@ -22,6 +22,7 @@ struct plat_serial8250_port { unsigned int uartclk; /* UART clock rate */ unsigned char regshift; /* register shift */ unsigned char iotype; /* UPIO_* */ + unsigned char hub6; unsigned int flags; /* UPF_* flags */ }; -- cgit v1.2.1 From 22e2c507c301c3dbbcf91b4948b88f78842ee6c9 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 27 Jun 2005 10:55:12 +0200 Subject: [PATCH] Update cfq io scheduler to time sliced design This updates the CFQ io scheduler to the new time sliced design (cfq v3). It provides full process fairness, while giving excellent aggregate system throughput even for many competing processes. It supports io priorities, either inherited from the cpu nice value or set directly with the ioprio_get/set syscalls. The latter closely mimic set/getpriority. This import is based on my latest from -mm. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds --- include/linux/bio.h | 14 ++++++++ include/linux/blkdev.h | 25 +++++++++----- include/linux/elevator.h | 8 ++--- include/linux/fs.h | 19 +++++++++++ include/linux/init_task.h | 2 ++ include/linux/ioprio.h | 87 +++++++++++++++++++++++++++++++++++++++++++++++ include/linux/sched.h | 6 +++- include/linux/writeback.h | 6 ++-- 8 files changed, 152 insertions(+), 15 deletions(-) create mode 100644 include/linux/ioprio.h (limited to 'include/linux') diff --git a/include/linux/bio.h b/include/linux/bio.h index 038022763f09..36ef29fa0d8b 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -22,6 +22,7 @@ #include #include +#include /* Platforms may set this to teach the BIO layer about IOMMU hardware. */ #include @@ -149,6 +150,19 @@ struct bio { #define BIO_RW_FAILFAST 3 #define BIO_RW_SYNC 4 +/* + * upper 16 bits of bi_rw define the io priority of this bio + */ +#define BIO_PRIO_SHIFT (8 * sizeof(unsigned long) - IOPRIO_BITS) +#define bio_prio(bio) ((bio)->bi_rw >> BIO_PRIO_SHIFT) +#define bio_prio_valid(bio) ioprio_valid(bio_prio(bio)) + +#define bio_set_prio(bio, prio) do { \ + WARN_ON(prio >= (1 << IOPRIO_BITS)); \ + (bio)->bi_rw &= ((1UL << BIO_PRIO_SHIFT) - 1); \ + (bio)->bi_rw |= ((unsigned long) (prio) << BIO_PRIO_SHIFT); \ +} while (0) + /* * various member access, note that bio_data should of course not be used * on highmem page vectors diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index b54a0348a890..21a8674cd149 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -54,16 +54,23 @@ struct as_io_context { struct cfq_queue; struct cfq_io_context { - void (*dtor)(struct cfq_io_context *); - void (*exit)(struct cfq_io_context *); - - struct io_context *ioc; - /* * circular list of cfq_io_contexts belonging to a process io context */ struct list_head list; struct cfq_queue *cfqq; + void *key; + + struct io_context *ioc; + + unsigned long last_end_request; + unsigned long last_queue; + unsigned long ttime_total; + unsigned long ttime_samples; + unsigned long ttime_mean; + + void (*dtor)(struct cfq_io_context *); + void (*exit)(struct cfq_io_context *); }; /* @@ -73,7 +80,9 @@ struct cfq_io_context { */ struct io_context { atomic_t refcount; - pid_t pid; + struct task_struct *task; + + int (*set_ioprio)(struct io_context *, unsigned int); /* * For request batching @@ -81,8 +90,6 @@ struct io_context { unsigned long last_waited; /* Time last woken after wait for request */ int nr_batch_requests; /* Number of requests left in the batch */ - spinlock_t lock; - struct as_io_context *aic; struct cfq_io_context *cic; }; @@ -134,6 +141,8 @@ struct request { void *elevator_private; + unsigned short ioprio; + int rq_status; /* should split this into a few status bits */ struct gendisk *rq_disk; int errors; diff --git a/include/linux/elevator.h b/include/linux/elevator.h index ee54f81faad5..ea6bbc2d7407 100644 --- a/include/linux/elevator.h +++ b/include/linux/elevator.h @@ -16,9 +16,9 @@ typedef void (elevator_remove_req_fn) (request_queue_t *, struct request *); typedef void (elevator_requeue_req_fn) (request_queue_t *, struct request *); typedef struct request *(elevator_request_list_fn) (request_queue_t *, struct request *); typedef void (elevator_completed_req_fn) (request_queue_t *, struct request *); -typedef int (elevator_may_queue_fn) (request_queue_t *, int); +typedef int (elevator_may_queue_fn) (request_queue_t *, int, struct bio *); -typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, int); +typedef int (elevator_set_req_fn) (request_queue_t *, struct request *, struct bio *, int); typedef void (elevator_put_req_fn) (request_queue_t *, struct request *); typedef void (elevator_deactivate_req_fn) (request_queue_t *, struct request *); @@ -96,9 +96,9 @@ extern struct request *elv_former_request(request_queue_t *, struct request *); extern struct request *elv_latter_request(request_queue_t *, struct request *); extern int elv_register_queue(request_queue_t *q); extern void elv_unregister_queue(request_queue_t *q); -extern int elv_may_queue(request_queue_t *, int); +extern int elv_may_queue(request_queue_t *, int, struct bio *); extern void elv_completed_request(request_queue_t *, struct request *); -extern int elv_set_request(request_queue_t *, struct request *, int); +extern int elv_set_request(request_queue_t *, struct request *, struct bio *, int); extern void elv_put_request(request_queue_t *, struct request *); /* diff --git a/include/linux/fs.h b/include/linux/fs.h index 3ae8e37bdfc8..047bde30836a 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -213,6 +213,7 @@ extern int dir_notify_enable; #include #include #include +#include #include #include @@ -822,16 +823,34 @@ enum { #define vfs_check_frozen(sb, level) \ wait_event((sb)->s_wait_unfrozen, ((sb)->s_frozen < (level))) +static inline void get_fs_excl(void) +{ + atomic_inc(¤t->fs_excl); +} + +static inline void put_fs_excl(void) +{ + atomic_dec(¤t->fs_excl); +} + +static inline int has_fs_excl(void) +{ + return atomic_read(¤t->fs_excl); +} + + /* * Superblock locking. */ static inline void lock_super(struct super_block * sb) { + get_fs_excl(); down(&sb->s_lock); } static inline void unlock_super(struct super_block * sb) { + put_fs_excl(); up(&sb->s_lock); } diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 03206a425d7a..c727c195a91a 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -81,6 +81,7 @@ extern struct group_info init_groups; .mm = NULL, \ .active_mm = &init_mm, \ .run_list = LIST_HEAD_INIT(tsk.run_list), \ + .ioprio = 0, \ .time_slice = HZ, \ .tasks = LIST_HEAD_INIT(tsk.tasks), \ .ptrace_children= LIST_HEAD_INIT(tsk.ptrace_children), \ @@ -110,6 +111,7 @@ extern struct group_info init_groups; .proc_lock = SPIN_LOCK_UNLOCKED, \ .journal_info = NULL, \ .cpu_timers = INIT_CPU_TIMERS(tsk.cpu_timers), \ + .fs_excl = ATOMIC_INIT(0), \ } diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h new file mode 100644 index 000000000000..7811300d88ee --- /dev/null +++ b/include/linux/ioprio.h @@ -0,0 +1,87 @@ +#ifndef IOPRIO_H +#define IOPRIO_H + +#include + +/* + * Gives us 8 prio classes with 13-bits of data for each class + */ +#define IOPRIO_BITS (16) +#define IOPRIO_CLASS_SHIFT (13) +#define IOPRIO_PRIO_MASK ((1UL << IOPRIO_CLASS_SHIFT) - 1) + +#define IOPRIO_PRIO_CLASS(mask) ((mask) >> IOPRIO_CLASS_SHIFT) +#define IOPRIO_PRIO_DATA(mask) ((mask) & IOPRIO_PRIO_MASK) + +#define ioprio_valid(mask) (IOPRIO_PRIO_CLASS((mask)) != IOPRIO_CLASS_NONE) + +/* + * These are the io priority groups as implemented by CFQ. RT is the realtime + * class, it always gets premium service. BE is the best-effort scheduling + * class, the default for any process. IDLE is the idle scheduling class, it + * is only served when no one else is using the disk. + */ +enum { + IOPRIO_CLASS_NONE, + IOPRIO_CLASS_RT, + IOPRIO_CLASS_BE, + IOPRIO_CLASS_IDLE, +}; + +/* + * 8 best effort priority levels are supported + */ +#define IOPRIO_BE_NR (8) + +asmlinkage int sys_ioprio_set(int, int, int); +asmlinkage int sys_ioprio_get(int, int); + +enum { + IOPRIO_WHO_PROCESS = 1, + IOPRIO_WHO_PGRP, + IOPRIO_WHO_USER, +}; + +/* + * if process has set io priority explicitly, use that. if not, convert + * the cpu scheduler nice value to an io priority + */ +#define IOPRIO_NORM (4) +static inline int task_ioprio(struct task_struct *task) +{ + WARN_ON(!ioprio_valid(task->ioprio)); + return IOPRIO_PRIO_DATA(task->ioprio); +} + +static inline int task_nice_ioprio(struct task_struct *task) +{ + return (task_nice(task) + 20) / 5; +} + +/* + * For inheritance, return the highest of the two given priorities + */ +static inline int ioprio_best(unsigned short aprio, unsigned short bprio) +{ + unsigned short aclass = IOPRIO_PRIO_CLASS(aprio); + unsigned short bclass = IOPRIO_PRIO_CLASS(bprio); + + if (!ioprio_valid(aprio)) + return bprio; + if (!ioprio_valid(bprio)) + return aprio; + + if (aclass == IOPRIO_CLASS_NONE) + aclass = IOPRIO_CLASS_BE; + if (bclass == IOPRIO_CLASS_NONE) + bclass = IOPRIO_CLASS_BE; + + if (aclass == bclass) + return min(aprio, bprio); + if (aclass > bclass) + return bprio; + else + return aprio; +} + +#endif diff --git a/include/linux/sched.h b/include/linux/sched.h index 9530b1903160..ff48815bd3a2 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -608,6 +608,8 @@ struct task_struct { struct list_head run_list; prio_array_t *array; + unsigned short ioprio; + unsigned long sleep_avg; unsigned long long timestamp, last_ran; unsigned long long sched_time; /* sched_clock time spent running */ @@ -763,6 +765,7 @@ struct task_struct { nodemask_t mems_allowed; int cpuset_mems_generation; #endif + atomic_t fs_excl; /* holding fs exclusive resources */ }; static inline pid_t process_group(struct task_struct *tsk) @@ -1112,7 +1115,8 @@ extern void unhash_process(struct task_struct *p); /* * Protects ->fs, ->files, ->mm, ->ptrace, ->group_info, ->comm, keyring - * subscriptions and synchronises with wait4(). Also used in procfs. + * subscriptions and synchronises with wait4(). Also used in procfs. Also + * pins the final release of task.io_context. * * Nests both inside and outside of read_lock(&tasklist_lock). * It must not be nested with write_lock_irq(&tasklist_lock), diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 1262cb43c3ab..d5c3fe1bf33d 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -14,11 +14,13 @@ extern struct list_head inode_unused; * Yes, writeback.h requires sched.h * No, sched.h is not included from here. */ -static inline int current_is_pdflush(void) +static inline int task_is_pdflush(struct task_struct *task) { - return current->flags & PF_FLUSHER; + return task->flags & PF_FLUSHER; } +#define current_is_pdflush() task_is_pdflush(current) + /* * fs/fs-writeback.c */ -- cgit v1.2.1 From 3b18152c327707ae6a2eeba4cfb66457143753bc Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 27 Jun 2005 10:56:24 +0200 Subject: [PATCH] CFQ io scheduler updates - Adjust slice values - Instead of one async queue, one is defined per priority level. This prevents kernel threads (such as reiserfs/x and others) that run at higher io priority from conflicting with others. Previously, it was a coin toss what io prio the async queue got, it was defined by who first set up the queue. - Let a time slice only begin, when the previous slice is completely done. Previously we could be somewhat unfair to a new sync slice, if the previous slice was async and had several ios queued. This might need a little tweaking if throughput suffers a little due to this, allowing perhaps an overlap of a single request or so. - Optimize the calling of kblockd_schedule_work() by doing it only when it is strictly necessary (no requests in driver and work left to do). - Correct sync vs async logic. A 'normal' process can be purely async as well, and a flusher can be purely sync as well. Sync or async is now a property of the class defined and requests pending. Previously writers could be considered sync, when they were really async. - Get rid of the bit fields in cfqq and crq, use flags instead. - Various other cleanups and fixes Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds --- include/linux/ioprio.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h index 7811300d88ee..8a453a0b5e4b 100644 --- a/include/linux/ioprio.h +++ b/include/linux/ioprio.h @@ -12,6 +12,7 @@ #define IOPRIO_PRIO_CLASS(mask) ((mask) >> IOPRIO_CLASS_SHIFT) #define IOPRIO_PRIO_DATA(mask) ((mask) & IOPRIO_PRIO_MASK) +#define IOPRIO_PRIO_VALUE(class, data) (((class) << IOPRIO_CLASS_SHIFT) | data) #define ioprio_valid(mask) (IOPRIO_PRIO_CLASS((mask)) != IOPRIO_CLASS_NONE) -- cgit v1.2.1 From 4808a1c0261176f9c7e28e7f108d41a381a7d0fc Mon Sep 17 00:00:00 2001 From: Olav Kongas Date: Sat, 9 Apr 2005 22:57:39 +0300 Subject: [PATCH] USB: Add isp116x-hcd USB host controller driver This patch provides an "isp116x-hcd" driver for Philips' ISP1160/ISP1161 USB host controllers. The driver: - is relatively small, meant for use on embedded platforms. - runs usbtests 1-14 without problems for days. - has been in use by 6-7 different people on ARM and PPC platforms, running a range of devices including USB hubs. - supports suspend/resume of both the platform device and the root hub; supports remote wakeup of the root hub (but NOT the platform device) by USB devices. - does NOT support ISO transfers (nobody has asked for them). - is PIO-only. Signed-off-by: Olav Kongas Signed-off-by: Greg Kroah-Hartman --- include/linux/usb_isp116x.h | 47 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 47 insertions(+) create mode 100644 include/linux/usb_isp116x.h (limited to 'include/linux') diff --git a/include/linux/usb_isp116x.h b/include/linux/usb_isp116x.h new file mode 100644 index 000000000000..5f5a9d9bd6c2 --- /dev/null +++ b/include/linux/usb_isp116x.h @@ -0,0 +1,47 @@ + +/* + * Board initialization code should put one of these into dev->platform_data + * and place the isp116x onto platform_bus. + */ + +struct isp116x_platform_data { + /* Enable internal resistors on downstream ports */ + unsigned sel15Kres:1; + /* Chip's internal clock won't be stopped in suspended state. + Setting/unsetting this bit takes effect only if + 'remote_wakeup_enable' below is not set. */ + unsigned clknotstop:1; + /* On-chip overcurrent protection */ + unsigned oc_enable:1; + /* INT output polarity */ + unsigned int_act_high:1; + /* INT edge or level triggered */ + unsigned int_edge_triggered:1; + /* WAKEUP pin connected - NOT SUPPORTED */ + /* unsigned remote_wakeup_connected:1; */ + /* Wakeup by devices on usb bus enabled */ + unsigned remote_wakeup_enable:1; + /* Switch or not to switch (keep always powered) */ + unsigned no_power_switching:1; + /* Ganged port power switching (0) or individual port + power switching (1) */ + unsigned power_switching_mode:1; + /* Given port_power, msec/2 after power on till power good */ + u8 potpg; + /* Hardware reset set/clear. If implemented, this function must: + if set == 0, deassert chip's HW reset pin + otherwise, assert chip's HW reset pin */ + void (*reset) (struct device * dev, int set); + /* Hardware clock start/stop. If implemented, this function must: + if start == 0, stop the external clock + otherwise, start the external clock + */ + void (*clock) (struct device * dev, int start); + /* Inter-io delay (ns). The chip is picky about access timings; it + expects at least: + 150ns delay between consecutive accesses to DATA_REG, + 300ns delay between access to ADDR_REG and DATA_REG + OE, WE MUST NOT be changed during these intervals + */ + void (*delay) (struct device * dev, int delay); +}; -- cgit v1.2.1 From 1bbc169621cbe502b9143a27eb12802a0f1d43a0 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Sat, 7 May 2005 13:05:13 -0700 Subject: [PATCH] USB: gadget driver updates (SETUP api change) This updates most of the gadget framework to expect SETUP packets use USB byteorder (matching the annotation in and usage in the host side stack): - definition in - gadget drivers: Ethernet/RNDIS, serial/ACM, file_storage, gadgetfs. - dummy_hcd It also includes some other similar changes as suggested by "sparse", which was used to detect byteorder bugs. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- include/linux/usb_gadget.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/usb_gadget.h b/include/linux/usb_gadget.h index 9bba9997947b..b00f127cb447 100644 --- a/include/linux/usb_gadget.h +++ b/include/linux/usb_gadget.h @@ -711,7 +711,7 @@ usb_gadget_disconnect (struct usb_gadget *gadget) * the hardware level driver. Most calls must be handled by * the gadget driver, including descriptor and configuration * management. The 16 bit members of the setup data are in - * cpu order. Called in_interrupt; this may not sleep. Driver + * USB byte order. Called in_interrupt; this may not sleep. Driver * queues a response to ep0, or returns negative to stall. * @disconnect: Invoked after all transfers have been stopped, * when the host is disconnected. May be called in_interrupt; this -- cgit v1.2.1 From 5da0106f0b9b13afa4a902c01d4c98b002df55ff Mon Sep 17 00:00:00 2001 From: David Brownell Date: Tue, 31 May 2005 10:21:11 -0700 Subject: [PATCH] USB: wireless usb declarations This provides declarations for new requests, descriptors, and bitfields as defined in the Wireless USB 1.0 spec. Device support will involve a new "Wire Adapter" device class, connecting a USB Host to a cluster of wireless USB devices. There will be two adapter types: * Host Wireless Adapter (HWA): the downstream link is wireless, which connects a wireless USB host to wireless USB devices (not unlike like a hub) including to the second type of adapter. * Device Wireless Adapter (DWA): the upstream link is wireless, for connecting existing USB devices through wired links into the cluser. All wireless USB devices will need persistent (and secure!) key storage, and it's probable that Linux -- or device firmware -- will need to be involved with that to bootstrap the initial secure key exchange. Some user interface is required in that initial key exchange, and since the most "hands-off" one is a wired USB link, I suspect wireless operation will usually not be the only mode for wireless USB devices. (Plus, devices can recharge batteries using wired USB...) All other key exchange protocols need error prone user interactions, like copying and/or verifying keys. It'll likely be a while before we have commercial Wireless USB hardware, much less Linux implementations that know how to use it. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- include/linux/usb_ch9.h | 183 ++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 176 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/usb_ch9.h b/include/linux/usb_ch9.h index f5fe94e09a03..39e7ff4ffd28 100644 --- a/include/linux/usb_ch9.h +++ b/include/linux/usb_ch9.h @@ -6,11 +6,14 @@ * * - the master/host side Linux-USB kernel driver API; * - the "usbfs" user space API; and - * - (eventually) a Linux "gadget" slave/device side driver API. + * - the Linux "gadget" slave/device/peripheral side driver API. * * USB 2.0 adds an additional "On The Go" (OTG) mode, which lets systems * act either as a USB master/host or as a USB slave/device. That means - * the master and slave side APIs will benefit from working well together. + * the master and slave side APIs benefit from working well together. + * + * There's also "Wireless USB", using low power short range radios for + * peripheral interconnection but otherwise building on the USB framework. */ #ifndef __LINUX_USB_CH9_H @@ -68,6 +71,18 @@ #define USB_REQ_SET_INTERFACE 0x0B #define USB_REQ_SYNCH_FRAME 0x0C +#define USB_REQ_SET_ENCRYPTION 0x0D /* Wireless USB */ +#define USB_REQ_GET_ENCRYPTION 0x0E +#define USB_REQ_SET_HANDSHAKE 0x0F +#define USB_REQ_GET_HANDSHAKE 0x10 +#define USB_REQ_SET_CONNECTION 0x11 +#define USB_REQ_SET_SECURITY_DATA 0x12 +#define USB_REQ_GET_SECURITY_DATA 0x13 +#define USB_REQ_SET_WUSB_DATA 0x14 +#define USB_REQ_LOOPBACK_DATA_WRITE 0x15 +#define USB_REQ_LOOPBACK_DATA_READ 0x16 +#define USB_REQ_SET_INTERFACE_DS 0x17 + /* * USB feature flags are written using USB_REQ_{CLEAR,SET}_FEATURE, and * are read as a bit array returned by USB_REQ_GET_STATUS. (So there @@ -75,10 +90,12 @@ */ #define USB_DEVICE_SELF_POWERED 0 /* (read only) */ #define USB_DEVICE_REMOTE_WAKEUP 1 /* dev may initiate wakeup */ -#define USB_DEVICE_TEST_MODE 2 /* (high speed only) */ -#define USB_DEVICE_B_HNP_ENABLE 3 /* dev may initiate HNP */ -#define USB_DEVICE_A_HNP_SUPPORT 4 /* RH port supports HNP */ -#define USB_DEVICE_A_ALT_HNP_SUPPORT 5 /* other RH port does */ +#define USB_DEVICE_TEST_MODE 2 /* (wired high speed only) */ +#define USB_DEVICE_BATTERY 2 /* (wireless) */ +#define USB_DEVICE_B_HNP_ENABLE 3 /* (otg) dev may initiate HNP */ +#define USB_DEVICE_WUSB_DEVICE 3 /* (wireless)*/ +#define USB_DEVICE_A_HNP_SUPPORT 4 /* (otg) RH port supports HNP */ +#define USB_DEVICE_A_ALT_HNP_SUPPORT 5 /* (otg) other RH port does */ #define USB_DEVICE_DEBUG_MODE 6 /* (special devices only) */ #define USB_ENDPOINT_HALT 0 /* IN/OUT will STALL */ @@ -135,6 +152,13 @@ struct usb_ctrlrequest { #define USB_DT_OTG 0x09 #define USB_DT_DEBUG 0x0a #define USB_DT_INTERFACE_ASSOCIATION 0x0b +/* these are from the Wireless USB spec */ +#define USB_DT_SECURITY 0x0c +#define USB_DT_KEY 0x0d +#define USB_DT_ENCRYPTION_TYPE 0x0e +#define USB_DT_BOS 0x0f +#define USB_DT_DEVICE_CAPABILITY 0x10 +#define USB_DT_WIRELESS_ENDPOINT_COMP 0x11 /* conventional codes for class-specific descriptors */ #define USB_DT_CS_DEVICE 0x21 @@ -192,6 +216,7 @@ struct usb_device_descriptor { #define USB_CLASS_CSCID 0x0b /* chip+ smart card */ #define USB_CLASS_CONTENT_SEC 0x0d /* content security */ #define USB_CLASS_VIDEO 0x0e +#define USB_CLASS_WIRELESS_CONTROLLER 0xe0 #define USB_CLASS_APP_SPEC 0xfe #define USB_CLASS_VENDOR_SPEC 0xff @@ -223,6 +248,7 @@ struct usb_config_descriptor { #define USB_CONFIG_ATT_ONE (1 << 7) /* must be set */ #define USB_CONFIG_ATT_SELFPOWER (1 << 6) /* self powered */ #define USB_CONFIG_ATT_WAKEUP (1 << 5) /* can wakeup */ +#define USB_CONFIG_ATT_BATTERY (1 << 4) /* battery powered */ /*-------------------------------------------------------------------------*/ @@ -289,6 +315,7 @@ struct usb_endpoint_descriptor { #define USB_ENDPOINT_XFER_ISOC 1 #define USB_ENDPOINT_XFER_BULK 2 #define USB_ENDPOINT_XFER_INT 3 +#define USB_ENDPOINT_MAX_ADJUSTABLE 0x80 /*-------------------------------------------------------------------------*/ @@ -350,6 +377,147 @@ struct usb_interface_assoc_descriptor { } __attribute__ ((packed)); +/*-------------------------------------------------------------------------*/ + +/* USB_DT_SECURITY: group of wireless security descriptors, including + * encryption types available for setting up a CC/association. + */ +struct usb_security_descriptor { + __u8 bLength; + __u8 bDescriptorType; + + __le16 wTotalLength; + __u8 bNumEncryptionTypes; +}; + +/*-------------------------------------------------------------------------*/ + +/* USB_DT_KEY: used with {GET,SET}_SECURITY_DATA; only public keys + * may be retrieved. + */ +struct usb_key_descriptor { + __u8 bLength; + __u8 bDescriptorType; + + __u8 tTKID[3]; + __u8 bReserved; + __u8 bKeyData[0]; +}; + +/*-------------------------------------------------------------------------*/ + +/* USB_DT_ENCRYPTION_TYPE: bundled in DT_SECURITY groups */ +struct usb_encryption_descriptor { + __u8 bLength; + __u8 bDescriptorType; + + __u8 bEncryptionType; +#define USB_ENC_TYPE_UNSECURE 0 +#define USB_ENC_TYPE_WIRED 1 /* non-wireless mode */ +#define USB_ENC_TYPE_CCM_1 2 /* aes128/cbc session */ +#define USB_ENC_TYPE_RSA_1 3 /* rsa3072/sha1 auth */ + __u8 bEncryptionValue; /* use in SET_ENCRYPTION */ + __u8 bAuthKeyIndex; +}; + + +/*-------------------------------------------------------------------------*/ + +/* USB_DT_BOS: group of wireless capabilities */ +struct usb_bos_descriptor { + __u8 bLength; + __u8 bDescriptorType; + + __le16 wTotalLength; + __u8 bNumDeviceCaps; +}; + +/*-------------------------------------------------------------------------*/ + +/* USB_DT_DEVICE_CAPABILITY: grouped with BOS */ +struct usb_dev_cap_header { + __u8 bLength; + __u8 bDescriptorType; + __u8 bDevCapabilityType; +}; + +#define USB_CAP_TYPE_WIRELESS_USB 1 + +struct usb_wireless_cap_descriptor { /* Ultra Wide Band */ + __u8 bLength; + __u8 bDescriptorType; + __u8 bDevCapabilityType; + + __u8 bmAttributes; +#define USB_WIRELESS_P2P_DRD (1 << 1) +#define USB_WIRELESS_BEACON_MASK (3 << 2) +#define USB_WIRELESS_BEACON_SELF (1 << 2) +#define USB_WIRELESS_BEACON_DIRECTED (2 << 2) +#define USB_WIRELESS_BEACON_NONE (3 << 2) + __le16 wPHYRates; /* bit rates, Mbps */ +#define USB_WIRELESS_PHY_53 (1 << 0) /* always set */ +#define USB_WIRELESS_PHY_80 (1 << 1) +#define USB_WIRELESS_PHY_107 (1 << 2) /* always set */ +#define USB_WIRELESS_PHY_160 (1 << 3) +#define USB_WIRELESS_PHY_200 (1 << 4) /* always set */ +#define USB_WIRELESS_PHY_320 (1 << 5) +#define USB_WIRELESS_PHY_400 (1 << 6) +#define USB_WIRELESS_PHY_480 (1 << 7) + __u8 bmTFITXPowerInfo; /* TFI power levels */ + __u8 bmFFITXPowerInfo; /* FFI power levels */ + __le16 bmBandGroup; + __u8 bReserved; +}; + +/*-------------------------------------------------------------------------*/ + +/* USB_DT_WIRELESS_ENDPOINT_COMP: companion descriptor associated with + * each endpoint descriptor for a wireless device + */ +struct usb_wireless_ep_comp_descriptor { + __u8 bLength; + __u8 bDescriptorType; + + __u8 bMaxBurst; + __u8 bMaxSequence; + __le16 wMaxStreamDelay; + __le16 wOverTheAirPacketSize; + __u8 bOverTheAirInterval; + __u8 bmCompAttributes; +#define USB_ENDPOINT_SWITCH_MASK 0x03 /* in bmCompAttributes */ +#define USB_ENDPOINT_SWITCH_NO 0 +#define USB_ENDPOINT_SWITCH_SWITCH 1 +#define USB_ENDPOINT_SWITCH_SCALE 2 +}; + +/*-------------------------------------------------------------------------*/ + +/* USB_REQ_SET_HANDSHAKE is a four-way handshake used between a wireless + * host and a device for connection set up, mutual authentication, and + * exchanging short lived session keys. The handshake depends on a CC. + */ +struct usb_handshake { + __u8 bMessageNumber; + __u8 bStatus; + __u8 tTKID[3]; + __u8 bReserved; + __u8 CDID[16]; + __u8 nonce[16]; + __u8 MIC[8]; +}; + +/*-------------------------------------------------------------------------*/ + +/* USB_REQ_SET_CONNECTION modifies or revokes a connection context (CC). + * A CC may also be set up using non-wireless secure channels (including + * wired USB!), and some devices may support CCs with multiple hosts. + */ +struct usb_connection_context { + __u8 CHID[16]; /* persistent host id */ + __u8 CDID[16]; /* device id (unique w/in host context) */ + __u8 CK[16]; /* connection key */ +}; + /*-------------------------------------------------------------------------*/ /* USB 2.0 defines three speeds, here's how Linux identifies them */ @@ -357,7 +525,8 @@ struct usb_interface_assoc_descriptor { enum usb_device_speed { USB_SPEED_UNKNOWN = 0, /* enumerating */ USB_SPEED_LOW, USB_SPEED_FULL, /* usb 1.1 */ - USB_SPEED_HIGH /* usb 2.0 */ + USB_SPEED_HIGH, /* usb 2.0 */ + USB_SPEED_VARIABLE, /* wireless (usb 2.5) */ }; enum usb_device_state { -- cgit v1.2.1 From 8c8709334cec803368a432a33e0f2e116d48fe07 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Mon, 27 Jun 2005 14:36:34 -0700 Subject: [PATCH] ppc32: Remove CONFIG_PMAC_PBOOK This patch removes CONFIG_PMAC_PBOOK (PowerBook support). This is now split into CONFIG_PMAC_MEDIABAY for the actual hotswap bay that some powerbooks have, CONFIG_PM for power management related code, and just left out of any CONFIG_* option for some generally useful stuff that can be used on non-laptops as well. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pmu.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pmu.h b/include/linux/pmu.h index 6d73eada277e..373bd3b9b330 100644 --- a/include/linux/pmu.h +++ b/include/linux/pmu.h @@ -166,7 +166,7 @@ extern int pmu_i2c_simple_read(int bus, int addr, u8* data, int len); extern int pmu_i2c_simple_write(int bus, int addr, u8* data, int len); -#ifdef CONFIG_PMAC_PBOOK +#ifdef CONFIG_PM /* * Stuff for putting the powerbook to sleep and waking it again. * @@ -208,6 +208,8 @@ struct pmu_sleep_notifier int pmu_register_sleep_notifier(struct pmu_sleep_notifier* notifier); int pmu_unregister_sleep_notifier(struct pmu_sleep_notifier* notifier); +#endif /* CONFIG_PM */ + #define PMU_MAX_BATTERIES 2 /* values for pmu_power_flags */ @@ -235,6 +237,4 @@ extern int pmu_battery_count; extern struct pmu_battery_info pmu_batteries[PMU_MAX_BATTERIES]; extern unsigned int pmu_power_flags; -#endif /* CONFIG_PMAC_PBOOK */ - #endif /* __KERNEL__ */ -- cgit v1.2.1 From ffaa8bd6c904d1ab79b677905067349a5ff51d84 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Mon, 27 Jun 2005 14:36:36 -0700 Subject: [PATCH] seccomp: tsc disable I believe at least for seccomp it's worth to turn off the tsc, not just for HT but for the L2 cache too. So it's up to you, either you turn it off completely (which isn't very nice IMHO) or I recommend to apply this below patch. This has been tested successfully on x86-64 against current cogito repository (i686 compiles so I didn't bother testing ;). People selling the cpu through cpushare may appreciate this bit for a peace of mind. There's no way to get any timing info anymore with this applied (gettimeofday is forbidden of course). The seccomp environment is completely deterministic so it can't be allowed to get timing info, it has to be deterministic so in the future I can enable a computing mode that does a parallel computing for each task with server side transparent checkpointing and verification that the output is the same from all the 2/3 seller computers for each task, without the buyer even noticing (for now the verification is left to the buyer client side and there's no checkpointing, since that would require more kernel changes to track the dirty bits but it'll be easy to extend once the basic mode is finished). Eliminating a cold-cache read of the cr4 global variable will save one cacheline during the tlb flush while making the code per-cpu-safe at the same time. Thanks to Mikael Pettersson for noticing the tlb flush wasn't per-cpu-safe. The global tlb flush can run from irq (IPI calling do_flush_tlb_all) but it'll be transparent to the switch_to code since the IPI won't make any change to the cr4 contents from the point of view of the interrupted code and since it's now all per-cpu stuff, it will not race. So no need to disable irqs in switch_to slow path. Signed-off-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/seccomp.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/linux') diff --git a/include/linux/seccomp.h b/include/linux/seccomp.h index 3a2702bbb1d6..dc89116bb1ca 100644 --- a/include/linux/seccomp.h +++ b/include/linux/seccomp.h @@ -19,6 +19,11 @@ static inline void secure_computing(int this_syscall) __secure_computing(this_syscall); } +static inline int has_secure_computing(struct thread_info *ti) +{ + return unlikely(test_ti_thread_flag(ti, TIF_SECCOMP)); +} + #else /* CONFIG_SECCOMP */ #if (__GNUC__ > 2) @@ -28,6 +33,11 @@ static inline void secure_computing(int this_syscall) #endif #define secure_computing(x) do { } while (0) +/* static inline to preserve typechecking */ +static inline int has_secure_computing(struct thread_info *ti) +{ + return 0; +} #endif /* CONFIG_SECCOMP */ -- cgit v1.2.1 From 3de0a70bd926ff974adb27a38d4fd1049f05e54e Mon Sep 17 00:00:00 2001 From: Mike Miller Date: Mon, 27 Jun 2005 14:36:48 -0700 Subject: [PATCH] cciss: pci id fix This patch fixes a PCI ID I got wrong before. It also adds support for another new SAS controller due out this summer. I didn't have a marketing name prior to my last submission. Also modifies the copyright date range. Signed-off-by: Mike Miller Acked-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pci_ids.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 3af7450278b7..a3961e1d5183 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -712,8 +712,9 @@ #define PCI_DEVICE_ID_HP_DIVA_AUX 0x1290 #define PCI_DEVICE_ID_HP_DIVA_RMP3 0x1301 #define PCI_DEVICE_ID_HP_CISSA 0x3220 -#define PCI_DEVICE_ID_HP_CISSB 0x3230 +#define PCI_DEVICE_ID_HP_CISSB 0x3222 #define PCI_DEVICE_ID_HP_ZX2_IOC 0x4031 +#define PCI_DEVICE_ID_HP_CISSC 0x3230 #define PCI_VENDOR_ID_PCTECH 0x1042 #define PCI_DEVICE_ID_PCTECH_RZ1000 0x1000 -- cgit v1.2.1 From cd6fb584cf7f18ec6b221192b57d712ecc8c1859 Mon Sep 17 00:00:00 2001 From: Mike Miller Date: Mon, 27 Jun 2005 14:36:49 -0700 Subject: [PATCH] cciss: pci domain info pass 2 This is pass 2 of my patch to add pci domain info to an existing ioctl. This time I insert the domain between dev_fn and board_id as Willy suggested and change the var to unsigned short to ease Christoph's concerns. Although I thought unsigned int was the correct var type for this. I also thought it didn't matter where I inserted it in the structure. Signed-off-by: Mike Miller Acked-by: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cciss_ioctl.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/cciss_ioctl.h b/include/linux/cciss_ioctl.h index ee0c6e8995da..424d5e622b43 100644 --- a/include/linux/cciss_ioctl.h +++ b/include/linux/cciss_ioctl.h @@ -10,6 +10,7 @@ typedef struct _cciss_pci_info_struct { unsigned char bus; + unsigned short domain; unsigned char dev_fn; __u32 board_id; } cciss_pci_info_struct; -- cgit v1.2.1 From 9ec4b1f356b3bad928ae8e2aa9caebfa737d52df Mon Sep 17 00:00:00 2001 From: Ananth N Mavinakayanahalli Date: Mon, 27 Jun 2005 15:17:01 -0700 Subject: [PATCH] kprobes: fix single-step out of line - take2 Now that PPC64 has no-execute support, here is a second try to fix the single step out of line during kprobe execution. Kprobes on x86_64 already solved this problem by allocating an executable page and using it as the scratch area for stepping out of line. Reuse that. Signed-off-by: Ananth N Mavinakayanahalli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kprobes.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index 5e1a7b0d7b3f..d304d4579856 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -177,6 +177,8 @@ extern void arch_arm_kprobe(struct kprobe *p); extern void arch_disarm_kprobe(struct kprobe *p); extern void arch_remove_kprobe(struct kprobe *p); extern void show_registers(struct pt_regs *regs); +extern kprobe_opcode_t *get_insn_slot(void); +extern void free_insn_slot(kprobe_opcode_t *slot); /* Get the kprobe at this addr (if any). Must have called lock_kprobes */ struct kprobe *get_kprobe(void *addr); -- cgit v1.2.1 From 802eae7c800fb7f583e6c06afa363585af2bef00 Mon Sep 17 00:00:00 2001 From: Rusty Lynch Date: Mon, 27 Jun 2005 15:17:08 -0700 Subject: [PATCH] Return probe redesign: architecture independent changes The following is the second version of the function return probe patches I sent out earlier this week. Changes since my last submission include: * Fix in ppc64 code removing an unneeded call to re-enable preemption * Fix a build problem in ia64 when kprobes was turned off * Added another BUG_ON check to each of the architecture trampoline handlers My initial patch description ==> From my experiences with adding return probes to x86_64 and ia64, and the feedback on LKML to those patches, I think we can simplify the design for return probes. The following patch tweaks the original design such that: * Instead of storing the stack address in the return probe instance, the task pointer is stored. This gives us all we need in order to: - find the correct return probe instance when we enter the trampoline (even if we are recursing) - find all left-over return probe instances when the task is going away This has the side effect of simplifying the implementation since more work can be done in kernel/kprobes.c since architecture specific knowledge of the stack layout is no longer required. Specifically, we no longer have: - arch_get_kprobe_task() - arch_kprobe_flush_task() - get_rp_inst_tsk() - get_rp_inst() - trampoline_post_handler() * Instead of splitting the return probe handling and cleanup logic across the pre and post trampoline handlers, all the work is pushed into the pre function (trampoline_probe_handler), and then we skip single stepping the original function. In this case the original instruction to be single stepped was just a NOP, and we can do without the extra interruption. The new flow of events to having a return probe handler execute when a target function exits is: * At system initialization time, a kprobe is inserted at the beginning of kretprobe_trampoline. kernel/kprobes.c use to handle this on it's own, but ia64 needed to do this a little differently (i.e. a function pointer is really a pointer to a structure containing the instruction pointer and a global pointer), so I added the notion of arch_init(), so that kernel/kprobes.c:init_kprobes() now allows architecture specific initialization by calling arch_init() before exiting. Each architecture now registers a kprobe on it's own trampoline function. * register_kretprobe() will insert a kprobe at the beginning of the targeted function with the kprobe pre_handler set to arch_prepare_kretprobe (still no change) * When the target function is entered, the kprobe is fired, calling arch_prepare_kretprobe (still no change) * In arch_prepare_kretprobe() we try to get a free instance and if one is available then we fill out the instance with a pointer to the return probe, the original return address, and a pointer to the task structure (instead of the stack address.) Just like before we change the return address to the trampoline function and mark the instance as used. If multiple return probes are registered for a given target function, then arch_prepare_kretprobe() will get called multiple times for the same task (since our kprobe implementation is able to handle multiple kprobes at the same address.) Past the first call to arch_prepare_kretprobe, we end up with the original address stored in the return probe instance pointing to our trampoline function. (This is a significant difference from the original arch_prepare_kretprobe design.) * Target function executes like normal and then returns to kretprobe_trampoline. * kprobe inserted on the first instruction of kretprobe_trampoline is fired and calls trampoline_probe_handler() (no change here) * trampoline_probe_handler() consumes each of the instances associated with the current task by calling the registered handler function and marking the instance as unused until an instance is found that has a return address different then the trampoline function. (change similar to my previous ia64 RFC) * If the task is killed with some left-over return probe instances (meaning that a target function was entered, but never returned), then we just free any instances associated with the task. (Not much different other then we can handle this without calling architecture specific functions.) There is a known problem that this patch does not yet solve where registering a return probe flush_old_exec or flush_thread will put us in a bad state. Most likely the best way to handle this is to not allow registering return probes on these two functions. (Significant change) This patch series applies to the 2.6.12-rc6-mm1 kernel, and provides: * kernel/kprobes.c changes * i386 patch of existing return probes implementation * x86_64 patch of existing return probe implementation * ia64 implementation * ppc64 implementation (provided by Ananth) This patch implements the architecture independant changes for a reworking of the kprobes based function return probes design. Changes include: * Removing functions for querying a return probe instance off a stack address * Removing the stack_addr field from the kretprobe_instance definition, and adding a task pointer * Adding architecture specific initialization via arch_init() * Removing extern definitions for the architecture trampoline functions (this isn't needed anymore since the architecture handles the initialization of the kprobe in the return probe trampoline function.) Signed-off-by: Rusty Lynch Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kprobes.h | 28 +++------------------------- 1 file changed, 3 insertions(+), 25 deletions(-) (limited to 'include/linux') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index d304d4579856..b7a194c4362a 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -104,33 +104,12 @@ struct jprobe { }; #ifdef ARCH_SUPPORTS_KRETPROBES -extern int trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs); -extern void trampoline_post_handler(struct kprobe *p, struct pt_regs *regs, - unsigned long flags); -extern struct task_struct *arch_get_kprobe_task(void *ptr); extern void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs); -extern void arch_kprobe_flush_task(struct task_struct *tk); #else /* ARCH_SUPPORTS_KRETPROBES */ -static inline void kretprobe_trampoline(void) -{ -} -static inline int trampoline_probe_handler(struct kprobe *p, - struct pt_regs *regs) -{ - return 0; -} -static inline void trampoline_post_handler(struct kprobe *p, - struct pt_regs *regs, unsigned long flags) -{ -} static inline void arch_prepare_kretprobe(struct kretprobe *rp, struct pt_regs *regs) { } -static inline void arch_kprobe_flush_task(struct task_struct *tk) -{ -} -#define arch_get_kprobe_task(ptr) ((struct task_struct *)NULL) #endif /* ARCH_SUPPORTS_KRETPROBES */ /* * Function-return probe - @@ -155,8 +134,8 @@ struct kretprobe_instance { struct hlist_node uflist; /* either on free list or used list */ struct hlist_node hlist; struct kretprobe *rp; - void *ret_addr; - void *stack_addr; + kprobe_opcode_t *ret_addr; + struct task_struct *task; }; #ifdef CONFIG_KPROBES @@ -176,6 +155,7 @@ extern void arch_copy_kprobe(struct kprobe *p); extern void arch_arm_kprobe(struct kprobe *p); extern void arch_disarm_kprobe(struct kprobe *p); extern void arch_remove_kprobe(struct kprobe *p); +extern int arch_init(void); extern void show_registers(struct pt_regs *regs); extern kprobe_opcode_t *get_insn_slot(void); extern void free_insn_slot(kprobe_opcode_t *slot); @@ -196,8 +176,6 @@ int register_kretprobe(struct kretprobe *rp); void unregister_kretprobe(struct kretprobe *rp); struct kretprobe_instance *get_free_rp_inst(struct kretprobe *rp); -struct kretprobe_instance *get_rp_inst(void *sara); -struct kretprobe_instance *get_rp_inst_tsk(struct task_struct *tk); void add_rp_inst(struct kretprobe_instance *ri); void kprobe_flush_task(struct task_struct *tk); void recycle_rp_inst(struct kretprobe_instance *ri); -- cgit v1.2.1 From da9091ee3b5f9808c64abb925cefe7b100018614 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 27 Jun 2005 15:24:30 -0700 Subject: [PATCH] ide: it8212 backport for Bartlomiej IDE This lets you throw out the iteraid stuff that has ended up back in due to stupid goings on in the IDE world. Its the same heavily tested code shipped in Fedora/Red Hat products but without the other dependancies on the Bartlomiej IDE layer. Pre-requisite: the ide-disk patch I sent to handle pure LBA devices. Obviously you lose things like hot unplug with the Bartlomiej IDE layer at the moment but that won't matter to most users. The patch does the following - Add IT8211/12 to pci_ids.h - Add Makefile/Kconfig entry - Add it8212 driver No core IDE code is touched by this diff Embedded system testing and the ability to force raid mode off by David Howells Made possible by the ite reference code, documentation and also several clarifications and pieces of assistance provided by ITE themselves Signed-off-by: Alan Cox Acked-by: Bartlomiej Zolnierkiewicz Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pci_ids.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index a3961e1d5183..1e0bc6a8d653 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1815,6 +1815,8 @@ #define PCI_VENDOR_ID_ITE 0x1283 #define PCI_DEVICE_ID_ITE_IT8172G 0x8172 #define PCI_DEVICE_ID_ITE_IT8172G_AUDIO 0x0801 +#define PCI_DEVICE_ID_ITE_8211 0x8211 +#define PCI_DEVICE_ID_ITE_8212 0x8212 #define PCI_DEVICE_ID_ITE_8872 0x8872 #define PCI_DEVICE_ID_ITE_IT8330G_0 0xe886 -- cgit v1.2.1 From 1ad275e3e7d253d44f03868e85977c908e334fed Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Mon, 27 Jun 2005 16:28:06 -0700 Subject: [PATCH] pcmcia: device and driver matching The actual matching of pcmcia drivers and pcmcia devices. The original version of this was written by David Woodhouse. Signed-off-by: Dominik Brodowski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mod_devicetable.h | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h index d6eb7b2efc04..e9651cd8310c 100644 --- a/include/linux/mod_devicetable.h +++ b/include/linux/mod_devicetable.h @@ -175,4 +175,37 @@ struct serio_device_id { }; +/* PCMCIA */ + +struct pcmcia_device_id { + __u16 match_flags; + + __u16 manf_id; + __u16 card_id; + + __u8 func_id; + + /* for real multi-function devices */ + __u8 function; + + /* for pseude multi-function devices */ + __u8 device_no; + + const char * prod_id[4]; + __u32 prod_id_hash[4]; + + /* not matched against */ + kernel_ulong_t driver_info; +}; + +#define PCMCIA_DEV_ID_MATCH_MANF_ID 0x0001 +#define PCMCIA_DEV_ID_MATCH_CARD_ID 0x0002 +#define PCMCIA_DEV_ID_MATCH_FUNC_ID 0x0004 +#define PCMCIA_DEV_ID_MATCH_FUNCTION 0x0008 +#define PCMCIA_DEV_ID_MATCH_PROD_ID1 0x0010 +#define PCMCIA_DEV_ID_MATCH_PROD_ID2 0x0020 +#define PCMCIA_DEV_ID_MATCH_PROD_ID3 0x0040 +#define PCMCIA_DEV_ID_MATCH_PROD_ID4 0x0080 +#define PCMCIA_DEV_ID_MATCH_DEVICE_NO 0x0100 + #endif /* LINUX_MOD_DEVICETABLE_H */ -- cgit v1.2.1 From ea7b38825bba66a81745a706da70a1c81adc95bd Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Mon, 27 Jun 2005 16:28:07 -0700 Subject: [PATCH] pcmcia: match for fake CIS Add another match flag for devices needing a CIS override. The driver will only probe/attach if the CIS has been replaced before. Signed-off-by: Dominik Brodowski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mod_devicetable.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h index e9651cd8310c..c0106d68bb6d 100644 --- a/include/linux/mod_devicetable.h +++ b/include/linux/mod_devicetable.h @@ -196,6 +196,7 @@ struct pcmcia_device_id { /* not matched against */ kernel_ulong_t driver_info; + char * cisfile; }; #define PCMCIA_DEV_ID_MATCH_MANF_ID 0x0001 @@ -207,5 +208,6 @@ struct pcmcia_device_id { #define PCMCIA_DEV_ID_MATCH_PROD_ID3 0x0040 #define PCMCIA_DEV_ID_MATCH_PROD_ID4 0x0080 #define PCMCIA_DEV_ID_MATCH_DEVICE_NO 0x0100 +#define PCMCIA_DEV_ID_MATCH_FAKE_CIS 0x0200 #endif /* LINUX_MOD_DEVICETABLE_H */ -- cgit v1.2.1 From f602ff7eb4e44e7245bfeeba4d078144703fcd76 Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Mon, 27 Jun 2005 16:28:09 -0700 Subject: [PATCH] pcmcia: match "anonymous" cards If a card doesn't provide _any_ information about itself, assume it is a so-called "anonymous" card. pcmciamtd will bind to it if it is configured to do so. Signed-off-by: Dominik Brodowski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mod_devicetable.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h index c0106d68bb6d..8a8dc82a941d 100644 --- a/include/linux/mod_devicetable.h +++ b/include/linux/mod_devicetable.h @@ -209,5 +209,6 @@ struct pcmcia_device_id { #define PCMCIA_DEV_ID_MATCH_PROD_ID4 0x0080 #define PCMCIA_DEV_ID_MATCH_DEVICE_NO 0x0100 #define PCMCIA_DEV_ID_MATCH_FAKE_CIS 0x0200 +#define PCMCIA_DEV_ID_MATCH_ANONYMOUS 0x0400 #endif /* LINUX_MOD_DEVICETABLE_H */ -- cgit v1.2.1 From aecab27aeabaa897d69fc082686df314329830de Mon Sep 17 00:00:00 2001 From: Dominik Brodowski Date: Mon, 27 Jun 2005 16:28:56 -0700 Subject: [PATCH] pcmcia: mod_devicetable.h fix for different sizes in kernel- and userspace The size of pointers may differ between (userspace) modpost and (kernelspace) modules -- so fix mod_devicetable.h to reflect this possibility. Signed-off-by: Dominik Brodowski Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mod_devicetable.h | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h index 8a8dc82a941d..9b6d05172ed4 100644 --- a/include/linux/mod_devicetable.h +++ b/include/linux/mod_devicetable.h @@ -191,12 +191,22 @@ struct pcmcia_device_id { /* for pseude multi-function devices */ __u8 device_no; - const char * prod_id[4]; __u32 prod_id_hash[4]; + /* not matched against in kernelspace*/ +#ifdef __KERNEL__ + const char * prod_id[4]; +#else + kernel_ulong_t prod_id[4]; +#endif + /* not matched against */ kernel_ulong_t driver_info; +#ifdef __KERNEL__ char * cisfile; +#else + kernel_ulong_t cisfile; +#endif }; #define PCMCIA_DEV_ID_MATCH_MANF_ID 0x0001 -- cgit v1.2.1 From a5fe736eaf9bae1b45317313de04b564441b94f2 Mon Sep 17 00:00:00 2001 From: Jeff Garzik Date: Mon, 27 Jun 2005 22:47:18 -0400 Subject: Update is_multicast_ether_addr() definition; net/ieee80211.h cleanups. --- include/linux/etherdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index a1478258d002..8a2df4dfbc59 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -65,7 +65,7 @@ static inline int is_zero_ether_addr(const u8 *addr) */ static inline int is_multicast_ether_addr(const u8 *addr) { - return addr[0] & 0x01; + return ((addr[0] != 0xff) && (0x01 & addr[0])); } /** -- cgit v1.2.1 From c431ada45d65b305a6aab4557067e564b23ce5a5 Mon Sep 17 00:00:00 2001 From: Rajesh Shah Date: Thu, 28 Apr 2005 00:25:45 -0700 Subject: [PATCH] acpi bridge hotadd: ACPI based root bridge hot-add When you hot-plug a (root) bridge hierarchy, it may have p2p bridges and devices attached to it that have not been configured by firmware. In this case, we need to configure the devices before starting them. This patch separates device start from device scan so that we can introduce the configuration step in the middle. I kept the existing semantics for pci_scan_bus() since there are a huge number of callers to that function. Also, I have no way of testing the changes I made to the parisc files, so this needs review by those folks. Sorry for the massive cross-post, this touches files in many different places. Signed-off-by: Rajesh Shah Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- include/linux/pci.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index b5238bd18830..0e9844929fe3 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -734,16 +734,20 @@ void pcibios_update_irq(struct pci_dev *, int irq); /* Generic PCI functions used internally */ extern struct pci_bus *pci_find_bus(int domain, int busnr); +void pci_bus_add_devices(struct pci_bus *bus); struct pci_bus *pci_scan_bus_parented(struct device *parent, int bus, struct pci_ops *ops, void *sysdata); static inline struct pci_bus *pci_scan_bus(int bus, struct pci_ops *ops, void *sysdata) { - return pci_scan_bus_parented(NULL, bus, ops, sysdata); + struct pci_bus *root_bus; + root_bus = pci_scan_bus_parented(NULL, bus, ops, sysdata); + if (root_bus) + pci_bus_add_devices(root_bus); + return root_bus; } int pci_scan_slot(struct pci_bus *bus, int devfn); struct pci_dev * pci_scan_single_device(struct pci_bus *bus, int devfn); unsigned int pci_scan_child_bus(struct pci_bus *bus); void pci_bus_add_device(struct pci_dev *dev); -void pci_bus_add_devices(struct pci_bus *bus); void pci_name_device(struct pci_dev *dev); char *pci_class_name(u32 class); void pci_read_bridge_bases(struct pci_bus *child); -- cgit v1.2.1 From b1bb248a5d2230a3d8ef42199c742194a8580b15 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Thu, 28 Apr 2005 00:25:58 -0700 Subject: [PATCH] ACPI based I/O APIC hot-plug: add interfaces This patch adds the following new interfaces for I/O xAPIC hotplug. The implementation of these interfaces depends on each architecture. o int acpi_register_ioapic(acpi_handle handle, u64 phys_addr, u32 gsi_base); This new interface is to add a new I/O xAPIC specified by phys_addr and gsi_base pair. phys_addr is the physical address to which the I/O xAPIC is mapped and gsi_base is global system interrupt base of the I/O xAPIC. acpi_register_ioapic returns 0 on success, or negative value on error. o int acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base); This new interface is to remove a I/O xAPIC specified by gsi_base. acpi_unregister_ioapic returns 0 on success, or negative value on error. Signed-off-by: Kenji Kaneshige Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- include/linux/acpi.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index b123cc08773d..f5bc298707e1 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -407,6 +407,9 @@ int acpi_map_lsapic(acpi_handle handle, int *pcpu); int acpi_unmap_lsapic(int cpu); #endif /* CONFIG_ACPI_HOTPLUG_CPU */ +int acpi_register_ioapic(acpi_handle handle, u64 phys_addr, u32 gsi_base); +int acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base); + extern int acpi_mp_config; extern u32 pci_mmcfg_base_addr; -- cgit v1.2.1 From a0d399a808916d22c1c222c6b5ca4e8edd6d91a9 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Thu, 28 Apr 2005 00:25:59 -0700 Subject: [PATCH] ACPI based I/O APIC hot-plug: acpiphp support This patch adds PCI based I/O xAPIC hot-add support to ACPIPHP driver. When PCI root bridge is hot-added, all PCI based I/O xAPICs under the root bridge are hot-added by this patch. Hot-remove support is TBD. Signed-off-by: Kenji Kaneshige Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- include/linux/pci_ids.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index bf608808a60c..810bbbcee404 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -62,6 +62,8 @@ #define PCI_BASE_CLASS_SYSTEM 0x08 #define PCI_CLASS_SYSTEM_PIC 0x0800 +#define PCI_CLASS_SYSTEM_PIC_IOAPIC 0x080010 +#define PCI_CLASS_SYSTEM_PIC_IOXAPIC 0x080020 #define PCI_CLASS_SYSTEM_DMA 0x0801 #define PCI_CLASS_SYSTEM_TIMER 0x0802 #define PCI_CLASS_SYSTEM_RTC 0x0803 -- cgit v1.2.1 From 2311b1f2bbd36fa5f366a7448c718b2556e0f02c Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Fri, 13 May 2005 17:44:10 +1000 Subject: [PATCH] PCI: fix-pci-mmap-on-ppc-and-ppc64.patch This is an updated version of Ben's fix-pci-mmap-on-ppc-and-ppc64.patch which is in 2.6.12-rc4-mm1. It fixes the patch to work on PPC iSeries, removes some debug printks at Ben's request, and incorporates your fix-pci-mmap-on-ppc-and-ppc64-fix.patch also. Originally from Benjamin Herrenschmidt This patch was discussed at length on linux-pci and so far, the last iteration of it didn't raise any comment. It's effect is a nop on architecture that don't define the new pci_resource_to_user() callback anyway. It allows architecture like ppc who put weird things inside of PCI resource structures to convert to some different value for user visible ones. It also fixes mmap'ing of IO space on those archs. Signed-off-by: Benjamin Herrenschmidt Signed-off-by: Michael Ellerman Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- include/linux/pci.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 0e9844929fe3..cfa1455848f4 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1020,6 +1020,20 @@ static inline char *pci_name(struct pci_dev *pdev) #define pci_pretty_name(dev) "" #endif + +/* Some archs don't want to expose struct resource to userland as-is + * in sysfs and /proc + */ +#ifndef HAVE_ARCH_PCI_RESOURCE_TO_USER +static inline void pci_resource_to_user(const struct pci_dev *dev, int bar, + const struct resource *rsrc, u64 *start, u64 *end) +{ + *start = rsrc->start; + *end = rsrc->end; +} +#endif /* HAVE_ARCH_PCI_RESOURCE_TO_USER */ + + /* * The world is not perfect and supplies us with broken PCI devices. * For at least a part of these bugs we need a work-around, so both -- cgit v1.2.1 From e24c2d963a604d9eaa560c90371fa387d3eec8f1 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 2 Jun 2005 12:55:50 -0700 Subject: [PATCH] PCI: DMA bursting advice After seeing, at best, "guesses" as to the following kind of information in several drivers, I decided that we really need a way for platforms to specifically give advice in this area for what works best with their PCI controller implementation. Basically, this new interface gives DMA bursting advice on PCI. There are three forms of the advice: 1) Burst as much as possible, it is not necessary to end bursts on some particular boundary for best performance. 2) Burst on some byte count multiple. A DMA burst to some multiple of number of bytes may be done, but it is important to end the burst on an exact multiple for best performance. The best example of this I am aware of are the PPC64 PCI controllers, where if you end a burst mid-cacheline then chip has to refetch the data and the IOMMU translations which hurts performance a lot. 3) Burst on a single byte count multiple. Bursts shall end exactly on the next multiple boundary for best performance. Sparc64 and Alpha's PCI controllers operate this way. They disconnect any device which tries to burst across a cacheline boundary. Actually, newer sparc64 PCI controllers do not have this behavior. That is why the "pdev" is passed into the interface, so I can add code later to check which PCI controller the system is using and give advice accordingly. Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- include/linux/pci.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index cfa1455848f4..9ce4f1be093f 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -874,6 +874,15 @@ int pci_scan_bridge(struct pci_bus *bus, struct pci_dev * dev, int max, int pass #define pci_pool_alloc(pool, flags, handle) dma_pool_alloc(pool, flags, handle) #define pci_pool_free(pool, vaddr, addr) dma_pool_free(pool, vaddr, addr) +enum pci_dma_burst_strategy { + PCI_DMA_BURST_INFINITY, /* make bursts as large as possible, + strategy_parameter is N/A */ + PCI_DMA_BURST_BOUNDARY, /* disconnect at every strategy_parameter + byte boundaries */ + PCI_DMA_BURST_MULTIPLE, /* disconnect at some multiple of + strategy_parameter byte boundaries */ +}; + #if defined(CONFIG_ISA) || defined(CONFIG_EISA) extern struct pci_dev *isa_bridge; #endif -- cgit v1.2.1 From bb4a61b6eaee01707f24deeefc5d7136f25f75c5 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 6 Jun 2005 23:07:46 -0700 Subject: [PATCH] PCI: fix up errors after dma bursting patch and CONFIG_PCI=n With CONFIG_PCI=n: In file included from include/linux/pci.h:917, from lib/iomap.c:6: include/asm/pci.h:104: warning: `enum pci_dma_burst_strategy' declared inside parameter list include/asm/pci.h:104: warning: its scope is only this definition or declaration, which is probably not what you want. include/asm/pci.h: In function `pci_dma_burst_advice': include/asm/pci.h:106: dereferencing pointer to incomplete type include/asm/pci.h:106: `PCI_DMA_BURST_INFINITY' undeclared (first use in this function) include/asm/pci.h:106: (Each undeclared identifier is reported only once include/asm/pci.h:106: for each function it appears in.) make[1]: *** [lib/iomap.o] Error 1 Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- include/linux/pci.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 9ce4f1be093f..66798b46f308 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -985,6 +985,8 @@ static inline int pci_proc_domain(struct pci_bus *bus) } #endif +#define pci_dma_burst_advice(pdev, strat, strategy_parameter) do { } while (0) + #endif /* !CONFIG_PCI */ /* these helpers provide future and backwards compatibility -- cgit v1.2.1 From 545493917dc90298e1c38f018ad893f5518928e7 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 23 Jun 2005 17:35:56 -0700 Subject: [PATCH] PCI: add proper MCFG table parsing to ACPI core. This patch is the first step in properly handling the MCFG PCI table. It defines the structures properly, and saves off the table so that the pci mmconfig code can access it. It moves the parsing of the table a little later in the boot process, but still before the information is needed. Signed-off-by: Greg Kroah-Hartman --- include/linux/acpi.h | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index f5bc298707e1..ef8483673aa3 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -342,11 +342,19 @@ struct acpi_table_ecdt { /* PCI MMCONFIG */ +/* Defined in PCI Firmware Specification 3.0 */ +struct acpi_table_mcfg_config { + u32 base_address; + u32 base_reserved; + u16 pci_segment_group_number; + u8 start_bus_number; + u8 end_bus_number; + u8 reserved[4]; +} __attribute__ ((packed)); struct acpi_table_mcfg { struct acpi_table_header header; u8 reserved[8]; - u32 base_address; - u32 base_reserved; + struct acpi_table_mcfg_config config[0]; } __attribute__ ((packed)); /* Table Handlers */ @@ -391,6 +399,7 @@ int acpi_table_parse (enum acpi_table_id id, acpi_table_handler handler); int acpi_get_table_header_early (enum acpi_table_id id, struct acpi_table_header **header); int acpi_table_parse_madt (enum acpi_madt_entry_id id, acpi_madt_entry_handler handler, unsigned int max_entries); int acpi_table_parse_srat (enum acpi_srat_entry_id id, acpi_madt_entry_handler handler, unsigned int max_entries); +int acpi_parse_mcfg (unsigned long phys_addr, unsigned long size); void acpi_table_print (struct acpi_table_header *header, unsigned long phys_addr); void acpi_table_print_madt_entry (acpi_table_entry_header *madt); void acpi_table_print_srat_entry (acpi_table_entry_header *srat); @@ -412,7 +421,8 @@ int acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base); extern int acpi_mp_config; -extern u32 pci_mmcfg_base_addr; +extern struct acpi_table_mcfg_config *pci_mmcfg_config; +extern int pci_mmcfg_config_num; extern int sbf_port ; -- cgit v1.2.1 From b3563c4fbff906991a1b4ef4609f99cca2a0de6a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 28 Jun 2005 12:54:43 -0700 Subject: [NETLINK]: Clear padding in netlink messages Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 + include/linux/rtnetlink.h | 5 ++++- 2 files changed, 5 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 3029cad63a01..27e4d164a108 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -168,6 +168,7 @@ __nlmsg_put(struct sk_buff *skb, u32 pid, u32 seq, int type, int len, int flags) nlh->nlmsg_flags = flags; nlh->nlmsg_pid = pid; nlh->nlmsg_seq = seq; + memset(NLMSG_DATA(nlh) + len, 0, NLMSG_ALIGN(size) - size); return nlh; } diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index d021888b58f1..dc26e82ba0fd 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -898,7 +898,9 @@ extern void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const voi memcpy(skb_put(skb, attrlen), data, attrlen); }) #define RTA_PUT_NOHDR(skb, attrlen, data) \ - RTA_APPEND(skb, RTA_ALIGN(attrlen), data) +({ RTA_APPEND(skb, RTA_ALIGN(attrlen), data); \ + memset(skb->tail - (RTA_ALIGN(attrlen) - attrlen), 0, \ + RTA_ALIGN(attrlen) - attrlen); }) #define RTA_PUT_U8(skb, attrtype, value) \ ({ u8 _tmp = (value); \ @@ -978,6 +980,7 @@ __rta_reserve(struct sk_buff *skb, int attrtype, int attrlen) rta = (struct rtattr*)skb_put(skb, RTA_ALIGN(size)); rta->rta_type = attrtype; rta->rta_len = size; + memset(RTA_DATA(rta) + attrlen, 0, RTA_ALIGN(size) - size); return rta; } -- cgit v1.2.1 From 8a47077a0b5aa2649751c46e7a27884e6686ccbf Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 28 Jun 2005 12:56:45 -0700 Subject: [NETLINK]: Missing padding fields in dumped structures Plug holes with padding fields and initialized them to zero. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/pkt_cls.h | 1 + include/linux/pkt_sched.h | 9 ++++++--- include/linux/rtnetlink.h | 5 +++++ 3 files changed, 12 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h index 25d2d67c1faf..bd2c5a2bbbf5 100644 --- a/include/linux/pkt_cls.h +++ b/include/linux/pkt_cls.h @@ -276,6 +276,7 @@ struct tc_rsvp_pinfo __u8 protocol; __u8 tunnelid; __u8 tunnelhdr; + __u8 pad; }; /* ROUTE filter */ diff --git a/include/linux/pkt_sched.h b/include/linux/pkt_sched.h index 1d9da36eb9db..60ffcb9c5791 100644 --- a/include/linux/pkt_sched.h +++ b/include/linux/pkt_sched.h @@ -221,9 +221,11 @@ struct tc_gred_qopt /* gred setup */ struct tc_gred_sopt { - __u32 DPs; - __u32 def_DP; - __u8 grio; + __u32 DPs; + __u32 def_DP; + __u8 grio; + __u8 pad1; + __u16 pad2; }; /* HTB section */ @@ -351,6 +353,7 @@ struct tc_cbq_ovl #define TC_CBQ_OVL_DROP 3 #define TC_CBQ_OVL_RCLASSIC 4 unsigned char priority2; + __u16 pad; __u32 penalty; }; diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index dc26e82ba0fd..657c05ab8f9e 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -363,6 +363,8 @@ enum struct rta_session { __u8 proto; + __u8 pad1; + __u16 pad2; union { struct { @@ -635,10 +637,13 @@ struct ifinfomsg struct prefixmsg { unsigned char prefix_family; + unsigned char prefix_pad1; + unsigned short prefix_pad2; int prefix_ifindex; unsigned char prefix_type; unsigned char prefix_len; unsigned char prefix_flags; + unsigned char prefix_pad3; }; enum -- cgit v1.2.1 From 2f85a42964dd43fed3a339701db046bee5a8b903 Mon Sep 17 00:00:00 2001 From: Vlad Yasevich Date: Tue, 28 Jun 2005 13:24:23 -0700 Subject: [SCTP] Make init & delayed sack timeouts configurable by user. Signed-off-by: Vlad Yasevich Signed-off-by: Sridhar Samudrala Signed-off-by: David S. Miller --- include/linux/sysctl.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index ebfe1250f0a4..5b5f434ac9a0 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -641,6 +641,7 @@ enum { NET_SCTP_ADDIP_ENABLE = 13, NET_SCTP_PRSCTP_ENABLE = 14, NET_SCTP_SNDBUF_POLICY = 15, + NET_SCTP_SACK_TIMEOUT = 16, }; /* /proc/sys/net/bridge */ -- cgit v1.2.1 From 7fe40f73d7591b38f129fe6a9c0fa46e0b192d09 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 28 Jun 2005 15:46:24 -0700 Subject: [IPV6]: remove more unused IPV6_AUTHHDR things. Remove two more unused IPV6_AUTHHDR option things, which I failed to remove them last time, plus, mark IPV6_AUTHHDR obsolete. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- include/linux/in6.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/in6.h b/include/linux/in6.h index f8256c582845..dcf5720ffcbb 100644 --- a/include/linux/in6.h +++ b/include/linux/in6.h @@ -156,7 +156,7 @@ struct in6_flowlabel_req #define IPV6_CHECKSUM 7 #define IPV6_HOPLIMIT 8 #define IPV6_NEXTHOP 9 -#define IPV6_AUTHHDR 10 +#define IPV6_AUTHHDR 10 /* obsolete */ #define IPV6_FLOWINFO 11 #define IPV6_UNICAST_HOPS 16 -- cgit v1.2.1 From bcd61272db5e643b6d9c01c9d5085b914d9f19df Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 28 Jun 2005 15:58:50 -0700 Subject: [NET]: Add missing include to linux/netdevice.h linux/etherdevice.h can't be included standalone at the moment, which is required in order to sort the header files in the recommended alphabetic order. This patch fixes that and is needed to build spider_net. Signed-off-by: Arnd Bergmann Signed-off-by: David S. Miller --- include/linux/etherdevice.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index 8a2df4dfbc59..cf3847edc50f 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -25,6 +25,7 @@ #define _LINUX_ETHERDEVICE_H #include +#include #include #ifdef __KERNEL__ -- cgit v1.2.1 From 05133fc498e788e1c1ca4e906f9e05d9779fd63b Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 28 Jun 2005 20:44:54 -0700 Subject: [PATCH] swabb.h warning fixes In file included from drivers/media/dvb/ttpci/av7110_hw.c:38: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type In file included from drivers/media/dvb/ttpci/av7110_v4l.c:36: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type In file included from drivers/media/dvb/ttpci/av7110_av.c:37: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type drivers/isdn/icn/icn.c:719:4: warning: #warning TODO test headroom or use skb->nb to flag ACK In file included from drivers/media/dvb/ttpci/av7110_ca.c:39: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type In file included from drivers/media/dvb/ttpci/av7110.c:41: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type Does declaring a function to return a const value actually mean something to gcc? Dunno. Kill it and replace sone `__inline__'s with `inline' too. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/byteorder/swabb.h | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/byteorder/swabb.h b/include/linux/byteorder/swabb.h index d28d9a804d3b..d5f2a3205109 100644 --- a/include/linux/byteorder/swabb.h +++ b/include/linux/byteorder/swabb.h @@ -92,29 +92,32 @@ #endif /* OPTIMIZE */ -static __inline__ __const__ __u32 __fswahw32(__u32 x) +static inline __u32 __fswahw32(__u32 x) { return __arch__swahw32(x); } -static __inline__ __u32 __swahw32p(__u32 *x) + +static inline __u32 __swahw32p(__u32 *x) { return __arch__swahw32p(x); } -static __inline__ void __swahw32s(__u32 *addr) + +static inline void __swahw32s(__u32 *addr) { __arch__swahw32s(addr); } - -static __inline__ __const__ __u32 __fswahb32(__u32 x) +static inline __u32 __fswahb32(__u32 x) { return __arch__swahb32(x); } -static __inline__ __u32 __swahb32p(__u32 *x) + +static inline __u32 __swahb32p(__u32 *x) { return __arch__swahb32p(x); } -static __inline__ void __swahb32s(__u32 *addr) + +static inline void __swahb32s(__u32 *addr) { __arch__swahb32s(addr); } -- cgit v1.2.1 From 687a21cee17000177b1935896b9b475acf136678 Mon Sep 17 00:00:00 2001 From: Pekka J Enberg Date: Tue, 28 Jun 2005 20:44:55 -0700 Subject: [PATCH] rename wakeup_bdflush to wakeup_pdflush Signed-off-by: Pekka Enberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/writeback.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/writeback.h b/include/linux/writeback.h index d5c3fe1bf33d..542dbaee6512 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -85,7 +85,7 @@ static inline void wait_on_inode(struct inode *inode) /* * mm/page-writeback.c */ -int wakeup_bdflush(long nr_pages); +int wakeup_pdflush(long nr_pages); void laptop_io_completion(void); void laptop_sync_completion(void); void throttle_vm_writeout(void); -- cgit v1.2.1 From 334a13ec3d01a1a4b4f2249735b793105cb4a519 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 28 Jun 2005 20:44:58 -0700 Subject: [PATCH] really remove xattr_acl.h Looks like it sneaked back with the NFS ACL merge.. Signed-off-by: Christoph Hellwig Cc: Trond Myklebust Cc: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/xattr_acl.h | 50 ----------------------------------------------- 1 file changed, 50 deletions(-) delete mode 100644 include/linux/xattr_acl.h (limited to 'include/linux') diff --git a/include/linux/xattr_acl.h b/include/linux/xattr_acl.h deleted file mode 100644 index 7a1f9b93a45f..000000000000 --- a/include/linux/xattr_acl.h +++ /dev/null @@ -1,50 +0,0 @@ -/* - File: linux/xattr_acl.h - - (extended attribute representation of access control lists) - - (C) 2000 Andreas Gruenbacher, -*/ - -#ifndef _LINUX_XATTR_ACL_H -#define _LINUX_XATTR_ACL_H - -#include - -#define XATTR_NAME_ACL_ACCESS "system.posix_acl_access" -#define XATTR_NAME_ACL_DEFAULT "system.posix_acl_default" - -#define XATTR_ACL_VERSION 0x0002 - -typedef struct { - __u16 e_tag; - __u16 e_perm; - __u32 e_id; -} xattr_acl_entry; - -typedef struct { - __u32 a_version; - xattr_acl_entry a_entries[0]; -} xattr_acl_header; - -static inline size_t xattr_acl_size(int count) -{ - return sizeof(xattr_acl_header) + count * sizeof(xattr_acl_entry); -} - -static inline int xattr_acl_count(size_t size) -{ - if (size < sizeof(xattr_acl_header)) - return -1; - size -= sizeof(xattr_acl_header); - if (size % sizeof(xattr_acl_entry)) - return -1; - return size / sizeof(xattr_acl_entry); -} - -struct posix_acl * posix_acl_from_xattr(const void *value, size_t size); -int posix_acl_to_xattr(const struct posix_acl *acl, void *buffer, size_t size); - - - -#endif /* _LINUX_XATTR_ACL_H */ -- cgit v1.2.1 From 3607d1dfc80dcfbd3a6f236c70aa0d8eb7292278 Mon Sep 17 00:00:00 2001 From: GOTO Masanori Date: Tue, 28 Jun 2005 20:45:04 -0700 Subject: [PATCH] headers: include linux/compiler.h for __user This patch lets i2c-dev.h include linux/compiler.h so that __user is defined. Signed-off-by: GOTO Masanori Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/i2c-dev.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/i2c-dev.h b/include/linux/i2c-dev.h index d228230ffe5d..541695679762 100644 --- a/include/linux/i2c-dev.h +++ b/include/linux/i2c-dev.h @@ -25,6 +25,7 @@ #define _LINUX_I2C_DEV_H #include +#include /* Some IOCTL commands are defined in */ /* Note: 10-bit addresses are NOT supported! */ -- cgit v1.2.1 From 4cceb4d13abaedbd52e54053367c793ed4aedb6b Mon Sep 17 00:00:00 2001 From: GOTO Masanori Date: Tue, 28 Jun 2005 20:45:05 -0700 Subject: [PATCH] headers: include linux/types.h for usb_ch9.h This patch for usb_ch9.h includes linux/types.h instead of asm/types.h so that __le16 and so on is explicitly defined. It also cleans up non standard // comment. Signed-off-by: GOTO Masanori Cc: Greg KH Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/usb_ch9.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/usb_ch9.h b/include/linux/usb_ch9.h index 39e7ff4ffd28..ee21e6bf3867 100644 --- a/include/linux/usb_ch9.h +++ b/include/linux/usb_ch9.h @@ -19,7 +19,7 @@ #ifndef __LINUX_USB_CH9_H #define __LINUX_USB_CH9_H -#include /* __u8 etc */ +#include /* __u8 etc */ /*-------------------------------------------------------------------------*/ @@ -294,8 +294,8 @@ struct usb_endpoint_descriptor { __le16 wMaxPacketSize; __u8 bInterval; - // NOTE: these two are _only_ in audio endpoints. - // use USB_DT_ENDPOINT*_SIZE in bLength, not sizeof. + /* NOTE: these two are _only_ in audio endpoints. */ + /* use USB_DT_ENDPOINT*_SIZE in bLength, not sizeof. */ __u8 bRefresh; __u8 bSynchAddress; } __attribute__ ((packed)); -- cgit v1.2.1 From fb3cc4320e1fd87143683b540e459a2e20fdc9bb Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Tue, 28 Jun 2005 20:45:15 -0700 Subject: [PATCH] blk: light iocontext ops get_io_context needlessly turned off interrupts and checked for racing io context creations. Both of which aren't needed, because the io context can only be created while in process context of the current process. Also, split the function in 2. A light version, current_io_context does not elevate the reference count specifically, but can be used when in process context, because the process holds a reference itself. Signed-off-by: Nick Piggin Cc: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/blkdev.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 21a8674cd149..0881b5cdee3d 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -96,6 +96,7 @@ struct io_context { void put_io_context(struct io_context *ioc); void exit_io_context(void); +struct io_context *current_io_context(int gfp_flags); struct io_context *get_io_context(int gfp_flags); void copy_io_context(struct io_context **pdst, struct io_context **psrc); void swap_io_context(struct io_context **ioc1, struct io_context **ioc2); -- cgit v1.2.1 From 200803dfe4ff772740d63db725ab2f1b185ccf92 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Tue, 28 Jun 2005 20:45:18 -0700 Subject: [PATCH] irqpoll Anyone reporting a stuck IRQ should try these options. Its effectiveness varies we've found in the Fedora case. Quite a few systems with misdescribed IRQ routing just work when you use irqpoll. It also fixes up the VIA systems although thats now fixed with the VIA quirk (which we could just make default as its what Redmond OS does but Linus didn't like it historically). A small number of systems have jammed IRQ sources or misdescribes that cause an IRQ that we have no handler registered anywhere for. In those cases it doesn't help. Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/irq.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/irq.h b/include/linux/irq.h index 12277799c007..069d3b84d311 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -85,9 +85,10 @@ extern int no_irq_affinity; extern int noirqdebug_setup(char *str); extern fastcall int handle_IRQ_event(unsigned int irq, struct pt_regs *regs, - struct irqaction *action); + struct irqaction *action); extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs); -extern void note_interrupt(unsigned int irq, irq_desc_t *desc, int action_ret); +extern void note_interrupt(unsigned int irq, irq_desc_t *desc, + int action_ret, struct pt_regs *regs); extern int can_request_irq(unsigned int irq, unsigned long irqflags); extern void init_irq_proc(void); -- cgit v1.2.1 From 115d6f3fd25991f2a7de1ff4d758086209b1ed12 Mon Sep 17 00:00:00 2001 From: Mauro Carvalho Chehab Date: Tue, 28 Jun 2005 20:45:27 -0700 Subject: [PATCH] V4L: API new webcam formats included Add Philips Webcam format. Signed-off-by: Mauro Carvalho Chehab Signed-off-by: Luc Saillard . Signed-off-by: Nickolay V Shmyrev Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/videodev2.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/videodev2.h b/include/linux/videodev2.h index 4e0edce53760..acbfc525576d 100644 --- a/include/linux/videodev2.h +++ b/include/linux/videodev2.h @@ -221,6 +221,8 @@ struct v4l2_pix_format /* Vendor-specific formats */ #define V4L2_PIX_FMT_WNVA v4l2_fourcc('W','N','V','A') /* Winnov hw compress */ #define V4L2_PIX_FMT_SN9C10X v4l2_fourcc('S','9','1','0') /* SN9C10x compression */ +#define V4L2_PIX_FMT_PWC1 v4l2_fourcc('P','W','C','1') /* pwc older webcam */ +#define V4L2_PIX_FMT_PWC2 v4l2_fourcc('P','W','C','2') /* pwc newer webcam */ /* * F O R M A T E N U M E R A T I O N -- cgit v1.2.1 From 0dfc62465ef92c7ddcb1ba223bf062453566fd0f Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 31 May 2005 20:39:20 +0100 Subject: [MTD] NAND: Reorganize chip locking The code was wrong in several aspects. The locking order was inconsistent, the device aquire code did not reset a variable after a wakeup and the wakeup handling was not working for applications where multiple chips are sharing a single hardware controller. When a hardware controller is available the locking is now reduced to the hardware controller lock and the waitqueue is moved to the hardware controller structure in order to avoid a wake_up_all(). The problem was pointed out by Ben Dooks, who also found the missing variable reset as main cause for his deadlock problem. Signed-off-by: Thomas Gleixner --- include/linux/mtd/nand.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h index bee78969cb21..9b5b76217584 100644 --- a/include/linux/mtd/nand.h +++ b/include/linux/mtd/nand.h @@ -5,7 +5,7 @@ * Steven J. Hill * Thomas Gleixner * - * $Id: nand.h,v 1.71 2005/02/09 12:12:59 gleixner Exp $ + * $Id: nand.h,v 1.73 2005/05/31 19:39:17 gleixner Exp $ * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -253,10 +253,13 @@ struct nand_chip; * struct nand_hw_control - Control structure for hardware controller (e.g ECC generator) shared among independend devices * @lock: protection lock * @active: the mtd device which holds the controller currently + * @wq: wait queue to sleep on if a NAND operation is in progress + * used instead of the per chip wait queue when a hw controller is available */ struct nand_hw_control { spinlock_t lock; struct nand_chip *active; + wait_queue_head_t wq; }; /** -- cgit v1.2.1 From 02b15e343aeefb49f8cac949be599d78250a568f Mon Sep 17 00:00:00 2001 From: Todd Poynor Date: Tue, 7 Jun 2005 00:04:39 +0100 Subject: [MTD] XIP for AMD CFI flash. Author: Vitaly Wool Signed-off-by: Todd Poynor Signed-off-by: Thomas Gleixner --- include/linux/mtd/cfi.h | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/cfi.h b/include/linux/mtd/cfi.h index 66e0a32efbac..e6b6a1c66bd5 100644 --- a/include/linux/mtd/cfi.h +++ b/include/linux/mtd/cfi.h @@ -1,7 +1,7 @@ /* Common Flash Interface structures * See http://support.intel.com/design/flash/technote/index.htm - * $Id: cfi.h,v 1.53 2005/03/15 19:03:13 gleixner Exp $ + * $Id: cfi.h,v 1.54 2005/06/06 23:04:36 tpoynor Exp $ */ #ifndef __MTD_CFI_H__ @@ -428,16 +428,6 @@ static inline void cfi_udelay(int us) } } -static inline void cfi_spin_lock(spinlock_t *mutex) -{ - spin_lock_bh(mutex); -} - -static inline void cfi_spin_unlock(spinlock_t *mutex) -{ - spin_unlock_bh(mutex); -} - struct cfi_extquery *cfi_read_pri(struct map_info *map, uint16_t adr, uint16_t size, const char* name); struct cfi_fixup { -- cgit v1.2.1 From bfabb98688e7089381baa0974f7ff6786ce2a2d0 Mon Sep 17 00:00:00 2001 From: Sean Young Date: Mon, 13 Jun 2005 14:08:48 +0100 Subject: [MTD] Use correct major number for INFTL inftl was assigned new major number 96, 94 is in use by dasd. See: http://www.ussg.iu.edu/hypermail/linux/kernel/0409.2/1220.html Signed-off-by: Sean Young Signed-off-by: Thomas Gleixner --- include/linux/mtd/inftl.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/inftl.h b/include/linux/mtd/inftl.h index b52c8cbd235c..0268125a6271 100644 --- a/include/linux/mtd/inftl.h +++ b/include/linux/mtd/inftl.h @@ -3,7 +3,7 @@ * * (C) Copyright 2002, Greg Ungerer (gerg@snapgear.com) * - * $Id: inftl.h,v 1.6 2004/06/30 14:49:00 dbrown Exp $ + * $Id: inftl.h,v 1.7 2005/06/13 13:08:45 sean Exp $ */ #ifndef __MTD_INFTL_H__ @@ -20,7 +20,7 @@ #include #ifndef INFTL_MAJOR -#define INFTL_MAJOR 94 +#define INFTL_MAJOR 96 #endif #define INFTL_PARTN_BITS 4 -- cgit v1.2.1 From 0edb586049e57c56e625536476931117a57671e9 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Wed, 22 Jun 2005 16:59:51 +0200 Subject: [PATCH] driver core: add bus_find_device & driver_find_device functions Add bus_find_device() and driver_find_device() which allow searching for a device in the bus's resp. the driver's klist and obtain a reference on it. Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman --- include/linux/device.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/device.h b/include/linux/device.h index 7b781a72b293..07222c531d37 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -80,6 +80,8 @@ extern struct bus_type * find_bus(char * name); int bus_for_each_dev(struct bus_type * bus, struct device * start, void * data, int (*fn)(struct device *, void *)); +struct device * bus_find_device(struct bus_type *bus, struct device *start, + void *data, int (*match)(struct device *, void *)); int bus_for_each_drv(struct bus_type * bus, struct device_driver * start, void * data, int (*fn)(struct device_driver *, void *)); @@ -142,6 +144,9 @@ extern void driver_remove_file(struct device_driver *, struct driver_attribute * extern int driver_for_each_device(struct device_driver * drv, struct device * start, void * data, int (*fn)(struct device *, void *)); +struct device * driver_find_device(struct device_driver *drv, + struct device *start, void *data, + int (*match)(struct device *, void *)); /* -- cgit v1.2.1 From 23d3d602cb96addd3c1158424fb01a49ea5e81b1 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 22 Jun 2005 16:09:05 -0700 Subject: [PATCH] driver core: change bus_rescan_devices to return void No one was looking at the return value of bus_rescan_devices, and it really wasn't anything that anyone in the kernel would ever care about. So change it which enabled some counting code to be removed also. Signed-off-by: Greg Kroah-Hartman --- include/linux/device.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/device.h b/include/linux/device.h index 07222c531d37..f378c846e6d5 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -69,7 +69,7 @@ struct bus_type { extern int bus_register(struct bus_type * bus); extern void bus_unregister(struct bus_type * bus); -extern int bus_rescan_devices(struct bus_type * bus); +extern void bus_rescan_devices(struct bus_type * bus); extern struct bus_type * get_bus(struct bus_type * bus); extern void put_bus(struct bus_type * bus); -- cgit v1.2.1 From 0048e6030d41453c2f5ce0e9aead910d46cfd448 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Thu, 30 Jun 2005 00:48:14 -0500 Subject: Input: uinput - use completions instead of events and manual wakeups in force feedback code. Signed-off-by: Dmitry Torokhov --- include/linux/uinput.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/uinput.h b/include/linux/uinput.h index 4c2c82336d10..84876077027f 100644 --- a/include/linux/uinput.h +++ b/include/linux/uinput.h @@ -42,8 +42,7 @@ struct uinput_request { int code; /* UI_FF_UPLOAD, UI_FF_ERASE */ int retval; - wait_queue_head_t waitq; - int completed; + struct completion done; union { int effect_id; @@ -62,7 +61,7 @@ struct uinput_device { struct uinput_request *requests[UINPUT_NUM_REQUESTS]; wait_queue_head_t requests_waitq; - struct semaphore requests_sem; + spinlock_t requests_lock; }; #endif /* __KERNEL__ */ -- cgit v1.2.1 From 16a334c0de5a94b1d10a1ac9a33f4dedac89a075 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Thu, 30 Jun 2005 00:49:08 -0500 Subject: Input: introduce usb_to_input_id() to uniformly produce struct input_id for USB input devices. Signed-off-by: Dmitry Torokhov --- include/linux/usb_input.h | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) create mode 100644 include/linux/usb_input.h (limited to 'include/linux') diff --git a/include/linux/usb_input.h b/include/linux/usb_input.h new file mode 100644 index 000000000000..716e0cc16043 --- /dev/null +++ b/include/linux/usb_input.h @@ -0,0 +1,25 @@ +#ifndef __USB_INPUT_H +#define __USB_INPUT_H + +/* + * Copyright (C) 2005 Dmitry Torokhov + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + */ + +#include +#include +#include + +static inline void +usb_to_input_id(const struct usb_device *dev, struct input_id *id) +{ + id->bustype = BUS_USB; + id->vendor = le16_to_cpu(dev->descriptor.idVendor); + id->product = le16_to_cpu(dev->descriptor.idProduct); + id->version = le16_to_cpu(dev->descriptor.bcdDevice); +} + +#endif -- cgit v1.2.1 From 5b6271bda42be8edb77fbd588621cc09199fa7fb Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Thu, 30 Jun 2005 00:50:38 -0500 Subject: Input: make name, phys and uniq be 'const char *' because once set noone should attempt to change them. Signed-off-by: Dmitry Torokhov --- include/linux/input.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/input.h b/include/linux/input.h index b9cc0ac71f44..bdc53c6cc962 100644 --- a/include/linux/input.h +++ b/include/linux/input.h @@ -811,9 +811,9 @@ struct input_dev { void *private; - char *name; - char *phys; - char *uniq; + const char *name; + const char *phys; + const char *uniq; struct input_id id; unsigned long evbit[NBITS(EV_MAX)]; -- cgit v1.2.1 From a03fa955576af50df80bec9127b46ef57e0877c0 Mon Sep 17 00:00:00 2001 From: "rajesh.shah@intel.com" Date: Thu, 2 Jun 2005 15:41:48 -0700 Subject: [PATCH] PCI: Increase the number of PCI bus resources This patch increases the number of resource pointers in the pci_bus structure. This is needed to store >4 resource ranges for host bridges and transparent PCI bridges. With this change, all PCI buses will have more resource pointers, but most PCI buses will only use the first 3 or 4, the remaining being NULL. The PCI core already deals with this correctly. Signed-off-by: Rajesh Shah Signed-off-by: Greg Kroah-Hartman --- include/linux/pci.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 66798b46f308..a46cabfd08c8 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -586,7 +586,7 @@ struct pci_dev { #define PCI_NUM_RESOURCES 11 #ifndef PCI_BUS_NUM_RESOURCES -#define PCI_BUS_NUM_RESOURCES 4 +#define PCI_BUS_NUM_RESOURCES 8 #endif #define PCI_REGION_FLAG_MASK 0x0fU /* These bits of resource flags tell us the PCI region flags */ -- cgit v1.2.1 From 75865858971add95809c5c9cd35dc4cfba08e33b Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 30 Jun 2005 02:18:12 -0700 Subject: [PATCH] PCI: clean up dynamic pci id logic The dynamic pci id logic has been bothering me for a while, and now that I started to look into how to move some of this to the driver core, I thought it was time to clean it all up. It ends up making the code smaller, and easier to follow, and fixes a few bugs at the same time (dynamic ids were not being matched everywhere, and so could be missed on some call paths for new devices, semaphore not needed to be grabbed when adding a new id and calling the driver core, etc.) I also renamed the function pci_match_device() to pci_match_id() as that's what it really does. Signed-off-by: Greg Kroah-Hartman --- include/linux/pci-dynids.h | 18 ------------------ include/linux/pci.h | 3 ++- 2 files changed, 2 insertions(+), 19 deletions(-) delete mode 100644 include/linux/pci-dynids.h (limited to 'include/linux') diff --git a/include/linux/pci-dynids.h b/include/linux/pci-dynids.h deleted file mode 100644 index 183b6b0de81c..000000000000 --- a/include/linux/pci-dynids.h +++ /dev/null @@ -1,18 +0,0 @@ -/* - * PCI defines and function prototypes - * Copyright 2003 Dell Inc. - * by Matt Domsch - */ - -#ifndef LINUX_PCI_DYNIDS_H -#define LINUX_PCI_DYNIDS_H - -#include -#include - -struct dynid { - struct list_head node; - struct pci_device_id id; -}; - -#endif diff --git a/include/linux/pci.h b/include/linux/pci.h index a46cabfd08c8..7ac14961ba22 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -860,7 +860,8 @@ int pci_register_driver(struct pci_driver *); void pci_unregister_driver(struct pci_driver *); void pci_remove_behind_bridge(struct pci_dev *); struct pci_driver *pci_dev_driver(const struct pci_dev *); -const struct pci_device_id *pci_match_device(const struct pci_device_id *ids, const struct pci_dev *dev); +const struct pci_device_id *pci_match_device(struct pci_driver *drv, struct pci_dev *dev); +const struct pci_device_id *pci_match_id(const struct pci_device_id *ids, struct pci_dev *dev); int pci_scan_bridge(struct pci_bus *bus, struct pci_dev * dev, int max, int pass); /* kmem_cache style wrapper around pci_alloc_consistent() */ -- cgit v1.2.1 From 21e2c01dc3e38d466eda5871645878d2c3a33261 Mon Sep 17 00:00:00 2001 From: Rob Punkunus Date: Sun, 3 Jul 2005 17:37:18 +0200 Subject: [PATCH] amd74xx: support MCP55 device IDs From: Rob Punkunus Rob Punkunus recently submitted a patch to enable support for MCP51/MCP55 in the amd74xx driver. This patch was whitespace-corrupted and didn't apply to 2.6.12 since MCP51 support was merged in the 2.6.12-rc series. Gentoo would like to support this hardware for our upcoming release media, so I fixed the patch, and here it is :) Signed-off-by: Daniel Drake Signed-off-by: Bartlomiej Zolnierkiewicz --- include/linux/pci_ids.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index c3ee1ae4545a..27348c22dacb 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1238,6 +1238,7 @@ #define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE 0x0265 #define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_SATA 0x0266 #define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_SATA2 0x0267 +#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE 0x036E #define PCI_DEVICE_ID_NVIDIA_NVENET_12 0x0268 #define PCI_DEVICE_ID_NVIDIA_NVENET_13 0x0269 #define PCI_DEVICE_ID_NVIDIA_MCP51_AUDIO 0x026B -- cgit v1.2.1 From e7270dec080002d8aa18256c756af6c32331ef48 Mon Sep 17 00:00:00 2001 From: Raphael Assenat Date: Mon, 4 Jul 2005 13:23:45 -0700 Subject: [SPARC64/COMPAT]: Add some compat ioctl for ppdev The following patch adds some ioctls to include/linux/compat_ioctl.h to allow using ppdev from the 32 bit user space on sparc64. This patch also adds the PPDEV option in the sparc64 menu, near Parallel printer support in the 'General machine setup' submenu. All those ioctls seem to be compatible, since (correct me if I'm wrong) they dont use the 'long' type. See include/linux/ppdev.h. The application I used to test the new ioctls only used the following: PPEXCL PPCLAIM PPNEGOT PPGETMODES PPRCONTROL PPWCONTROL PPDATADIR PPWDATA PPRDATA But I beleive that the other ioctls will work fine. Signed-off-by: David S. Miller --- include/linux/compat_ioctl.h | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h index 70a4ebb5d964..ecb0d39c0798 100644 --- a/include/linux/compat_ioctl.h +++ b/include/linux/compat_ioctl.h @@ -346,10 +346,27 @@ COMPATIBLE_IOCTL(PPPOEIOCDFWD) /* LP */ COMPATIBLE_IOCTL(LPGETSTATUS) /* ppdev */ +COMPATIBLE_IOCTL(PPSETMODE) +COMPATIBLE_IOCTL(PPRSTATUS) +COMPATIBLE_IOCTL(PPRCONTROL) +COMPATIBLE_IOCTL(PPWCONTROL) +COMPATIBLE_IOCTL(PPFCONTROL) +COMPATIBLE_IOCTL(PPRDATA) +COMPATIBLE_IOCTL(PPWDATA) COMPATIBLE_IOCTL(PPCLAIM) COMPATIBLE_IOCTL(PPRELEASE) -COMPATIBLE_IOCTL(PPEXCL) COMPATIBLE_IOCTL(PPYIELD) +COMPATIBLE_IOCTL(PPEXCL) +COMPATIBLE_IOCTL(PPDATADIR) +COMPATIBLE_IOCTL(PPNEGOT) +COMPATIBLE_IOCTL(PPWCTLONIRQ) +COMPATIBLE_IOCTL(PPCLRIRQ) +COMPATIBLE_IOCTL(PPSETPHASE) +COMPATIBLE_IOCTL(PPGETMODES) +COMPATIBLE_IOCTL(PPGETMODE) +COMPATIBLE_IOCTL(PPGETPHASE) +COMPATIBLE_IOCTL(PPGETFLAGS) +COMPATIBLE_IOCTL(PPSETFLAGS) /* CDROM stuff */ COMPATIBLE_IOCTL(CDROMPAUSE) COMPATIBLE_IOCTL(CDROMRESUME) -- cgit v1.2.1 From 55820ee2f8c767a2833b21bd365e5753f50bd8ce Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 5 Jul 2005 14:08:10 -0700 Subject: [NET]: Fix signedness issues in net/core/filter.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is the code to load packet data into a register: k = fentry->k; if (k < 0) { ... } else { u32 _tmp, *p; p = skb_header_pointer(skb, k, 4, &_tmp); if (p != NULL) { A = ntohl(*p); continue; } } skb_header_pointer checks if the requested data is within the linear area: int hlen = skb_headlen(skb); if (offset + len <= hlen) return skb->data + offset; When offset is within [INT_MAX-len+1..INT_MAX] the addition will result in a negative number which is <= hlen. I couldn't trigger a crash on my AMD64 with 2GB of memory, but a coworker tried on his x86 machine and it crashed immediately. This patch fixes the check in skb_header_pointer to handle large positive offsets similar to skb_copy_bits. Invalid data can still be accessed using negative offsets (also similar to skb_copy_bits), anyone using negative offsets needs to verify them himself. Thanks to Thomas Vögtle for verifying the problem by crashing his machine and providing me with an Oops. Signed-off-by: Patrick McHardy Acked-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/skbuff.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 416a2e4024b2..fbcb18651970 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1211,7 +1211,7 @@ static inline void *skb_header_pointer(const struct sk_buff *skb, int offset, { int hlen = skb_headlen(skb); - if (offset + len <= hlen) + if (hlen - offset >= len) return skb->data + offset; if (skb_copy_bits(skb, offset, buffer, len) < 0) -- cgit v1.2.1 From e176fe8954a5239c24afe79b1001ba3c29511963 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 5 Jul 2005 14:12:44 -0700 Subject: [NET]: Remove unused security member in sk_buff Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/skbuff.h | 4 +--- include/linux/tc_ematch/tc_em_meta.h | 2 +- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index fbcb18651970..1e6290f4f81e 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -183,7 +183,6 @@ struct skb_shared_info { * @priority: Packet queueing priority * @users: User count - see {datagram,tcp}.c * @protocol: Packet protocol from driver - * @security: Security level of packet * @truesize: Buffer size * @head: Head of buffer * @data: Data head pointer @@ -255,8 +254,7 @@ struct sk_buff { pkt_type, ip_summed; __u32 priority; - unsigned short protocol, - security; + unsigned short protocol; void (*destructor)(struct sk_buff *skb); #ifdef CONFIG_NETFILTER diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index a6b2cc530af5..bcb762d93123 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -45,7 +45,7 @@ enum TCF_META_ID_REALDEV, TCF_META_ID_PRIORITY, TCF_META_ID_PROTOCOL, - TCF_META_ID_SECURITY, + TCF_META_ID_SECURITY, /* obsolete */ TCF_META_ID_PKTTYPE, TCF_META_ID_PKTLEN, TCF_META_ID_DATALEN, -- cgit v1.2.1 From 1cbb3380ef683f742876f48e3739b3df4ea9e168 Mon Sep 17 00:00:00 2001 From: Thomas Graf Date: Tue, 5 Jul 2005 14:13:41 -0700 Subject: [NET]: Reduce size of sk_buff by 4 bytes Reduce local_df to a bit field and ip_summed to a 2 bits field thus saving 13 bits. Move bit fields, packet type, and protocol into the spare area between the priority and the destructor. Saves 4 bytes on both, 32bit and 64bit architectures. Signed-off-by: Thomas Graf Signed-off-by: David S. Miller --- include/linux/skbuff.h | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 1e6290f4f81e..14b950413495 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -248,17 +248,18 @@ struct sk_buff { data_len, mac_len, csum; - unsigned char local_df, - cloned:1, - nohdr:1, - pkt_type, - ip_summed; __u32 priority; - unsigned short protocol; + __u8 local_df:1, + cloned:1, + ip_summed:2, + nohdr:1; + /* 3 bits spare */ + __u8 pkt_type; + __u16 protocol; void (*destructor)(struct sk_buff *skb); #ifdef CONFIG_NETFILTER - unsigned long nfmark; + unsigned long nfmark; __u32 nfcache; __u32 nfctinfo; struct nf_conntrack *nfct; -- cgit v1.2.1 From bc971dee6ece1fd0d431948924becd9c50e7b778 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 5 Jul 2005 15:03:46 -0700 Subject: [SHAPER]: Switch to spinlocks. Dave, you were right and the sleeping locks in shaper were broken. Markus Kanet noticed this and also tested the patch below that switches locking to spinlocks. Signed-off-by: Christoph Hellwig Signed-off-by: David S. Miller --- include/linux/if_shaper.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/if_shaper.h b/include/linux/if_shaper.h index 004e6f09a6e2..68c896a36a34 100644 --- a/include/linux/if_shaper.h +++ b/include/linux/if_shaper.h @@ -23,7 +23,7 @@ struct shaper __u32 shapeclock; unsigned long recovery; /* Time we can next clock a packet out on an empty queue */ - struct semaphore sem; + spinlock_t lock; struct net_device_stats stats; struct net_device *dev; int (*hard_start_xmit) (struct sk_buff *skb, -- cgit v1.2.1 From c1b4a7e69576d65efc31a8cea0714173c2841244 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 5 Jul 2005 15:24:38 -0700 Subject: [TCP]: Move to new TSO segmenting scheme. Make TSO segment transmit size decisions at send time not earlier. The basic scheme is that we try to build as large a TSO frame as possible when pulling in the user data, but the size of the TSO frame output to the card is determined at transmit time. This is guided by tp->xmit_size_goal. It is always set to a multiple of MSS and tells sendmsg/sendpage how large an SKB to try and build. Later, tcp_write_xmit() and tcp_push_one() chop up the packet if necessary and conditions warrant. These routines can also decide to "defer" in order to wait for more ACKs to arrive and thus allow larger TSO frames to be emitted. A general observation is that TSO elongates the pipe, thus requiring a larger congestion window and larger buffering especially at the sender side. Therefore, it is important that applications 1) get a large enough socket send buffer (this is accomplished by our dynamic send buffer expansion code) 2) do large enough writes. Signed-off-by: David S. Miller --- include/linux/tcp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index dfd93d03f5d2..e4fd82e42104 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -286,7 +286,7 @@ struct tcp_sock { __u32 max_window; /* Maximal window ever seen from peer */ __u32 pmtu_cookie; /* Last pmtu seen by socket */ __u32 mss_cache; /* Cached effective mss, not including SACKS */ - __u16 mss_cache_std; /* Like mss_cache, but without TSO */ + __u16 xmit_size_goal; /* Goal for segmenting output packets */ __u16 ext_header_len; /* Network protocol overhead (IP/IPv6 options) */ __u8 ca_state; /* State of fast-retransmit machine */ __u8 retransmits; /* Number of unrecovered RTO timeouts. */ -- cgit v1.2.1 From 6772926bef3c9f0ec761b39e5702535471fff70b Mon Sep 17 00:00:00 2001 From: Rusty Lynch Date: Tue, 5 Jul 2005 18:54:50 -0700 Subject: [PATCH] kprobes: fix namespace problem and sparc64 build The following renames arch_init, a kprobes function for performing any architecture specific initialization, to arch_init_kprobes in order to cleanup the namespace. Also, this patch adds arch_init_kprobes to sparc64 to fix the sparc64 kprobes build from the last return probe patch. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/kprobes.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index b7a194c4362a..e050fc2d4c26 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -155,7 +155,7 @@ extern void arch_copy_kprobe(struct kprobe *p); extern void arch_arm_kprobe(struct kprobe *p); extern void arch_disarm_kprobe(struct kprobe *p); extern void arch_remove_kprobe(struct kprobe *p); -extern int arch_init(void); +extern int arch_init_kprobes(void); extern void show_registers(struct pt_regs *regs); extern kprobe_opcode_t *get_insn_slot(void); extern void free_insn_slot(kprobe_opcode_t *slot); -- cgit v1.2.1 From 5e6557722e69840506eb8bc5a1edcdb4e447a917 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Wed, 6 Jul 2005 15:44:41 -0400 Subject: [PATCH] openfirmware: generate device table for userspace This converts the usage of struct of_match to struct of_device_id, similar to pci_device_id. This allows a device table to be generated, which can be parsed by depmod(8) to generate a map file for module loading. In order for hotplug to work with macio devices, patches to module-init-tools and hotplug must be applied. Those patches are available at: ftp://ftp.suse.com/pub/people/jeffm/linux/macio-hotplug/ Signed-off-by: Jeff Mahoney Signed-off-by: Linus Torvalds --- include/linux/mod_devicetable.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h index 9b6d05172ed4..dce53ac1625d 100644 --- a/include/linux/mod_devicetable.h +++ b/include/linux/mod_devicetable.h @@ -174,6 +174,17 @@ struct serio_device_id { __u8 proto; }; +/* + * Struct used for matching a device + */ +struct of_device_id +{ + char name[32]; + char type[32]; + char compatible[128]; + void *data; +}; + /* PCMCIA */ -- cgit v1.2.1 From 40725181b74be6b0e3bdc8c05bd1e0b9873ec5cc Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 6 Jul 2005 13:51:52 -0700 Subject: [CRYPTO] Add support for low-level multi-block operations This patch adds hooks for cipher algorithms to implement multi-block ECB/CBC operations directly. This is expected to provide significant performance boots to the VIA Padlock. It could also be used for improving software implementations such as AES where operating on multiple blocks at a time may enable certain optimisations. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/crypto.h | 28 +++++++++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/crypto.h b/include/linux/crypto.h index 387da6a3e58c..26ce01c25745 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -61,6 +61,15 @@ #define CRYPTO_DIR_DECRYPT 0 struct scatterlist; +struct crypto_tfm; + +struct cipher_desc { + struct crypto_tfm *tfm; + void (*crfn)(void *ctx, u8 *dst, const u8 *src); + unsigned int (*prfn)(const struct cipher_desc *desc, u8 *dst, + const u8 *src, unsigned int nbytes); + void *info; +}; /* * Algorithms: modular crypto algorithm implementations, managed @@ -73,6 +82,19 @@ struct cipher_alg { unsigned int keylen, u32 *flags); void (*cia_encrypt)(void *ctx, u8 *dst, const u8 *src); void (*cia_decrypt)(void *ctx, u8 *dst, const u8 *src); + + unsigned int (*cia_encrypt_ecb)(const struct cipher_desc *desc, + u8 *dst, const u8 *src, + unsigned int nbytes); + unsigned int (*cia_decrypt_ecb)(const struct cipher_desc *desc, + u8 *dst, const u8 *src, + unsigned int nbytes); + unsigned int (*cia_encrypt_cbc)(const struct cipher_desc *desc, + u8 *dst, const u8 *src, + unsigned int nbytes); + unsigned int (*cia_decrypt_cbc)(const struct cipher_desc *desc, + u8 *dst, const u8 *src, + unsigned int nbytes); }; struct digest_alg { @@ -136,7 +158,6 @@ static inline int crypto_alg_available(const char *name, u32 flags) * and core processing logic. Managed via crypto_alloc_tfm() and * crypto_free_tfm(), as well as the various helpers below. */ -struct crypto_tfm; struct cipher_tfm { void *cit_iv; @@ -266,6 +287,11 @@ static inline unsigned int crypto_tfm_alg_digestsize(struct crypto_tfm *tfm) return tfm->__crt_alg->cra_digest.dia_digestsize; } +static inline void *crypto_tfm_ctx(struct crypto_tfm *tfm) +{ + return (void *)&tfm[1]; +} + /* * API wrappers. */ -- cgit v1.2.1 From 95477377995aefa2ec1654a9a3777bd57ea99146 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 6 Jul 2005 13:52:09 -0700 Subject: [CRYPTO] Add alignmask for low-level cipher implementations The VIA Padlock device requires the input and output buffers to be aligned on 16-byte boundaries. This patch adds the alignmask attribute for low-level cipher implementations to indicate their alignment requirements. The mid-level crypt() function will copy the input/output buffers if they are not aligned correctly before they are passed to the low-level implementation. Strictly speaking, some of the software implementations require the buffers to be aligned on 4-byte boundaries as they do 32-bit loads. However, it is not clear whether it is better to copy the buffers or pay the penalty for unaligned loads/stores. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/crypto.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/crypto.h b/include/linux/crypto.h index 26ce01c25745..ac9d49beecd3 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -124,6 +124,7 @@ struct crypto_alg { u32 cra_flags; unsigned int cra_blocksize; unsigned int cra_ctxsize; + unsigned int cra_alignmask; const char cra_name[CRYPTO_MAX_ALG_NAME]; union { -- cgit v1.2.1 From fbdae9f3e7fb57c07cb0d973f113eb25da2e8ff2 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 6 Jul 2005 13:53:29 -0700 Subject: [CRYPTO] Ensure cit_iv is aligned correctly This patch ensures that cit_iv is aligned according to cra_alignmask by allocating it as part of the tfm structure. As a side effect the crypto layer will also guarantee that the tfm ctx area has enough space to be aligned by cra_alignmask. This allows us to remove the extra space reservation from the Padlock driver. Signed-off-by: Herbert Xu Signed-off-by: David S. Miller --- include/linux/crypto.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/crypto.h b/include/linux/crypto.h index ac9d49beecd3..5e2bcc636a02 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -288,6 +288,11 @@ static inline unsigned int crypto_tfm_alg_digestsize(struct crypto_tfm *tfm) return tfm->__crt_alg->cra_digest.dia_digestsize; } +static inline unsigned int crypto_tfm_alg_alignmask(struct crypto_tfm *tfm) +{ + return tfm->__crt_alg->cra_alignmask; +} + static inline void *crypto_tfm_ctx(struct crypto_tfm *tfm) { return (void *)&tfm[1]; -- cgit v1.2.1 From 97f927a4d7dbccde0a854a62c3ea54d90bae8679 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 7 Jul 2005 16:50:16 +0200 Subject: [MTD] XIP cleanup Move the architecture dependend code into include/asm/mtd-xip.h Signed-off-by: Thomas Gleixner --- include/linux/mtd/xip.h | 31 +++++++++++++------------------ 1 file changed, 13 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mtd/xip.h b/include/linux/mtd/xip.h index fc071125cbcc..7b7deef6b180 100644 --- a/include/linux/mtd/xip.h +++ b/include/linux/mtd/xip.h @@ -58,22 +58,16 @@ * returned value is <= the real elapsed time. * note 2: this should be able to cope with a few seconds without * overflowing. + * + * xip_iprefetch() + * + * Macro to fill instruction prefetch + * e.g. a series of nops: asm volatile (".rep 8; nop; .endr"); */ -#if defined(CONFIG_ARCH_SA1100) || defined(CONFIG_ARCH_PXA) - -#include -#ifdef CONFIG_ARCH_PXA -#include -#endif - -#define xip_irqpending() (ICIP & ICMR) - -/* we sample OSCR and convert desired delta to usec (1/4 ~= 1000000/3686400) */ -#define xip_currtime() (OSCR) -#define xip_elapsed_since(x) (signed)((OSCR - (x)) / 4) +#include -#else +#ifndef xip_irqpending #warning "missing IRQ and timer primitives for XIP MTD support" #warning "some of the XIP MTD support code will be disabled" @@ -85,16 +79,17 @@ #endif +#ifndef xip_iprefetch +#define xip_iprefetch() do { } while (0) +#endif + /* * xip_cpu_idle() is used when waiting for a delay equal or larger than * the system timer tick period. This should put the CPU into idle mode * to save power and to be woken up only when some interrupts are pending. - * As above, this should not rely upon standard kernel code. + * This should not rely upon standard kernel code. */ - -#if defined(CONFIG_CPU_XSCALE) -#define xip_cpu_idle() asm volatile ("mcr p14, 0, %0, c7, c0, 0" :: "r" (1)) -#else +#ifndef xip_cpu_idle #define xip_cpu_idle() do { } while (0) #endif -- cgit v1.2.1 From cb2c0233755429037462e16ea0d5497a0092738c Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Thu, 7 Jul 2005 17:56:03 -0700 Subject: [PATCH] export generic_drop_inode() to modules OCFS2 wants to mark an inode which has been orphaned by another node so that during final iput it takes the correct path through the VFS and can pass through the OCFS2 delete_inode callback. Since i_nlink can get out of date with other nodes, the best way I see to accomplish this is by clearing i_nlink on those inodes at drop_inode time. Other than this small amount of work, nothing different needs to happen, so I think it would be cleanest to be able to just call generic_drop_inode at the end of the OCFS2 drop_inode callback. Signed-off-by: Mark Fasheh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 047bde30836a..302ec20838ca 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1435,6 +1435,7 @@ extern struct inode * igrab(struct inode *); extern ino_t iunique(struct super_block *, ino_t); extern int inode_needs_sync(struct inode *inode); extern void generic_delete_inode(struct inode *inode); +extern void generic_drop_inode(struct inode *inode); extern struct inode *ilookup5(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data); -- cgit v1.2.1 From 79b9ce311e192e9a31fd9f3cf1ee4a4edf9e2650 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Thu, 7 Jul 2005 17:56:04 -0700 Subject: [PATCH] print order information when OOM killing Dump the current allocation order when OOM killing. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/swap.h b/include/linux/swap.h index 2343f999e6e1..c75954f2d868 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -148,7 +148,7 @@ struct swap_list_t { #define vm_swap_full() (nr_swap_pages*2 < total_swap_pages) /* linux/mm/oom_kill.c */ -extern void out_of_memory(unsigned int __nocast gfp_mask); +extern void out_of_memory(unsigned int __nocast gfp_mask, int order); /* linux/mm/memory.c */ extern void swapin_readahead(swp_entry_t, unsigned long, struct vm_area_struct *); -- cgit v1.2.1 From cf36680887d6d942d2119c1ff1dfb2428b0f21f4 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 7 Jul 2005 17:56:13 -0700 Subject: [PATCH] move ioprio syscalls into syscalls.h - Make ioprio syscalls return long, like set/getpriority syscalls. - Move function prototypes into syscalls.h so we can pick them up in the 32/64bit compat code. Signed-off-by: Anton Blanchard Acked-by: Jens Axboe Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ioprio.h | 3 --- include/linux/syscalls.h | 3 +++ 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h index 8a453a0b5e4b..88d5961f7a3f 100644 --- a/include/linux/ioprio.h +++ b/include/linux/ioprio.h @@ -34,9 +34,6 @@ enum { */ #define IOPRIO_BE_NR (8) -asmlinkage int sys_ioprio_set(int, int, int); -asmlinkage int sys_ioprio_get(int, int); - enum { IOPRIO_WHO_PROCESS = 1, IOPRIO_WHO_PGRP, diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 52830b6d94e5..425f58c8ea4a 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -506,4 +506,7 @@ asmlinkage long sys_request_key(const char __user *_type, asmlinkage long sys_keyctl(int cmd, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5); +asmlinkage long sys_ioprio_set(int which, int who, int ioprio); +asmlinkage long sys_ioprio_get(int which, int who); + #endif -- cgit v1.2.1 From e00d9967e3addea86dded46deefc5daec5d52e5a Mon Sep 17 00:00:00 2001 From: Bernard Blackham Date: Thu, 7 Jul 2005 17:56:42 -0700 Subject: [PATCH] pm: fix u32 vs. pm_message_t confusion in cpufreq Fix u32 vs pm_message_t confusion in cpufreq. Signed-off-by: Bernard Blackham Signed-off-by: Pavel Machek Cc: Dave Jones Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cpufreq.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index 927daa86c9b3..ff7f80f48df1 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -201,7 +201,7 @@ struct cpufreq_driver { /* optional */ int (*exit) (struct cpufreq_policy *policy); - int (*suspend) (struct cpufreq_policy *policy, u32 state); + int (*suspend) (struct cpufreq_policy *policy, pm_message_t pmsg); int (*resume) (struct cpufreq_policy *policy); struct freq_attr **attr; }; -- cgit v1.2.1 From a39722034ae37f80a1803bf781fe3fe1b03e20bc Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Thu, 7 Jul 2005 17:56:56 -0700 Subject: [PATCH] page_uptodate locking scalability Use a bit spin lock in the first buffer of the page to synchronise asynch IO buffer completions, instead of the global page_uptodate_lock, which is showing some scalabilty problems. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/buffer_head.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h index 802c91e9b3da..90828493791f 100644 --- a/include/linux/buffer_head.h +++ b/include/linux/buffer_head.h @@ -19,6 +19,9 @@ enum bh_state_bits { BH_Dirty, /* Is dirty */ BH_Lock, /* Is locked */ BH_Req, /* Has been submitted for I/O */ + BH_Uptodate_Lock,/* Used by the first bh in a page, to serialise + * IO completion of other buffers in the page + */ BH_Mapped, /* Has a disk mapping */ BH_New, /* Disk mapping was newly created by get_block */ -- cgit v1.2.1 From 0db925af1db5f3dfe1691c35b39496e2baaff9c9 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 7 Jul 2005 17:56:58 -0700 Subject: [PATCH] propagate __nocast annotations Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 4 ++-- include/linux/slab.h | 4 ++-- include/linux/string.h | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 8d6bf608b199..7c7400137e97 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -12,8 +12,8 @@ struct vm_area_struct; * GFP bitmasks.. */ /* Zone modifiers in GFP_ZONEMASK (see linux/mmzone.h - low two bits) */ -#define __GFP_DMA 0x01 -#define __GFP_HIGHMEM 0x02 +#define __GFP_DMA 0x01u +#define __GFP_HIGHMEM 0x02u /* * Action modifiers - doesn't change the zoning diff --git a/include/linux/slab.h b/include/linux/slab.h index 76cf7e60216c..4c8e552471b0 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -65,7 +65,7 @@ extern void *kmem_cache_alloc(kmem_cache_t *, unsigned int __nocast); extern void kmem_cache_free(kmem_cache_t *, void *); extern unsigned int kmem_cache_size(kmem_cache_t *); extern const char *kmem_cache_name(kmem_cache_t *); -extern kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags); +extern kmem_cache_t *kmem_find_general_cachep(size_t size, unsigned int __nocast gfpflags); /* Size description struct for general caches. */ struct cache_sizes { @@ -105,7 +105,7 @@ extern unsigned int ksize(const void *); #ifdef CONFIG_NUMA extern void *kmem_cache_alloc_node(kmem_cache_t *, int flags, int node); -extern void *kmalloc_node(size_t size, int flags, int node); +extern void *kmalloc_node(size_t size, unsigned int __nocast flags, int node); #else static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int node) { diff --git a/include/linux/string.h b/include/linux/string.h index 93994c613095..dab2652acbd8 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -88,7 +88,7 @@ extern int memcmp(const void *,const void *,__kernel_size_t); extern void * memchr(const void *,int,__kernel_size_t); #endif -extern char *kstrdup(const char *s, int gfp); +extern char *kstrdup(const char *s, unsigned int __nocast gfp); #ifdef __cplusplus } -- cgit v1.2.1 From 6c036527a630720063b67d9a65455e8caca2c8fa Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 7 Jul 2005 17:56:59 -0700 Subject: [PATCH] mostly_read data section Add a new section called ".data.read_mostly" for data items that are read frequently and rarely written to like cpumaps etc. If these maps are placed in the .data section then these frequenly read items may end up in cachelines with data is is frequently updated. In that case all processors in an SMP system must needlessly reload the cachelines again and again containing elements of those frequently used variables. The ability to share these cachelines will allow each cpu in an SMP system to keep local copies of those shared cachelines thereby optimizing performance. Signed-off-by: Alok N Kataria Signed-off-by: Shobhit Dayal Signed-off-by: Christoph Lameter Signed-off-by: Shai Fultheim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cache.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/cache.h b/include/linux/cache.h index 4d767b93738a..2b66a36d85f0 100644 --- a/include/linux/cache.h +++ b/include/linux/cache.h @@ -13,6 +13,12 @@ #define SMP_CACHE_BYTES L1_CACHE_BYTES #endif +#ifdef CONFIG_X86 +#define __read_mostly __attribute__((__section__(".data.read_mostly"))) +#else +#define __read_mostly +#endif + #ifndef ____cacheline_aligned #define ____cacheline_aligned __attribute__((__aligned__(SMP_CACHE_BYTES))) #endif -- cgit v1.2.1 From 1ce88cf466f7b6078b14d67d186a3d7c19dd5609 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Thu, 7 Jul 2005 17:57:24 -0700 Subject: [PATCH] namespace.c: fix race in mark_mounts_for_expiry() This patch fixes a race found by Ram in mark_mounts_for_expiry() in fs/namespace.c. The bug can only be triggered with simultaneous exiting of a process having a private namespace, and expiry of a mount from within that namespace. It's practically impossible to trigger, and I haven't even tried. But still, a bug is a bug. The race happens when put_namespace() is called by another task, while mark_mounts_for_expiry() is between atomic_read() and get_namespace(). In that case get_namespace() will be called on an already dead namespace with unforeseeable results. The solution was suggested by Al Viro, with his own words: Instead of screwing with atomic_read() in there, why don't we simply do the following: a) atomic_dec_and_lock() in put_namespace() b) __put_namespace() called without dropping lock c) the first thing done by __put_namespace would be struct vfsmount *root = namespace->root; namespace->root = NULL; spin_unlock(...); .... umount_tree(root); ... d) check in mark_... would be simply namespace && namespace->root. And we are all set; no screwing around with atomic_read(), no magic at all. Dying namespace gets NULL ->root. All changes of ->root happen under spinlock. If under a spinlock we see non-NULL ->mnt_namespace, it won't be freed until we drop the lock (we will set ->mnt_namespace to NULL under that lock before we get to freeing namespace). If under a spinlock we see non-NULL ->mnt_namespace and ->mnt_namespace->root, we can grab a reference to namespace and be sure that it won't go away. Signed-off-by: Miklos Szeredi Acked-by: Al Viro Acked-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/namespace.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/namespace.h b/include/linux/namespace.h index 697991b69f9b..0e5a86f13b2f 100644 --- a/include/linux/namespace.h +++ b/include/linux/namespace.h @@ -17,7 +17,8 @@ extern void __put_namespace(struct namespace *namespace); static inline void put_namespace(struct namespace *namespace) { - if (atomic_dec_and_test(&namespace->count)) + if (atomic_dec_and_lock(&namespace->count, &vfsmount_lock)) + /* releases vfsmount_lock */ __put_namespace(namespace); } -- cgit v1.2.1 From 55e700b924f9e0ba24e3a071d1097d050b05abe6 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Thu, 7 Jul 2005 17:57:30 -0700 Subject: [PATCH] namespace: rename mnt_fslink to mnt_expire This patch renames vfsmount->mnt_fslink to something a little more descriptive: vfsmount->mnt_expire. Signed-off-by: Mike Waychison Signed-off-by: Miklos Szeredi Acked-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mount.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mount.h b/include/linux/mount.h index 8b8d3b9beefd..196d2d6de4a3 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -34,7 +34,7 @@ struct vfsmount int mnt_expiry_mark; /* true if marked for expiry */ char *mnt_devname; /* Name of device e.g. /dev/dsk/hda1 */ struct list_head mnt_list; - struct list_head mnt_fslink; /* link in fs-specific expiry list */ + struct list_head mnt_expire; /* link in fs-specific expiry list */ struct namespace *mnt_namespace; /* containing namespace */ }; -- cgit v1.2.1 From 751c404b8f63e8199d5f2f8f2bcfd69b41d11caa Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Thu, 7 Jul 2005 17:57:30 -0700 Subject: [PATCH] namespace: rename _mntput to mntput_no_expire This patch renames _mntput() to something a little more descriptive: mntput_no_expire(). Signed-off-by: Miklos Szeredi Acked-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mount.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mount.h b/include/linux/mount.h index 196d2d6de4a3..74b4727a4e30 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -47,7 +47,7 @@ static inline struct vfsmount *mntget(struct vfsmount *mnt) extern void __mntput(struct vfsmount *mnt); -static inline void _mntput(struct vfsmount *mnt) +static inline void mntput_no_expire(struct vfsmount *mnt) { if (mnt) { if (atomic_dec_and_test(&mnt->mnt_count)) @@ -59,7 +59,7 @@ static inline void mntput(struct vfsmount *mnt) { if (mnt) { mnt->mnt_expiry_mark = 0; - _mntput(mnt); + mntput_no_expire(mnt); } } -- cgit v1.2.1 From a6ccbbb8865101d83c2e716f08feae1da1c48584 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 7 Jul 2005 17:59:11 -0700 Subject: [PATCH] nfsd4: fix sync'ing of recovery directory We need to fsync the recovery directory after writing to it, but we weren't doing this correctly. (For example, we weren't taking the i_sem when calling ->fsync().) Just reuse the existing nfsd fsync code instead. Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/nfsd.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index 5791dfd30dd0..c2da1b62d416 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -124,6 +124,7 @@ int nfsd_statfs(struct svc_rqst *, struct svc_fh *, int nfsd_notify_change(struct inode *, struct iattr *); int nfsd_permission(struct svc_export *, struct dentry *, int); +void nfsd_sync_dir(struct dentry *dp); #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) #ifdef CONFIG_NFSD_V2_ACL -- cgit v1.2.1 From 7fb64cee34f5dc743f697041717cafda8a94b5ac Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 7 Jul 2005 17:59:20 -0700 Subject: [PATCH] nfsd4: seqid comments Add some comments on the use of so_seqid, in an attempt to avoid some of the confusion outlined in the previous patch.... Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index a84a3fa99be1..2d19431f47ea 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -203,7 +203,9 @@ struct nfs4_stateowner { int so_is_open_owner; /* 1=openowner,0=lockowner */ u32 so_id; struct nfs4_client * so_client; - u32 so_seqid; + /* after increment in ENCODE_SEQID_OP_TAIL, represents the next + * sequence id expected from the client: */ + u32 so_seqid; struct xdr_netobj so_owner; /* open owner name */ int so_confirmed; /* successful OPEN_CONFIRM? */ struct nfs4_replay so_replay; -- cgit v1.2.1 From b700949b781480819e53bdc38a53f053226dd75e Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 7 Jul 2005 17:59:23 -0700 Subject: [PATCH] nfsd4: return better error on io incompatible with open mode from RFC 3530: "Share reservations are established by OPEN operations and by their nature are mandatory in that when the OPEN denies READ or WRITE operations, that denial results in such operations being rejected with error NFS4ERR_LOCKED." (Note that share_denied is really only a legal error for OPEN.) Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/nfsd.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h index c2da1b62d416..6d5a24f3fc6d 100644 --- a/include/linux/nfsd/nfsd.h +++ b/include/linux/nfsd/nfsd.h @@ -231,6 +231,7 @@ void nfsd_lockd_shutdown(void); #define nfserr_reclaim_bad __constant_htonl(NFSERR_RECLAIM_BAD) #define nfserr_badname __constant_htonl(NFSERR_BADNAME) #define nfserr_cb_path_down __constant_htonl(NFSERR_CB_PATH_DOWN) +#define nfserr_locked __constant_htonl(NFSERR_LOCKED) /* error codes for internal use */ /* if a request fails due to kmalloc failure, it gets dropped. -- cgit v1.2.1 From 4c4cd222ee329025840bc2f8cebf71d36c62440c Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 7 Jul 2005 17:59:27 -0700 Subject: [PATCH] nfsd4: check lock type against openmode. We shouldn't be allowing, e.g., write locks on files not open for read. To enforce this, we add a pointer from the lock stateid back to the open stateid it came from, so that the check will continue to be correct even after the open is upgraded or downgraded. Signed-off-by: Andy Adamson Signed-off-by: J. Bruce Fields Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfsd/state.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/nfsd/state.h b/include/linux/nfsd/state.h index 2d19431f47ea..8bf23cf8b603 100644 --- a/include/linux/nfsd/state.h +++ b/include/linux/nfsd/state.h @@ -237,6 +237,10 @@ struct nfs4_file { * st_perlockowner: (open stateid) list of lock nfs4_stateowners * st_access_bmap: used only for open stateid * st_deny_bmap: used only for open stateid +* st_openstp: open stateid lock stateid was derived from +* +* XXX: open stateids and lock stateids have diverged sufficiently that +* we should consider defining separate structs for the two cases. */ struct nfs4_stateid { @@ -250,6 +254,7 @@ struct nfs4_stateid { struct file * st_vfs_file; unsigned long st_access_bmap; unsigned long st_deny_bmap; + struct nfs4_stateid * st_openstp; }; /* flags for preprocess_seqid_op() */ -- cgit v1.2.1 From 86a76caf8705e3524e15f343f3c4806939a06dc8 Mon Sep 17 00:00:00 2001 From: Victor Fusco Date: Fri, 8 Jul 2005 14:57:47 -0700 Subject: [NET]: Fix sparse warnings From: Victor Fusco Fix the sparse warning "implicit cast to nocast type" Signed-off-by: Victor Fusco Signed-off-by: Domen Puncer Signed-off-by: David S. Miller --- include/linux/skbuff.h | 29 ++++++++++++++++++----------- 1 file changed, 18 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 14b950413495..5d4a990d5577 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -300,20 +300,26 @@ struct sk_buff { #include extern void __kfree_skb(struct sk_buff *skb); -extern struct sk_buff *alloc_skb(unsigned int size, int priority); +extern struct sk_buff *alloc_skb(unsigned int size, + unsigned int __nocast priority); extern struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, - unsigned int size, int priority); + unsigned int size, + unsigned int __nocast priority); extern void kfree_skbmem(struct sk_buff *skb); -extern struct sk_buff *skb_clone(struct sk_buff *skb, int priority); -extern struct sk_buff *skb_copy(const struct sk_buff *skb, int priority); -extern struct sk_buff *pskb_copy(struct sk_buff *skb, int gfp_mask); +extern struct sk_buff *skb_clone(struct sk_buff *skb, + unsigned int __nocast priority); +extern struct sk_buff *skb_copy(const struct sk_buff *skb, + unsigned int __nocast priority); +extern struct sk_buff *pskb_copy(struct sk_buff *skb, + unsigned int __nocast gfp_mask); extern int pskb_expand_head(struct sk_buff *skb, - int nhead, int ntail, int gfp_mask); + int nhead, int ntail, + unsigned int __nocast gfp_mask); extern struct sk_buff *skb_realloc_headroom(struct sk_buff *skb, unsigned int headroom); extern struct sk_buff *skb_copy_expand(const struct sk_buff *skb, int newheadroom, int newtailroom, - int priority); + unsigned int __nocast priority); extern struct sk_buff * skb_pad(struct sk_buff *skb, int pad); #define dev_kfree_skb(a) kfree_skb(a) extern void skb_over_panic(struct sk_buff *skb, int len, @@ -464,7 +470,8 @@ static inline int skb_shared(const struct sk_buff *skb) * * NULL is returned on a memory allocation failure. */ -static inline struct sk_buff *skb_share_check(struct sk_buff *skb, int pri) +static inline struct sk_buff *skb_share_check(struct sk_buff *skb, + unsigned int __nocast pri) { might_sleep_if(pri & __GFP_WAIT); if (skb_shared(skb)) { @@ -1001,7 +1008,7 @@ static inline void __skb_queue_purge(struct sk_buff_head *list) * %NULL is returned in there is no free memory. */ static inline struct sk_buff *__dev_alloc_skb(unsigned int length, - int gfp_mask) + unsigned int __nocast gfp_mask) { struct sk_buff *skb = alloc_skb(length + 16, gfp_mask); if (likely(skb)) @@ -1114,8 +1121,8 @@ static inline int skb_can_coalesce(struct sk_buff *skb, int i, * If there is no free memory -ENOMEM is returned, otherwise zero * is returned and the old skb data released. */ -extern int __skb_linearize(struct sk_buff *skb, int gfp); -static inline int skb_linearize(struct sk_buff *skb, int gfp) +extern int __skb_linearize(struct sk_buff *skb, unsigned int __nocast gfp); +static inline int skb_linearize(struct sk_buff *skb, unsigned int __nocast gfp) { return __skb_linearize(skb, gfp); } -- cgit v1.2.1 From ca9b907d140a5f249250d19f956129dbbbf84f73 Mon Sep 17 00:00:00 2001 From: David L Stevens Date: Fri, 8 Jul 2005 17:38:07 -0700 Subject: [IPV4]: multicast API "join" issues This patch corrects a few problems with the IP_ADD_MEMBERSHIP socket option: 1) The existing code makes an attempt at reference counting joins when using the ip_mreqn/imr_ifindex interface. Joining the same group on the same socket is an error, whatever the API. This leads to unexpected results when mixing ip_mreqn by index with ip_mreqn by address, ip_mreq, or other API's. For example, ip_mreq followed by ip_mreqn of the same group will "work" while the same two reversed will not. Fixed to always return EADDRINUSE on a duplicate join and removed the (now unused) reference count in ip_mc_socklist. 2) The group-search list in ip_mc_join_group() is comparing a full ip_mreqn structure and all of it must match for it to find the group. This doesn't correctly match a group that was joined with ip_mreq or ip_mreqn with an address (with or without an index). It also doesn't match groups that are joined by different addresses on the same interface. All of these are the same multicast group, which is identified by group address and interface index. Fixed the check to correctly match groups so we don't get duplicate group entries on the ip_mc_socklist. 3) The old code allocates a multicast address before searching for duplicates requiring it to free in various error cases. This patch moves the allocate until after the search and igmp_max_memberships check, so never a need to allocate, then free an entry. Signed-off-by: David L Stevens Signed-off-by: David S. Miller --- include/linux/igmp.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/igmp.h b/include/linux/igmp.h index 390e760a96d3..0c31ef0b5bad 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -148,7 +148,6 @@ struct ip_sf_socklist struct ip_mc_socklist { struct ip_mc_socklist *next; - int count; struct ip_mreqn multi; unsigned int sfmode; /* MCAST_{INCLUDE,EXCLUDE} */ struct ip_sf_socklist *sflist; -- cgit v1.2.1 From d369ddd2fc00fc3f46e9052d1017cbf407e3cdf7 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 10 Jul 2005 15:45:11 -0700 Subject: [SPARC64]: Add __read_mostly support. Signed-off-by: David S. Miller --- include/linux/cache.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cache.h b/include/linux/cache.h index 2b66a36d85f0..f6b5a46c5f82 100644 --- a/include/linux/cache.h +++ b/include/linux/cache.h @@ -13,7 +13,7 @@ #define SMP_CACHE_BYTES L1_CACHE_BYTES #endif -#ifdef CONFIG_X86 +#if defined(CONFIG_X86) || defined(CONFIG_SPARC64) #define __read_mostly __attribute__((__section__(".data.read_mostly"))) #else #define __read_mostly -- cgit v1.2.1 From f7ceba360cce9af3fbc4e5a5b1bd40b570b7021c Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 10 Jul 2005 19:29:45 -0700 Subject: [SPARC64]: Add syscall auditing support. Signed-off-by: David S. Miller --- include/linux/audit.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/audit.h b/include/linux/audit.h index bf2ad3ba72eb..68aba0c02e49 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -165,7 +165,7 @@ #define AUDIT_ARCH_SH64 (EM_SH|__AUDIT_ARCH_64BIT) #define AUDIT_ARCH_SHEL64 (EM_SH|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) #define AUDIT_ARCH_SPARC (EM_SPARC) -#define AUDIT_ARCH_SPARC64 (EM_SPARC64|__AUDIT_ARCH_64BIT) +#define AUDIT_ARCH_SPARC64 (EM_SPARCV9|__AUDIT_ARCH_64BIT) #define AUDIT_ARCH_V850 (EM_V850|__AUDIT_ARCH_LE) #define AUDIT_ARCH_X86_64 (EM_X86_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) -- cgit v1.2.1 From 63522f7fdb624adef20cb9d90c7effcd5b6301b2 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 11 Jul 2005 14:29:11 -0700 Subject: [NETLINK]: Reserve NETLINK_NETFILTER. Signed-off-by: David S. Miller --- include/linux/netlink.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 27e4d164a108..2f0c085f2c7d 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -16,6 +16,7 @@ #define NETLINK_AUDIT 9 /* auditing */ #define NETLINK_FIB_LOOKUP 10 #define NETLINK_ROUTE6 11 /* af_inet6 route comm channel */ +#define NETLINK_NETFILTER 12 /* netfilter subsystem */ #define NETLINK_IP6_FW 13 #define NETLINK_DNRTMSG 14 /* DECnet routing messages */ #define NETLINK_KOBJECT_UEVENT 15 /* Kernel messages to userspace */ -- cgit v1.2.1 From e2a5b420f716cd1a46674b1a90389612eced916f Mon Sep 17 00:00:00 2001 From: Alexey Starikovskiy Date: Fri, 18 Mar 2005 16:20:46 -0500 Subject: [ACPI] ACPI poweroff fix Register an "acpi" system device to be notified of shutdown preparation. This depends on CONFIG_PM http://bugzilla.kernel.org/show_bug.cgi?id=4041 Signed-off-by: Alexey Starikovskiy Signed-off-by: Andrew Morton Signed-off-by: Len Brown --- include/linux/pm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/pm.h b/include/linux/pm.h index ed2b76e75199..da88851266b8 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -175,7 +175,7 @@ struct pm_ops { }; extern void pm_set_ops(struct pm_ops *); - +extern struct pm_ops *pm_ops; extern int pm_suspend(suspend_state_t state); -- cgit v1.2.1 From 4e10d12a3d88c88fba3258809aa42d14fd8cf1d1 Mon Sep 17 00:00:00 2001 From: David Shaohua Li Date: Fri, 18 Mar 2005 18:45:35 -0500 Subject: [ACPI] Bind PCI devices with ACPI devices Implement the framework for binding physical devices with ACPI devices. A physical bus like PCI bus should create a 'acpi_bus_type', with: .find_device: For device which has parent such as normal PCI devices. .find_bridge: It's for special devices, such as PCI root bridge or IDE controller. Such devices generally haven't a parent or ->bus. We use the special method to get an ACPI handle. Uses new field in struct device: firmware_data http://bugzilla.kernel.org/show_bug.cgi?id=4277 Signed-off-by: David Shaohua Li Signed-off-by: Len Brown --- include/linux/device.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/device.h b/include/linux/device.h index df94c0de53f2..de2d6fe349de 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -269,8 +269,10 @@ struct device { struct device_driver *driver; /* which driver has allocated this device */ void *driver_data; /* data private to the driver */ - void *platform_data; /* Platform specific data (e.g. ACPI, - BIOS data relevant to device) */ + void *platform_data; /* Platform specific data, device + core doesn't touch it */ + void *firmware_data; /* Firmware specific data (e.g. ACPI, + BIOS data),reserved for device core*/ struct dev_pm_info power; u64 *dma_mask; /* dma mask (if dma'able device) */ -- cgit v1.2.1 From 55e59c511cea3c6c721971467c707e9955922bc2 Mon Sep 17 00:00:00 2001 From: Ashok Raj Date: Thu, 31 Mar 2005 22:51:10 -0500 Subject: [ACPI] Evaluate CPEI Processor Override flag ACPI 3.0 added a Correctable Platform Error Interrupt (CPEI) Processor Overide flag to MADT.Platform_Interrupt_Source. Record the processor that was provided as hint from ACPI. Signed-off-by: Ashok Raj Signed-off-by: Len Brown --- include/linux/acpi.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index b123cc08773d..70b3c52b75d7 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -206,7 +206,10 @@ struct acpi_table_plat_int_src { u8 eid; u8 iosapic_vector; u32 global_irq; - u32 reserved; + struct { + u32 cpei_override_flag:1; + u32 reserved:31; + } plint_flags; } __attribute__ ((packed)); enum acpi_interrupt_id { -- cgit v1.2.1 From c9c3e457de24cca2ca688fa397d93a241f472048 Mon Sep 17 00:00:00 2001 From: David Shaohua Li Date: Fri, 1 Apr 2005 00:07:31 -0500 Subject: [ACPI] PNPACPI vs sound IRQ http://bugme.osdl.org/show_bug.cgi?id=4016 Written-by: David Shaohua Li Acked-by: Adam Belay Signed-off-by: Len Brown --- include/linux/acpi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 70b3c52b75d7..9c14959bcfa0 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -465,7 +465,7 @@ struct acpi_prt_list { struct pci_dev; int acpi_pci_irq_enable (struct pci_dev *dev); -void acpi_penalize_isa_irq(int irq); +void acpi_penalize_isa_irq(int irq, int active); #ifdef CONFIG_ACPI_DEALLOCATE_IRQ void acpi_pci_irq_disable (struct pci_dev *dev); -- cgit v1.2.1 From 5db539e49fc7471e23bf3c94ca304f008cb7b7f3 Mon Sep 17 00:00:00 2001 From: Olav Kongas Date: Thu, 23 Jun 2005 20:25:36 +0300 Subject: [PATCH] USB: Fix kmalloc's flags type in USB Greg, This patch fixes the kmalloc() flags argument type in USB subsystem; hopefully all of its occurences. The patch was made against patch-2.6.12-git2 from Jun 20. Cleanup of flags for kmalloc() in USB subsystem. Signed-off-by: Olav Kongas Signed-off-by: Greg Kroah-Hartman --- include/linux/usb.h | 8 ++++---- include/linux/usb_gadget.h | 12 ++++++------ 2 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/usb.h b/include/linux/usb.h index eb282b581546..724637792996 100644 --- a/include/linux/usb.h +++ b/include/linux/usb.h @@ -938,17 +938,17 @@ static inline void usb_fill_int_urb (struct urb *urb, } extern void usb_init_urb(struct urb *urb); -extern struct urb *usb_alloc_urb(int iso_packets, int mem_flags); +extern struct urb *usb_alloc_urb(int iso_packets, unsigned mem_flags); extern void usb_free_urb(struct urb *urb); #define usb_put_urb usb_free_urb extern struct urb *usb_get_urb(struct urb *urb); -extern int usb_submit_urb(struct urb *urb, int mem_flags); +extern int usb_submit_urb(struct urb *urb, unsigned mem_flags); extern int usb_unlink_urb(struct urb *urb); extern void usb_kill_urb(struct urb *urb); #define HAVE_USB_BUFFERS void *usb_buffer_alloc (struct usb_device *dev, size_t size, - int mem_flags, dma_addr_t *dma); + unsigned mem_flags, dma_addr_t *dma); void usb_buffer_free (struct usb_device *dev, size_t size, void *addr, dma_addr_t dma); @@ -1055,7 +1055,7 @@ int usb_sg_init ( struct scatterlist *sg, int nents, size_t length, - int mem_flags + unsigned mem_flags ); void usb_sg_cancel (struct usb_sg_request *io); void usb_sg_wait (struct usb_sg_request *io); diff --git a/include/linux/usb_gadget.h b/include/linux/usb_gadget.h index b00f127cb447..71e608607324 100644 --- a/include/linux/usb_gadget.h +++ b/include/linux/usb_gadget.h @@ -107,18 +107,18 @@ struct usb_ep_ops { int (*disable) (struct usb_ep *ep); struct usb_request *(*alloc_request) (struct usb_ep *ep, - int gfp_flags); + unsigned gfp_flags); void (*free_request) (struct usb_ep *ep, struct usb_request *req); void *(*alloc_buffer) (struct usb_ep *ep, unsigned bytes, - dma_addr_t *dma, int gfp_flags); + dma_addr_t *dma, unsigned gfp_flags); void (*free_buffer) (struct usb_ep *ep, void *buf, dma_addr_t dma, unsigned bytes); // NOTE: on 2.6, drivers may also use dma_map() and // dma_sync_single_*() to directly manage dma overhead. int (*queue) (struct usb_ep *ep, struct usb_request *req, - int gfp_flags); + unsigned gfp_flags); int (*dequeue) (struct usb_ep *ep, struct usb_request *req); int (*set_halt) (struct usb_ep *ep, int value); @@ -214,7 +214,7 @@ usb_ep_disable (struct usb_ep *ep) * Returns the request, or null if one could not be allocated. */ static inline struct usb_request * -usb_ep_alloc_request (struct usb_ep *ep, int gfp_flags) +usb_ep_alloc_request (struct usb_ep *ep, unsigned gfp_flags) { return ep->ops->alloc_request (ep, gfp_flags); } @@ -254,7 +254,7 @@ usb_ep_free_request (struct usb_ep *ep, struct usb_request *req) */ static inline void * usb_ep_alloc_buffer (struct usb_ep *ep, unsigned len, dma_addr_t *dma, - int gfp_flags) + unsigned gfp_flags) { return ep->ops->alloc_buffer (ep, len, dma, gfp_flags); } @@ -330,7 +330,7 @@ usb_ep_free_buffer (struct usb_ep *ep, void *buf, dma_addr_t dma, unsigned len) * reported when the usb peripheral is disconnected. */ static inline int -usb_ep_queue (struct usb_ep *ep, struct usb_request *req, int gfp_flags) +usb_ep_queue (struct usb_ep *ep, struct usb_request *req, unsigned gfp_flags) { return ep->ops->queue (ep, req, gfp_flags); } -- cgit v1.2.1 From 00ab997dd24fff82900665449f859e23a78ad5f4 Mon Sep 17 00:00:00 2001 From: "david-b@pacbell.net" Date: Wed, 29 Jun 2005 07:04:14 -0700 Subject: [PATCH] USB: another cdc descriptor This adds another CDC descriptor type to ; the main claim to fame for this is that some Motorola phones include it. It's not currently needed by any driver code; included for completeness. Signed-off-by: David Brownell Signed-off-by: Greg Kroah-Hartman --- include/linux/usb_cdc.h | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'include/linux') diff --git a/include/linux/usb_cdc.h b/include/linux/usb_cdc.h index f22d6beecc73..ba617c372455 100644 --- a/include/linux/usb_cdc.h +++ b/include/linux/usb_cdc.h @@ -34,6 +34,7 @@ #define USB_CDC_ACM_TYPE 0x02 /* acm_descriptor */ #define USB_CDC_UNION_TYPE 0x06 /* union_desc */ #define USB_CDC_COUNTRY_TYPE 0x07 +#define USB_CDC_NETWORK_TERMINAL_TYPE 0x0a /* network_terminal_desc */ #define USB_CDC_ETHERNET_TYPE 0x0f /* ether_desc */ #define USB_CDC_WHCM_TYPE 0x11 #define USB_CDC_MDLM_TYPE 0x12 /* mdlm_desc */ @@ -83,6 +84,18 @@ struct usb_cdc_union_desc { /* ... and there could be other slave interfaces */ } __attribute__ ((packed)); +/* "Network Channel Terminal Functional Descriptor" from CDC spec 5.2.3.11 */ +struct usb_cdc_network_terminal_desc { + __u8 bLength; + __u8 bDescriptorType; + __u8 bDescriptorSubType; + + __u8 bEntityId; + __u8 iName; + __u8 bChannelIndex; + __u8 bPhysicalInterface; +} __attribute__ ((packed)); + /* "Ethernet Networking Functional Descriptor" from CDC spec 5.2.3.16 */ struct usb_cdc_ether_desc { __u8 bLength; -- cgit v1.2.1 From ab611487d8ada506e511d2b8f22fb8e7be9939b9 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Tue, 12 Jul 2005 12:08:43 -0700 Subject: [NET]: __be'ify *_type_trans() tr_type_trans(), hippi_type_trans() left as-is. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/etherdevice.h | 2 +- include/linux/fddidevice.h | 2 +- include/linux/hdlc.h | 4 ++-- include/linux/wanrouter.h | 3 +-- 4 files changed, 5 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index cf3847edc50f..ce8518e658b6 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -33,7 +33,7 @@ extern int eth_header(struct sk_buff *skb, struct net_device *dev, unsigned short type, void *daddr, void *saddr, unsigned len); extern int eth_rebuild_header(struct sk_buff *skb); -extern unsigned short eth_type_trans(struct sk_buff *skb, struct net_device *dev); +extern __be16 eth_type_trans(struct sk_buff *skb, struct net_device *dev); extern void eth_header_cache_update(struct hh_cache *hh, struct net_device *dev, unsigned char * haddr); extern int eth_header_cache(struct neighbour *neigh, diff --git a/include/linux/fddidevice.h b/include/linux/fddidevice.h index 002f6367697d..e61e42dfd317 100644 --- a/include/linux/fddidevice.h +++ b/include/linux/fddidevice.h @@ -25,7 +25,7 @@ #include #ifdef __KERNEL__ -extern unsigned short fddi_type_trans(struct sk_buff *skb, +extern __be16 fddi_type_trans(struct sk_buff *skb, struct net_device *dev); extern struct net_device *alloc_fddidev(int sizeof_priv); #endif diff --git a/include/linux/hdlc.h b/include/linux/hdlc.h index ed2927ef1ff7..df695e9ae327 100644 --- a/include/linux/hdlc.h +++ b/include/linux/hdlc.h @@ -242,8 +242,8 @@ static __inline__ struct net_device_stats *hdlc_stats(struct net_device *dev) } -static __inline__ unsigned short hdlc_type_trans(struct sk_buff *skb, - struct net_device *dev) +static __inline__ __be16 hdlc_type_trans(struct sk_buff *skb, + struct net_device *dev) { hdlc_device *hdlc = dev_to_hdlc(dev); diff --git a/include/linux/wanrouter.h b/include/linux/wanrouter.h index 3e89f0f15f49..1b6b76a4eb54 100644 --- a/include/linux/wanrouter.h +++ b/include/linux/wanrouter.h @@ -516,8 +516,7 @@ struct wan_device { /* Public functions available for device drivers */ extern int register_wan_device(struct wan_device *wandev); extern int unregister_wan_device(char *name); -unsigned short wanrouter_type_trans(struct sk_buff *skb, - struct net_device *dev); +__be16 wanrouter_type_trans(struct sk_buff *skb, struct net_device *dev); int wanrouter_encapsulate(struct sk_buff *skb, struct net_device *dev, unsigned short type); -- cgit v1.2.1 From d53d9f16ea95a91ad4aa114809dcde486ca4000d Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 12 Jul 2005 13:58:07 -0700 Subject: [PATCH] name_to_dev_t warning fix kernel/power/disk.c needs a declaration of name_to_dev_t() in scope. mount.h seems like an appropriate choice. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mount.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/mount.h b/include/linux/mount.h index 74b4727a4e30..f8f39937e301 100644 --- a/include/linux/mount.h +++ b/include/linux/mount.h @@ -12,6 +12,7 @@ #define _LINUX_MOUNT_H #ifdef __KERNEL__ +#include #include #include #include @@ -76,6 +77,7 @@ extern int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd, extern void mark_mounts_for_expiry(struct list_head *mounts); extern spinlock_t vfsmount_lock; +extern dev_t name_to_dev_t(char *name); #endif #endif /* _LINUX_MOUNT_H */ -- cgit v1.2.1 From 08c6a96fd77836856c090ebb39beadc81cb8484d Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 12 Jul 2005 13:58:28 -0700 Subject: [PATCH] ext3: fix options parsing Fix a problem with ext3 mount option parsing. When remount of a filesystem fails, old options are now restored. Signed-off-by: Jan Kara Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ext3_fs.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) (limited to 'include/linux') diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h index 4b6e1ab216a5..c16662836c58 100644 --- a/include/linux/ext3_fs.h +++ b/include/linux/ext3_fs.h @@ -238,6 +238,20 @@ struct ext3_new_group_data { #define EXT3_IOC_GETRSVSZ _IOR('f', 5, long) #define EXT3_IOC_SETRSVSZ _IOW('f', 6, long) +/* + * Mount options + */ +struct ext3_mount_options { + unsigned long s_mount_opt; + uid_t s_resuid; + gid_t s_resgid; + unsigned long s_commit_interval; +#ifdef CONFIG_QUOTA + int s_jquota_fmt; + char *s_qf_names[MAXQUOTAS]; +#endif +}; + /* * Structure of an inode on the disk */ -- cgit v1.2.1 From 542d1c88bd7f73e2e59d41b12e4a9041deea89e4 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 12 Jul 2005 13:58:31 -0700 Subject: [PATCH] tlb.h warning fix free_pages_and_swap_cache() and free_page_and_swap_cache() use release_pages() and page_cache_release() respectively, so make sure that we have the declarations in scope. Cc: Olaf Hering Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/swap.h b/include/linux/swap.h index c75954f2d868..239f520cc49e 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -7,6 +7,8 @@ #include #include #include +#include + #include #include -- cgit v1.2.1 From 67bc4eb0b1140a4bf364f2dcca152be659ed9057 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 12 Jul 2005 13:58:36 -0700 Subject: [PATCH] hardirq uses preempt hardirq.h uses preempt_count() from preempt.h Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/hardirq.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index 8336dba18971..5912874ca83c 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -2,6 +2,7 @@ #define LINUX_HARDIRQ_H #include +#include #include #include #include -- cgit v1.2.1 From bd4c625c061c2a38568d0add3478f59172455159 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 12 Jul 2005 20:21:28 -0700 Subject: reiserfs: run scripts/Lindent on reiserfs code This was a pure indentation change, using: scripts/Lindent fs/reiserfs/*.c include/linux/reiserfs_*.h to make reiserfs match the regular Linux indentation style. As Jeff Mahoney writes: The ReiserFS code is a mix of a number of different coding styles, sometimes different even from line-to-line. Since the code has been relatively stable for quite some time and there are few outstanding patches to be applied, it is time to reformat the code to conform to the Linux style standard outlined in Documentation/CodingStyle. This patch contains the result of running scripts/Lindent against fs/reiserfs/*.c and include/linux/reiserfs_*.h. There are places where the code can be made to look better, but I'd rather keep those patches separate so that there isn't a subtle by-hand hand accident in the middle of a huge patch. To be clear: This patch is reformatting *only*. A number of patches may follow that continue to make the code more consistent with the Linux coding style. Hans wasn't particularly enthusiastic about these patches, but said he wouldn't really oppose them either. Signed-off-by: Linus Torvalds --- include/linux/reiserfs_acl.h | 52 +- include/linux/reiserfs_fs.h | 1595 ++++++++++++++++++++-------------------- include/linux/reiserfs_fs_i.h | 59 +- include/linux/reiserfs_fs_sb.h | 616 ++++++++-------- include/linux/reiserfs_xattr.h | 126 ++-- 5 files changed, 1207 insertions(+), 1241 deletions(-) (limited to 'include/linux') diff --git a/include/linux/reiserfs_acl.h b/include/linux/reiserfs_acl.h index 0760507a545b..0a3605099c44 100644 --- a/include/linux/reiserfs_acl.h +++ b/include/linux/reiserfs_acl.h @@ -4,29 +4,29 @@ #define REISERFS_ACL_VERSION 0x0001 typedef struct { - __le16 e_tag; - __le16 e_perm; - __le32 e_id; + __le16 e_tag; + __le16 e_perm; + __le32 e_id; } reiserfs_acl_entry; typedef struct { - __le16 e_tag; - __le16 e_perm; + __le16 e_tag; + __le16 e_perm; } reiserfs_acl_entry_short; typedef struct { - __le32 a_version; + __le32 a_version; } reiserfs_acl_header; static inline size_t reiserfs_acl_size(int count) { if (count <= 4) { return sizeof(reiserfs_acl_header) + - count * sizeof(reiserfs_acl_entry_short); + count * sizeof(reiserfs_acl_entry_short); } else { return sizeof(reiserfs_acl_header) + - 4 * sizeof(reiserfs_acl_entry_short) + - (count - 4) * sizeof(reiserfs_acl_entry); + 4 * sizeof(reiserfs_acl_entry_short) + + (count - 4) * sizeof(reiserfs_acl_entry); } } @@ -46,14 +46,14 @@ static inline int reiserfs_acl_count(size_t size) } } - #ifdef CONFIG_REISERFS_FS_POSIX_ACL -struct posix_acl * reiserfs_get_acl(struct inode *inode, int type); -int reiserfs_acl_chmod (struct inode *inode); -int reiserfs_inherit_default_acl (struct inode *dir, struct dentry *dentry, struct inode *inode); -int reiserfs_cache_default_acl (struct inode *dir); -extern int reiserfs_xattr_posix_acl_init (void) __init; -extern int reiserfs_xattr_posix_acl_exit (void); +struct posix_acl *reiserfs_get_acl(struct inode *inode, int type); +int reiserfs_acl_chmod(struct inode *inode); +int reiserfs_inherit_default_acl(struct inode *dir, struct dentry *dentry, + struct inode *inode); +int reiserfs_cache_default_acl(struct inode *dir); +extern int reiserfs_xattr_posix_acl_init(void) __init; +extern int reiserfs_xattr_posix_acl_exit(void); extern struct reiserfs_xattr_handler posix_acl_default_handler; extern struct reiserfs_xattr_handler posix_acl_access_handler; #else @@ -61,28 +61,26 @@ extern struct reiserfs_xattr_handler posix_acl_access_handler; #define reiserfs_get_acl NULL #define reiserfs_cache_default_acl(inode) 0 -static inline int -reiserfs_xattr_posix_acl_init (void) +static inline int reiserfs_xattr_posix_acl_init(void) { - return 0; + return 0; } -static inline int -reiserfs_xattr_posix_acl_exit (void) +static inline int reiserfs_xattr_posix_acl_exit(void) { - return 0; + return 0; } -static inline int -reiserfs_acl_chmod (struct inode *inode) +static inline int reiserfs_acl_chmod(struct inode *inode) { - return 0; + return 0; } static inline int -reiserfs_inherit_default_acl (const struct inode *dir, struct dentry *dentry, struct inode *inode) +reiserfs_inherit_default_acl(const struct inode *dir, struct dentry *dentry, + struct inode *inode) { - return 0; + return 0; } #endif diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h index 4c7c5689ad93..17e458e17e2b 100644 --- a/include/linux/reiserfs_fs.h +++ b/include/linux/reiserfs_fs.h @@ -3,11 +3,10 @@ */ /* this file has an amazingly stupid - name, yura please fix it to be - reiserfs.h, and merge all the rest - of our .h files that are in this - directory into it. */ - + name, yura please fix it to be + reiserfs.h, and merge all the rest + of our .h files that are in this + directory into it. */ #ifndef _LINUX_REISER_FS_H #define _LINUX_REISER_FS_H @@ -74,9 +73,9 @@ /* debug levels. Right now, CONFIG_REISERFS_CHECK means print all debug ** messages. */ -#define REISERFS_DEBUG_CODE 5 /* extra messages to help find/debug errors */ +#define REISERFS_DEBUG_CODE 5 /* extra messages to help find/debug errors */ -void reiserfs_warning (struct super_block *s, const char * fmt, ...); +void reiserfs_warning(struct super_block *s, const char *fmt, ...); /* assertions handling */ /** always check a condition and panic if it's false. */ @@ -105,82 +104,78 @@ if( !( cond ) ) \ * Structure of super block on disk, a version of which in RAM is often accessed as REISERFS_SB(s)->s_rs * the version in RAM is part of a larger structure containing fields never written to disk. */ -#define UNSET_HASH 0 // read_super will guess about, what hash names - // in directories were sorted with +#define UNSET_HASH 0 // read_super will guess about, what hash names + // in directories were sorted with #define TEA_HASH 1 #define YURA_HASH 2 #define R5_HASH 3 #define DEFAULT_HASH R5_HASH - struct journal_params { - __le32 jp_journal_1st_block; /* where does journal start from on its - * device */ - __le32 jp_journal_dev; /* journal device st_rdev */ - __le32 jp_journal_size; /* size of the journal */ - __le32 jp_journal_trans_max; /* max number of blocks in a transaction. */ - __le32 jp_journal_magic; /* random value made on fs creation (this - * was sb_journal_block_count) */ - __le32 jp_journal_max_batch; /* max number of blocks to batch into a - * trans */ - __le32 jp_journal_max_commit_age; /* in seconds, how old can an async - * commit be */ - __le32 jp_journal_max_trans_age; /* in seconds, how old can a transaction - * be */ + __le32 jp_journal_1st_block; /* where does journal start from on its + * device */ + __le32 jp_journal_dev; /* journal device st_rdev */ + __le32 jp_journal_size; /* size of the journal */ + __le32 jp_journal_trans_max; /* max number of blocks in a transaction. */ + __le32 jp_journal_magic; /* random value made on fs creation (this + * was sb_journal_block_count) */ + __le32 jp_journal_max_batch; /* max number of blocks to batch into a + * trans */ + __le32 jp_journal_max_commit_age; /* in seconds, how old can an async + * commit be */ + __le32 jp_journal_max_trans_age; /* in seconds, how old can a transaction + * be */ }; /* this is the super from 3.5.X, where X >= 10 */ -struct reiserfs_super_block_v1 -{ - __le32 s_block_count; /* blocks count */ - __le32 s_free_blocks; /* free blocks count */ - __le32 s_root_block; /* root block number */ - struct journal_params s_journal; - __le16 s_blocksize; /* block size */ - __le16 s_oid_maxsize; /* max size of object id array, see - * get_objectid() commentary */ - __le16 s_oid_cursize; /* current size of object id array */ - __le16 s_umount_state; /* this is set to 1 when filesystem was - * umounted, to 2 - when not */ - char s_magic[10]; /* reiserfs magic string indicates that - * file system is reiserfs: - * "ReIsErFs" or "ReIsEr2Fs" or "ReIsEr3Fs" */ - __le16 s_fs_state; /* it is set to used by fsck to mark which - * phase of rebuilding is done */ - __le32 s_hash_function_code; /* indicate, what hash function is being use - * to sort names in a directory*/ - __le16 s_tree_height; /* height of disk tree */ - __le16 s_bmap_nr; /* amount of bitmap blocks needed to address - * each block of file system */ - __le16 s_version; /* this field is only reliable on filesystem - * with non-standard journal */ - __le16 s_reserved_for_journal; /* size in blocks of journal area on main - * device, we need to keep after - * making fs with non-standard journal */ +struct reiserfs_super_block_v1 { + __le32 s_block_count; /* blocks count */ + __le32 s_free_blocks; /* free blocks count */ + __le32 s_root_block; /* root block number */ + struct journal_params s_journal; + __le16 s_blocksize; /* block size */ + __le16 s_oid_maxsize; /* max size of object id array, see + * get_objectid() commentary */ + __le16 s_oid_cursize; /* current size of object id array */ + __le16 s_umount_state; /* this is set to 1 when filesystem was + * umounted, to 2 - when not */ + char s_magic[10]; /* reiserfs magic string indicates that + * file system is reiserfs: + * "ReIsErFs" or "ReIsEr2Fs" or "ReIsEr3Fs" */ + __le16 s_fs_state; /* it is set to used by fsck to mark which + * phase of rebuilding is done */ + __le32 s_hash_function_code; /* indicate, what hash function is being use + * to sort names in a directory*/ + __le16 s_tree_height; /* height of disk tree */ + __le16 s_bmap_nr; /* amount of bitmap blocks needed to address + * each block of file system */ + __le16 s_version; /* this field is only reliable on filesystem + * with non-standard journal */ + __le16 s_reserved_for_journal; /* size in blocks of journal area on main + * device, we need to keep after + * making fs with non-standard journal */ } __attribute__ ((__packed__)); #define SB_SIZE_V1 (sizeof(struct reiserfs_super_block_v1)) /* this is the on disk super block */ -struct reiserfs_super_block -{ - struct reiserfs_super_block_v1 s_v1; - __le32 s_inode_generation; - __le32 s_flags; /* Right now used only by inode-attributes, if enabled */ - unsigned char s_uuid[16]; /* filesystem unique identifier */ - unsigned char s_label[16]; /* filesystem volume label */ - char s_unused[88] ; /* zero filled by mkreiserfs and - * reiserfs_convert_objectid_map_v1() - * so any additions must be updated - * there as well. */ -} __attribute__ ((__packed__)); +struct reiserfs_super_block { + struct reiserfs_super_block_v1 s_v1; + __le32 s_inode_generation; + __le32 s_flags; /* Right now used only by inode-attributes, if enabled */ + unsigned char s_uuid[16]; /* filesystem unique identifier */ + unsigned char s_label[16]; /* filesystem volume label */ + char s_unused[88]; /* zero filled by mkreiserfs and + * reiserfs_convert_objectid_map_v1() + * so any additions must be updated + * there as well. */ +} __attribute__ ((__packed__)); #define SB_SIZE (sizeof(struct reiserfs_super_block)) #define REISERFS_VERSION_1 0 #define REISERFS_VERSION_2 2 - // on-disk super block fields converted to cpu form #define SB_DISK_SUPER_BLOCK(s) (REISERFS_SB(s)->s_rs) #define SB_V1_DISK_SUPER_BLOCK(s) (&(SB_DISK_SUPER_BLOCK(s)->s_v1)) @@ -210,13 +205,12 @@ struct reiserfs_super_block #define PUT_SB_TREE_HEIGHT(s, val) \ do { SB_V1_DISK_SUPER_BLOCK(s)->s_tree_height = cpu_to_le16(val); } while (0) #define PUT_SB_REISERFS_STATE(s, val) \ - do { SB_V1_DISK_SUPER_BLOCK(s)->s_umount_state = cpu_to_le16(val); } while (0) + do { SB_V1_DISK_SUPER_BLOCK(s)->s_umount_state = cpu_to_le16(val); } while (0) #define PUT_SB_VERSION(s, val) \ do { SB_V1_DISK_SUPER_BLOCK(s)->s_version = cpu_to_le16(val); } while (0) #define PUT_SB_BMAP_NR(s, val) \ do { SB_V1_DISK_SUPER_BLOCK(s)->s_bmap_nr = cpu_to_le16 (val); } while (0) - #define SB_ONDISK_JP(s) (&SB_V1_DISK_SUPER_BLOCK(s)->s_journal) #define SB_ONDISK_JOURNAL_SIZE(s) \ le32_to_cpu ((SB_ONDISK_JP(s)->jp_journal_size)) @@ -231,21 +225,19 @@ struct reiserfs_super_block block >= SB_JOURNAL_1st_RESERVED_BLOCK(s) \ && block < SB_JOURNAL_1st_RESERVED_BLOCK(s) + \ ((!is_reiserfs_jr(SB_DISK_SUPER_BLOCK(s)) ? \ - SB_ONDISK_JOURNAL_SIZE(s) + 1 : SB_ONDISK_RESERVED_FOR_JOURNAL(s))) - - + SB_ONDISK_JOURNAL_SIZE(s) + 1 : SB_ONDISK_RESERVED_FOR_JOURNAL(s))) /* used by gcc */ #define REISERFS_SUPER_MAGIC 0x52654973 /* used by file system utilities that - look at the superblock, etc. */ + look at the superblock, etc. */ #define REISERFS_SUPER_MAGIC_STRING "ReIsErFs" #define REISER2FS_SUPER_MAGIC_STRING "ReIsEr2Fs" #define REISER2FS_JR_SUPER_MAGIC_STRING "ReIsEr3Fs" -int is_reiserfs_3_5 (struct reiserfs_super_block * rs); -int is_reiserfs_3_6 (struct reiserfs_super_block * rs); -int is_reiserfs_jr (struct reiserfs_super_block * rs); +int is_reiserfs_3_5(struct reiserfs_super_block *rs); +int is_reiserfs_3_6(struct reiserfs_super_block *rs); +int is_reiserfs_jr(struct reiserfs_super_block *rs); /* ReiserFS leaves the first 64k unused, so that partition labels have enough space. If someone wants to write a fancy bootloader that @@ -272,8 +264,8 @@ typedef __u32 b_blocknr_t; typedef __le32 unp_t; struct unfm_nodeinfo { - unp_t unfm_nodenum; - unsigned short unfm_freespace; + unp_t unfm_nodenum; + unsigned short unfm_freespace; }; /* there are two formats of keys: 3.5 and 3.6 @@ -285,7 +277,6 @@ struct unfm_nodeinfo { #define STAT_DATA_V1 0 #define STAT_DATA_V2 1 - static inline struct reiserfs_inode_info *REISERFS_I(const struct inode *inode) { return container_of(inode, struct reiserfs_inode_info, vfs_inode); @@ -343,15 +334,13 @@ static inline struct reiserfs_sb_info *REISERFS_SB(const struct super_block *sb) file would fit into one DIRECT item. Primary intention for this one is to increase performance by decreasing seeking. -*/ +*/ #define STORE_TAIL_IN_UNFM_S2(n_file_size,n_tail_size,n_block_size) \ (\ (!(n_tail_size)) || \ (((n_file_size) > MAX_DIRECT_ITEM_LEN(n_block_size)) ) \ ) - - /* * values for s_umount_state field */ @@ -364,9 +353,9 @@ static inline struct reiserfs_sb_info *REISERFS_SB(const struct super_block *sb) #define TYPE_STAT_DATA 0 #define TYPE_INDIRECT 1 #define TYPE_DIRECT 2 -#define TYPE_DIRENTRY 3 -#define TYPE_MAXTYPE 3 -#define TYPE_ANY 15 // FIXME: comment is required +#define TYPE_DIRENTRY 3 +#define TYPE_MAXTYPE 3 +#define TYPE_ANY 15 // FIXME: comment is required /***************************************************************************/ /* KEY & ITEM HEAD */ @@ -376,60 +365,62 @@ static inline struct reiserfs_sb_info *REISERFS_SB(const struct super_block *sb) // directories use this key as well as old files // struct offset_v1 { - __le32 k_offset; - __le32 k_uniqueness; + __le32 k_offset; + __le32 k_uniqueness; } __attribute__ ((__packed__)); struct offset_v2 { __le64 v; } __attribute__ ((__packed__)); -static inline __u16 offset_v2_k_type( const struct offset_v2 *v2 ) +static inline __u16 offset_v2_k_type(const struct offset_v2 *v2) { __u8 type = le64_to_cpu(v2->v) >> 60; - return (type <= TYPE_MAXTYPE)?type:TYPE_ANY; + return (type <= TYPE_MAXTYPE) ? type : TYPE_ANY; } - -static inline void set_offset_v2_k_type( struct offset_v2 *v2, int type ) + +static inline void set_offset_v2_k_type(struct offset_v2 *v2, int type) { - v2->v = (v2->v & cpu_to_le64(~0ULL>>4)) | cpu_to_le64((__u64)type<<60); + v2->v = + (v2->v & cpu_to_le64(~0ULL >> 4)) | cpu_to_le64((__u64) type << 60); } - -static inline loff_t offset_v2_k_offset( const struct offset_v2 *v2 ) + +static inline loff_t offset_v2_k_offset(const struct offset_v2 *v2) { - return le64_to_cpu(v2->v) & (~0ULL>>4); + return le64_to_cpu(v2->v) & (~0ULL >> 4); } -static inline void set_offset_v2_k_offset( struct offset_v2 *v2, loff_t offset ){ - offset &= (~0ULL>>4); - v2->v = (v2->v & cpu_to_le64(15ULL<<60)) | cpu_to_le64(offset); +static inline void set_offset_v2_k_offset(struct offset_v2 *v2, loff_t offset) +{ + offset &= (~0ULL >> 4); + v2->v = (v2->v & cpu_to_le64(15ULL << 60)) | cpu_to_le64(offset); } /* Key of an item determines its location in the S+tree, and is composed of 4 components */ struct reiserfs_key { - __le32 k_dir_id; /* packing locality: by default parent - directory object id */ - __le32 k_objectid; /* object identifier */ - union { - struct offset_v1 k_offset_v1; - struct offset_v2 k_offset_v2; - } __attribute__ ((__packed__)) u; + __le32 k_dir_id; /* packing locality: by default parent + directory object id */ + __le32 k_objectid; /* object identifier */ + union { + struct offset_v1 k_offset_v1; + struct offset_v2 k_offset_v2; + } __attribute__ ((__packed__)) u; } __attribute__ ((__packed__)); struct in_core_key { - __u32 k_dir_id; /* packing locality: by default parent - directory object id */ - __u32 k_objectid; /* object identifier */ - __u64 k_offset; - __u8 k_type; + __u32 k_dir_id; /* packing locality: by default parent + directory object id */ + __u32 k_objectid; /* object identifier */ + __u64 k_offset; + __u8 k_type; }; struct cpu_key { - struct in_core_key on_disk_key; - int version; - int key_length; /* 3 in all cases but direct2indirect and - indirect2direct conversion */ + struct in_core_key on_disk_key; + int version; + int key_length; /* 3 in all cases but direct2indirect and + indirect2direct conversion */ }; /* Our function for comparing keys can compare keys of different @@ -475,8 +466,7 @@ struct cpu_key { indirect items) and specifies the location of the item itself within the block. */ -struct item_head -{ +struct item_head { /* Everything in the tree is found by searching for it based on * its key.*/ struct reiserfs_key ih_key; @@ -492,13 +482,13 @@ struct item_head number of directory entries in the directory item. */ __le16 ih_entry_count; } __attribute__ ((__packed__)) u; - __le16 ih_item_len; /* total size of the item body */ - __le16 ih_item_location; /* an offset to the item body - * within the block */ - __le16 ih_version; /* 0 for all old items, 2 for new - ones. Highest bit is set by fsck - temporary, cleaned after all - done */ + __le16 ih_item_len; /* total size of the item body */ + __le16 ih_item_location; /* an offset to the item body + * within the block */ + __le16 ih_version; /* 0 for all old items, 2 for new + ones. Highest bit is set by fsck + temporary, cleaned after all + done */ } __attribute__ ((__packed__)); /* size of item header */ #define IH_SIZE (sizeof(struct item_head)) @@ -515,7 +505,6 @@ struct item_head #define put_ih_location(ih, val) do { (ih)->ih_item_location = cpu_to_le16(val); } while (0) #define put_ih_item_len(ih, val) do { (ih)->ih_item_len = cpu_to_le16(val); } while (0) - #define unreachable_item(ih) (ih_version(ih) & (1 << 15)) #define get_ih_free_space(ih) (ih_version (ih) == KEY_FORMAT_3_6 ? 0 : ih_free_space (ih)) @@ -537,40 +526,48 @@ struct item_head #define V1_INDIRECT_UNIQUENESS 0xfffffffe #define V1_DIRECT_UNIQUENESS 0xffffffff #define V1_DIRENTRY_UNIQUENESS 500 -#define V1_ANY_UNIQUENESS 555 // FIXME: comment is required +#define V1_ANY_UNIQUENESS 555 // FIXME: comment is required // // here are conversion routines // -static inline int uniqueness2type (__u32 uniqueness) CONSTF; -static inline int uniqueness2type (__u32 uniqueness) +static inline int uniqueness2type(__u32 uniqueness) CONSTF; +static inline int uniqueness2type(__u32 uniqueness) { - switch ((int)uniqueness) { - case V1_SD_UNIQUENESS: return TYPE_STAT_DATA; - case V1_INDIRECT_UNIQUENESS: return TYPE_INDIRECT; - case V1_DIRECT_UNIQUENESS: return TYPE_DIRECT; - case V1_DIRENTRY_UNIQUENESS: return TYPE_DIRENTRY; - default: - reiserfs_warning (NULL, "vs-500: unknown uniqueness %d", - uniqueness); + switch ((int)uniqueness) { + case V1_SD_UNIQUENESS: + return TYPE_STAT_DATA; + case V1_INDIRECT_UNIQUENESS: + return TYPE_INDIRECT; + case V1_DIRECT_UNIQUENESS: + return TYPE_DIRECT; + case V1_DIRENTRY_UNIQUENESS: + return TYPE_DIRENTRY; + default: + reiserfs_warning(NULL, "vs-500: unknown uniqueness %d", + uniqueness); case V1_ANY_UNIQUENESS: - return TYPE_ANY; - } + return TYPE_ANY; + } } -static inline __u32 type2uniqueness (int type) CONSTF; -static inline __u32 type2uniqueness (int type) +static inline __u32 type2uniqueness(int type) CONSTF; +static inline __u32 type2uniqueness(int type) { - switch (type) { - case TYPE_STAT_DATA: return V1_SD_UNIQUENESS; - case TYPE_INDIRECT: return V1_INDIRECT_UNIQUENESS; - case TYPE_DIRECT: return V1_DIRECT_UNIQUENESS; - case TYPE_DIRENTRY: return V1_DIRENTRY_UNIQUENESS; - default: - reiserfs_warning (NULL, "vs-501: unknown type %d", type); + switch (type) { + case TYPE_STAT_DATA: + return V1_SD_UNIQUENESS; + case TYPE_INDIRECT: + return V1_INDIRECT_UNIQUENESS; + case TYPE_DIRECT: + return V1_DIRECT_UNIQUENESS; + case TYPE_DIRENTRY: + return V1_DIRENTRY_UNIQUENESS; + default: + reiserfs_warning(NULL, "vs-501: unknown type %d", type); case TYPE_ANY: - return V1_ANY_UNIQUENESS; - } + return V1_ANY_UNIQUENESS; + } } // @@ -578,57 +575,56 @@ static inline __u32 type2uniqueness (int type) // there is no way to get version of object from key, so, provide // version to these defines // -static inline loff_t le_key_k_offset (int version, const struct reiserfs_key * key) +static inline loff_t le_key_k_offset(int version, + const struct reiserfs_key *key) { - return (version == KEY_FORMAT_3_5) ? - le32_to_cpu( key->u.k_offset_v1.k_offset ) : - offset_v2_k_offset( &(key->u.k_offset_v2) ); + return (version == KEY_FORMAT_3_5) ? + le32_to_cpu(key->u.k_offset_v1.k_offset) : + offset_v2_k_offset(&(key->u.k_offset_v2)); } -static inline loff_t le_ih_k_offset (const struct item_head * ih) +static inline loff_t le_ih_k_offset(const struct item_head *ih) { - return le_key_k_offset (ih_version (ih), &(ih->ih_key)); + return le_key_k_offset(ih_version(ih), &(ih->ih_key)); } -static inline loff_t le_key_k_type (int version, const struct reiserfs_key * key) +static inline loff_t le_key_k_type(int version, const struct reiserfs_key *key) { - return (version == KEY_FORMAT_3_5) ? - uniqueness2type( le32_to_cpu( key->u.k_offset_v1.k_uniqueness)) : - offset_v2_k_type( &(key->u.k_offset_v2) ); + return (version == KEY_FORMAT_3_5) ? + uniqueness2type(le32_to_cpu(key->u.k_offset_v1.k_uniqueness)) : + offset_v2_k_type(&(key->u.k_offset_v2)); } -static inline loff_t le_ih_k_type (const struct item_head * ih) +static inline loff_t le_ih_k_type(const struct item_head *ih) { - return le_key_k_type (ih_version (ih), &(ih->ih_key)); + return le_key_k_type(ih_version(ih), &(ih->ih_key)); } - -static inline void set_le_key_k_offset (int version, struct reiserfs_key * key, loff_t offset) +static inline void set_le_key_k_offset(int version, struct reiserfs_key *key, + loff_t offset) { - (version == KEY_FORMAT_3_5) ? - (void)(key->u.k_offset_v1.k_offset = cpu_to_le32 (offset)) : /* jdm check */ - (void)(set_offset_v2_k_offset( &(key->u.k_offset_v2), offset )); + (version == KEY_FORMAT_3_5) ? (void)(key->u.k_offset_v1.k_offset = cpu_to_le32(offset)) : /* jdm check */ + (void)(set_offset_v2_k_offset(&(key->u.k_offset_v2), offset)); } - -static inline void set_le_ih_k_offset (struct item_head * ih, loff_t offset) +static inline void set_le_ih_k_offset(struct item_head *ih, loff_t offset) { - set_le_key_k_offset (ih_version (ih), &(ih->ih_key), offset); + set_le_key_k_offset(ih_version(ih), &(ih->ih_key), offset); } - -static inline void set_le_key_k_type (int version, struct reiserfs_key * key, int type) +static inline void set_le_key_k_type(int version, struct reiserfs_key *key, + int type) { - (version == KEY_FORMAT_3_5) ? - (void)(key->u.k_offset_v1.k_uniqueness = cpu_to_le32(type2uniqueness(type))): - (void)(set_offset_v2_k_type( &(key->u.k_offset_v2), type )); + (version == KEY_FORMAT_3_5) ? + (void)(key->u.k_offset_v1.k_uniqueness = + cpu_to_le32(type2uniqueness(type))) + : (void)(set_offset_v2_k_type(&(key->u.k_offset_v2), type)); } -static inline void set_le_ih_k_type (struct item_head * ih, int type) +static inline void set_le_ih_k_type(struct item_head *ih, int type) { - set_le_key_k_type (ih_version (ih), &(ih->ih_key), type); + set_le_key_k_type(ih_version(ih), &(ih->ih_key), type); } - #define is_direntry_le_key(version,key) (le_key_k_type (version, key) == TYPE_DIRENTRY) #define is_direct_le_key(version,key) (le_key_k_type (version, key) == TYPE_DIRECT) #define is_indirect_le_key(version,key) (le_key_k_type (version, key) == TYPE_INDIRECT) @@ -642,34 +638,32 @@ static inline void set_le_ih_k_type (struct item_head * ih, int type) #define is_indirect_le_ih(ih) is_indirect_le_key (ih_version(ih), &((ih)->ih_key)) #define is_statdata_le_ih(ih) is_statdata_le_key (ih_version (ih), &((ih)->ih_key)) - - // // key is pointer to cpu key, result is cpu // -static inline loff_t cpu_key_k_offset (const struct cpu_key * key) +static inline loff_t cpu_key_k_offset(const struct cpu_key *key) { - return key->on_disk_key.k_offset; + return key->on_disk_key.k_offset; } -static inline loff_t cpu_key_k_type (const struct cpu_key * key) +static inline loff_t cpu_key_k_type(const struct cpu_key *key) { - return key->on_disk_key.k_type; + return key->on_disk_key.k_type; } -static inline void set_cpu_key_k_offset (struct cpu_key * key, loff_t offset) +static inline void set_cpu_key_k_offset(struct cpu_key *key, loff_t offset) { key->on_disk_key.k_offset = offset; } -static inline void set_cpu_key_k_type (struct cpu_key * key, int type) +static inline void set_cpu_key_k_type(struct cpu_key *key, int type) { key->on_disk_key.k_type = type; } -static inline void cpu_key_k_offset_dec (struct cpu_key * key) +static inline void cpu_key_k_offset_dec(struct cpu_key *key) { - key->on_disk_key.k_offset --; + key->on_disk_key.k_offset--; } #define is_direntry_cpu_key(key) (cpu_key_k_type (key) == TYPE_DIRENTRY) @@ -677,34 +671,25 @@ static inline void cpu_key_k_offset_dec (struct cpu_key * key) #define is_indirect_cpu_key(key) (cpu_key_k_type (key) == TYPE_INDIRECT) #define is_statdata_cpu_key(key) (cpu_key_k_type (key) == TYPE_STAT_DATA) - /* are these used ? */ #define is_direntry_cpu_ih(ih) (is_direntry_cpu_key (&((ih)->ih_key))) #define is_direct_cpu_ih(ih) (is_direct_cpu_key (&((ih)->ih_key))) #define is_indirect_cpu_ih(ih) (is_indirect_cpu_key (&((ih)->ih_key))) #define is_statdata_cpu_ih(ih) (is_statdata_cpu_key (&((ih)->ih_key))) - - - - #define I_K_KEY_IN_ITEM(p_s_ih, p_s_key, n_blocksize) \ ( ! COMP_SHORT_KEYS(p_s_ih, p_s_key) && \ I_OFF_BYTE_IN_ITEM(p_s_ih, k_offset (p_s_key), n_blocksize) ) -/* maximal length of item */ +/* maximal length of item */ #define MAX_ITEM_LEN(block_size) (block_size - BLKH_SIZE - IH_SIZE) #define MIN_ITEM_LEN 1 - /* object identifier for root dir */ #define REISERFS_ROOT_OBJECTID 2 #define REISERFS_ROOT_PARENT_OBJECTID 1 extern struct reiserfs_key root_key; - - - /* * Picture represents a leaf of the S+tree * ______________________________________________________ @@ -716,13 +701,13 @@ extern struct reiserfs_key root_key; /* Header of a disk block. More precisely, header of a formatted leaf or internal node, and not the header of an unformatted node. */ -struct block_head { - __le16 blk_level; /* Level of a block in the tree. */ - __le16 blk_nr_item; /* Number of keys/items in a block. */ - __le16 blk_free_space; /* Block free space in bytes. */ - __le16 blk_reserved; - /* dump this in v4/planA */ - struct reiserfs_key blk_right_delim_key; /* kept only for compatibility */ +struct block_head { + __le16 blk_level; /* Level of a block in the tree. */ + __le16 blk_nr_item; /* Number of keys/items in a block. */ + __le16 blk_free_space; /* Block free space in bytes. */ + __le16 blk_reserved; + /* dump this in v4/planA */ + struct reiserfs_key blk_right_delim_key; /* kept only for compatibility */ }; #define BLKH_SIZE (sizeof(struct block_head)) @@ -741,12 +726,12 @@ struct block_head { * values for blk_level field of the struct block_head */ -#define FREE_LEVEL 0 /* when node gets removed from the tree its - blk_level is set to FREE_LEVEL. It is then - used to see whether the node is still in the - tree */ +#define FREE_LEVEL 0 /* when node gets removed from the tree its + blk_level is set to FREE_LEVEL. It is then + used to see whether the node is still in the + tree */ -#define DISK_LEAF_NODE_LEVEL 1 /* Leaf node level.*/ +#define DISK_LEAF_NODE_LEVEL 1 /* Leaf node level. */ /* Given the buffer head of a formatted node, resolve to the block head of that node. */ #define B_BLK_HEAD(p_s_bh) ((struct block_head *)((p_s_bh)->b_data)) @@ -759,7 +744,6 @@ struct block_head { #define PUT_B_LEVEL(p_s_bh,val) do { set_blkh_level(B_BLK_HEAD(p_s_bh),val); } while (0) #define PUT_B_FREE_SPACE(p_s_bh,val) do { set_blkh_free_space(B_BLK_HEAD(p_s_bh),val); } while (0) - /* Get right delimiting key. -- little endian */ #define B_PRIGHT_DELIM_KEY(p_s_bh) (&(blk_right_delim_key(B_BLK_HEAD(p_s_bh)) @@ -770,41 +754,36 @@ struct block_head { #define B_IS_KEYS_LEVEL(p_s_bh) (B_LEVEL(p_s_bh) > DISK_LEAF_NODE_LEVEL \ && B_LEVEL(p_s_bh) <= MAX_HEIGHT) - - - /***************************************************************************/ /* STAT DATA */ /***************************************************************************/ - // // old stat data is 32 bytes long. We are going to distinguish new one by // different size // -struct stat_data_v1 -{ - __le16 sd_mode; /* file type, permissions */ - __le16 sd_nlink; /* number of hard links */ - __le16 sd_uid; /* owner */ - __le16 sd_gid; /* group */ - __le32 sd_size; /* file size */ - __le32 sd_atime; /* time of last access */ - __le32 sd_mtime; /* time file was last modified */ - __le32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ - union { - __le32 sd_rdev; - __le32 sd_blocks; /* number of blocks file uses */ - } __attribute__ ((__packed__)) u; - __le32 sd_first_direct_byte; /* first byte of file which is stored - in a direct item: except that if it - equals 1 it is a symlink and if it - equals ~(__u32)0 there is no - direct item. The existence of this - field really grates on me. Let's - replace it with a macro based on - sd_size and our tail suppression - policy. Someday. -Hans */ +struct stat_data_v1 { + __le16 sd_mode; /* file type, permissions */ + __le16 sd_nlink; /* number of hard links */ + __le16 sd_uid; /* owner */ + __le16 sd_gid; /* group */ + __le32 sd_size; /* file size */ + __le32 sd_atime; /* time of last access */ + __le32 sd_mtime; /* time file was last modified */ + __le32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ + union { + __le32 sd_rdev; + __le32 sd_blocks; /* number of blocks file uses */ + } __attribute__ ((__packed__)) u; + __le32 sd_first_direct_byte; /* first byte of file which is stored + in a direct item: except that if it + equals 1 it is a symlink and if it + equals ~(__u32)0 there is no + direct item. The existence of this + field really grates on me. Let's + replace it with a macro based on + sd_size and our tail suppression + policy. Someday. -Hans */ } __attribute__ ((__packed__)); #define SD_V1_SIZE (sizeof(struct stat_data_v1)) @@ -862,29 +841,29 @@ struct stat_data_v1 /* Stat Data on disk (reiserfs version of UFS disk inode minus the address blocks) */ struct stat_data { - __le16 sd_mode; /* file type, permissions */ - __le16 sd_attrs; /* persistent inode flags */ - __le32 sd_nlink; /* number of hard links */ - __le64 sd_size; /* file size */ - __le32 sd_uid; /* owner */ - __le32 sd_gid; /* group */ - __le32 sd_atime; /* time of last access */ - __le32 sd_mtime; /* time file was last modified */ - __le32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ - __le32 sd_blocks; - union { - __le32 sd_rdev; - __le32 sd_generation; - //__le32 sd_first_direct_byte; - /* first byte of file which is stored in a - direct item: except that if it equals 1 - it is a symlink and if it equals - ~(__u32)0 there is no direct item. The - existence of this field really grates - on me. Let's replace it with a macro - based on sd_size and our tail - suppression policy? */ - } __attribute__ ((__packed__)) u; + __le16 sd_mode; /* file type, permissions */ + __le16 sd_attrs; /* persistent inode flags */ + __le32 sd_nlink; /* number of hard links */ + __le64 sd_size; /* file size */ + __le32 sd_uid; /* owner */ + __le32 sd_gid; /* group */ + __le32 sd_atime; /* time of last access */ + __le32 sd_mtime; /* time file was last modified */ + __le32 sd_ctime; /* time inode (stat data) was last changed (except changes to sd_atime and sd_mtime) */ + __le32 sd_blocks; + union { + __le32 sd_rdev; + __le32 sd_generation; + //__le32 sd_first_direct_byte; + /* first byte of file which is stored in a + direct item: except that if it equals 1 + it is a symlink and if it equals + ~(__u32)0 there is no direct item. The + existence of this field really grates + on me. Let's replace it with a macro + based on sd_size and our tail + suppression policy? */ + } __attribute__ ((__packed__)) u; } __attribute__ ((__packed__)); // // this is 44 bytes long @@ -919,7 +898,6 @@ struct stat_data { #define sd_v2_attrs(sdp) (le16_to_cpu((sdp)->sd_attrs)) #define set_sd_v2_attrs(sdp,v) ((sdp)->sd_attrs = cpu_to_le16(v)) - /***************************************************************************/ /* DIRECTORY STRUCTURE */ /***************************************************************************/ @@ -954,17 +932,14 @@ struct stat_data { /* NOT IMPLEMENTED: Directory will someday contain stat data of object */ - - -struct reiserfs_de_head -{ - __le32 deh_offset; /* third component of the directory entry key */ - __le32 deh_dir_id; /* objectid of the parent directory of the object, that is referenced - by directory entry */ - __le32 deh_objectid; /* objectid of the object, that is referenced by directory entry */ - __le16 deh_location; /* offset of name in the whole item */ - __le16 deh_state; /* whether 1) entry contains stat data (for future), and 2) whether - entry is hidden (unlinked) */ +struct reiserfs_de_head { + __le32 deh_offset; /* third component of the directory entry key */ + __le32 deh_dir_id; /* objectid of the parent directory of the object, that is referenced + by directory entry */ + __le32 deh_objectid; /* objectid of the object, that is referenced by directory entry */ + __le16 deh_location; /* offset of name in the whole item */ + __le16 deh_state; /* whether 1) entry contains stat data (for future), and 2) whether + entry is hidden (unlinked) */ } __attribute__ ((__packed__)); #define DEH_SIZE sizeof(struct reiserfs_de_head) #define deh_offset(p_deh) (le32_to_cpu((p_deh)->deh_offset)) @@ -986,7 +961,7 @@ struct reiserfs_de_head /* old format directories have this size when empty */ #define EMPTY_DIR_SIZE_V1 (DEH_SIZE * 2 + 3) -#define DEH_Statdata 0 /* not used now */ +#define DEH_Statdata 0 /* not used now */ #define DEH_Visible 2 /* 64 bit systems (and the S/390) need to be aligned explicitly -jdm */ @@ -1023,10 +998,10 @@ struct reiserfs_de_head #define de_visible(deh) test_bit_unaligned (DEH_Visible, &((deh)->deh_state)) #define de_hidden(deh) !test_bit_unaligned (DEH_Visible, &((deh)->deh_state)) -extern void make_empty_dir_item_v1 (char * body, __le32 dirid, __le32 objid, - __le32 par_dirid, __le32 par_objid); -extern void make_empty_dir_item (char * body, __le32 dirid, __le32 objid, - __le32 par_dirid, __le32 par_objid); +extern void make_empty_dir_item_v1(char *body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid); +extern void make_empty_dir_item(char *body, __le32 dirid, __le32 objid, + __le32 par_dirid, __le32 par_objid); /* array of the entry headers */ /* get item body */ @@ -1043,53 +1018,48 @@ extern void make_empty_dir_item (char * body, __le32 dirid, __le32 objid, #define I_DEH_N_ENTRY_LENGTH(ih,deh,i) \ ((i) ? (deh_location((deh)-1) - deh_location((deh))) : (ih_item_len((ih)) - deh_location((deh)))) */ -static inline int entry_length (const struct buffer_head * bh, - const struct item_head * ih, int pos_in_item) +static inline int entry_length(const struct buffer_head *bh, + const struct item_head *ih, int pos_in_item) { - struct reiserfs_de_head * deh; + struct reiserfs_de_head *deh; - deh = B_I_DEH (bh, ih) + pos_in_item; - if (pos_in_item) - return deh_location(deh-1) - deh_location(deh); + deh = B_I_DEH(bh, ih) + pos_in_item; + if (pos_in_item) + return deh_location(deh - 1) - deh_location(deh); - return ih_item_len(ih) - deh_location(deh); + return ih_item_len(ih) - deh_location(deh); } - - /* number of entries in the directory item, depends on ENTRY_COUNT being at the start of directory dynamic data. */ #define I_ENTRY_COUNT(ih) (ih_entry_count((ih))) - /* name by bh, ih and entry_num */ #define B_I_E_NAME(bh,ih,entry_num) ((char *)(bh->b_data + ih_location(ih) + deh_location(B_I_DEH(bh,ih)+(entry_num)))) // two entries per block (at least) #define REISERFS_MAX_NAME(block_size) 255 - /* this structure is used for operations on directory entries. It is not a disk structure. */ /* When reiserfs_find_entry or search_by_entry_key find directory entry, they return filled reiserfs_dir_entry structure */ -struct reiserfs_dir_entry -{ - struct buffer_head * de_bh; - int de_item_num; - struct item_head * de_ih; - int de_entry_num; - struct reiserfs_de_head * de_deh; - int de_entrylen; - int de_namelen; - char * de_name; - char * de_gen_number_bit_string; - - __u32 de_dir_id; - __u32 de_objectid; - - struct cpu_key de_entry_key; +struct reiserfs_dir_entry { + struct buffer_head *de_bh; + int de_item_num; + struct item_head *de_ih; + int de_entry_num; + struct reiserfs_de_head *de_deh; + int de_entrylen; + int de_namelen; + char *de_name; + char *de_gen_number_bit_string; + + __u32 de_dir_id; + __u32 de_objectid; + + struct cpu_key de_entry_key; }; - + /* these defines are useful when a particular member of a reiserfs_dir_entry is needed */ /* pointer to file name, stored in entry */ @@ -1099,8 +1069,6 @@ struct reiserfs_dir_entry #define I_DEH_N_ENTRY_FILE_NAME_LENGTH(ih,deh,entry_num) \ (I_DEH_N_ENTRY_LENGTH (ih, deh, entry_num) - (de_with_sd (deh) ? SD_SIZE : 0)) - - /* hash value occupies bits from 7 up to 30 */ #define GET_HASH_VALUE(offset) ((offset) & 0x7fffff80LL) /* generation number occupies 7 bits starting from 0 up to 6 */ @@ -1109,7 +1077,6 @@ struct reiserfs_dir_entry #define SET_GENERATION_NUMBER(offset,gen_number) (GET_HASH_VALUE(offset)|(gen_number)) - /* * Picture represents an internal node of the reiserfs tree * ______________________________________________________ @@ -1125,9 +1092,9 @@ struct reiserfs_dir_entry /* Disk child pointer: The pointer from an internal node of the tree to a node that is on disk. */ struct disk_child { - __le32 dc_block_number; /* Disk child's block number. */ - __le16 dc_size; /* Disk child's used space. */ - __le16 dc_reserved; + __le32 dc_block_number; /* Disk child's block number. */ + __le16 dc_size; /* Disk child's used space. */ + __le16 dc_reserved; }; #define DC_SIZE (sizeof(struct disk_child)) @@ -1144,7 +1111,7 @@ struct disk_child { #define B_N_CHILD_NUM(p_s_bh,n_pos) (dc_block_number(B_N_CHILD(p_s_bh,n_pos))) #define PUT_B_N_CHILD_NUM(p_s_bh,n_pos, val) (put_dc_block_number(B_N_CHILD(p_s_bh,n_pos), val )) - /* maximal value of field child_size in structure disk_child */ + /* maximal value of field child_size in structure disk_child */ /* child size is the combined size of all items and their headers */ #define MAX_CHILD_SIZE(bh) ((int)( (bh)->b_size - BLKH_SIZE )) @@ -1159,7 +1126,6 @@ struct disk_child { /* PATH STRUCTURES AND DEFINES */ /***************************************************************************/ - /* Search_by_key fills up the path from the root to the leaf as it descends the tree looking for the key. It uses reiserfs_bread to try to find buffers in the cache given their block number. If it does not find them in the cache it reads them from disk. For each node search_by_key finds using @@ -1168,20 +1134,18 @@ struct disk_child { is looking through a leaf node bin_search will find the position of the item which has key either equal to given key, or which is the maximal key less than the given key. */ -struct path_element { - struct buffer_head * pe_buffer; /* Pointer to the buffer at the path in the tree. */ - int pe_position; /* Position in the tree node which is placed in the */ - /* buffer above. */ +struct path_element { + struct buffer_head *pe_buffer; /* Pointer to the buffer at the path in the tree. */ + int pe_position; /* Position in the tree node which is placed in the */ + /* buffer above. */ }; -#define MAX_HEIGHT 5 /* maximal height of a tree. don't change this without changing JOURNAL_PER_BALANCE_CNT */ -#define EXTENDED_MAX_HEIGHT 7 /* Must be equals MAX_HEIGHT + FIRST_PATH_ELEMENT_OFFSET */ -#define FIRST_PATH_ELEMENT_OFFSET 2 /* Must be equal to at least 2. */ - -#define ILLEGAL_PATH_ELEMENT_OFFSET 1 /* Must be equal to FIRST_PATH_ELEMENT_OFFSET - 1 */ -#define MAX_FEB_SIZE 6 /* this MUST be MAX_HEIGHT + 1. See about FEB below */ - +#define MAX_HEIGHT 5 /* maximal height of a tree. don't change this without changing JOURNAL_PER_BALANCE_CNT */ +#define EXTENDED_MAX_HEIGHT 7 /* Must be equals MAX_HEIGHT + FIRST_PATH_ELEMENT_OFFSET */ +#define FIRST_PATH_ELEMENT_OFFSET 2 /* Must be equal to at least 2. */ +#define ILLEGAL_PATH_ELEMENT_OFFSET 1 /* Must be equal to FIRST_PATH_ELEMENT_OFFSET - 1 */ +#define MAX_FEB_SIZE 6 /* this MUST be MAX_HEIGHT + 1. See about FEB below */ /* We need to keep track of who the ancestors of nodes are. When we perform a search we record which nodes were visited while @@ -1200,14 +1164,14 @@ excessive effort to avoid disturbing the precious VFS code.:-( The gods only know how we are going to SMP the code that uses them. znodes are the way! */ -#define PATH_READA 0x1 /* do read ahead */ -#define PATH_READA_BACK 0x2 /* read backwards */ +#define PATH_READA 0x1 /* do read ahead */ +#define PATH_READA_BACK 0x2 /* read backwards */ -struct path { - int path_length; /* Length of the array above. */ - int reada; - struct path_element path_elements[EXTENDED_MAX_HEIGHT]; /* Array of the path elements. */ - int pos_in_item; +struct path { + int path_length; /* Length of the array above. */ + int reada; + struct path_element path_elements[EXTENDED_MAX_HEIGHT]; /* Array of the path elements. */ + int pos_in_item; }; #define pos_in_item(path) ((path)->pos_in_item) @@ -1224,25 +1188,23 @@ struct path var = {.path_length = ILLEGAL_PATH_ELEMENT_OFFSET, .reada = 0,} /* Get position in the element at the path by path and path position. */ #define PATH_OFFSET_POSITION(p_s_path,n_offset) (PATH_OFFSET_PELEMENT(p_s_path,n_offset)->pe_position) - #define PATH_PLAST_BUFFER(p_s_path) (PATH_OFFSET_PBUFFER((p_s_path), (p_s_path)->path_length)) /* you know, to the person who didn't - write this the macro name does not - at first suggest what it does. - Maybe POSITION_FROM_PATH_END? Or - maybe we should just focus on - dumping paths... -Hans */ + write this the macro name does not + at first suggest what it does. + Maybe POSITION_FROM_PATH_END? Or + maybe we should just focus on + dumping paths... -Hans */ #define PATH_LAST_POSITION(p_s_path) (PATH_OFFSET_POSITION((p_s_path), (p_s_path)->path_length)) - #define PATH_PITEM_HEAD(p_s_path) B_N_PITEM_HEAD(PATH_PLAST_BUFFER(p_s_path),PATH_LAST_POSITION(p_s_path)) /* in do_balance leaf has h == 0 in contrast with path structure, where root has level == 0. That is why we need these defines */ #define PATH_H_PBUFFER(p_s_path, h) PATH_OFFSET_PBUFFER (p_s_path, p_s_path->path_length - (h)) /* tb->S[h] */ -#define PATH_H_PPARENT(path, h) PATH_H_PBUFFER (path, (h) + 1) /* tb->F[h] or tb->S[0]->b_parent */ -#define PATH_H_POSITION(path, h) PATH_OFFSET_POSITION (path, path->path_length - (h)) -#define PATH_H_B_ITEM_ORDER(path, h) PATH_H_POSITION(path, h + 1) /* tb->S[h]->b_item_order */ +#define PATH_H_PPARENT(path, h) PATH_H_PBUFFER (path, (h) + 1) /* tb->F[h] or tb->S[0]->b_parent */ +#define PATH_H_POSITION(path, h) PATH_OFFSET_POSITION (path, path->path_length - (h)) +#define PATH_H_B_ITEM_ORDER(path, h) PATH_H_POSITION(path, h + 1) /* tb->S[h]->b_item_order */ #define PATH_H_PATH_OFFSET(p_s_path, n_h) ((p_s_path)->path_length - (n_h)) @@ -1253,7 +1215,6 @@ struct path var = {.path_length = ILLEGAL_PATH_ELEMENT_OFFSET, .reada = 0,} #define item_moved(ih,path) comp_items(ih, path) #define path_changed(ih,path) comp_items (ih, path) - /***************************************************************************/ /* MISC */ /***************************************************************************/ @@ -1272,30 +1233,26 @@ struct path var = {.path_length = ILLEGAL_PATH_ELEMENT_OFFSET, .reada = 0,} // reiserfs version 2 has max offset 60 bits. Version 1 - 32 bit offset #define U32_MAX (~(__u32)0) -static inline loff_t max_reiserfs_offset (struct inode * inode) +static inline loff_t max_reiserfs_offset(struct inode *inode) { - if (get_inode_item_key_version(inode) == KEY_FORMAT_3_5) - return (loff_t)U32_MAX; + if (get_inode_item_key_version(inode) == KEY_FORMAT_3_5) + return (loff_t) U32_MAX; - return (loff_t)((~(__u64)0) >> 4); + return (loff_t) ((~(__u64) 0) >> 4); } - /*#define MAX_KEY_UNIQUENESS MAX_UL_INT*/ #define MAX_KEY_OBJECTID MAX_UL_INT - #define MAX_B_NUM MAX_UL_INT #define MAX_FC_NUM MAX_US_INT - /* the purpose is to detect overflow of an unsigned short */ #define REISERFS_LINK_MAX (MAX_US_INT - 1000) - /* The following defines are used in reiserfs_insert_item and reiserfs_append_item */ -#define REISERFS_KERNEL_MEM 0 /* reiserfs kernel memory mode */ -#define REISERFS_USER_MEM 1 /* reiserfs user memory mode */ +#define REISERFS_KERNEL_MEM 0 /* reiserfs kernel memory mode */ +#define REISERFS_USER_MEM 1 /* reiserfs user memory mode */ #define fs_generation(s) (REISERFS_SB(s)->s_generation_counter) #define get_generation(s) atomic_read (&fs_generation(s)) @@ -1303,7 +1260,6 @@ static inline loff_t max_reiserfs_offset (struct inode * inode) #define __fs_changed(gen,s) (gen != get_generation (s)) #define fs_changed(gen,s) ({cond_resched(); __fs_changed(gen, s);}) - /***************************************************************************/ /* FIXATE NODES */ /***************************************************************************/ @@ -1324,38 +1280,34 @@ static inline loff_t max_reiserfs_offset (struct inode * inode) calculating what we can shift to neighbors and how many nodes we have to have if we do not any shiftings, if we shift to left/right neighbor or to both. */ -struct virtual_item -{ - int vi_index; // index in the array of item operations - unsigned short vi_type; // left/right mergeability - unsigned short vi_item_len; /* length of item that it will have after balancing */ - struct item_head * vi_ih; - const char * vi_item; // body of item (old or new) - const void * vi_new_data; // 0 always but paste mode - void * vi_uarea; // item specific area +struct virtual_item { + int vi_index; // index in the array of item operations + unsigned short vi_type; // left/right mergeability + unsigned short vi_item_len; /* length of item that it will have after balancing */ + struct item_head *vi_ih; + const char *vi_item; // body of item (old or new) + const void *vi_new_data; // 0 always but paste mode + void *vi_uarea; // item specific area }; - -struct virtual_node -{ - char * vn_free_ptr; /* this is a pointer to the free space in the buffer */ - unsigned short vn_nr_item; /* number of items in virtual node */ - short vn_size; /* size of node , that node would have if it has unlimited size and no balancing is performed */ - short vn_mode; /* mode of balancing (paste, insert, delete, cut) */ - short vn_affected_item_num; - short vn_pos_in_item; - struct item_head * vn_ins_ih; /* item header of inserted item, 0 for other modes */ - const void * vn_data; - struct virtual_item * vn_vi; /* array of items (including a new one, excluding item to be deleted) */ +struct virtual_node { + char *vn_free_ptr; /* this is a pointer to the free space in the buffer */ + unsigned short vn_nr_item; /* number of items in virtual node */ + short vn_size; /* size of node , that node would have if it has unlimited size and no balancing is performed */ + short vn_mode; /* mode of balancing (paste, insert, delete, cut) */ + short vn_affected_item_num; + short vn_pos_in_item; + struct item_head *vn_ins_ih; /* item header of inserted item, 0 for other modes */ + const void *vn_data; + struct virtual_item *vn_vi; /* array of items (including a new one, excluding item to be deleted) */ }; /* used by directory items when creating virtual nodes */ struct direntry_uarea { - int flags; - __u16 entry_count; - __u16 entry_sizes[1]; -} __attribute__ ((__packed__)) ; - + int flags; + __u16 entry_count; + __u16 entry_sizes[1]; +} __attribute__ ((__packed__)); /***************************************************************************/ /* TREE BALANCE */ @@ -1378,73 +1330,72 @@ struct direntry_uarea { #define MAX_AMOUNT_NEEDED 2 /* someday somebody will prefix every field in this struct with tb_ */ -struct tree_balance -{ - int tb_mode; - int need_balance_dirty; - struct super_block * tb_sb; - struct reiserfs_transaction_handle *transaction_handle ; - struct path * tb_path; - struct buffer_head * L[MAX_HEIGHT]; /* array of left neighbors of nodes in the path */ - struct buffer_head * R[MAX_HEIGHT]; /* array of right neighbors of nodes in the path*/ - struct buffer_head * FL[MAX_HEIGHT]; /* array of fathers of the left neighbors */ - struct buffer_head * FR[MAX_HEIGHT]; /* array of fathers of the right neighbors */ - struct buffer_head * CFL[MAX_HEIGHT]; /* array of common parents of center node and its left neighbor */ - struct buffer_head * CFR[MAX_HEIGHT]; /* array of common parents of center node and its right neighbor */ - - struct buffer_head * FEB[MAX_FEB_SIZE]; /* array of empty buffers. Number of buffers in array equals - cur_blknum. */ - struct buffer_head * used[MAX_FEB_SIZE]; - struct buffer_head * thrown[MAX_FEB_SIZE]; - int lnum[MAX_HEIGHT]; /* array of number of items which must be - shifted to the left in order to balance the - current node; for leaves includes item that - will be partially shifted; for internal - nodes, it is the number of child pointers - rather than items. It includes the new item - being created. The code sometimes subtracts - one to get the number of wholly shifted - items for other purposes. */ - int rnum[MAX_HEIGHT]; /* substitute right for left in comment above */ - int lkey[MAX_HEIGHT]; /* array indexed by height h mapping the key delimiting L[h] and - S[h] to its item number within the node CFL[h] */ - int rkey[MAX_HEIGHT]; /* substitute r for l in comment above */ - int insert_size[MAX_HEIGHT]; /* the number of bytes by we are trying to add or remove from - S[h]. A negative value means removing. */ - int blknum[MAX_HEIGHT]; /* number of nodes that will replace node S[h] after - balancing on the level h of the tree. If 0 then S is - being deleted, if 1 then S is remaining and no new nodes - are being created, if 2 or 3 then 1 or 2 new nodes is - being created */ - - /* fields that are used only for balancing leaves of the tree */ - int cur_blknum; /* number of empty blocks having been already allocated */ - int s0num; /* number of items that fall into left most node when S[0] splits */ - int s1num; /* number of items that fall into first new node when S[0] splits */ - int s2num; /* number of items that fall into second new node when S[0] splits */ - int lbytes; /* number of bytes which can flow to the left neighbor from the left */ - /* most liquid item that cannot be shifted from S[0] entirely */ - /* if -1 then nothing will be partially shifted */ - int rbytes; /* number of bytes which will flow to the right neighbor from the right */ - /* most liquid item that cannot be shifted from S[0] entirely */ - /* if -1 then nothing will be partially shifted */ - int s1bytes; /* number of bytes which flow to the first new node when S[0] splits */ - /* note: if S[0] splits into 3 nodes, then items do not need to be cut */ - int s2bytes; - struct buffer_head * buf_to_free[MAX_FREE_BLOCK]; /* buffers which are to be freed after do_balance finishes by unfix_nodes */ - char * vn_buf; /* kmalloced memory. Used to create +struct tree_balance { + int tb_mode; + int need_balance_dirty; + struct super_block *tb_sb; + struct reiserfs_transaction_handle *transaction_handle; + struct path *tb_path; + struct buffer_head *L[MAX_HEIGHT]; /* array of left neighbors of nodes in the path */ + struct buffer_head *R[MAX_HEIGHT]; /* array of right neighbors of nodes in the path */ + struct buffer_head *FL[MAX_HEIGHT]; /* array of fathers of the left neighbors */ + struct buffer_head *FR[MAX_HEIGHT]; /* array of fathers of the right neighbors */ + struct buffer_head *CFL[MAX_HEIGHT]; /* array of common parents of center node and its left neighbor */ + struct buffer_head *CFR[MAX_HEIGHT]; /* array of common parents of center node and its right neighbor */ + + struct buffer_head *FEB[MAX_FEB_SIZE]; /* array of empty buffers. Number of buffers in array equals + cur_blknum. */ + struct buffer_head *used[MAX_FEB_SIZE]; + struct buffer_head *thrown[MAX_FEB_SIZE]; + int lnum[MAX_HEIGHT]; /* array of number of items which must be + shifted to the left in order to balance the + current node; for leaves includes item that + will be partially shifted; for internal + nodes, it is the number of child pointers + rather than items. It includes the new item + being created. The code sometimes subtracts + one to get the number of wholly shifted + items for other purposes. */ + int rnum[MAX_HEIGHT]; /* substitute right for left in comment above */ + int lkey[MAX_HEIGHT]; /* array indexed by height h mapping the key delimiting L[h] and + S[h] to its item number within the node CFL[h] */ + int rkey[MAX_HEIGHT]; /* substitute r for l in comment above */ + int insert_size[MAX_HEIGHT]; /* the number of bytes by we are trying to add or remove from + S[h]. A negative value means removing. */ + int blknum[MAX_HEIGHT]; /* number of nodes that will replace node S[h] after + balancing on the level h of the tree. If 0 then S is + being deleted, if 1 then S is remaining and no new nodes + are being created, if 2 or 3 then 1 or 2 new nodes is + being created */ + + /* fields that are used only for balancing leaves of the tree */ + int cur_blknum; /* number of empty blocks having been already allocated */ + int s0num; /* number of items that fall into left most node when S[0] splits */ + int s1num; /* number of items that fall into first new node when S[0] splits */ + int s2num; /* number of items that fall into second new node when S[0] splits */ + int lbytes; /* number of bytes which can flow to the left neighbor from the left */ + /* most liquid item that cannot be shifted from S[0] entirely */ + /* if -1 then nothing will be partially shifted */ + int rbytes; /* number of bytes which will flow to the right neighbor from the right */ + /* most liquid item that cannot be shifted from S[0] entirely */ + /* if -1 then nothing will be partially shifted */ + int s1bytes; /* number of bytes which flow to the first new node when S[0] splits */ + /* note: if S[0] splits into 3 nodes, then items do not need to be cut */ + int s2bytes; + struct buffer_head *buf_to_free[MAX_FREE_BLOCK]; /* buffers which are to be freed after do_balance finishes by unfix_nodes */ + char *vn_buf; /* kmalloced memory. Used to create virtual node and keep map of dirtied bitmap blocks */ - int vn_buf_size; /* size of the vn_buf */ - struct virtual_node * tb_vn; /* VN starts after bitmap of bitmap blocks */ + int vn_buf_size; /* size of the vn_buf */ + struct virtual_node *tb_vn; /* VN starts after bitmap of bitmap blocks */ - int fs_gen; /* saved value of `reiserfs_generation' counter - see FILESYSTEM_CHANGED() macro in reiserfs_fs.h */ + int fs_gen; /* saved value of `reiserfs_generation' counter + see FILESYSTEM_CHANGED() macro in reiserfs_fs.h */ #ifdef DISPLACE_NEW_PACKING_LOCALITIES - struct in_core_key key; /* key pointer, to pass to block allocator or - another low-level subsystem */ + struct in_core_key key; /* key pointer, to pass to block allocator or + another low-level subsystem */ #endif -} ; +}; /* These are modes of balancing */ @@ -1479,13 +1430,12 @@ struct tree_balance /* used in do_balance for passing parent of node information that has been gotten from tb struct */ struct buffer_info { - struct tree_balance * tb; - struct buffer_head * bi_bh; - struct buffer_head * bi_parent; - int bi_position; + struct tree_balance *tb; + struct buffer_head *bi_bh; + struct buffer_head *bi_parent; + int bi_position; }; - /* there are 4 types of items: stat data, directory item, indirect, direct. +-------------------+------------+--------------+------------+ | | k_offset | k_uniqueness | mergeable? | @@ -1503,24 +1453,24 @@ struct buffer_info { */ struct item_operations { - int (*bytes_number) (struct item_head * ih, int block_size); - void (*decrement_key) (struct cpu_key *); - int (*is_left_mergeable) (struct reiserfs_key * ih, unsigned long bsize); - void (*print_item) (struct item_head *, char * item); - void (*check_item) (struct item_head *, char * item); - - int (*create_vi) (struct virtual_node * vn, struct virtual_item * vi, - int is_affected, int insert_size); - int (*check_left) (struct virtual_item * vi, int free, - int start_skip, int end_skip); - int (*check_right) (struct virtual_item * vi, int free); - int (*part_size) (struct virtual_item * vi, int from, int to); - int (*unit_num) (struct virtual_item * vi); - void (*print_vi) (struct virtual_item * vi); + int (*bytes_number) (struct item_head * ih, int block_size); + void (*decrement_key) (struct cpu_key *); + int (*is_left_mergeable) (struct reiserfs_key * ih, + unsigned long bsize); + void (*print_item) (struct item_head *, char *item); + void (*check_item) (struct item_head *, char *item); + + int (*create_vi) (struct virtual_node * vn, struct virtual_item * vi, + int is_affected, int insert_size); + int (*check_left) (struct virtual_item * vi, int free, + int start_skip, int end_skip); + int (*check_right) (struct virtual_item * vi, int free); + int (*part_size) (struct virtual_item * vi, int from, int to); + int (*unit_num) (struct virtual_item * vi); + void (*print_vi) (struct virtual_item * vi); }; - -extern struct item_operations * item_ops [TYPE_ANY + 1]; +extern struct item_operations *item_ops[TYPE_ANY + 1]; #define op_bytes_number(ih,bsize) item_ops[le_ih_k_type (ih)]->bytes_number (ih, bsize) #define op_is_left_mergeable(key,bsize) item_ops[le_key_k_type (le_key_version (key), key)]->is_left_mergeable (key, bsize) @@ -1533,8 +1483,6 @@ extern struct item_operations * item_ops [TYPE_ANY + 1]; #define op_unit_num(vi) item_ops[(vi)->vi_index]->unit_num (vi) #define op_print_vi(vi) item_ops[(vi)->vi_index]->print_vi (vi) - - #define COMP_SHORT_KEYS comp_short_keys /* number of blocks pointed to by the indirect item */ @@ -1545,8 +1493,7 @@ extern struct item_operations * item_ops [TYPE_ANY + 1]; /* number of bytes contained by the direct item or the unformatted nodes the indirect item points to */ - -/* get the item header */ +/* get the item header */ #define B_N_PITEM_HEAD(bh,item_num) ( (struct item_head * )((bh)->b_data + BLKH_SIZE) + (item_num) ) /* get key */ @@ -1577,9 +1524,9 @@ extern struct item_operations * item_ops [TYPE_ANY + 1]; #define PUT_B_I_POS_UNFM_POINTER(bh,ih,pos, val) do {*(((unp_t *)B_I_PITEM(bh,ih)) + (pos)) = cpu_to_le32(val); } while (0) struct reiserfs_iget_args { - __u32 objectid ; - __u32 dirid ; -} ; + __u32 objectid; + __u32 dirid; +}; /***************************************************************************/ /* FUNCTION DECLARATIONS */ @@ -1595,11 +1542,11 @@ struct reiserfs_iget_args { /* first block written in a commit. */ struct reiserfs_journal_desc { - __le32 j_trans_id ; /* id of commit */ - __le32 j_len ; /* length of commit. len +1 is the commit block */ - __le32 j_mount_id ; /* mount id of this trans*/ - __le32 j_realblock[1] ; /* real locations for each block */ -} ; + __le32 j_trans_id; /* id of commit */ + __le32 j_len; /* length of commit. len +1 is the commit block */ + __le32 j_mount_id; /* mount id of this trans */ + __le32 j_realblock[1]; /* real locations for each block */ +}; #define get_desc_trans_id(d) le32_to_cpu((d)->j_trans_id) #define get_desc_trans_len(d) le32_to_cpu((d)->j_len) @@ -1611,10 +1558,10 @@ struct reiserfs_journal_desc { /* last block written in a commit */ struct reiserfs_journal_commit { - __le32 j_trans_id ; /* must match j_trans_id from the desc block */ - __le32 j_len ; /* ditto */ - __le32 j_realblock[1] ; /* real locations for each block */ -} ; + __le32 j_trans_id; /* must match j_trans_id from the desc block */ + __le32 j_len; /* ditto */ + __le32 j_realblock[1]; /* real locations for each block */ +}; #define get_commit_trans_id(c) le32_to_cpu((c)->j_trans_id) #define get_commit_trans_len(c) le32_to_cpu((c)->j_len) @@ -1628,19 +1575,19 @@ struct reiserfs_journal_commit { ** and this transaction does not need to be replayed. */ struct reiserfs_journal_header { - __le32 j_last_flush_trans_id ; /* id of last fully flushed transaction */ - __le32 j_first_unflushed_offset ; /* offset in the log of where to start replay after a crash */ - __le32 j_mount_id ; - /* 12 */ struct journal_params jh_journal; -} ; + __le32 j_last_flush_trans_id; /* id of last fully flushed transaction */ + __le32 j_first_unflushed_offset; /* offset in the log of where to start replay after a crash */ + __le32 j_mount_id; + /* 12 */ struct journal_params jh_journal; +}; /* biggest tunable defines are right here */ -#define JOURNAL_BLOCK_COUNT 8192 /* number of blocks in the journal */ -#define JOURNAL_TRANS_MAX_DEFAULT 1024 /* biggest possible single transaction, don't change for now (8/3/99) */ +#define JOURNAL_BLOCK_COUNT 8192 /* number of blocks in the journal */ +#define JOURNAL_TRANS_MAX_DEFAULT 1024 /* biggest possible single transaction, don't change for now (8/3/99) */ #define JOURNAL_TRANS_MIN_DEFAULT 256 -#define JOURNAL_MAX_BATCH_DEFAULT 900 /* max blocks to batch into one transaction, don't make this any bigger than 900 */ +#define JOURNAL_MAX_BATCH_DEFAULT 900 /* max blocks to batch into one transaction, don't make this any bigger than 900 */ #define JOURNAL_MIN_RATIO 2 -#define JOURNAL_MAX_COMMIT_AGE 30 +#define JOURNAL_MAX_COMMIT_AGE 30 #define JOURNAL_MAX_TRANS_AGE 30 #define JOURNAL_PER_BALANCE_CNT (3 * (MAX_HEIGHT-2) + 9) #ifdef CONFIG_QUOTA @@ -1664,10 +1611,10 @@ struct reiserfs_journal_header { ** the current number of nodes is > max, the node is freed, otherwise, ** it is put on a free list for faster use later. */ -#define REISERFS_MIN_BITMAP_NODES 10 -#define REISERFS_MAX_BITMAP_NODES 100 +#define REISERFS_MIN_BITMAP_NODES 10 +#define REISERFS_MAX_BITMAP_NODES 100 -#define JBH_HASH_SHIFT 13 /* these are based on journal hash size of 8192 */ +#define JBH_HASH_SHIFT 13 /* these are based on journal hash size of 8192 */ #define JBH_HASH_MASK 8191 #define _jhashfn(sb,block) \ @@ -1681,14 +1628,14 @@ struct reiserfs_journal_header { #define journal_bread(s, block) __bread(SB_JOURNAL(s)->j_dev_bd, block, s->s_blocksize) enum reiserfs_bh_state_bits { - BH_JDirty = BH_PrivateStart, /* buffer is in current transaction */ - BH_JDirty_wait, - BH_JNew, /* disk block was taken off free list before - * being in a finished transaction, or - * written to disk. Can be reused immed. */ - BH_JPrepared, - BH_JRestore_dirty, - BH_JTest, // debugging only will go away + BH_JDirty = BH_PrivateStart, /* buffer is in current transaction */ + BH_JDirty_wait, + BH_JNew, /* disk block was taken off free list before + * being in a finished transaction, or + * written to disk. Can be reused immed. */ + BH_JPrepared, + BH_JRestore_dirty, + BH_JTest, // debugging only will go away }; BUFFER_FNS(JDirty, journaled); @@ -1708,175 +1655,192 @@ TAS_BUFFER_FNS(JTest, journal_test); ** transaction handle which is passed around for all journal calls */ struct reiserfs_transaction_handle { - struct super_block *t_super ; /* super for this FS when journal_begin was - called. saves calls to reiserfs_get_super - also used by nested transactions to make - sure they are nesting on the right FS - _must_ be first in the handle - */ - int t_refcount; - int t_blocks_logged ; /* number of blocks this writer has logged */ - int t_blocks_allocated ; /* number of blocks this writer allocated */ - unsigned long t_trans_id ; /* sanity check, equals the current trans id */ - void *t_handle_save ; /* save existing current->journal_info */ - unsigned displace_new_blocks:1; /* if new block allocation occurres, that block - should be displaced from others */ - struct list_head t_list; -} ; + struct super_block *t_super; /* super for this FS when journal_begin was + called. saves calls to reiserfs_get_super + also used by nested transactions to make + sure they are nesting on the right FS + _must_ be first in the handle + */ + int t_refcount; + int t_blocks_logged; /* number of blocks this writer has logged */ + int t_blocks_allocated; /* number of blocks this writer allocated */ + unsigned long t_trans_id; /* sanity check, equals the current trans id */ + void *t_handle_save; /* save existing current->journal_info */ + unsigned displace_new_blocks:1; /* if new block allocation occurres, that block + should be displaced from others */ + struct list_head t_list; +}; /* used to keep track of ordered and tail writes, attached to the buffer * head through b_journal_head. */ struct reiserfs_jh { - struct reiserfs_journal_list *jl; - struct buffer_head *bh; - struct list_head list; + struct reiserfs_journal_list *jl; + struct buffer_head *bh; + struct list_head list; }; void reiserfs_free_jh(struct buffer_head *bh); int reiserfs_add_tail_list(struct inode *inode, struct buffer_head *bh); int reiserfs_add_ordered_list(struct inode *inode, struct buffer_head *bh); -int journal_mark_dirty(struct reiserfs_transaction_handle *, struct super_block *, struct buffer_head *bh) ; - -static inline int -reiserfs_file_data_log(struct inode *inode) { - if (reiserfs_data_log(inode->i_sb) || - (REISERFS_I(inode)->i_flags & i_data_log)) - return 1 ; - return 0 ; +int journal_mark_dirty(struct reiserfs_transaction_handle *, + struct super_block *, struct buffer_head *bh); + +static inline int reiserfs_file_data_log(struct inode *inode) +{ + if (reiserfs_data_log(inode->i_sb) || + (REISERFS_I(inode)->i_flags & i_data_log)) + return 1; + return 0; } -static inline int reiserfs_transaction_running(struct super_block *s) { - struct reiserfs_transaction_handle *th = current->journal_info ; - if (th && th->t_super == s) - return 1 ; - if (th && th->t_super == NULL) - BUG(); - return 0 ; +static inline int reiserfs_transaction_running(struct super_block *s) +{ + struct reiserfs_transaction_handle *th = current->journal_info; + if (th && th->t_super == s) + return 1; + if (th && th->t_super == NULL) + BUG(); + return 0; } int reiserfs_async_progress_wait(struct super_block *s); -struct reiserfs_transaction_handle * -reiserfs_persistent_transaction(struct super_block *, int count); +struct reiserfs_transaction_handle *reiserfs_persistent_transaction(struct + super_block + *, + int count); int reiserfs_end_persistent_transaction(struct reiserfs_transaction_handle *); int reiserfs_commit_page(struct inode *inode, struct page *page, - unsigned from, unsigned to); + unsigned from, unsigned to); int reiserfs_flush_old_commits(struct super_block *); -int reiserfs_commit_for_inode(struct inode *) ; -int reiserfs_inode_needs_commit(struct inode *) ; -void reiserfs_update_inode_transaction(struct inode *) ; -void reiserfs_wait_on_write_block(struct super_block *s) ; -void reiserfs_block_writes(struct reiserfs_transaction_handle *th) ; -void reiserfs_allow_writes(struct super_block *s) ; -void reiserfs_check_lock_depth(struct super_block *s, char *caller) ; -int reiserfs_prepare_for_journal(struct super_block *, struct buffer_head *bh, int wait) ; -void reiserfs_restore_prepared_buffer(struct super_block *, struct buffer_head *bh) ; -int journal_init(struct super_block *, const char * j_dev_name, int old_format, unsigned int) ; -int journal_release(struct reiserfs_transaction_handle*, struct super_block *) ; -int journal_release_error(struct reiserfs_transaction_handle*, struct super_block *) ; -int journal_end(struct reiserfs_transaction_handle *, struct super_block *, unsigned long) ; -int journal_end_sync(struct reiserfs_transaction_handle *, struct super_block *, unsigned long) ; -int journal_mark_freed(struct reiserfs_transaction_handle *, struct super_block *, b_blocknr_t blocknr) ; -int journal_transaction_should_end(struct reiserfs_transaction_handle *, int) ; -int reiserfs_in_journal(struct super_block *p_s_sb, int bmap_nr, int bit_nr, int searchall, b_blocknr_t *next) ; -int journal_begin(struct reiserfs_transaction_handle *, struct super_block *p_s_sb, unsigned long) ; -int journal_join_abort(struct reiserfs_transaction_handle *, struct super_block *p_s_sb, unsigned long) ; -void reiserfs_journal_abort (struct super_block *sb, int errno); -void reiserfs_abort (struct super_block *sb, int errno, const char *fmt, ...); -int reiserfs_allocate_list_bitmaps(struct super_block *s, struct reiserfs_list_bitmap *, int) ; - -void add_save_link (struct reiserfs_transaction_handle * th, - struct inode * inode, int truncate); -int remove_save_link (struct inode * inode, int truncate); +int reiserfs_commit_for_inode(struct inode *); +int reiserfs_inode_needs_commit(struct inode *); +void reiserfs_update_inode_transaction(struct inode *); +void reiserfs_wait_on_write_block(struct super_block *s); +void reiserfs_block_writes(struct reiserfs_transaction_handle *th); +void reiserfs_allow_writes(struct super_block *s); +void reiserfs_check_lock_depth(struct super_block *s, char *caller); +int reiserfs_prepare_for_journal(struct super_block *, struct buffer_head *bh, + int wait); +void reiserfs_restore_prepared_buffer(struct super_block *, + struct buffer_head *bh); +int journal_init(struct super_block *, const char *j_dev_name, int old_format, + unsigned int); +int journal_release(struct reiserfs_transaction_handle *, struct super_block *); +int journal_release_error(struct reiserfs_transaction_handle *, + struct super_block *); +int journal_end(struct reiserfs_transaction_handle *, struct super_block *, + unsigned long); +int journal_end_sync(struct reiserfs_transaction_handle *, struct super_block *, + unsigned long); +int journal_mark_freed(struct reiserfs_transaction_handle *, + struct super_block *, b_blocknr_t blocknr); +int journal_transaction_should_end(struct reiserfs_transaction_handle *, int); +int reiserfs_in_journal(struct super_block *p_s_sb, int bmap_nr, int bit_nr, + int searchall, b_blocknr_t * next); +int journal_begin(struct reiserfs_transaction_handle *, + struct super_block *p_s_sb, unsigned long); +int journal_join_abort(struct reiserfs_transaction_handle *, + struct super_block *p_s_sb, unsigned long); +void reiserfs_journal_abort(struct super_block *sb, int errno); +void reiserfs_abort(struct super_block *sb, int errno, const char *fmt, ...); +int reiserfs_allocate_list_bitmaps(struct super_block *s, + struct reiserfs_list_bitmap *, int); + +void add_save_link(struct reiserfs_transaction_handle *th, + struct inode *inode, int truncate); +int remove_save_link(struct inode *inode, int truncate); /* objectid.c */ -__u32 reiserfs_get_unused_objectid (struct reiserfs_transaction_handle *th); -void reiserfs_release_objectid (struct reiserfs_transaction_handle *th, __u32 objectid_to_release); -int reiserfs_convert_objectid_map_v1(struct super_block *) ; +__u32 reiserfs_get_unused_objectid(struct reiserfs_transaction_handle *th); +void reiserfs_release_objectid(struct reiserfs_transaction_handle *th, + __u32 objectid_to_release); +int reiserfs_convert_objectid_map_v1(struct super_block *); /* stree.c */ int B_IS_IN_TREE(const struct buffer_head *); -extern void copy_item_head(struct item_head * p_v_to, - const struct item_head * p_v_from); +extern void copy_item_head(struct item_head *p_v_to, + const struct item_head *p_v_from); // first key is in cpu form, second - le -extern int comp_short_keys (const struct reiserfs_key * le_key, - const struct cpu_key * cpu_key); -extern void le_key2cpu_key (struct cpu_key * to, const struct reiserfs_key * from); +extern int comp_short_keys(const struct reiserfs_key *le_key, + const struct cpu_key *cpu_key); +extern void le_key2cpu_key(struct cpu_key *to, const struct reiserfs_key *from); // both are in le form -extern int comp_le_keys (const struct reiserfs_key *, const struct reiserfs_key *); -extern int comp_short_le_keys (const struct reiserfs_key *, const struct reiserfs_key *); +extern int comp_le_keys(const struct reiserfs_key *, + const struct reiserfs_key *); +extern int comp_short_le_keys(const struct reiserfs_key *, + const struct reiserfs_key *); // // get key version from on disk key - kludge // -static inline int le_key_version (const struct reiserfs_key * key) +static inline int le_key_version(const struct reiserfs_key *key) { - int type; - - type = offset_v2_k_type( &(key->u.k_offset_v2)); - if (type != TYPE_DIRECT && type != TYPE_INDIRECT && type != TYPE_DIRENTRY) - return KEY_FORMAT_3_5; - - return KEY_FORMAT_3_6; - -} + int type; + type = offset_v2_k_type(&(key->u.k_offset_v2)); + if (type != TYPE_DIRECT && type != TYPE_INDIRECT + && type != TYPE_DIRENTRY) + return KEY_FORMAT_3_5; + + return KEY_FORMAT_3_6; -static inline void copy_key (struct reiserfs_key *to, const struct reiserfs_key *from) -{ - memcpy (to, from, KEY_SIZE); } +static inline void copy_key(struct reiserfs_key *to, + const struct reiserfs_key *from) +{ + memcpy(to, from, KEY_SIZE); +} -int comp_items (const struct item_head * stored_ih, const struct path * p_s_path); -const struct reiserfs_key * get_rkey (const struct path * p_s_chk_path, - const struct super_block * p_s_sb); -int search_by_key (struct super_block *, const struct cpu_key *, - struct path *, int); +int comp_items(const struct item_head *stored_ih, const struct path *p_s_path); +const struct reiserfs_key *get_rkey(const struct path *p_s_chk_path, + const struct super_block *p_s_sb); +int search_by_key(struct super_block *, const struct cpu_key *, + struct path *, int); #define search_item(s,key,path) search_by_key (s, key, path, DISK_LEAF_NODE_LEVEL) -int search_for_position_by_key (struct super_block * p_s_sb, - const struct cpu_key * p_s_cpu_key, - struct path * p_s_search_path); -extern void decrement_bcount (struct buffer_head * p_s_bh); -void decrement_counters_in_path (struct path * p_s_search_path); -void pathrelse (struct path * p_s_search_path); -int reiserfs_check_path(struct path *p) ; -void pathrelse_and_restore (struct super_block *s, struct path * p_s_search_path); - -int reiserfs_insert_item (struct reiserfs_transaction_handle *th, - struct path * path, - const struct cpu_key * key, - struct item_head * ih, - struct inode *inode, const char * body); - -int reiserfs_paste_into_item (struct reiserfs_transaction_handle *th, - struct path * path, - const struct cpu_key * key, - struct inode *inode, - const char * body, int paste_size); - -int reiserfs_cut_from_item (struct reiserfs_transaction_handle *th, - struct path * path, - struct cpu_key * key, - struct inode * inode, - struct page *page, - loff_t new_file_size); - -int reiserfs_delete_item (struct reiserfs_transaction_handle *th, - struct path * path, - const struct cpu_key * key, - struct inode * inode, - struct buffer_head * p_s_un_bh); - -void reiserfs_delete_solid_item (struct reiserfs_transaction_handle *th, - struct inode *inode, struct reiserfs_key * key); -int reiserfs_delete_object (struct reiserfs_transaction_handle *th, struct inode * p_s_inode); -int reiserfs_do_truncate (struct reiserfs_transaction_handle *th, - struct inode * p_s_inode, struct page *, - int update_timestamps); +int search_for_position_by_key(struct super_block *p_s_sb, + const struct cpu_key *p_s_cpu_key, + struct path *p_s_search_path); +extern void decrement_bcount(struct buffer_head *p_s_bh); +void decrement_counters_in_path(struct path *p_s_search_path); +void pathrelse(struct path *p_s_search_path); +int reiserfs_check_path(struct path *p); +void pathrelse_and_restore(struct super_block *s, struct path *p_s_search_path); + +int reiserfs_insert_item(struct reiserfs_transaction_handle *th, + struct path *path, + const struct cpu_key *key, + struct item_head *ih, + struct inode *inode, const char *body); + +int reiserfs_paste_into_item(struct reiserfs_transaction_handle *th, + struct path *path, + const struct cpu_key *key, + struct inode *inode, + const char *body, int paste_size); + +int reiserfs_cut_from_item(struct reiserfs_transaction_handle *th, + struct path *path, + struct cpu_key *key, + struct inode *inode, + struct page *page, loff_t new_file_size); + +int reiserfs_delete_item(struct reiserfs_transaction_handle *th, + struct path *path, + const struct cpu_key *key, + struct inode *inode, struct buffer_head *p_s_un_bh); + +void reiserfs_delete_solid_item(struct reiserfs_transaction_handle *th, + struct inode *inode, struct reiserfs_key *key); +int reiserfs_delete_object(struct reiserfs_transaction_handle *th, + struct inode *p_s_inode); +int reiserfs_do_truncate(struct reiserfs_transaction_handle *th, + struct inode *p_s_inode, struct page *, + int update_timestamps); #define i_block_size(inode) ((inode)->i_sb->s_blocksize) #define file_size(inode) ((inode)->i_size) @@ -1885,66 +1849,67 @@ int reiserfs_do_truncate (struct reiserfs_transaction_handle *th, #define tail_has_to_be_packed(inode) (have_large_tails ((inode)->i_sb)?\ !STORE_TAIL_IN_UNFM_S1(file_size (inode), tail_size(inode), inode->i_sb->s_blocksize):have_small_tails ((inode)->i_sb)?!STORE_TAIL_IN_UNFM_S2(file_size (inode), tail_size(inode), inode->i_sb->s_blocksize):0 ) -void padd_item (char * item, int total_length, int length); +void padd_item(char *item, int total_length, int length); /* inode.c */ /* args for the create parameter of reiserfs_get_block */ -#define GET_BLOCK_NO_CREATE 0 /* don't create new blocks or convert tails */ -#define GET_BLOCK_CREATE 1 /* add anything you need to find block */ -#define GET_BLOCK_NO_HOLE 2 /* return -ENOENT for file holes */ -#define GET_BLOCK_READ_DIRECT 4 /* read the tail if indirect item not found */ -#define GET_BLOCK_NO_ISEM 8 /* i_sem is not held, don't preallocate */ -#define GET_BLOCK_NO_DANGLE 16 /* don't leave any transactions running */ - -int restart_transaction(struct reiserfs_transaction_handle *th, struct inode *inode, struct path *path); -void reiserfs_read_locked_inode(struct inode * inode, struct reiserfs_iget_args *args) ; -int reiserfs_find_actor(struct inode * inode, void *p) ; -int reiserfs_init_locked_inode(struct inode * inode, void *p) ; -void reiserfs_delete_inode (struct inode * inode); -int reiserfs_write_inode (struct inode * inode, int) ; -int reiserfs_get_block (struct inode * inode, sector_t block, struct buffer_head * bh_result, int create); -struct dentry *reiserfs_get_dentry(struct super_block *, void *) ; -struct dentry *reiserfs_decode_fh(struct super_block *sb, __u32 *data, - int len, int fhtype, - int (*acceptable)(void *contect, struct dentry *de), - void *context) ; -int reiserfs_encode_fh( struct dentry *dentry, __u32 *data, int *lenp, - int connectable ); - -int reiserfs_truncate_file(struct inode *, int update_timestamps) ; -void make_cpu_key (struct cpu_key * cpu_key, struct inode * inode, loff_t offset, - int type, int key_length); -void make_le_item_head (struct item_head * ih, const struct cpu_key * key, - int version, - loff_t offset, int type, int length, int entry_count); -struct inode * reiserfs_iget (struct super_block * s, - const struct cpu_key * key); - - -int reiserfs_new_inode (struct reiserfs_transaction_handle *th, - struct inode * dir, int mode, - const char * symname, loff_t i_size, - struct dentry *dentry, struct inode *inode); - -void reiserfs_update_sd_size (struct reiserfs_transaction_handle *th, - struct inode * inode, loff_t size); +#define GET_BLOCK_NO_CREATE 0 /* don't create new blocks or convert tails */ +#define GET_BLOCK_CREATE 1 /* add anything you need to find block */ +#define GET_BLOCK_NO_HOLE 2 /* return -ENOENT for file holes */ +#define GET_BLOCK_READ_DIRECT 4 /* read the tail if indirect item not found */ +#define GET_BLOCK_NO_ISEM 8 /* i_sem is not held, don't preallocate */ +#define GET_BLOCK_NO_DANGLE 16 /* don't leave any transactions running */ + +int restart_transaction(struct reiserfs_transaction_handle *th, + struct inode *inode, struct path *path); +void reiserfs_read_locked_inode(struct inode *inode, + struct reiserfs_iget_args *args); +int reiserfs_find_actor(struct inode *inode, void *p); +int reiserfs_init_locked_inode(struct inode *inode, void *p); +void reiserfs_delete_inode(struct inode *inode); +int reiserfs_write_inode(struct inode *inode, int); +int reiserfs_get_block(struct inode *inode, sector_t block, + struct buffer_head *bh_result, int create); +struct dentry *reiserfs_get_dentry(struct super_block *, void *); +struct dentry *reiserfs_decode_fh(struct super_block *sb, __u32 * data, + int len, int fhtype, + int (*acceptable) (void *contect, + struct dentry * de), + void *context); +int reiserfs_encode_fh(struct dentry *dentry, __u32 * data, int *lenp, + int connectable); + +int reiserfs_truncate_file(struct inode *, int update_timestamps); +void make_cpu_key(struct cpu_key *cpu_key, struct inode *inode, loff_t offset, + int type, int key_length); +void make_le_item_head(struct item_head *ih, const struct cpu_key *key, + int version, + loff_t offset, int type, int length, int entry_count); +struct inode *reiserfs_iget(struct super_block *s, const struct cpu_key *key); + +int reiserfs_new_inode(struct reiserfs_transaction_handle *th, + struct inode *dir, int mode, + const char *symname, loff_t i_size, + struct dentry *dentry, struct inode *inode); + +void reiserfs_update_sd_size(struct reiserfs_transaction_handle *th, + struct inode *inode, loff_t size); static inline void reiserfs_update_sd(struct reiserfs_transaction_handle *th, - struct inode *inode) + struct inode *inode) { - reiserfs_update_sd_size(th, inode, inode->i_size) ; + reiserfs_update_sd_size(th, inode, inode->i_size); } -void sd_attrs_to_i_attrs( __u16 sd_attrs, struct inode *inode ); -void i_attrs_to_sd_attrs( struct inode *inode, __u16 *sd_attrs ); +void sd_attrs_to_i_attrs(__u16 sd_attrs, struct inode *inode); +void i_attrs_to_sd_attrs(struct inode *inode, __u16 * sd_attrs); int reiserfs_setattr(struct dentry *dentry, struct iattr *attr); /* namei.c */ -void set_de_name_and_namelen (struct reiserfs_dir_entry * de); -int search_by_entry_key (struct super_block * sb, const struct cpu_key * key, - struct path * path, - struct reiserfs_dir_entry * de); -struct dentry *reiserfs_get_parent(struct dentry *) ; +void set_de_name_and_namelen(struct reiserfs_dir_entry *de); +int search_by_entry_key(struct super_block *sb, const struct cpu_key *key, + struct path *path, struct reiserfs_dir_entry *de); +struct dentry *reiserfs_get_parent(struct dentry *); /* procfs.c */ #if defined( CONFIG_PROC_FS ) && defined( CONFIG_REISERFS_PROC_INFO ) @@ -1953,15 +1918,15 @@ struct dentry *reiserfs_get_parent(struct dentry *) ; #undef REISERFS_PROC_INFO #endif -int reiserfs_proc_info_init( struct super_block *sb ); -int reiserfs_proc_info_done( struct super_block *sb ); -struct proc_dir_entry *reiserfs_proc_register_global( char *name, - read_proc_t *func ); -void reiserfs_proc_unregister_global( const char *name ); -int reiserfs_proc_info_global_init( void ); -int reiserfs_proc_info_global_done( void ); -int reiserfs_global_version_in_proc( char *buffer, char **start, off_t offset, - int count, int *eof, void *data ); +int reiserfs_proc_info_init(struct super_block *sb); +int reiserfs_proc_info_done(struct super_block *sb); +struct proc_dir_entry *reiserfs_proc_register_global(char *name, + read_proc_t * func); +void reiserfs_proc_unregister_global(const char *name); +int reiserfs_proc_info_global_init(void); +int reiserfs_proc_info_global_done(void); +int reiserfs_global_version_in_proc(char *buffer, char **start, off_t offset, + int count, int *eof, void *data); #if defined( REISERFS_PROC_INFO ) @@ -1993,123 +1958,132 @@ extern struct inode_operations reiserfs_special_inode_operations; extern struct file_operations reiserfs_dir_operations; /* tail_conversion.c */ -int direct2indirect (struct reiserfs_transaction_handle *, struct inode *, struct path *, struct buffer_head *, loff_t); -int indirect2direct (struct reiserfs_transaction_handle *, struct inode *, struct page *, struct path *, const struct cpu_key *, loff_t, char *); -void reiserfs_unmap_buffer(struct buffer_head *) ; - +int direct2indirect(struct reiserfs_transaction_handle *, struct inode *, + struct path *, struct buffer_head *, loff_t); +int indirect2direct(struct reiserfs_transaction_handle *, struct inode *, + struct page *, struct path *, const struct cpu_key *, + loff_t, char *); +void reiserfs_unmap_buffer(struct buffer_head *); /* file.c */ extern struct inode_operations reiserfs_file_inode_operations; extern struct file_operations reiserfs_file_operations; -extern struct address_space_operations reiserfs_address_space_operations ; +extern struct address_space_operations reiserfs_address_space_operations; /* fix_nodes.c */ #ifdef CONFIG_REISERFS_CHECK -void * reiserfs_kmalloc (size_t size, int flags, struct super_block * s); -void reiserfs_kfree (const void * vp, size_t size, struct super_block * s); +void *reiserfs_kmalloc(size_t size, int flags, struct super_block *s); +void reiserfs_kfree(const void *vp, size_t size, struct super_block *s); #else static inline void *reiserfs_kmalloc(size_t size, int flags, - struct super_block *s) + struct super_block *s) { return kmalloc(size, flags); } static inline void reiserfs_kfree(const void *vp, size_t size, - struct super_block *s) + struct super_block *s) { kfree(vp); } #endif -int fix_nodes (int n_op_mode, struct tree_balance * p_s_tb, - struct item_head * p_s_ins_ih, const void *); -void unfix_nodes (struct tree_balance *); - +int fix_nodes(int n_op_mode, struct tree_balance *p_s_tb, + struct item_head *p_s_ins_ih, const void *); +void unfix_nodes(struct tree_balance *); /* prints.c */ -void reiserfs_panic (struct super_block * s, const char * fmt, ...) __attribute__ ( ( noreturn ) ); -void reiserfs_info (struct super_block *s, const char * fmt, ...); -void reiserfs_debug (struct super_block *s, int level, const char * fmt, ...); -void print_indirect_item (struct buffer_head * bh, int item_num); -void store_print_tb (struct tree_balance * tb); -void print_cur_tb (char * mes); -void print_de (struct reiserfs_dir_entry * de); -void print_bi (struct buffer_info * bi, char * mes); -#define PRINT_LEAF_ITEMS 1 /* print all items */ -#define PRINT_DIRECTORY_ITEMS 2 /* print directory items */ -#define PRINT_DIRECT_ITEMS 4 /* print contents of direct items */ -void print_block (struct buffer_head * bh, ...); -void print_bmap (struct super_block * s, int silent); -void print_bmap_block (int i, char * data, int size, int silent); +void reiserfs_panic(struct super_block *s, const char *fmt, ...) + __attribute__ ((noreturn)); +void reiserfs_info(struct super_block *s, const char *fmt, ...); +void reiserfs_debug(struct super_block *s, int level, const char *fmt, ...); +void print_indirect_item(struct buffer_head *bh, int item_num); +void store_print_tb(struct tree_balance *tb); +void print_cur_tb(char *mes); +void print_de(struct reiserfs_dir_entry *de); +void print_bi(struct buffer_info *bi, char *mes); +#define PRINT_LEAF_ITEMS 1 /* print all items */ +#define PRINT_DIRECTORY_ITEMS 2 /* print directory items */ +#define PRINT_DIRECT_ITEMS 4 /* print contents of direct items */ +void print_block(struct buffer_head *bh, ...); +void print_bmap(struct super_block *s, int silent); +void print_bmap_block(int i, char *data, int size, int silent); /*void print_super_block (struct super_block * s, char * mes);*/ -void print_objectid_map (struct super_block * s); -void print_block_head (struct buffer_head * bh, char * mes); -void check_leaf (struct buffer_head * bh); -void check_internal (struct buffer_head * bh); -void print_statistics (struct super_block * s); -char * reiserfs_hashname(int code); +void print_objectid_map(struct super_block *s); +void print_block_head(struct buffer_head *bh, char *mes); +void check_leaf(struct buffer_head *bh); +void check_internal(struct buffer_head *bh); +void print_statistics(struct super_block *s); +char *reiserfs_hashname(int code); /* lbalance.c */ -int leaf_move_items (int shift_mode, struct tree_balance * tb, int mov_num, int mov_bytes, struct buffer_head * Snew); -int leaf_shift_left (struct tree_balance * tb, int shift_num, int shift_bytes); -int leaf_shift_right (struct tree_balance * tb, int shift_num, int shift_bytes); -void leaf_delete_items (struct buffer_info * cur_bi, int last_first, int first, int del_num, int del_bytes); -void leaf_insert_into_buf (struct buffer_info * bi, int before, - struct item_head * inserted_item_ih, const char * inserted_item_body, int zeros_number); -void leaf_paste_in_buffer (struct buffer_info * bi, int pasted_item_num, - int pos_in_item, int paste_size, const char * body, int zeros_number); -void leaf_cut_from_buffer (struct buffer_info * bi, int cut_item_num, int pos_in_item, - int cut_size); -void leaf_paste_entries (struct buffer_head * bh, int item_num, int before, - int new_entry_count, struct reiserfs_de_head * new_dehs, const char * records, int paste_size); +int leaf_move_items(int shift_mode, struct tree_balance *tb, int mov_num, + int mov_bytes, struct buffer_head *Snew); +int leaf_shift_left(struct tree_balance *tb, int shift_num, int shift_bytes); +int leaf_shift_right(struct tree_balance *tb, int shift_num, int shift_bytes); +void leaf_delete_items(struct buffer_info *cur_bi, int last_first, int first, + int del_num, int del_bytes); +void leaf_insert_into_buf(struct buffer_info *bi, int before, + struct item_head *inserted_item_ih, + const char *inserted_item_body, int zeros_number); +void leaf_paste_in_buffer(struct buffer_info *bi, int pasted_item_num, + int pos_in_item, int paste_size, const char *body, + int zeros_number); +void leaf_cut_from_buffer(struct buffer_info *bi, int cut_item_num, + int pos_in_item, int cut_size); +void leaf_paste_entries(struct buffer_head *bh, int item_num, int before, + int new_entry_count, struct reiserfs_de_head *new_dehs, + const char *records, int paste_size); /* ibalance.c */ -int balance_internal (struct tree_balance * , int, int, struct item_head * , - struct buffer_head **); +int balance_internal(struct tree_balance *, int, int, struct item_head *, + struct buffer_head **); /* do_balance.c */ -void do_balance_mark_leaf_dirty (struct tree_balance * tb, - struct buffer_head * bh, int flag); +void do_balance_mark_leaf_dirty(struct tree_balance *tb, + struct buffer_head *bh, int flag); #define do_balance_mark_internal_dirty do_balance_mark_leaf_dirty #define do_balance_mark_sb_dirty do_balance_mark_leaf_dirty -void do_balance (struct tree_balance * tb, struct item_head * ih, - const char * body, int flag); -void reiserfs_invalidate_buffer (struct tree_balance * tb, struct buffer_head * bh); +void do_balance(struct tree_balance *tb, struct item_head *ih, + const char *body, int flag); +void reiserfs_invalidate_buffer(struct tree_balance *tb, + struct buffer_head *bh); -int get_left_neighbor_position (struct tree_balance * tb, int h); -int get_right_neighbor_position (struct tree_balance * tb, int h); -void replace_key (struct tree_balance * tb, struct buffer_head *, int, struct buffer_head *, int); -void make_empty_node (struct buffer_info *); -struct buffer_head * get_FEB (struct tree_balance *); +int get_left_neighbor_position(struct tree_balance *tb, int h); +int get_right_neighbor_position(struct tree_balance *tb, int h); +void replace_key(struct tree_balance *tb, struct buffer_head *, int, + struct buffer_head *, int); +void make_empty_node(struct buffer_info *); +struct buffer_head *get_FEB(struct tree_balance *); /* bitmap.c */ /* structure contains hints for block allocator, and it is a container for * arguments, such as node, search path, transaction_handle, etc. */ - struct __reiserfs_blocknr_hint { - struct inode * inode; /* inode passed to allocator, if we allocate unf. nodes */ - long block; /* file offset, in blocks */ - struct in_core_key key; - struct path * path; /* search path, used by allocator to deternine search_start by - * various ways */ - struct reiserfs_transaction_handle * th; /* transaction handle is needed to log super blocks and - * bitmap blocks changes */ - b_blocknr_t beg, end; - b_blocknr_t search_start; /* a field used to transfer search start value (block number) +struct __reiserfs_blocknr_hint { + struct inode *inode; /* inode passed to allocator, if we allocate unf. nodes */ + long block; /* file offset, in blocks */ + struct in_core_key key; + struct path *path; /* search path, used by allocator to deternine search_start by + * various ways */ + struct reiserfs_transaction_handle *th; /* transaction handle is needed to log super blocks and + * bitmap blocks changes */ + b_blocknr_t beg, end; + b_blocknr_t search_start; /* a field used to transfer search start value (block number) * between different block allocator procedures * (determine_search_start() and others) */ - int prealloc_size; /* is set in determine_prealloc_size() function, used by underlayed - * function that do actual allocation */ + int prealloc_size; /* is set in determine_prealloc_size() function, used by underlayed + * function that do actual allocation */ - unsigned formatted_node:1; /* the allocator uses different polices for getting disk space for + unsigned formatted_node:1; /* the allocator uses different polices for getting disk space for * formatted/unformatted blocks with/without preallocation */ - unsigned preallocate:1; + unsigned preallocate:1; }; typedef struct __reiserfs_blocknr_hint reiserfs_blocknr_hint_t; -int reiserfs_parse_alloc_options (struct super_block *, char *); -void reiserfs_init_alloc_options (struct super_block *s); +int reiserfs_parse_alloc_options(struct super_block *, char *); +void reiserfs_init_alloc_options(struct super_block *s); /* * given a directory, this will tell you what packing locality @@ -2118,68 +2092,72 @@ void reiserfs_init_alloc_options (struct super_block *s); */ __le32 reiserfs_choose_packing(struct inode *dir); -int is_reusable (struct super_block * s, b_blocknr_t block, int bit_value); -void reiserfs_free_block (struct reiserfs_transaction_handle *th, struct inode *, b_blocknr_t, int for_unformatted); -int reiserfs_allocate_blocknrs(reiserfs_blocknr_hint_t *, b_blocknr_t * , int, int); -extern inline int reiserfs_new_form_blocknrs (struct tree_balance * tb, - b_blocknr_t *new_blocknrs, int amount_needed) +int is_reusable(struct super_block *s, b_blocknr_t block, int bit_value); +void reiserfs_free_block(struct reiserfs_transaction_handle *th, struct inode *, + b_blocknr_t, int for_unformatted); +int reiserfs_allocate_blocknrs(reiserfs_blocknr_hint_t *, b_blocknr_t *, int, + int); +extern inline int reiserfs_new_form_blocknrs(struct tree_balance *tb, + b_blocknr_t * new_blocknrs, + int amount_needed) { - reiserfs_blocknr_hint_t hint = { - .th = tb->transaction_handle, - .path = tb->tb_path, - .inode = NULL, - .key = tb->key, - .block = 0, - .formatted_node = 1 - }; - return reiserfs_allocate_blocknrs(&hint, new_blocknrs, amount_needed, 0); + reiserfs_blocknr_hint_t hint = { + .th = tb->transaction_handle, + .path = tb->tb_path, + .inode = NULL, + .key = tb->key, + .block = 0, + .formatted_node = 1 + }; + return reiserfs_allocate_blocknrs(&hint, new_blocknrs, amount_needed, + 0); } -extern inline int reiserfs_new_unf_blocknrs (struct reiserfs_transaction_handle *th, - struct inode *inode, - b_blocknr_t *new_blocknrs, - struct path * path, long block) +extern inline int reiserfs_new_unf_blocknrs(struct reiserfs_transaction_handle + *th, struct inode *inode, + b_blocknr_t * new_blocknrs, + struct path *path, long block) { - reiserfs_blocknr_hint_t hint = { - .th = th, - .path = path, - .inode = inode, - .block = block, - .formatted_node = 0, - .preallocate = 0 - }; - return reiserfs_allocate_blocknrs(&hint, new_blocknrs, 1, 0); + reiserfs_blocknr_hint_t hint = { + .th = th, + .path = path, + .inode = inode, + .block = block, + .formatted_node = 0, + .preallocate = 0 + }; + return reiserfs_allocate_blocknrs(&hint, new_blocknrs, 1, 0); } #ifdef REISERFS_PREALLOCATE -extern inline int reiserfs_new_unf_blocknrs2(struct reiserfs_transaction_handle *th, - struct inode * inode, - b_blocknr_t *new_blocknrs, - struct path * path, long block) +extern inline int reiserfs_new_unf_blocknrs2(struct reiserfs_transaction_handle + *th, struct inode *inode, + b_blocknr_t * new_blocknrs, + struct path *path, long block) { - reiserfs_blocknr_hint_t hint = { - .th = th, - .path = path, - .inode = inode, - .block = block, - .formatted_node = 0, - .preallocate = 1 - }; - return reiserfs_allocate_blocknrs(&hint, new_blocknrs, 1, 0); + reiserfs_blocknr_hint_t hint = { + .th = th, + .path = path, + .inode = inode, + .block = block, + .formatted_node = 0, + .preallocate = 1 + }; + return reiserfs_allocate_blocknrs(&hint, new_blocknrs, 1, 0); } -void reiserfs_discard_prealloc (struct reiserfs_transaction_handle *th, - struct inode * inode); -void reiserfs_discard_all_prealloc (struct reiserfs_transaction_handle *th); +void reiserfs_discard_prealloc(struct reiserfs_transaction_handle *th, + struct inode *inode); +void reiserfs_discard_all_prealloc(struct reiserfs_transaction_handle *th); #endif -void reiserfs_claim_blocks_to_be_allocated( struct super_block *sb, int blocks); -void reiserfs_release_claimed_blocks( struct super_block *sb, int blocks); +void reiserfs_claim_blocks_to_be_allocated(struct super_block *sb, int blocks); +void reiserfs_release_claimed_blocks(struct super_block *sb, int blocks); int reiserfs_can_fit_pages(struct super_block *sb); /* hashes.c */ -__u32 keyed_hash (const signed char *msg, int len); -__u32 yura_hash (const signed char *msg, int len); -__u32 r5_hash (const signed char *msg, int len); +__u32 keyed_hash(const signed char *msg, int len); +__u32 yura_hash(const signed char *msg, int len); +__u32 r5_hash(const signed char *msg, int len); /* the ext2 bit routines adjust for big or little endian as ** appropriate for the arch, so in our laziness we use them rather @@ -2199,11 +2177,10 @@ __u32 r5_hash (const signed char *msg, int len); absolutely safe */ #define SPARE_SPACE 500 - /* prototypes from ioctl.c */ -int reiserfs_ioctl (struct inode * inode, struct file * filp, - unsigned int cmd, unsigned long arg); - +int reiserfs_ioctl(struct inode *inode, struct file *filp, + unsigned int cmd, unsigned long arg); + /* ioctl's command */ #define REISERFS_IOC_UNPACK _IOW(0xCD,1,long) /* define following flags to be the same as in ext2, so that chattr(1), @@ -2218,10 +2195,8 @@ int reiserfs_ioctl (struct inode * inode, struct file * filp, would evolve into real per-fs locks */ #define reiserfs_write_lock( sb ) lock_kernel() #define reiserfs_write_unlock( sb ) unlock_kernel() - + /* xattr stuff */ #define REISERFS_XATTR_DIR_SEM(s) (REISERFS_SB(s)->xattr_dir_sem) -#endif /* _LINUX_REISER_FS_H */ - - +#endif /* _LINUX_REISER_FS_H */ diff --git a/include/linux/reiserfs_fs_i.h b/include/linux/reiserfs_fs_i.h index e321eb050d65..149be8d9a0c9 100644 --- a/include/linux/reiserfs_fs_i.h +++ b/include/linux/reiserfs_fs_i.h @@ -10,54 +10,53 @@ typedef enum { /** this says what format of key do all items (but stat data) of an object have. If this is set, that format is 3.6 otherwise - 3.5 */ - i_item_key_version_mask = 0x0001, + i_item_key_version_mask = 0x0001, /** If this is unset, object has 3.5 stat data, otherwise, it has 3.6 stat data with 64bit size, 32bit nlink etc. */ - i_stat_data_version_mask = 0x0002, + i_stat_data_version_mask = 0x0002, /** file might need tail packing on close */ - i_pack_on_close_mask = 0x0004, + i_pack_on_close_mask = 0x0004, /** don't pack tail of file */ - i_nopack_mask = 0x0008, + i_nopack_mask = 0x0008, /** If those is set, "safe link" was created for this file during truncate or unlink. Safe link is used to avoid leakage of disk space on crash with some files open, but unlinked. */ - i_link_saved_unlink_mask = 0x0010, - i_link_saved_truncate_mask = 0x0020, - i_has_xattr_dir = 0x0040, - i_data_log = 0x0080, + i_link_saved_unlink_mask = 0x0010, + i_link_saved_truncate_mask = 0x0020, + i_has_xattr_dir = 0x0040, + i_data_log = 0x0080, } reiserfs_inode_flags; - struct reiserfs_inode_info { - __u32 i_key [4];/* key is still 4 32 bit integers */ + __u32 i_key[4]; /* key is still 4 32 bit integers */ /** transient inode flags that are never stored on disk. Bitmasks for this field are defined above. */ - __u32 i_flags; + __u32 i_flags; - __u32 i_first_direct_byte; // offset of first byte stored in direct item. + __u32 i_first_direct_byte; // offset of first byte stored in direct item. - /* copy of persistent inode flags read from sd_attrs. */ - __u32 i_attrs; + /* copy of persistent inode flags read from sd_attrs. */ + __u32 i_attrs; - int i_prealloc_block; /* first unused block of a sequence of unused blocks */ - int i_prealloc_count; /* length of that sequence */ - struct list_head i_prealloc_list; /* per-transaction list of inodes which - * have preallocated blocks */ + int i_prealloc_block; /* first unused block of a sequence of unused blocks */ + int i_prealloc_count; /* length of that sequence */ + struct list_head i_prealloc_list; /* per-transaction list of inodes which + * have preallocated blocks */ - unsigned new_packing_locality:1; /* new_packig_locality is created; new blocks - * for the contents of this directory should be - * displaced */ + unsigned new_packing_locality:1; /* new_packig_locality is created; new blocks + * for the contents of this directory should be + * displaced */ - /* we use these for fsync or O_SYNC to decide which transaction - ** needs to be committed in order for this inode to be properly - ** flushed */ - unsigned long i_trans_id ; - struct reiserfs_journal_list *i_jl; + /* we use these for fsync or O_SYNC to decide which transaction + ** needs to be committed in order for this inode to be properly + ** flushed */ + unsigned long i_trans_id; + struct reiserfs_journal_list *i_jl; - struct posix_acl *i_acl_access; - struct posix_acl *i_acl_default; - struct rw_semaphore xattr_sem; - struct inode vfs_inode; + struct posix_acl *i_acl_access; + struct posix_acl *i_acl_default; + struct rw_semaphore xattr_sem; + struct inode vfs_inode; }; #endif diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h index 31c709d0fe18..3e68592e52e9 100644 --- a/include/linux/reiserfs_fs_sb.h +++ b/include/linux/reiserfs_fs_sb.h @@ -10,7 +10,7 @@ #endif typedef enum { - reiserfs_attrs_cleared = 0x00000001, + reiserfs_attrs_cleared = 0x00000001, } reiserfs_super_block_flags; /* struct reiserfs_super_block accessors/mutators @@ -61,7 +61,7 @@ typedef enum { #define sb_umount_state(sbp) (le16_to_cpu((sbp)->s_v1.s_umount_state)) #define set_sb_umount_state(sbp,v) ((sbp)->s_v1.s_umount_state = cpu_to_le16(v)) #define sb_fs_state(sbp) (le16_to_cpu((sbp)->s_v1.s_fs_state)) -#define set_sb_fs_state(sbp,v) ((sbp)->s_v1.s_fs_state = cpu_to_le16(v)) +#define set_sb_fs_state(sbp,v) ((sbp)->s_v1.s_fs_state = cpu_to_le16(v)) #define sb_hash_function_code(sbp) \ (le32_to_cpu((sbp)->s_v1.s_hash_function_code)) #define set_sb_hash_function_code(sbp,v) \ @@ -103,10 +103,10 @@ typedef enum { /* don't mess with these for a while */ /* we have a node size define somewhere in reiserfs_fs.h. -Hans */ -#define JOURNAL_BLOCK_SIZE 4096 /* BUG gotta get rid of this */ -#define JOURNAL_MAX_CNODE 1500 /* max cnodes to allocate. */ -#define JOURNAL_HASH_SIZE 8192 -#define JOURNAL_NUM_BITMAPS 5 /* number of copies of the bitmaps to have floating. Must be >= 2 */ +#define JOURNAL_BLOCK_SIZE 4096 /* BUG gotta get rid of this */ +#define JOURNAL_MAX_CNODE 1500 /* max cnodes to allocate. */ +#define JOURNAL_HASH_SIZE 8192 +#define JOURNAL_NUM_BITMAPS 5 /* number of copies of the bitmaps to have floating. Must be >= 2 */ /* One of these for every block in every transaction ** Each one is in two hash tables. First, a hash of the current transaction, and after journal_end, a @@ -117,27 +117,27 @@ typedef enum { ** to a given transaction. */ struct reiserfs_journal_cnode { - struct buffer_head *bh ; /* real buffer head */ - struct super_block *sb ; /* dev of real buffer head */ - __u32 blocknr ; /* block number of real buffer head, == 0 when buffer on disk */ - long state ; - struct reiserfs_journal_list *jlist ; /* journal list this cnode lives in */ - struct reiserfs_journal_cnode *next ; /* next in transaction list */ - struct reiserfs_journal_cnode *prev ; /* prev in transaction list */ - struct reiserfs_journal_cnode *hprev ; /* prev in hash list */ - struct reiserfs_journal_cnode *hnext ; /* next in hash list */ + struct buffer_head *bh; /* real buffer head */ + struct super_block *sb; /* dev of real buffer head */ + __u32 blocknr; /* block number of real buffer head, == 0 when buffer on disk */ + long state; + struct reiserfs_journal_list *jlist; /* journal list this cnode lives in */ + struct reiserfs_journal_cnode *next; /* next in transaction list */ + struct reiserfs_journal_cnode *prev; /* prev in transaction list */ + struct reiserfs_journal_cnode *hprev; /* prev in hash list */ + struct reiserfs_journal_cnode *hnext; /* next in hash list */ }; struct reiserfs_bitmap_node { - int id ; - char *data ; - struct list_head list ; -} ; + int id; + char *data; + struct list_head list; +}; struct reiserfs_list_bitmap { - struct reiserfs_journal_list *journal_list ; - struct reiserfs_bitmap_node **bitmaps ; -} ; + struct reiserfs_journal_list *journal_list; + struct reiserfs_bitmap_node **bitmaps; +}; /* ** one of these for each transaction. The most important part here is the j_realblock. @@ -146,273 +146,269 @@ struct reiserfs_list_bitmap { ** and to make sure every real block in a transaction is on disk before allowing the log area ** to be overwritten */ struct reiserfs_journal_list { - unsigned long j_start ; - unsigned long j_state; - unsigned long j_len ; - atomic_t j_nonzerolen ; - atomic_t j_commit_left ; - atomic_t j_older_commits_done ; /* all commits older than this on disk*/ - struct semaphore j_commit_lock; - unsigned long j_trans_id ; - time_t j_timestamp ; - struct reiserfs_list_bitmap *j_list_bitmap ; - struct buffer_head *j_commit_bh ; /* commit buffer head */ - struct reiserfs_journal_cnode *j_realblock ; - struct reiserfs_journal_cnode *j_freedlist ; /* list of buffers that were freed during this trans. free each of these on flush */ - /* time ordered list of all active transactions */ - struct list_head j_list; - - /* time ordered list of all transactions we haven't tried to flush yet */ - struct list_head j_working_list; - - /* list of tail conversion targets in need of flush before commit */ - struct list_head j_tail_bh_list; - /* list of data=ordered buffers in need of flush before commit */ - struct list_head j_bh_list; - int j_refcount; -} ; + unsigned long j_start; + unsigned long j_state; + unsigned long j_len; + atomic_t j_nonzerolen; + atomic_t j_commit_left; + atomic_t j_older_commits_done; /* all commits older than this on disk */ + struct semaphore j_commit_lock; + unsigned long j_trans_id; + time_t j_timestamp; + struct reiserfs_list_bitmap *j_list_bitmap; + struct buffer_head *j_commit_bh; /* commit buffer head */ + struct reiserfs_journal_cnode *j_realblock; + struct reiserfs_journal_cnode *j_freedlist; /* list of buffers that were freed during this trans. free each of these on flush */ + /* time ordered list of all active transactions */ + struct list_head j_list; + + /* time ordered list of all transactions we haven't tried to flush yet */ + struct list_head j_working_list; + + /* list of tail conversion targets in need of flush before commit */ + struct list_head j_tail_bh_list; + /* list of data=ordered buffers in need of flush before commit */ + struct list_head j_bh_list; + int j_refcount; +}; struct reiserfs_journal { - struct buffer_head ** j_ap_blocks ; /* journal blocks on disk */ - struct reiserfs_journal_cnode *j_last ; /* newest journal block */ - struct reiserfs_journal_cnode *j_first ; /* oldest journal block. start here for traverse */ - - struct file *j_dev_file; - struct block_device *j_dev_bd; - int j_1st_reserved_block; /* first block on s_dev of reserved area journal */ - - long j_state ; - unsigned long j_trans_id ; - unsigned long j_mount_id ; - unsigned long j_start ; /* start of current waiting commit (index into j_ap_blocks) */ - unsigned long j_len ; /* lenght of current waiting commit */ - unsigned long j_len_alloc ; /* number of buffers requested by journal_begin() */ - atomic_t j_wcount ; /* count of writers for current commit */ - unsigned long j_bcount ; /* batch count. allows turning X transactions into 1 */ - unsigned long j_first_unflushed_offset ; /* first unflushed transactions offset */ - unsigned long j_last_flush_trans_id ; /* last fully flushed journal timestamp */ - struct buffer_head *j_header_bh ; - - time_t j_trans_start_time ; /* time this transaction started */ - struct semaphore j_lock; - struct semaphore j_flush_sem; - wait_queue_head_t j_join_wait ; /* wait for current transaction to finish before starting new one */ - atomic_t j_jlock ; /* lock for j_join_wait */ - int j_list_bitmap_index ; /* number of next list bitmap to use */ - int j_must_wait ; /* no more journal begins allowed. MUST sleep on j_join_wait */ - int j_next_full_flush ; /* next journal_end will flush all journal list */ - int j_next_async_flush ; /* next journal_end will flush all async commits */ - - int j_cnode_used ; /* number of cnodes on the used list */ - int j_cnode_free ; /* number of cnodes on the free list */ - - unsigned int j_trans_max ; /* max number of blocks in a transaction. */ - unsigned int j_max_batch ; /* max number of blocks to batch into a trans */ - unsigned int j_max_commit_age ; /* in seconds, how old can an async commit be */ - unsigned int j_max_trans_age ; /* in seconds, how old can a transaction be */ - unsigned int j_default_max_commit_age ; /* the default for the max commit age */ - - struct reiserfs_journal_cnode *j_cnode_free_list ; - struct reiserfs_journal_cnode *j_cnode_free_orig ; /* orig pointer returned from vmalloc */ - - struct reiserfs_journal_list *j_current_jl; - int j_free_bitmap_nodes ; - int j_used_bitmap_nodes ; - - int j_num_lists; /* total number of active transactions */ - int j_num_work_lists; /* number that need attention from kreiserfsd */ - - /* debugging to make sure things are flushed in order */ - int j_last_flush_id; - - /* debugging to make sure things are committed in order */ - int j_last_commit_id; - - struct list_head j_bitmap_nodes ; - struct list_head j_dirty_buffers ; - spinlock_t j_dirty_buffers_lock ; /* protects j_dirty_buffers */ - - /* list of all active transactions */ - struct list_head j_journal_list; - /* lists that haven't been touched by writeback attempts */ - struct list_head j_working_list; - - struct reiserfs_list_bitmap j_list_bitmap[JOURNAL_NUM_BITMAPS] ; /* array of bitmaps to record the deleted blocks */ - struct reiserfs_journal_cnode *j_hash_table[JOURNAL_HASH_SIZE] ; /* hash table for real buffer heads in current trans */ - struct reiserfs_journal_cnode *j_list_hash_table[JOURNAL_HASH_SIZE] ; /* hash table for all the real buffer heads in all - the transactions */ - struct list_head j_prealloc_list; /* list of inodes which have preallocated blocks */ - int j_persistent_trans; - unsigned long j_max_trans_size ; - unsigned long j_max_batch_size ; - - int j_errno; - - /* when flushing ordered buffers, throttle new ordered writers */ - struct work_struct j_work; - atomic_t j_async_throttle; + struct buffer_head **j_ap_blocks; /* journal blocks on disk */ + struct reiserfs_journal_cnode *j_last; /* newest journal block */ + struct reiserfs_journal_cnode *j_first; /* oldest journal block. start here for traverse */ + + struct file *j_dev_file; + struct block_device *j_dev_bd; + int j_1st_reserved_block; /* first block on s_dev of reserved area journal */ + + long j_state; + unsigned long j_trans_id; + unsigned long j_mount_id; + unsigned long j_start; /* start of current waiting commit (index into j_ap_blocks) */ + unsigned long j_len; /* lenght of current waiting commit */ + unsigned long j_len_alloc; /* number of buffers requested by journal_begin() */ + atomic_t j_wcount; /* count of writers for current commit */ + unsigned long j_bcount; /* batch count. allows turning X transactions into 1 */ + unsigned long j_first_unflushed_offset; /* first unflushed transactions offset */ + unsigned long j_last_flush_trans_id; /* last fully flushed journal timestamp */ + struct buffer_head *j_header_bh; + + time_t j_trans_start_time; /* time this transaction started */ + struct semaphore j_lock; + struct semaphore j_flush_sem; + wait_queue_head_t j_join_wait; /* wait for current transaction to finish before starting new one */ + atomic_t j_jlock; /* lock for j_join_wait */ + int j_list_bitmap_index; /* number of next list bitmap to use */ + int j_must_wait; /* no more journal begins allowed. MUST sleep on j_join_wait */ + int j_next_full_flush; /* next journal_end will flush all journal list */ + int j_next_async_flush; /* next journal_end will flush all async commits */ + + int j_cnode_used; /* number of cnodes on the used list */ + int j_cnode_free; /* number of cnodes on the free list */ + + unsigned int j_trans_max; /* max number of blocks in a transaction. */ + unsigned int j_max_batch; /* max number of blocks to batch into a trans */ + unsigned int j_max_commit_age; /* in seconds, how old can an async commit be */ + unsigned int j_max_trans_age; /* in seconds, how old can a transaction be */ + unsigned int j_default_max_commit_age; /* the default for the max commit age */ + + struct reiserfs_journal_cnode *j_cnode_free_list; + struct reiserfs_journal_cnode *j_cnode_free_orig; /* orig pointer returned from vmalloc */ + + struct reiserfs_journal_list *j_current_jl; + int j_free_bitmap_nodes; + int j_used_bitmap_nodes; + + int j_num_lists; /* total number of active transactions */ + int j_num_work_lists; /* number that need attention from kreiserfsd */ + + /* debugging to make sure things are flushed in order */ + int j_last_flush_id; + + /* debugging to make sure things are committed in order */ + int j_last_commit_id; + + struct list_head j_bitmap_nodes; + struct list_head j_dirty_buffers; + spinlock_t j_dirty_buffers_lock; /* protects j_dirty_buffers */ + + /* list of all active transactions */ + struct list_head j_journal_list; + /* lists that haven't been touched by writeback attempts */ + struct list_head j_working_list; + + struct reiserfs_list_bitmap j_list_bitmap[JOURNAL_NUM_BITMAPS]; /* array of bitmaps to record the deleted blocks */ + struct reiserfs_journal_cnode *j_hash_table[JOURNAL_HASH_SIZE]; /* hash table for real buffer heads in current trans */ + struct reiserfs_journal_cnode *j_list_hash_table[JOURNAL_HASH_SIZE]; /* hash table for all the real buffer heads in all + the transactions */ + struct list_head j_prealloc_list; /* list of inodes which have preallocated blocks */ + int j_persistent_trans; + unsigned long j_max_trans_size; + unsigned long j_max_batch_size; + + int j_errno; + + /* when flushing ordered buffers, throttle new ordered writers */ + struct work_struct j_work; + atomic_t j_async_throttle; }; enum journal_state_bits { - J_WRITERS_BLOCKED = 1, /* set when new writers not allowed */ - J_WRITERS_QUEUED, /* set when log is full due to too many writers */ - J_ABORTED, /* set when log is aborted */ + J_WRITERS_BLOCKED = 1, /* set when new writers not allowed */ + J_WRITERS_QUEUED, /* set when log is full due to too many writers */ + J_ABORTED, /* set when log is aborted */ }; +#define JOURNAL_DESC_MAGIC "ReIsErLB" /* ick. magic string to find desc blocks in the journal */ -#define JOURNAL_DESC_MAGIC "ReIsErLB" /* ick. magic string to find desc blocks in the journal */ +typedef __u32(*hashf_t) (const signed char *, int); -typedef __u32 (*hashf_t) (const signed char *, int); - -struct reiserfs_bitmap_info -{ - // FIXME: Won't work with block sizes > 8K - __u16 first_zero_hint; - __u16 free_count; - struct buffer_head *bh; /* the actual bitmap */ +struct reiserfs_bitmap_info { + // FIXME: Won't work with block sizes > 8K + __u16 first_zero_hint; + __u16 free_count; + struct buffer_head *bh; /* the actual bitmap */ }; struct proc_dir_entry; #if defined( CONFIG_PROC_FS ) && defined( CONFIG_REISERFS_PROC_INFO ) typedef unsigned long int stat_cnt_t; -typedef struct reiserfs_proc_info_data -{ - spinlock_t lock; - int exiting; - int max_hash_collisions; - - stat_cnt_t breads; - stat_cnt_t bread_miss; - stat_cnt_t search_by_key; - stat_cnt_t search_by_key_fs_changed; - stat_cnt_t search_by_key_restarted; - - stat_cnt_t insert_item_restarted; - stat_cnt_t paste_into_item_restarted; - stat_cnt_t cut_from_item_restarted; - stat_cnt_t delete_solid_item_restarted; - stat_cnt_t delete_item_restarted; - - stat_cnt_t leaked_oid; - stat_cnt_t leaves_removable; - - /* balances per level. Use explicit 5 as MAX_HEIGHT is not visible yet. */ - stat_cnt_t balance_at[ 5 ]; /* XXX */ - /* sbk == search_by_key */ - stat_cnt_t sbk_read_at[ 5 ]; /* XXX */ - stat_cnt_t sbk_fs_changed[ 5 ]; - stat_cnt_t sbk_restarted[ 5 ]; - stat_cnt_t items_at[ 5 ]; /* XXX */ - stat_cnt_t free_at[ 5 ]; /* XXX */ - stat_cnt_t can_node_be_removed[ 5 ]; /* XXX */ - long int lnum[ 5 ]; /* XXX */ - long int rnum[ 5 ]; /* XXX */ - long int lbytes[ 5 ]; /* XXX */ - long int rbytes[ 5 ]; /* XXX */ - stat_cnt_t get_neighbors[ 5 ]; - stat_cnt_t get_neighbors_restart[ 5 ]; - stat_cnt_t need_l_neighbor[ 5 ]; - stat_cnt_t need_r_neighbor[ 5 ]; - - stat_cnt_t free_block; - struct __scan_bitmap_stats { - stat_cnt_t call; - stat_cnt_t wait; - stat_cnt_t bmap; - stat_cnt_t retry; - stat_cnt_t in_journal_hint; - stat_cnt_t in_journal_nohint; - stat_cnt_t stolen; - } scan_bitmap; - struct __journal_stats { - stat_cnt_t in_journal; - stat_cnt_t in_journal_bitmap; - stat_cnt_t in_journal_reusable; - stat_cnt_t lock_journal; - stat_cnt_t lock_journal_wait; - stat_cnt_t journal_being; - stat_cnt_t journal_relock_writers; - stat_cnt_t journal_relock_wcount; - stat_cnt_t mark_dirty; - stat_cnt_t mark_dirty_already; - stat_cnt_t mark_dirty_notjournal; - stat_cnt_t restore_prepared; - stat_cnt_t prepare; - stat_cnt_t prepare_retry; - } journal; +typedef struct reiserfs_proc_info_data { + spinlock_t lock; + int exiting; + int max_hash_collisions; + + stat_cnt_t breads; + stat_cnt_t bread_miss; + stat_cnt_t search_by_key; + stat_cnt_t search_by_key_fs_changed; + stat_cnt_t search_by_key_restarted; + + stat_cnt_t insert_item_restarted; + stat_cnt_t paste_into_item_restarted; + stat_cnt_t cut_from_item_restarted; + stat_cnt_t delete_solid_item_restarted; + stat_cnt_t delete_item_restarted; + + stat_cnt_t leaked_oid; + stat_cnt_t leaves_removable; + + /* balances per level. Use explicit 5 as MAX_HEIGHT is not visible yet. */ + stat_cnt_t balance_at[5]; /* XXX */ + /* sbk == search_by_key */ + stat_cnt_t sbk_read_at[5]; /* XXX */ + stat_cnt_t sbk_fs_changed[5]; + stat_cnt_t sbk_restarted[5]; + stat_cnt_t items_at[5]; /* XXX */ + stat_cnt_t free_at[5]; /* XXX */ + stat_cnt_t can_node_be_removed[5]; /* XXX */ + long int lnum[5]; /* XXX */ + long int rnum[5]; /* XXX */ + long int lbytes[5]; /* XXX */ + long int rbytes[5]; /* XXX */ + stat_cnt_t get_neighbors[5]; + stat_cnt_t get_neighbors_restart[5]; + stat_cnt_t need_l_neighbor[5]; + stat_cnt_t need_r_neighbor[5]; + + stat_cnt_t free_block; + struct __scan_bitmap_stats { + stat_cnt_t call; + stat_cnt_t wait; + stat_cnt_t bmap; + stat_cnt_t retry; + stat_cnt_t in_journal_hint; + stat_cnt_t in_journal_nohint; + stat_cnt_t stolen; + } scan_bitmap; + struct __journal_stats { + stat_cnt_t in_journal; + stat_cnt_t in_journal_bitmap; + stat_cnt_t in_journal_reusable; + stat_cnt_t lock_journal; + stat_cnt_t lock_journal_wait; + stat_cnt_t journal_being; + stat_cnt_t journal_relock_writers; + stat_cnt_t journal_relock_wcount; + stat_cnt_t mark_dirty; + stat_cnt_t mark_dirty_already; + stat_cnt_t mark_dirty_notjournal; + stat_cnt_t restore_prepared; + stat_cnt_t prepare; + stat_cnt_t prepare_retry; + } journal; } reiserfs_proc_info_data_t; #else -typedef struct reiserfs_proc_info_data -{} reiserfs_proc_info_data_t; +typedef struct reiserfs_proc_info_data { +} reiserfs_proc_info_data_t; #endif /* reiserfs union of in-core super block data */ -struct reiserfs_sb_info -{ - struct buffer_head * s_sbh; /* Buffer containing the super block */ - /* both the comment and the choice of - name are unclear for s_rs -Hans */ - struct reiserfs_super_block * s_rs; /* Pointer to the super block in the buffer */ - struct reiserfs_bitmap_info * s_ap_bitmap; - struct reiserfs_journal *s_journal ; /* pointer to journal information */ - unsigned short s_mount_state; /* reiserfs state (valid, invalid) */ - - /* Comment? -Hans */ - void (*end_io_handler)(struct buffer_head *, int); - hashf_t s_hash_function; /* pointer to function which is used - to sort names in directory. Set on - mount */ - unsigned long s_mount_opt; /* reiserfs's mount options are set - here (currently - NOTAIL, NOLOG, - REPLAYONLY) */ - - struct { /* This is a structure that describes block allocator options */ - unsigned long bits; /* Bitfield for enable/disable kind of options */ - unsigned long large_file_size; /* size started from which we consider file to be a large one(in blocks) */ - int border; /* percentage of disk, border takes */ - int preallocmin; /* Minimal file size (in blocks) starting from which we do preallocations */ - int preallocsize; /* Number of blocks we try to prealloc when file - reaches preallocmin size (in blocks) or - prealloc_list is empty. */ - } s_alloc_options; - - /* Comment? -Hans */ - wait_queue_head_t s_wait; - /* To be obsoleted soon by per buffer seals.. -Hans */ - atomic_t s_generation_counter; // increased by one every time the - // tree gets re-balanced - unsigned long s_properties; /* File system properties. Currently holds - on-disk FS format */ - - /* session statistics */ - int s_kmallocs; - int s_disk_reads; - int s_disk_writes; - int s_fix_nodes; - int s_do_balance; - int s_unneeded_left_neighbor; - int s_good_search_by_key_reada; - int s_bmaps; - int s_bmaps_without_search; - int s_direct2indirect; - int s_indirect2direct; +struct reiserfs_sb_info { + struct buffer_head *s_sbh; /* Buffer containing the super block */ + /* both the comment and the choice of + name are unclear for s_rs -Hans */ + struct reiserfs_super_block *s_rs; /* Pointer to the super block in the buffer */ + struct reiserfs_bitmap_info *s_ap_bitmap; + struct reiserfs_journal *s_journal; /* pointer to journal information */ + unsigned short s_mount_state; /* reiserfs state (valid, invalid) */ + + /* Comment? -Hans */ + void (*end_io_handler) (struct buffer_head *, int); + hashf_t s_hash_function; /* pointer to function which is used + to sort names in directory. Set on + mount */ + unsigned long s_mount_opt; /* reiserfs's mount options are set + here (currently - NOTAIL, NOLOG, + REPLAYONLY) */ + + struct { /* This is a structure that describes block allocator options */ + unsigned long bits; /* Bitfield for enable/disable kind of options */ + unsigned long large_file_size; /* size started from which we consider file to be a large one(in blocks) */ + int border; /* percentage of disk, border takes */ + int preallocmin; /* Minimal file size (in blocks) starting from which we do preallocations */ + int preallocsize; /* Number of blocks we try to prealloc when file + reaches preallocmin size (in blocks) or + prealloc_list is empty. */ + } s_alloc_options; + + /* Comment? -Hans */ + wait_queue_head_t s_wait; + /* To be obsoleted soon by per buffer seals.. -Hans */ + atomic_t s_generation_counter; // increased by one every time the + // tree gets re-balanced + unsigned long s_properties; /* File system properties. Currently holds + on-disk FS format */ + + /* session statistics */ + int s_kmallocs; + int s_disk_reads; + int s_disk_writes; + int s_fix_nodes; + int s_do_balance; + int s_unneeded_left_neighbor; + int s_good_search_by_key_reada; + int s_bmaps; + int s_bmaps_without_search; + int s_direct2indirect; + int s_indirect2direct; /* set up when it's ok for reiserfs_read_inode2() to read from disk inode with nlink==0. Currently this is only used during finish_unfinished() processing at mount time */ - int s_is_unlinked_ok; - reiserfs_proc_info_data_t s_proc_info_data; - struct proc_dir_entry *procdir; - int reserved_blocks; /* amount of blocks reserved for further allocations */ - spinlock_t bitmap_lock; /* this lock on now only used to protect reserved_blocks variable */ - struct dentry *priv_root; /* root of /.reiserfs_priv */ - struct dentry *xattr_root; /* root of /.reiserfs_priv/.xa */ - struct rw_semaphore xattr_dir_sem; - - int j_errno; + int s_is_unlinked_ok; + reiserfs_proc_info_data_t s_proc_info_data; + struct proc_dir_entry *procdir; + int reserved_blocks; /* amount of blocks reserved for further allocations */ + spinlock_t bitmap_lock; /* this lock on now only used to protect reserved_blocks variable */ + struct dentry *priv_root; /* root of /.reiserfs_priv */ + struct dentry *xattr_root; /* root of /.reiserfs_priv/.xa */ + struct rw_semaphore xattr_dir_sem; + + int j_errno; #ifdef CONFIG_QUOTA - char *s_qf_names[MAXQUOTAS]; - int s_jquota_fmt; + char *s_qf_names[MAXQUOTAS]; + int s_jquota_fmt; #endif }; @@ -422,14 +418,14 @@ struct reiserfs_sb_info enum reiserfs_mount_options { /* Mount options */ - REISERFS_LARGETAIL, /* large tails will be created in a session */ - REISERFS_SMALLTAIL, /* small (for files less than block size) tails will be created in a session */ - REPLAYONLY, /* replay journal and return 0. Use by fsck */ - REISERFS_CONVERT, /* -o conv: causes conversion of old - format super block to the new - format. If not specified - old - partition will be dealt with in a - manner of 3.5.x */ + REISERFS_LARGETAIL, /* large tails will be created in a session */ + REISERFS_SMALLTAIL, /* small (for files less than block size) tails will be created in a session */ + REPLAYONLY, /* replay journal and return 0. Use by fsck */ + REISERFS_CONVERT, /* -o conv: causes conversion of old + format super block to the new + format. If not specified - old + partition will be dealt with in a + manner of 3.5.x */ /* -o hash={tea, rupasov, r5, detect} is meant for properly mounting ** reiserfs disks from 3.5.19 or earlier. 99% of the time, this option @@ -439,41 +435,41 @@ enum reiserfs_mount_options { ** the existing hash on the FS, so if you have a tea hash disk, and mount ** with -o hash=rupasov, the mount will fail. */ - FORCE_TEA_HASH, /* try to force tea hash on mount */ - FORCE_RUPASOV_HASH, /* try to force rupasov hash on mount */ - FORCE_R5_HASH, /* try to force rupasov hash on mount */ - FORCE_HASH_DETECT, /* try to detect hash function on mount */ + FORCE_TEA_HASH, /* try to force tea hash on mount */ + FORCE_RUPASOV_HASH, /* try to force rupasov hash on mount */ + FORCE_R5_HASH, /* try to force rupasov hash on mount */ + FORCE_HASH_DETECT, /* try to detect hash function on mount */ - REISERFS_DATA_LOG, - REISERFS_DATA_ORDERED, - REISERFS_DATA_WRITEBACK, + REISERFS_DATA_LOG, + REISERFS_DATA_ORDERED, + REISERFS_DATA_WRITEBACK, /* used for testing experimental features, makes benchmarking new features with and without more convenient, should never be used by users in any code shipped to users (ideally) */ - REISERFS_NO_BORDER, - REISERFS_NO_UNHASHED_RELOCATION, - REISERFS_HASHED_RELOCATION, - REISERFS_ATTRS, - REISERFS_XATTRS, - REISERFS_XATTRS_USER, - REISERFS_POSIXACL, - REISERFS_BARRIER_NONE, - REISERFS_BARRIER_FLUSH, - - /* Actions on error */ - REISERFS_ERROR_PANIC, - REISERFS_ERROR_RO, - REISERFS_ERROR_CONTINUE, - - REISERFS_QUOTA, /* Some quota option specified */ - - REISERFS_TEST1, - REISERFS_TEST2, - REISERFS_TEST3, - REISERFS_TEST4, - REISERFS_UNSUPPORTED_OPT, + REISERFS_NO_BORDER, + REISERFS_NO_UNHASHED_RELOCATION, + REISERFS_HASHED_RELOCATION, + REISERFS_ATTRS, + REISERFS_XATTRS, + REISERFS_XATTRS_USER, + REISERFS_POSIXACL, + REISERFS_BARRIER_NONE, + REISERFS_BARRIER_FLUSH, + + /* Actions on error */ + REISERFS_ERROR_PANIC, + REISERFS_ERROR_RO, + REISERFS_ERROR_CONTINUE, + + REISERFS_QUOTA, /* Some quota option specified */ + + REISERFS_TEST1, + REISERFS_TEST2, + REISERFS_TEST3, + REISERFS_TEST4, + REISERFS_UNSUPPORTED_OPT, }; #define reiserfs_r5_hash(s) (REISERFS_SB(s)->s_mount_opt & (1 << FORCE_R5_HASH)) @@ -504,18 +500,17 @@ enum reiserfs_mount_options { #define reiserfs_error_panic(s) (REISERFS_SB(s)->s_mount_opt & (1 << REISERFS_ERROR_PANIC)) #define reiserfs_error_ro(s) (REISERFS_SB(s)->s_mount_opt & (1 << REISERFS_ERROR_RO)) -void reiserfs_file_buffer (struct buffer_head * bh, int list); +void reiserfs_file_buffer(struct buffer_head *bh, int list); extern struct file_system_type reiserfs_fs_type; -int reiserfs_resize(struct super_block *, unsigned long) ; +int reiserfs_resize(struct super_block *, unsigned long); #define CARRY_ON 0 #define SCHEDULE_OCCURRED 1 - #define SB_BUFFER_WITH_SB(s) (REISERFS_SB(s)->s_sbh) #define SB_JOURNAL(s) (REISERFS_SB(s)->s_journal) #define SB_JOURNAL_1st_RESERVED_BLOCK(s) (SB_JOURNAL(s)->j_1st_reserved_block) -#define SB_JOURNAL_LEN_FREE(s) (SB_JOURNAL(s)->j_journal_len_free) +#define SB_JOURNAL_LEN_FREE(s) (SB_JOURNAL(s)->j_journal_len_free) #define SB_AP_BITMAP(s) (REISERFS_SB(s)->s_ap_bitmap) #define SB_DISK_JOURNAL_HEAD(s) (SB_JOURNAL(s)->j_header_bh->) @@ -525,13 +520,14 @@ int reiserfs_resize(struct super_block *, unsigned long) ; */ static inline char *reiserfs_bdevname(struct super_block *s) { - return (s == NULL) ? "Null superblock" : s -> s_id; + return (s == NULL) ? "Null superblock" : s->s_id; } #define reiserfs_is_journal_aborted(journal) (unlikely (__reiserfs_is_journal_aborted (journal))) -static inline int __reiserfs_is_journal_aborted (struct reiserfs_journal *journal) +static inline int __reiserfs_is_journal_aborted(struct reiserfs_journal + *journal) { - return test_bit (J_ABORTED, &journal->j_state); + return test_bit(J_ABORTED, &journal->j_state); } -#endif /* _LINUX_REISER_FS_SB */ +#endif /* _LINUX_REISER_FS_SB */ diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h index 9244c5748820..c84354e8374c 100644 --- a/include/linux/reiserfs_xattr.h +++ b/include/linux/reiserfs_xattr.h @@ -7,48 +7,48 @@ #include /* Magic value in header */ -#define REISERFS_XATTR_MAGIC 0x52465841 /* "RFXA" */ +#define REISERFS_XATTR_MAGIC 0x52465841 /* "RFXA" */ struct reiserfs_xattr_header { - __le32 h_magic; /* magic number for identification */ - __le32 h_hash; /* hash of the value */ + __le32 h_magic; /* magic number for identification */ + __le32 h_hash; /* hash of the value */ }; #ifdef __KERNEL__ struct reiserfs_xattr_handler { char *prefix; - int (*init)(void); - void (*exit)(void); - int (*get)(struct inode *inode, const char *name, void *buffer, - size_t size); - int (*set)(struct inode *inode, const char *name, const void *buffer, - size_t size, int flags); - int (*del)(struct inode *inode, const char *name); - int (*list)(struct inode *inode, const char *name, int namelen, char *out); - struct list_head handlers; + int (*init) (void); + void (*exit) (void); + int (*get) (struct inode * inode, const char *name, void *buffer, + size_t size); + int (*set) (struct inode * inode, const char *name, const void *buffer, + size_t size, int flags); + int (*del) (struct inode * inode, const char *name); + int (*list) (struct inode * inode, const char *name, int namelen, + char *out); + struct list_head handlers; }; - #ifdef CONFIG_REISERFS_FS_XATTR #define is_reiserfs_priv_object(inode) IS_PRIVATE(inode) #define has_xattr_dir(inode) (REISERFS_I(inode)->i_flags & i_has_xattr_dir) -ssize_t reiserfs_getxattr (struct dentry *dentry, const char *name, - void *buffer, size_t size); -int reiserfs_setxattr (struct dentry *dentry, const char *name, - const void *value, size_t size, int flags); -ssize_t reiserfs_listxattr (struct dentry *dentry, char *buffer, size_t size); -int reiserfs_removexattr (struct dentry *dentry, const char *name); -int reiserfs_delete_xattrs (struct inode *inode); -int reiserfs_chown_xattrs (struct inode *inode, struct iattr *attrs); -int reiserfs_xattr_init (struct super_block *sb, int mount_flags); -int reiserfs_permission (struct inode *inode, int mask, struct nameidata *nd); -int reiserfs_permission_locked (struct inode *inode, int mask, struct nameidata *nd); - -int reiserfs_xattr_del (struct inode *, const char *); -int reiserfs_xattr_get (const struct inode *, const char *, void *, size_t); -int reiserfs_xattr_set (struct inode *, const char *, const void *, - size_t, int); +ssize_t reiserfs_getxattr(struct dentry *dentry, const char *name, + void *buffer, size_t size); +int reiserfs_setxattr(struct dentry *dentry, const char *name, + const void *value, size_t size, int flags); +ssize_t reiserfs_listxattr(struct dentry *dentry, char *buffer, size_t size); +int reiserfs_removexattr(struct dentry *dentry, const char *name); +int reiserfs_delete_xattrs(struct inode *inode); +int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs); +int reiserfs_xattr_init(struct super_block *sb, int mount_flags); +int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd); +int reiserfs_permission_locked(struct inode *inode, int mask, + struct nameidata *nd); + +int reiserfs_xattr_del(struct inode *, const char *); +int reiserfs_xattr_get(const struct inode *, const char *, void *, size_t); +int reiserfs_xattr_set(struct inode *, const char *, const void *, size_t, int); extern struct reiserfs_xattr_handler user_handler; extern struct reiserfs_xattr_handler trusted_handler; @@ -56,57 +56,48 @@ extern struct reiserfs_xattr_handler trusted_handler; extern struct reiserfs_xattr_handler security_handler; #endif -int reiserfs_xattr_register_handlers (void) __init; -void reiserfs_xattr_unregister_handlers (void); +int reiserfs_xattr_register_handlers(void) __init; +void reiserfs_xattr_unregister_handlers(void); -static inline void -reiserfs_write_lock_xattrs(struct super_block *sb) +static inline void reiserfs_write_lock_xattrs(struct super_block *sb) { - down_write (&REISERFS_XATTR_DIR_SEM(sb)); + down_write(&REISERFS_XATTR_DIR_SEM(sb)); } -static inline void -reiserfs_write_unlock_xattrs(struct super_block *sb) +static inline void reiserfs_write_unlock_xattrs(struct super_block *sb) { - up_write (&REISERFS_XATTR_DIR_SEM(sb)); + up_write(&REISERFS_XATTR_DIR_SEM(sb)); } -static inline void -reiserfs_read_lock_xattrs(struct super_block *sb) +static inline void reiserfs_read_lock_xattrs(struct super_block *sb) { - down_read (&REISERFS_XATTR_DIR_SEM(sb)); + down_read(&REISERFS_XATTR_DIR_SEM(sb)); } -static inline void -reiserfs_read_unlock_xattrs(struct super_block *sb) +static inline void reiserfs_read_unlock_xattrs(struct super_block *sb) { - up_read (&REISERFS_XATTR_DIR_SEM(sb)); + up_read(&REISERFS_XATTR_DIR_SEM(sb)); } -static inline void -reiserfs_write_lock_xattr_i(struct inode *inode) +static inline void reiserfs_write_lock_xattr_i(struct inode *inode) { - down_write (&REISERFS_I(inode)->xattr_sem); + down_write(&REISERFS_I(inode)->xattr_sem); } -static inline void -reiserfs_write_unlock_xattr_i(struct inode *inode) +static inline void reiserfs_write_unlock_xattr_i(struct inode *inode) { - up_write (&REISERFS_I(inode)->xattr_sem); + up_write(&REISERFS_I(inode)->xattr_sem); } -static inline void -reiserfs_read_lock_xattr_i(struct inode *inode) +static inline void reiserfs_read_lock_xattr_i(struct inode *inode) { - down_read (&REISERFS_I(inode)->xattr_sem); + down_read(&REISERFS_I(inode)->xattr_sem); } -static inline void -reiserfs_read_unlock_xattr_i(struct inode *inode) +static inline void reiserfs_read_unlock_xattr_i(struct inode *inode) { - up_read (&REISERFS_I(inode)->xattr_sem); + up_read(&REISERFS_I(inode)->xattr_sem); } -static inline void -reiserfs_mark_inode_private(struct inode *inode) +static inline void reiserfs_mark_inode_private(struct inode *inode) { - inode->i_flags |= S_PRIVATE; + inode->i_flags |= S_PRIVATE; } #else @@ -127,13 +118,20 @@ reiserfs_mark_inode_private(struct inode *inode) #define reiserfs_xattr_register_handlers() 0 #define reiserfs_xattr_unregister_handlers() -static inline int reiserfs_delete_xattrs (struct inode *inode) { return 0; }; -static inline int reiserfs_chown_xattrs (struct inode *inode, struct iattr *attrs) { return 0; }; -static inline int reiserfs_xattr_init (struct super_block *sb, int mount_flags) +static inline int reiserfs_delete_xattrs(struct inode *inode) +{ + return 0; +}; +static inline int reiserfs_chown_xattrs(struct inode *inode, + struct iattr *attrs) +{ + return 0; +}; +static inline int reiserfs_xattr_init(struct super_block *sb, int mount_flags) { - sb->s_flags = (sb->s_flags & ~MS_POSIXACL); /* to be sure */ - return 0; + sb->s_flags = (sb->s_flags & ~MS_POSIXACL); /* to be sure */ + return 0; }; #endif -#endif /* __KERNEL__ */ +#endif /* __KERNEL__ */ -- cgit v1.2.1 From 0eeca28300df110bd6ed54b31193c83b87921443 Mon Sep 17 00:00:00 2001 From: Robert Love Date: Tue, 12 Jul 2005 17:06:03 -0400 Subject: [PATCH] inotify inotify is intended to correct the deficiencies of dnotify, particularly its inability to scale and its terrible user interface: * dnotify requires the opening of one fd per each directory that you intend to watch. This quickly results in too many open files and pins removable media, preventing unmount. * dnotify is directory-based. You only learn about changes to directories. Sure, a change to a file in a directory affects the directory, but you are then forced to keep a cache of stat structures. * dnotify's interface to user-space is awful. Signals? inotify provides a more usable, simple, powerful solution to file change notification: * inotify's interface is a system call that returns a fd, not SIGIO. You get a single fd, which is select()-able. * inotify has an event that says "the filesystem that the item you were watching is on was unmounted." * inotify can watch directories or files. Inotify is currently used by Beagle (a desktop search infrastructure), Gamin (a FAM replacement), and other projects. See Documentation/filesystems/inotify.txt. Signed-off-by: Robert Love Cc: John McCutchan Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 6 +- include/linux/fsnotify.h | 248 +++++++++++++++++++++++++++++++++++++++++++++++ include/linux/inotify.h | 108 +++++++++++++++++++++ include/linux/sched.h | 4 + include/linux/sysctl.h | 11 ++- 5 files changed, 375 insertions(+), 2 deletions(-) create mode 100644 include/linux/fsnotify.h create mode 100644 include/linux/inotify.h (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 302ec20838ca..c9bf3746a9fb 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -474,6 +474,11 @@ struct inode { struct dnotify_struct *i_dnotify; /* for directory notifications */ #endif +#ifdef CONFIG_INOTIFY + struct list_head inotify_watches; /* watches on this inode */ + struct semaphore inotify_sem; /* protects the watches list */ +#endif + unsigned long i_state; unsigned long dirtied_when; /* jiffies of first dirtying */ @@ -1393,7 +1398,6 @@ extern void emergency_remount(void); extern int do_remount_sb(struct super_block *sb, int flags, void *data, int force); extern sector_t bmap(struct inode *, sector_t); -extern int setattr_mask(unsigned int); extern int notify_change(struct dentry *, struct iattr *); extern int permission(struct inode *, int, struct nameidata *); extern int generic_permission(struct inode *, int, diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h new file mode 100644 index 000000000000..eb581b6cfca9 --- /dev/null +++ b/include/linux/fsnotify.h @@ -0,0 +1,248 @@ +#ifndef _LINUX_FS_NOTIFY_H +#define _LINUX_FS_NOTIFY_H + +/* + * include/linux/fsnotify.h - generic hooks for filesystem notification, to + * reduce in-source duplication from both dnotify and inotify. + * + * We don't compile any of this away in some complicated menagerie of ifdefs. + * Instead, we rely on the code inside to optimize away as needed. + * + * (C) Copyright 2005 Robert Love + */ + +#ifdef __KERNEL__ + +#include +#include + +/* + * fsnotify_move - file old_name at old_dir was moved to new_name at new_dir + */ +static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, + const char *old_name, const char *new_name, + int isdir) +{ + u32 cookie = inotify_get_cookie(); + + if (old_dir == new_dir) + inode_dir_notify(old_dir, DN_RENAME); + else { + inode_dir_notify(old_dir, DN_DELETE); + inode_dir_notify(new_dir, DN_CREATE); + } + + if (isdir) + isdir = IN_ISDIR; + inotify_inode_queue_event(old_dir, IN_MOVED_FROM|isdir,cookie,old_name); + inotify_inode_queue_event(new_dir, IN_MOVED_TO|isdir, cookie, new_name); +} + +/* + * fsnotify_unlink - file was unlinked + */ +static inline void fsnotify_unlink(struct dentry *dentry, struct inode *dir) +{ + struct inode *inode = dentry->d_inode; + + inode_dir_notify(dir, DN_DELETE); + inotify_inode_queue_event(dir, IN_DELETE, 0, dentry->d_name.name); + inotify_inode_queue_event(inode, IN_DELETE_SELF, 0, NULL); + + inotify_inode_is_dead(inode); +} + +/* + * fsnotify_rmdir - directory was removed + */ +static inline void fsnotify_rmdir(struct dentry *dentry, struct inode *inode, + struct inode *dir) +{ + inode_dir_notify(dir, DN_DELETE); + inotify_inode_queue_event(dir,IN_DELETE|IN_ISDIR,0,dentry->d_name.name); + inotify_inode_queue_event(inode, IN_DELETE_SELF | IN_ISDIR, 0, NULL); + inotify_inode_is_dead(inode); +} + +/* + * fsnotify_create - 'name' was linked in + */ +static inline void fsnotify_create(struct inode *inode, const char *name) +{ + inode_dir_notify(inode, DN_CREATE); + inotify_inode_queue_event(inode, IN_CREATE, 0, name); +} + +/* + * fsnotify_mkdir - directory 'name' was created + */ +static inline void fsnotify_mkdir(struct inode *inode, const char *name) +{ + inode_dir_notify(inode, DN_CREATE); + inotify_inode_queue_event(inode, IN_CREATE | IN_ISDIR, 0, name); +} + +/* + * fsnotify_access - file was read + */ +static inline void fsnotify_access(struct dentry *dentry) +{ + struct inode *inode = dentry->d_inode; + u32 mask = IN_ACCESS; + + if (S_ISDIR(inode->i_mode)) + mask |= IN_ISDIR; + + dnotify_parent(dentry, DN_ACCESS); + inotify_dentry_parent_queue_event(dentry, mask, 0, dentry->d_name.name); + inotify_inode_queue_event(inode, mask, 0, NULL); +} + +/* + * fsnotify_modify - file was modified + */ +static inline void fsnotify_modify(struct dentry *dentry) +{ + struct inode *inode = dentry->d_inode; + u32 mask = IN_MODIFY; + + if (S_ISDIR(inode->i_mode)) + mask |= IN_ISDIR; + + dnotify_parent(dentry, DN_MODIFY); + inotify_dentry_parent_queue_event(dentry, mask, 0, dentry->d_name.name); + inotify_inode_queue_event(inode, mask, 0, NULL); +} + +/* + * fsnotify_open - file was opened + */ +static inline void fsnotify_open(struct dentry *dentry) +{ + struct inode *inode = dentry->d_inode; + u32 mask = IN_OPEN; + + if (S_ISDIR(inode->i_mode)) + mask |= IN_ISDIR; + + inotify_inode_queue_event(inode, mask, 0, NULL); + inotify_dentry_parent_queue_event(dentry, mask, 0, dentry->d_name.name); +} + +/* + * fsnotify_close - file was closed + */ +static inline void fsnotify_close(struct file *file) +{ + struct dentry *dentry = file->f_dentry; + struct inode *inode = dentry->d_inode; + const char *name = dentry->d_name.name; + mode_t mode = file->f_mode; + u32 mask = (mode & FMODE_WRITE) ? IN_CLOSE_WRITE : IN_CLOSE_NOWRITE; + + if (S_ISDIR(inode->i_mode)) + mask |= IN_ISDIR; + + inotify_dentry_parent_queue_event(dentry, mask, 0, name); + inotify_inode_queue_event(inode, mask, 0, NULL); +} + +/* + * fsnotify_xattr - extended attributes were changed + */ +static inline void fsnotify_xattr(struct dentry *dentry) +{ + struct inode *inode = dentry->d_inode; + u32 mask = IN_ATTRIB; + + if (S_ISDIR(inode->i_mode)) + mask |= IN_ISDIR; + + inotify_dentry_parent_queue_event(dentry, mask, 0, dentry->d_name.name); + inotify_inode_queue_event(inode, mask, 0, NULL); +} + +/* + * fsnotify_change - notify_change event. file was modified and/or metadata + * was changed. + */ +static inline void fsnotify_change(struct dentry *dentry, unsigned int ia_valid) +{ + struct inode *inode = dentry->d_inode; + int dn_mask = 0; + u32 in_mask = 0; + + if (ia_valid & ATTR_UID) { + in_mask |= IN_ATTRIB; + dn_mask |= DN_ATTRIB; + } + if (ia_valid & ATTR_GID) { + in_mask |= IN_ATTRIB; + dn_mask |= DN_ATTRIB; + } + if (ia_valid & ATTR_SIZE) { + in_mask |= IN_MODIFY; + dn_mask |= DN_MODIFY; + } + /* both times implies a utime(s) call */ + if ((ia_valid & (ATTR_ATIME | ATTR_MTIME)) == (ATTR_ATIME | ATTR_MTIME)) + { + in_mask |= IN_ATTRIB; + dn_mask |= DN_ATTRIB; + } else if (ia_valid & ATTR_ATIME) { + in_mask |= IN_ACCESS; + dn_mask |= DN_ACCESS; + } else if (ia_valid & ATTR_MTIME) { + in_mask |= IN_MODIFY; + dn_mask |= DN_MODIFY; + } + if (ia_valid & ATTR_MODE) { + in_mask |= IN_ATTRIB; + dn_mask |= DN_ATTRIB; + } + + if (dn_mask) + dnotify_parent(dentry, dn_mask); + if (in_mask) { + if (S_ISDIR(inode->i_mode)) + in_mask |= IN_ISDIR; + inotify_inode_queue_event(inode, in_mask, 0, NULL); + inotify_dentry_parent_queue_event(dentry, in_mask, 0, + dentry->d_name.name); + } +} + +#ifdef CONFIG_INOTIFY /* inotify helpers */ + +/* + * fsnotify_oldname_init - save off the old filename before we change it + */ +static inline const char *fsnotify_oldname_init(const char *name) +{ + return kstrdup(name, GFP_KERNEL); +} + +/* + * fsnotify_oldname_free - free the name we got from fsnotify_oldname_init + */ +static inline void fsnotify_oldname_free(const char *old_name) +{ + kfree(old_name); +} + +#else /* CONFIG_INOTIFY */ + +static inline const char *fsnotify_oldname_init(const char *name) +{ + return NULL; +} + +static inline void fsnotify_oldname_free(const char *old_name) +{ +} + +#endif /* ! CONFIG_INOTIFY */ + +#endif /* __KERNEL__ */ + +#endif /* _LINUX_FS_NOTIFY_H */ diff --git a/include/linux/inotify.h b/include/linux/inotify.h new file mode 100644 index 000000000000..a40c2bf0408e --- /dev/null +++ b/include/linux/inotify.h @@ -0,0 +1,108 @@ +/* + * Inode based directory notification for Linux + * + * Copyright (C) 2005 John McCutchan + */ + +#ifndef _LINUX_INOTIFY_H +#define _LINUX_INOTIFY_H + +#include + +/* + * struct inotify_event - structure read from the inotify device for each event + * + * When you are watching a directory, you will receive the filename for events + * such as IN_CREATE, IN_DELETE, IN_OPEN, IN_CLOSE, ..., relative to the wd. + */ +struct inotify_event { + __s32 wd; /* watch descriptor */ + __u32 mask; /* watch mask */ + __u32 cookie; /* cookie to synchronize two events */ + __u32 len; /* length (including nulls) of name */ + char name[0]; /* stub for possible name */ +}; + +/* the following are legal, implemented events that user-space can watch for */ +#define IN_ACCESS 0x00000001 /* File was accessed */ +#define IN_MODIFY 0x00000002 /* File was modified */ +#define IN_ATTRIB 0x00000004 /* Metadata changed */ +#define IN_CLOSE_WRITE 0x00000008 /* Writtable file was closed */ +#define IN_CLOSE_NOWRITE 0x00000010 /* Unwrittable file closed */ +#define IN_OPEN 0x00000020 /* File was opened */ +#define IN_MOVED_FROM 0x00000040 /* File was moved from X */ +#define IN_MOVED_TO 0x00000080 /* File was moved to Y */ +#define IN_CREATE 0x00000100 /* Subfile was created */ +#define IN_DELETE 0x00000200 /* Subfile was deleted */ +#define IN_DELETE_SELF 0x00000400 /* Self was deleted */ + +/* the following are legal events. they are sent as needed to any watch */ +#define IN_UNMOUNT 0x00002000 /* Backing fs was unmounted */ +#define IN_Q_OVERFLOW 0x00004000 /* Event queued overflowed */ +#define IN_IGNORED 0x00008000 /* File was ignored */ + +/* helper events */ +#define IN_CLOSE (IN_CLOSE_WRITE | IN_CLOSE_NOWRITE) /* close */ +#define IN_MOVE (IN_MOVED_FROM | IN_MOVED_TO) /* moves */ + +/* special flags */ +#define IN_ISDIR 0x40000000 /* event occurred against dir */ +#define IN_ONESHOT 0x80000000 /* only send event once */ + +/* + * All of the events - we build the list by hand so that we can add flags in + * the future and not break backward compatibility. Apps will get only the + * events that they originally wanted. Be sure to add new events here! + */ +#define IN_ALL_EVENTS (IN_ACCESS | IN_MODIFY | IN_ATTRIB | IN_CLOSE_WRITE | \ + IN_CLOSE_NOWRITE | IN_OPEN | IN_MOVED_FROM | \ + IN_MOVED_TO | IN_DELETE | IN_CREATE | IN_DELETE_SELF) + +#ifdef __KERNEL__ + +#include +#include +#include + +#ifdef CONFIG_INOTIFY + +extern void inotify_inode_queue_event(struct inode *, __u32, __u32, + const char *); +extern void inotify_dentry_parent_queue_event(struct dentry *, __u32, __u32, + const char *); +extern void inotify_unmount_inodes(struct list_head *); +extern void inotify_inode_is_dead(struct inode *); +extern u32 inotify_get_cookie(void); + +#else + +static inline void inotify_inode_queue_event(struct inode *inode, + __u32 mask, __u32 cookie, + const char *filename) +{ +} + +static inline void inotify_dentry_parent_queue_event(struct dentry *dentry, + __u32 mask, __u32 cookie, + const char *filename) +{ +} + +static inline void inotify_unmount_inodes(struct list_head *list) +{ +} + +static inline void inotify_inode_is_dead(struct inode *inode) +{ +} + +static inline u32 inotify_get_cookie(void) +{ + return 0; +} + +#endif /* CONFIG_INOTIFY */ + +#endif /* __KERNEL __ */ + +#endif /* _LINUX_INOTIFY_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index ff48815bd3a2..dec5827c7742 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -410,6 +410,10 @@ struct user_struct { atomic_t processes; /* How many processes does this user have? */ atomic_t files; /* How many open files does this user have? */ atomic_t sigpending; /* How many pending signals does this user have? */ +#ifdef CONFIG_INOTIFY + atomic_t inotify_watches; /* How many inotify watches does this user have? */ + atomic_t inotify_devs; /* How many inotify devs does this user have opened? */ +#endif /* protected by mq_lock */ unsigned long mq_bytes; /* How many bytes can be allocated to mqueue? */ unsigned long locked_shm; /* How many pages of mlocked shm ? */ diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 5b5f434ac9a0..ce19a2aa0b21 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -61,7 +61,8 @@ enum CTL_DEV=7, /* Devices */ CTL_BUS=8, /* Busses */ CTL_ABI=9, /* Binary emulation */ - CTL_CPU=10 /* CPU stuff (speed scaling, etc) */ + CTL_CPU=10, /* CPU stuff (speed scaling, etc) */ + CTL_INOTIFY=11 /* Inotify */ }; /* CTL_BUS names: */ @@ -70,6 +71,14 @@ enum CTL_BUS_ISA=1 /* ISA */ }; +/* CTL_INOTIFY names: */ +enum +{ + INOTIFY_MAX_USER_DEVICES=1, /* max number of inotify device instances per user */ + INOTIFY_MAX_USER_WATCHES=2, /* max number of inotify watches per user */ + INOTIFY_MAX_QUEUED_EVENTS=3 /* Max number of queued events per inotify device instance */ +}; + /* CTL_KERN names: */ enum { -- cgit v1.2.1 From 0399cb08c54708db231d616f106f64d920e0b723 Mon Sep 17 00:00:00 2001 From: Robert Love Date: Wed, 13 Jul 2005 12:38:18 -0400 Subject: [PATCH] inotify: move sysctl This moves the inotify sysctl knobs to "/proc/sys/fs/inotify" from "/proc/sys/fs". Also some related cleanup. Signed-off-by: Robert Love Signed-off-by: Linus Torvalds --- include/linux/sysctl.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index ce19a2aa0b21..bfbbe94b297d 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -61,8 +61,7 @@ enum CTL_DEV=7, /* Devices */ CTL_BUS=8, /* Busses */ CTL_ABI=9, /* Binary emulation */ - CTL_CPU=10, /* CPU stuff (speed scaling, etc) */ - CTL_INOTIFY=11 /* Inotify */ + CTL_CPU=10 /* CPU stuff (speed scaling, etc) */ }; /* CTL_BUS names: */ @@ -71,12 +70,12 @@ enum CTL_BUS_ISA=1 /* ISA */ }; -/* CTL_INOTIFY names: */ +/* /proc/sys/fs/inotify/ */ enum { - INOTIFY_MAX_USER_DEVICES=1, /* max number of inotify device instances per user */ - INOTIFY_MAX_USER_WATCHES=2, /* max number of inotify watches per user */ - INOTIFY_MAX_QUEUED_EVENTS=3 /* Max number of queued events per inotify device instance */ + INOTIFY_MAX_USER_INSTANCES=1, /* max instances per user */ + INOTIFY_MAX_USER_WATCHES=2, /* max watches per user */ + INOTIFY_MAX_QUEUED_EVENTS=3 /* max queued events per instance */ }; /* CTL_KERN names: */ @@ -685,6 +684,7 @@ enum FS_XFS=17, /* struct: control xfs parameters */ FS_AIO_NR=18, /* current system-wide number of aio requests */ FS_AIO_MAX_NR=19, /* system-wide maximum number of aio requests */ + FS_INOTIFY=20, /* inotify submenu */ }; /* /proc/sys/fs/quota/ */ -- cgit v1.2.1 From 5995f16b4a464c8a57de7c9d5ddf4758dbacad41 Mon Sep 17 00:00:00 2001 From: Robert Love Date: Wed, 13 Jul 2005 12:38:39 -0400 Subject: [PATCH] inotify: event ordering This rearranges the event ordering for "open" to be consistent with the ordering of the other events. Signed-off-by: Robert Love Signed-off-by: Linus Torvalds --- include/linux/fsnotify.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index eb581b6cfca9..d07a92c94776 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -125,8 +125,8 @@ static inline void fsnotify_open(struct dentry *dentry) if (S_ISDIR(inode->i_mode)) mask |= IN_ISDIR; - inotify_inode_queue_event(inode, mask, 0, NULL); inotify_dentry_parent_queue_event(dentry, mask, 0, dentry->d_name.name); + inotify_inode_queue_event(inode, mask, 0, NULL); } /* -- cgit v1.2.1 From 88bd5121d635136e01369141367f315665534b3c Mon Sep 17 00:00:00 2001 From: Anton Altaparmakov Date: Wed, 13 Jul 2005 01:10:44 -0700 Subject: [PATCH] Fix soft lockup due to NTFS: VFS part and explanation Something has changed in the core kernel such that we now get concurrent inode write outs, one e.g via pdflush and one via sys_sync or whatever. This causes a nasty deadlock in ntfs. The only clean solution unfortunately requires a minor vfs api extension. First the deadlock analysis: Prerequisive knowledge: NTFS has a file $MFT (inode 0) loaded at mount time. The NTFS driver uses the page cache for storing the file contents as usual. More interestingly this file contains the table of on-disk inodes as a sequence of MFT_RECORDs. Thus NTFS driver accesses the on-disk inodes by accessing the MFT_RECORDs in the page cache pages of the loaded inode $MFT. The situation: VFS inode X on a mounted ntfs volume is dirty. For same inode X, the ntfs_inode is dirty and thus corresponding on-disk inode, which is as explained above in a dirty PAGE_CACHE_PAGE belonging to the table of inodes ($MFT, inode 0). What happens: Process 1: sys_sync()/umount()/whatever... calls __sync_single_inode() for $MFT -> do_writepages() -> write_page for the dirty page containing the on-disk inode X, the page is now locked -> ntfs_write_mst_block() which clears PageUptodate() on the page to prevent anyone else getting hold of it whilst it does the write out (this is necessary as the on-disk inode needs "fixups" applied before the write to disk which are removed again after the write and PageUptodate is then set again). It then analyses the page looking for dirty on-disk inodes and when it finds one it calls ntfs_may_write_mft_record() to see if it is safe to write this on-disk inode. This then calls ilookup5() to check if the corresponding VFS inode is in icache(). This in turn calls ifind() which waits on the inode lock via wait_on_inode whilst holding the global inode_lock. Process 2: pdflush results in a call to __sync_single_inode for the same VFS inode X on the ntfs volume. This locks the inode (I_LOCK) then calls write-inode -> ntfs_write_inode -> map_mft_record() -> read_cache_page() of the page (in page cache of table of inodes $MFT, inode 0) containing the on-disk inode. This page has PageUptodate() clear because of Process 1 (see above) so read_cache_page() blocks when tries to take the page lock for the page so it can call ntfs_read_page(). Thus Process 1 is holding the page lock on the page containing the on-disk inode X and it is waiting on the inode X to be unlocked in ifind() so it can write the page out and then unlock the page. And Process 2 is holding the inode lock on inode X and is waiting for the page to be unlocked so it can call ntfs_readpage() or discover that Process 1 set PageUptodate() again and use the page. Thus we have a deadlock due to ifind() waiting on the inode lock. The only sensible solution: NTFS does not care whether the VFS inode is locked or not when it calls ilookup5() (it doesn't use the VFS inode at all, it just uses it to find the corresponding ntfs_inode which is of course attached to the VFS inode (both are one single struct); and it uses the ntfs_inode which is subject to its own locking so I_LOCK is irrelevant) hence we want a modified ilookup5_nowait() which is the same as ilookup5() but it does not wait on the inode lock. Without such functionality I would have to keep my own ntfs_inode cache in the NTFS driver just so I can find ntfs_inodes independent of their VFS inodes which would be slow, memory and cpu cycle wasting, and incredibly stupid given the icache already exists in the VFS. Below is a patch that does the ilookup5_nowait() implementation in fs/inode.c and exports it. ilookup5_nowait.diff: Introduce ilookup5_nowait() which is basically the same as ilookup5() but it does not wait on the inode's lock (i.e. it omits the wait_on_inode() done in ifind()). This is needed to avoid a nasty deadlock in NTFS. Signed-off-by: Anton Altaparmakov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index c9bf3746a9fb..0f53e0124941 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1441,6 +1441,9 @@ extern int inode_needs_sync(struct inode *inode); extern void generic_delete_inode(struct inode *inode); extern void generic_drop_inode(struct inode *inode); +extern struct inode *ilookup5_nowait(struct super_block *sb, + unsigned long hashval, int (*test)(struct inode *, void *), + void *data); extern struct inode *ilookup5(struct super_block *sb, unsigned long hashval, int (*test)(struct inode *, void *), void *data); extern struct inode *ilookup(struct super_block *sb, unsigned long ino); -- cgit v1.2.1 From 068e1b94bbd268f375349f68531829c8b7c210bc Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 13 Jul 2005 01:10:46 -0700 Subject: [PATCH] s390: fadvise hint values. Add special case for the POSIX_FADV_DONTNEED and POSIX_FADV_NOREUSE hint values for s390-64. The user space values in the s390-64 glibc headers for these two defines have always been 6 and 7 instead of 4 and 5. All 64 bit applications therefore use the "wrong" values. To get these applications working without recompiling the kernel needs to accept the "wrong" values. Since the values for s390-31 are 4 and 5 the compat wrapper for fadvise64 and fadvise64_64 need to rewrite the values for 31 bit system calls. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fadvise.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fadvise.h b/include/linux/fadvise.h index 6fc656dfb93d..e8e747139b9a 100644 --- a/include/linux/fadvise.h +++ b/include/linux/fadvise.h @@ -5,7 +5,17 @@ #define POSIX_FADV_RANDOM 1 /* Expect random page references. */ #define POSIX_FADV_SEQUENTIAL 2 /* Expect sequential page references. */ #define POSIX_FADV_WILLNEED 3 /* Will need these pages. */ + +/* + * The advise values for POSIX_FADV_DONTNEED and POSIX_ADV_NOREUSE + * for s390-64 differ from the values for the rest of the world. + */ +#if defined(__s390x__) +#define POSIX_FADV_DONTNEED 6 /* Don't need these pages. */ +#define POSIX_FADV_NOREUSE 7 /* Data will be accessed once. */ +#else #define POSIX_FADV_DONTNEED 4 /* Don't need these pages. */ #define POSIX_FADV_NOREUSE 5 /* Data will be accessed once. */ +#endif #endif /* FADVISE_H_INCLUDED */ -- cgit v1.2.1 From ac96202ba096f8fc1ccaf45a2f159a52639ece29 Mon Sep 17 00:00:00 2001 From: Andrew Vasquez Date: Wed, 6 Jul 2005 10:30:16 -0700 Subject: [SCSI] qla2xxx: Add pci ids for new ISP types. Add pci ids for new ISP types. Move old definitions in local qla_def.h file to pci_ids.h as well. Signed-off-by: Andrew Vasquez Signed-off-by: James Bottomley --- include/linux/pci_ids.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index bf608808a60c..f4c2c393797d 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -908,6 +908,15 @@ #define PCI_DEVICE_ID_QLOGIC_ISP1022 0x1022 #define PCI_DEVICE_ID_QLOGIC_ISP2100 0x2100 #define PCI_DEVICE_ID_QLOGIC_ISP2200 0x2200 +#define PCI_DEVICE_ID_QLOGIC_ISP2300 0x2300 +#define PCI_DEVICE_ID_QLOGIC_ISP2312 0x2312 +#define PCI_DEVICE_ID_QLOGIC_ISP2322 0x2322 +#define PCI_DEVICE_ID_QLOGIC_ISP6312 0x6312 +#define PCI_DEVICE_ID_QLOGIC_ISP6322 0x6322 +#define PCI_DEVICE_ID_QLOGIC_ISP2422 0x2422 +#define PCI_DEVICE_ID_QLOGIC_ISP2432 0x2432 +#define PCI_DEVICE_ID_QLOGIC_ISP2512 0x2512 +#define PCI_DEVICE_ID_QLOGIC_ISP2522 0x2522 #define PCI_VENDOR_ID_CYRIX 0x1078 #define PCI_DEVICE_ID_CYRIX_5510 0x0000 -- cgit v1.2.1 From 6a806c510de490318846b53bbfec463d02ca274b Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Fri, 15 Jul 2005 03:56:35 -0700 Subject: [PATCH] md/raid1: clear bitmap when fullsync completes We need to be careful differentiating between a resync of a complete array, in which we can clear the bitmap, and a resync of a degraded array, in which we cannot. This patch cleans all that up. Cc: Paul Clements Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/raid/bitmap.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h index e24b74b11150..6213e976eade 100644 --- a/include/linux/raid/bitmap.h +++ b/include/linux/raid/bitmap.h @@ -262,7 +262,7 @@ void bitmap_write_all(struct bitmap *bitmap); int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors); void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long sectors, int success); -int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks); +int bitmap_start_sync(struct bitmap *bitmap, sector_t offset, int *blocks, int degraded); void bitmap_end_sync(struct bitmap *bitmap, sector_t offset, int *blocks, int aborted); void bitmap_close_sync(struct bitmap *bitmap); -- cgit v1.2.1 From 661f83a67c2e360d5a4d2406cc28379c909f94bf Mon Sep 17 00:00:00 2001 From: Russell King Date: Sat, 16 Jul 2005 09:30:53 +0100 Subject: [PATCH] Serial: Move deprecation of register_serial forward to September I think it's about time to make the build a little more vocal about the expiry of these functions. Due to recent discussions with problems in the console initialisation vs power manglement, I'd like to move the date forward to September. Signed-off-by: Russell King --- include/linux/serial.h | 6 ++++-- include/linux/serial_core.h | 5 +++-- 2 files changed, 7 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/serial.h b/include/linux/serial.h index 00145822fb74..9f2d85284d0b 100644 --- a/include/linux/serial.h +++ b/include/linux/serial.h @@ -174,9 +174,11 @@ struct serial_icounter_struct { #ifdef __KERNEL__ +#include + /* Export to allow PCMCIA to use this - Dave Hinds */ -extern int register_serial(struct serial_struct *req); -extern void unregister_serial(int line); +extern int __deprecated register_serial(struct serial_struct *req); +extern void __deprecated unregister_serial(int line); /* Allow architectures to override entries in serial8250_ports[] at run time: */ struct uart_port; /* forward declaration */ diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index d6025af7efac..30b64f3534f4 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -122,6 +122,7 @@ #ifdef __KERNEL__ #include +#include #include #include #include @@ -359,8 +360,8 @@ struct tty_driver *uart_console_device(struct console *co, int *index); */ int uart_register_driver(struct uart_driver *uart); void uart_unregister_driver(struct uart_driver *uart); -void uart_unregister_port(struct uart_driver *reg, int line); -int uart_register_port(struct uart_driver *reg, struct uart_port *port); +void __deprecated uart_unregister_port(struct uart_driver *reg, int line); +int __deprecated uart_register_port(struct uart_driver *reg, struct uart_port *port); int uart_add_one_port(struct uart_driver *reg, struct uart_port *port); int uart_remove_one_port(struct uart_driver *reg, struct uart_port *port); int uart_match_port(struct uart_port *port1, struct uart_port *port2); -- cgit v1.2.1 From 6d283d271674b1127881ebf082266a2c3fe6e0e4 Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Sat, 16 Jul 2005 09:59:00 +0100 Subject: [PATCH] Serial: Remove linux/version.h changing CONFIG_LOCALVERSION rebuilds too much, for no appearent reason. Signed-off-by: Olaf Hering Signed-off-by: Russell King --- include/linux/serialP.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/serialP.h b/include/linux/serialP.h index 2307f11d8a6b..2b2f35a64d75 100644 --- a/include/linux/serialP.h +++ b/include/linux/serialP.h @@ -19,7 +19,6 @@ * For definitions of the flags field, see tty.h */ -#include #include #include #include -- cgit v1.2.1 From fbc0dc0df54be06586d712ebf6958816e3b1b2b7 Mon Sep 17 00:00:00 2001 From: Andrey Panin Date: Mon, 18 Jul 2005 11:38:09 +0100 Subject: [PATCH] Serial: Add support for SIIG Quartet serial card Add support for SIIG Quartet Serial card. This card has Oxford Semiconducor 16954 quad UART which is clocked by 10x faster (18.432 MHz) quartz. Signed-off-by: Andrey Panin Signed-off-by: Andrew Morton Signed-off-by: Russell King --- include/linux/pci_ids.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 27348c22dacb..9a28b312eeb4 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1872,6 +1872,7 @@ #define PCI_DEVICE_ID_CBOARDS_DAS1602_16 0x0001 #define PCI_VENDOR_ID_SIIG 0x131f +#define PCI_SUBVENDOR_ID_SIIG 0x131f #define PCI_DEVICE_ID_SIIG_1S_10x_550 0x1000 #define PCI_DEVICE_ID_SIIG_1S_10x_650 0x1001 #define PCI_DEVICE_ID_SIIG_1S_10x_850 0x1002 @@ -1909,6 +1910,7 @@ #define PCI_DEVICE_ID_SIIG_2S1P_20x_550 0x2060 #define PCI_DEVICE_ID_SIIG_2S1P_20x_650 0x2061 #define PCI_DEVICE_ID_SIIG_2S1P_20x_850 0x2062 +#define PCI_SUBDEVICE_ID_SIIG_QUARTET_SERIAL 0x2050 #define PCI_VENDOR_ID_RADISYS 0x1331 #define PCI_DEVICE_ID_RADISYS_ENP2611 0x0030 -- cgit v1.2.1 From 23af27eb8fa9ea8614138c4cded7a16cb4197a55 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 18 Jul 2005 13:34:35 -0700 Subject: [PKT_SCHED]: Kill TCF_META_ID_TCCLASSID. Thomas Graf states: > I used to mark such ids as obsolete in the header but since > skb is on diet anyway and there has been no official > iproute2 release with the ematch bits included it might be > a better idea to remove the ids from the header completely. > Those that have picked up my patch on netdev shouldn't care > about a ABI breakage, actually I doubt that someone is using > it already. So here's the patch to remove it. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/tc_ematch/tc_em_meta.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index bcb762d93123..90ab9f95d43d 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -53,7 +53,6 @@ enum TCF_META_ID_NFMARK, TCF_META_ID_TCINDEX, TCF_META_ID_TCVERDICT, - TCF_META_ID_TCCLASSID, TCF_META_ID_RTCLASSID, TCF_META_ID_RTIIF, TCF_META_ID_SK_FAMILY, -- cgit v1.2.1 From e2bf521d9728bfae9b6c3d484614e5962d0b5afd Mon Sep 17 00:00:00 2001 From: Victor Fusco Date: Mon, 18 Jul 2005 13:36:38 -0700 Subject: [NET]: Fix "nocast type" warnings in skbuff.h From: Victor Fusco Fix the sparse warning "implicit cast to nocast type" Signed-off-by: Victor Fusco Signed-off-by: Domen Puncer Signed-off-by: David S. Miller --- include/linux/skbuff.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 5d4a990d5577..0061c9470482 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -502,7 +502,8 @@ static inline struct sk_buff *skb_share_check(struct sk_buff *skb, * * %NULL is returned on a memory allocation failure. */ -static inline struct sk_buff *skb_unshare(struct sk_buff *skb, int pri) +static inline struct sk_buff *skb_unshare(struct sk_buff *skb, + unsigned int __nocast pri) { might_sleep_if(pri & __GFP_WAIT); if (skb_cloned(skb)) { -- cgit v1.2.1 From 4acdbdbe5089c06d5e0c7e96783fcc4414ded00a Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 21 Jul 2005 13:14:46 -0700 Subject: [NETFILTER]: ip_conntrack_expect_related must not free expectation If a connection tracking helper tells us to expect a connection, and we're already expecting that connection, we simply free the one they gave us and return success. The problem is that NAT helpers (eg. FTP) have to allocate the expectation first (to see what port is available) then rewrite the packet. If that rewrite fails, they try to remove the expectation, but it was freed in ip_conntrack_expect_related. This is one example of a larger problem: having registered the expectation, the pointer is no longer ours to use. Reference counting is needed for ctnetlink anyway, so introduce it now. To have a single "put" path, we need to grab the reference to the connection on creation, rather than open-coding it in the caller. Signed-off-by: Rusty Russell Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_conntrack.h | 3 +++ include/linux/netfilter_ipv4/ip_conntrack_helper.h | 7 ++++--- 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index 3781192ce159..f8da7ddeff3a 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -197,6 +197,9 @@ struct ip_conntrack_expect /* Timer function; deletes the expectation. */ struct timer_list timeout; + /* Usage count. */ + atomic_t use; + #ifdef CONFIG_IP_NF_NAT_NEEDED /* This is the original per-proto part, used to map the * expected connection the way the recipient expects. */ diff --git a/include/linux/netfilter_ipv4/ip_conntrack_helper.h b/include/linux/netfilter_ipv4/ip_conntrack_helper.h index b1bbba0a12cb..3692daa93dec 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack_helper.h +++ b/include/linux/netfilter_ipv4/ip_conntrack_helper.h @@ -30,9 +30,10 @@ extern int ip_conntrack_helper_register(struct ip_conntrack_helper *); extern void ip_conntrack_helper_unregister(struct ip_conntrack_helper *); /* Allocate space for an expectation: this is mandatory before calling - ip_conntrack_expect_related. */ -extern struct ip_conntrack_expect *ip_conntrack_expect_alloc(void); -extern void ip_conntrack_expect_free(struct ip_conntrack_expect *exp); + ip_conntrack_expect_related. You will have to call put afterwards. */ +extern struct ip_conntrack_expect * +ip_conntrack_expect_alloc(struct ip_conntrack *master); +extern void ip_conntrack_expect_put(struct ip_conntrack_expect *exp); /* Add an expected connection: can have more than one per connection */ extern int ip_conntrack_expect_related(struct ip_conntrack_expect *exp); -- cgit v1.2.1 From 28e212fb360ce2568edd60b93d60683d5ad24146 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 22 Jul 2005 11:47:25 -0700 Subject: [PKT_SCHED]: Kill TCF_META_ID_REALDEV from meta ematch. It won't exist any longer when we shrink the SKB in 2.6.14, and we should kill this off before anyone in userspace starts using it. Signed-off-by: David S. Miller Acked-by: Thomas Graf --- include/linux/tc_ematch/tc_em_meta.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index 90ab9f95d43d..150efe07ff6b 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -42,7 +42,6 @@ enum TCF_META_ID_LOADAVG_2, TCF_META_ID_DEV, TCF_META_ID_INDEV, - TCF_META_ID_REALDEV, TCF_META_ID_PRIORITY, TCF_META_ID_PROTOCOL, TCF_META_ID_SECURITY, /* obsolete */ -- cgit v1.2.1 From 261688d01ec07d3a265b8ace6ec68310fbd96a96 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 22 Jul 2005 14:43:52 -0700 Subject: [PKT_SCHED]: em_meta: Kill TCF_META_ID_{INDEV,SECURITY,TCVERDICT} More unusable TCF_META_* match types that need to get eliminated before 2.6.13 goes out the door. Signed-off-by: David S. Miller Acked-by: Thomas Graf --- include/linux/tc_ematch/tc_em_meta.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/tc_ematch/tc_em_meta.h b/include/linux/tc_ematch/tc_em_meta.h index 150efe07ff6b..081b1ee8516e 100644 --- a/include/linux/tc_ematch/tc_em_meta.h +++ b/include/linux/tc_ematch/tc_em_meta.h @@ -41,17 +41,14 @@ enum TCF_META_ID_LOADAVG_1, TCF_META_ID_LOADAVG_2, TCF_META_ID_DEV, - TCF_META_ID_INDEV, TCF_META_ID_PRIORITY, TCF_META_ID_PROTOCOL, - TCF_META_ID_SECURITY, /* obsolete */ TCF_META_ID_PKTTYPE, TCF_META_ID_PKTLEN, TCF_META_ID_DATALEN, TCF_META_ID_MACLEN, TCF_META_ID_NFMARK, TCF_META_ID_TCINDEX, - TCF_META_ID_TCVERDICT, TCF_META_ID_RTCLASSID, TCF_META_ID_RTIIF, TCF_META_ID_SK_FAMILY, -- cgit v1.2.1 From 48647feed9f7a2d839c6ada12147b341833646e8 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 24 Jul 2005 19:30:28 -0700 Subject: [W1]: Do not use NFLOG netlink number. Use the reserved by never used NETLINK_SKIP value instead. Signed-off-by: David S. Miller --- include/linux/netlink.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 2f0c085f2c7d..70c2a9dc4b2b 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -5,7 +5,7 @@ #include #define NETLINK_ROUTE 0 /* Routing/device hook */ -#define NETLINK_SKIP 1 /* Reserved for ENskip */ +#define NETLINK_W1 1 /* 1-wire subsystem */ #define NETLINK_USERSOCK 2 /* Reserved for user mode socket protocols */ #define NETLINK_FIREWALL 3 /* Firewalling hook */ #define NETLINK_TCPDIAG 4 /* TCP socket monitoring */ -- cgit v1.2.1 From 4cf78e4fb678807e3f8265c9e9031a84f5c601f0 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Mon, 25 Jul 2005 12:29:19 -0700 Subject: [TG3]: add 5780 basic support Add 5780 PCI IDs, chip IDs, and other basic support. Signed-off-by: Michael Chan Signed-off-by: David S. Miller --- include/linux/pci_ids.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 9a28b312eeb4..9e1337783a31 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2098,6 +2098,8 @@ #define PCI_DEVICE_ID_TIGON3_5721 0x1659 #define PCI_DEVICE_ID_TIGON3_5705M 0x165d #define PCI_DEVICE_ID_TIGON3_5705M_2 0x165e +#define PCI_DEVICE_ID_TIGON3_5780 0x166a +#define PCI_DEVICE_ID_TIGON3_5780S 0x166b #define PCI_DEVICE_ID_TIGON3_5705F 0x166e #define PCI_DEVICE_ID_TIGON3_5750 0x1676 #define PCI_DEVICE_ID_TIGON3_5751 0x1677 -- cgit v1.2.1 From 4a00ea1e18228e5ef99d4780671fda97226bda30 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Tue, 26 Jul 2005 11:24:14 -0600 Subject: [PATCH] Refactor sys_reboot into reusable parts Because the factors of sys_reboot don't exist people calling into the reboot path duplicate the code badly, leading to inconsistent expectations of code in the reboot path. This patch should is just code motion. Signed-off-by: Eric W. Biederman Signed-off-by: Linus Torvalds --- include/linux/reboot.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/reboot.h b/include/linux/reboot.h index 2d4dd23168dd..828ba4f107d9 100644 --- a/include/linux/reboot.h +++ b/include/linux/reboot.h @@ -55,6 +55,15 @@ extern void machine_shutdown(void); struct pt_regs; extern void machine_crash_shutdown(struct pt_regs *); +/* + * Architecture independent implemenations of sys_reboot commands. + */ + +extern void kernel_restart(char *cmd); +extern void kernel_halt(void); +extern void kernel_power_off(void); +extern void kernel_kexec(void); + #endif #endif /* _LINUX_REBOOT_H */ -- cgit v1.2.1 From 7c9034735eccbf82608a4602c59aaf6053ea9416 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Tue, 26 Jul 2005 11:29:55 -0600 Subject: [PATCH] Add emergency_restart() When the kernel is working well and we want to restart cleanly kernel_restart is the function to use. But in many instances the kernel wants to reboot when thing are expected to be working very badly such as from panic or a software watchdog handler. This patch adds the function emergency_restart() so that callers can be clear what semantics they expect when calling restart. emergency_restart() is expected to be callable from interrupt context and possibly reliable in even more trying circumstances. This is an initial generic implementation for all architectures. Signed-off-by: Eric W. Biederman Signed-off-by: Linus Torvalds --- include/linux/reboot.h | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'include/linux') diff --git a/include/linux/reboot.h b/include/linux/reboot.h index 828ba4f107d9..3b3266ff1a95 100644 --- a/include/linux/reboot.h +++ b/include/linux/reboot.h @@ -64,6 +64,13 @@ extern void kernel_halt(void); extern void kernel_power_off(void); extern void kernel_kexec(void); +/* + * Emergency restart, callable from an interrupt handler. + */ + +extern void emergency_restart(void); +#include + #endif #endif /* _LINUX_REBOOT_H */ -- cgit v1.2.1 From cadf01c2fc0cd66dfef4956ef1a6482ed01c3150 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 26 Jul 2005 15:39:28 -0700 Subject: [NETFILTER]: Fix ip_conntrack_put() prototype. The function is not inline. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- include/linux/netfilter_ipv4/ip_conntrack.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netfilter_ipv4/ip_conntrack.h b/include/linux/netfilter_ipv4/ip_conntrack.h index f8da7ddeff3a..08fe5f7d14a0 100644 --- a/include/linux/netfilter_ipv4/ip_conntrack.h +++ b/include/linux/netfilter_ipv4/ip_conntrack.h @@ -239,7 +239,7 @@ ip_conntrack_get(const struct sk_buff *skb, enum ip_conntrack_info *ctinfo) } /* decrement reference count on a conntrack */ -extern inline void ip_conntrack_put(struct ip_conntrack *ct); +extern void ip_conntrack_put(struct ip_conntrack *ct); /* call to create an explicit dependency on ip_conntrack. */ extern void need_ip_conntrack(void); -- cgit v1.2.1 From c10b873695c6a1de0d8ebab40b525575ca576683 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 27 Jul 2005 11:43:27 -0700 Subject: [PATCH] Really __nocast-annotate kmalloc_node() One chunk was lost somewhere between my and Andrew's machine. Noticed by Victor Fusco. Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/slab.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/slab.h b/include/linux/slab.h index 4c8e552471b0..80b2dfde2e80 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -111,7 +111,7 @@ static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int n { return kmem_cache_alloc(cachep, flags); } -static inline void *kmalloc_node(size_t size, int flags, int node) +static inline void *kmalloc_node(size_t size, unsigned int __nocast flags, int node) { return kmalloc(size, flags); } -- cgit v1.2.1 From 49f29915856435ad8e34a4a3a907b09682a5826e Mon Sep 17 00:00:00 2001 From: Olivier Blin Date: Wed, 27 Jul 2005 11:43:47 -0700 Subject: [PATCH] i4l: add Olitec ISDN PCI card in hisax gazel driver This patch adds support for the Olitec ISDN PCI card in the hisax gazel driver. The gazel driver supports this card, but wasn't aware of its PCI ids. Users used to modify the PCI ids of a supported card in include/linux/pci_ids.h and recompile their kernel to get this card running, as said in most Howtos. This patch makes the hisax gazel driver recognize the PCI ids of the Olitec ISDN PCI card. Signed-off-by: Olivier Blin Signed-off-by: Karsten Keil Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/pci_ids.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index d2ad2c4f835a..bc4cc10fabe9 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1020,6 +1020,7 @@ #define PCI_DEVICE_ID_PLX_SPCOM200 0x1103 #define PCI_DEVICE_ID_PLX_DJINN_ITOO 0x1151 #define PCI_DEVICE_ID_PLX_R753 0x1152 +#define PCI_DEVICE_ID_PLX_OLITEC 0x1187 #define PCI_DEVICE_ID_PLX_9030 0x9030 #define PCI_DEVICE_ID_PLX_9050 0x9050 #define PCI_DEVICE_ID_PLX_9060 0x9060 -- cgit v1.2.1 From 4bfdf37830111321e2cd1fe0102dd776ce93194d Mon Sep 17 00:00:00 2001 From: Andrey Panin Date: Wed, 27 Jul 2005 11:43:58 -0700 Subject: [PATCH] consolidate CONFIG_WATCHDOG_NOWAYOUT handling Attached patch removes #ifdef CONFIG_WATCHDOG_NOWAYOUT mess duplicated in almost every watchdog driver and replaces it with common define in linux/watchdog.h. Signed-off-by: Andrey Panin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/watchdog.h | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'include/linux') diff --git a/include/linux/watchdog.h b/include/linux/watchdog.h index 88ba0d29f8c8..1192ed8f4fe8 100644 --- a/include/linux/watchdog.h +++ b/include/linux/watchdog.h @@ -47,4 +47,14 @@ struct watchdog_info { #define WDIOS_ENABLECARD 0x0002 /* Turn on the watchdog timer */ #define WDIOS_TEMPPANIC 0x0004 /* Kernel panic on temperature trip */ +#ifdef __KERNEL__ + +#ifdef CONFIG_WATCHDOG_NOWAYOUT +#define WATCHDOG_NOWAYOUT 1 +#else +#define WATCHDOG_NOWAYOUT 0 +#endif + +#endif /* __KERNEL__ */ + #endif /* ifndef _LINUX_WATCHDOG_H */ -- cgit v1.2.1 From 951f22d5b1f0eaae35dafc669e3774a0c2084d10 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Wed, 27 Jul 2005 11:44:57 -0700 Subject: [PATCH] s390: spin lock retry Split spin lock and r/w lock implementation into a single try which is done inline and an out of line function that repeatedly tries to get the lock before doing the cpu_relax(). Add a system control to set the number of retries before a cpu is yielded. The reason for the spin lock retry is that the diagnose 0x44 that is used to give up the virtual cpu is quite expensive. For spin locks that are held only for a short period of time the costs of the diagnoses outweights the savings for spin locks that are held for a longer timer. The default retry count is 1000. Signed-off-by: Martin Schwidefsky Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sysctl.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index bfbbe94b297d..e82be96d4906 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -145,6 +145,7 @@ enum KERN_BOOTLOADER_TYPE=67, /* int: boot loader type */ KERN_RANDOMIZE=68, /* int: randomize virtual address space */ KERN_SETUID_DUMPABLE=69, /* int: behaviour of dumps for setuid core */ + KERN_SPIN_RETRY=70, /* int: number of spinlock retries */ }; -- cgit v1.2.1 From c293621bbf678a3d85e3ed721c3921c8a670610d Mon Sep 17 00:00:00 2001 From: Peter Staubach Date: Wed, 27 Jul 2005 11:45:09 -0700 Subject: [PATCH] stale POSIX lock handling I believe that there is a problem with the handling of POSIX locks, which the attached patch should address. The problem appears to be a race between fcntl(2) and close(2). A multithreaded application could close a file descriptor at the same time as it is trying to acquire a lock using the same file descriptor. I would suggest that that multithreaded application is not providing the proper synchronization for itself, but the OS should still behave correctly. SUS3 (Single UNIX Specification Version 3, read: POSIX) indicates that when a file descriptor is closed, that all POSIX locks on the file, owned by the process which closed the file descriptor, should be released. The trick here is when those locks are released. The current code releases all locks which exist when close is processing, but any locks in progress are handled when the last reference to the open file is released. There are three cases to consider. One is the simple case, a multithreaded (mt) process has a file open and races to close it and acquire a lock on it. In this case, the close will release one reference to the open file and when the fcntl is done, it will release the other reference. For this situation, no locks should exist on the file when both the close and fcntl operations are done. The current system will handle this case because the last reference to the open file is being released. The second case is when the mt process has dup(2)'d the file descriptor. The close will release one reference to the file and the fcntl, when done, will release another, but there will still be at least one more reference to the open file. One could argue that the existence of a lock on the file after the close has completed is okay, because it was acquired after the close operation and there is still a way for the application to release the lock on the file, using an existing file descriptor. The third case is when the mt process has forked, after opening the file and either before or after becoming an mt process. In this case, each process would hold a reference to the open file. For each process, this degenerates to first case above. However, the lock continues to exist until both processes have released their references to the open file. This lock could block other lock requests. The changes to release the lock when the last reference to the open file aren't quite right because they would allow the lock to exist as long as there was a reference to the open file. This is too long. The new proposed solution is to add support in the fcntl code path to detect a race with close and then to release the lock which was just acquired when such as race is detected. This causes locks to be released in a timely fashion and for the system to conform to the POSIX semantic specification. This was tested by instrumenting a kernel to detect the handling locks and then running a program which generates case #3 above. A dangling lock could be reliably generated. When the changes to detect the close/fcntl race were added, a dangling lock could no longer be generated. Cc: Matthew Wilcox Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/fs.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index 0f53e0124941..f9adf75fd9b4 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -697,11 +697,13 @@ extern struct list_head file_lock_list; #include extern int fcntl_getlk(struct file *, struct flock __user *); -extern int fcntl_setlk(struct file *, unsigned int, struct flock __user *); +extern int fcntl_setlk(unsigned int, struct file *, unsigned int, + struct flock __user *); #if BITS_PER_LONG == 32 extern int fcntl_getlk64(struct file *, struct flock64 __user *); -extern int fcntl_setlk64(struct file *, unsigned int, struct flock64 __user *); +extern int fcntl_setlk64(unsigned int, struct file *, unsigned int, + struct flock64 __user *); #endif extern void send_sigio(struct fown_struct *fown, int fd, int band); -- cgit v1.2.1 From 8c52ab42c11b5a7fb44bb84c954d09968e90e9e7 Mon Sep 17 00:00:00 2001 From: Andreas Gruenbacher Date: Wed, 27 Jul 2005 11:45:15 -0700 Subject: [PATCH] mbcache: Remove unused mb_cache_shrink parameter The cache parameter to mb_cache_shrink isn't used. We may as well remove it. Signed-off-by: Andreas Gruenbacher Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mbcache.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/mbcache.h b/include/linux/mbcache.h index 8e5a10410a30..9263d2db2d67 100644 --- a/include/linux/mbcache.h +++ b/include/linux/mbcache.h @@ -29,7 +29,7 @@ struct mb_cache_op { struct mb_cache * mb_cache_create(const char *, struct mb_cache_op *, size_t, int, int); -void mb_cache_shrink(struct mb_cache *, struct block_device *); +void mb_cache_shrink(struct block_device *); void mb_cache_destroy(struct mb_cache *); /* Functions on cache entries */ -- cgit v1.2.1 From 44456d37b59d8e541936ed26d8b6e08d27e88ac1 Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Wed, 27 Jul 2005 11:45:17 -0700 Subject: [PATCH] turn many #if $undefined_string into #ifdef $undefined_string turn many #if $undefined_string into #ifdef $undefined_string to fix some warnings after -Wno-def was added to global CFLAGS Signed-off-by: Olaf Hering Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/ftape.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ftape.h b/include/linux/ftape.h index c6b38d5b9186..72faeec9f6e1 100644 --- a/include/linux/ftape.h +++ b/include/linux/ftape.h @@ -165,7 +165,7 @@ typedef union { # undef CONFIG_FT_FDC_DMA # define CONFIG_FT_FDC_DMA 2 # endif -#elif CONFIG_FT_ALT_FDC == 1 /* CONFIG_FT_MACH2 */ +#elif defined(CONFIG_FT_ALT_FDC) /* CONFIG_FT_MACH2 */ # if CONFIG_FT_FDC_BASE == 0 # undef CONFIG_FT_FDC_BASE # define CONFIG_FT_FDC_BASE 0x370 -- cgit v1.2.1 From e5c2d749172657ed51e20e4b5ab540447666cc50 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 27 Jul 2005 11:45:19 -0700 Subject: [PATCH] serial_core whitespace fix Use tabs for formatting like anywhere else in this file. Cc: Russell King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/serial_core.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index 30b64f3534f4..f6fca8f2f3ca 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -104,7 +104,7 @@ #define PORT_MPSC 63 /* TXX9 type number */ -#define PORT_TXX9 64 +#define PORT_TXX9 64 /* NEC VR4100 series SIU/DSIU */ #define PORT_VR41XX_SIU 65 -- cgit v1.2.1 From 79a8810221ee9ea96c4e5a5817afb88f22ea698c Mon Sep 17 00:00:00 2001 From: Richard Henderson Date: Thu, 28 Jul 2005 01:07:41 -0700 Subject: [PATCH] alpha: fix "statement with no effect" warnings Apparently gcc 4.0 complains about "({ 0; });", which leads to -Werror breakage in one of the alpha oprofile modules. One might could argue that this is a gcc bug, in that statement-expressions should be considered to be function-like rather than statement-like for the purposes of this warning. But it's just as easy to use an inline function in the first place, side-stepping the issue. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/smp.h | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/smp.h b/include/linux/smp.h index 9dfa3ee769ae..22b451d1b93f 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -94,11 +94,23 @@ void smp_prepare_boot_cpu(void); */ #define raw_smp_processor_id() 0 #define hard_smp_processor_id() 0 -#define smp_call_function(func,info,retry,wait) ({ 0; }) -#define on_each_cpu(func,info,retry,wait) ({ func(info); 0; }) -static inline void smp_send_reschedule(int cpu) { } #define num_booting_cpus() 1 -#define smp_prepare_boot_cpu() do {} while (0) + +static inline int smp_call_function(void (*func) (void *), void *info, + int retry, int wait) +{ + return 0; +} + +static inline int on_each_cpu(void (*func) (void *), void *info, + int retry, int wait) +{ + func(info); + return 0; +} + +static inline void smp_send_reschedule(int cpu) { } +static inline void smp_prepare_boot_cpu(void) { } #endif /* !SMP */ -- cgit v1.2.1 From 2ac6608c41f8c45371ea9dddae7f99bc2c15d5cf Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 28 Jul 2005 10:34:47 -0700 Subject: Revert broken "statement with no effect" warning fix It may shut up gcc, but it also incorrectly changes the semantics of the smp_call_function() helpers. You can fix the warning other ways if you are interested (create another inline function that takes no arguments and returns zero), but preferably gcc just shouldn't complain about unused return values from statement expressions in the first place. --- include/linux/smp.h | 20 ++++---------------- 1 file changed, 4 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/smp.h b/include/linux/smp.h index 22b451d1b93f..9dfa3ee769ae 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -94,23 +94,11 @@ void smp_prepare_boot_cpu(void); */ #define raw_smp_processor_id() 0 #define hard_smp_processor_id() 0 -#define num_booting_cpus() 1 - -static inline int smp_call_function(void (*func) (void *), void *info, - int retry, int wait) -{ - return 0; -} - -static inline int on_each_cpu(void (*func) (void *), void *info, - int retry, int wait) -{ - func(info); - return 0; -} - +#define smp_call_function(func,info,retry,wait) ({ 0; }) +#define on_each_cpu(func,info,retry,wait) ({ func(info); 0; }) static inline void smp_send_reschedule(int cpu) { } -static inline void smp_prepare_boot_cpu(void) { } +#define num_booting_cpus() 1 +#define smp_prepare_boot_cpu() do {} while (0) #endif /* !SMP */ -- cgit v1.2.1 From a46e812620bd7db457ce002544a1a6572c313d8a Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Fri, 29 Jul 2005 12:16:27 -0700 Subject: [PATCH] PCI: fix up errors after dma bursting patch and CONFIG_PCI=n -- bug? In the patch from: http://www.uwsg.iu.edu/hypermail/linux/kernel/0506.3/0985.html Is the the following line suppose inside the if CONFIG_PCI=n #define pci_dma_burst_advice(pdev, strat, strategy_parameter) do { } while (0) Signed-off-by: Kumar Gala Signed-off-by: Greg Kroah-Hartman Signed-off-by: Linus Torvalds --- include/linux/pci.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 7ac14961ba22..8621cf42b46f 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -971,6 +971,8 @@ static inline int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int en #define isa_bridge ((struct pci_dev *)NULL) +#define pci_dma_burst_advice(pdev, strat, strategy_parameter) do { } while (0) + #else /* @@ -985,9 +987,6 @@ static inline int pci_proc_domain(struct pci_bus *bus) return 0; } #endif - -#define pci_dma_burst_advice(pdev, strat, strategy_parameter) do { } while (0) - #endif /* !CONFIG_PCI */ /* these helpers provide future and backwards compatibility -- cgit v1.2.1 From 87bec66b9691522414862dd8d41e430b063735ef Mon Sep 17 00:00:00 2001 From: David Shaohua Li Date: Wed, 27 Jul 2005 23:02:00 -0400 Subject: [ACPI] suspend/resume ACPI PCI Interrupt Links Add reference count and disable ACPI PCI Interrupt Link when no device still uses it. Warn when drivers have not released Link at suspend time. http://bugzilla.kernel.org/show_bug.cgi?id=3469 Signed-off-by: David Shaohua Li Signed-off-by: Len Brown --- include/linux/acpi.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 9c14959bcfa0..ca0cd240cee0 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -440,9 +440,7 @@ int acpi_gsi_to_irq (u32 gsi, unsigned int *irq); * If this matches the last registration, any IRQ resources for gsi * are freed. */ -#ifdef CONFIG_ACPI_DEALLOCATE_IRQ void acpi_unregister_gsi (u32 gsi); -#endif #ifdef CONFIG_ACPI_PCI @@ -467,9 +465,7 @@ struct pci_dev; int acpi_pci_irq_enable (struct pci_dev *dev); void acpi_penalize_isa_irq(int irq, int active); -#ifdef CONFIG_ACPI_DEALLOCATE_IRQ void acpi_pci_irq_disable (struct pci_dev *dev); -#endif struct acpi_pci_driver { struct acpi_pci_driver *next; -- cgit v1.2.1 From 7544953685859875b5ac0260b6b1856066c092d6 Mon Sep 17 00:00:00 2001 From: John McCutchan Date: Mon, 1 Aug 2005 11:00:45 -0400 Subject: [PATCH] inotify: fix file deletion by rename detection When a file is moved over an existing file that you are watching, inotify won't send you a DELETE_SELF event and it won't unref the inode until the inotify instance is closed by the application. Signed-off-by: John McCutchan Signed-off-by: Robert Love Signed-off-by: Linus Torvalds --- include/linux/fsnotify.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index d07a92c94776..e96a4306ab3b 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -21,7 +21,7 @@ */ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, const char *old_name, const char *new_name, - int isdir) + int isdir, struct inode *target) { u32 cookie = inotify_get_cookie(); @@ -36,6 +36,11 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, isdir = IN_ISDIR; inotify_inode_queue_event(old_dir, IN_MOVED_FROM|isdir,cookie,old_name); inotify_inode_queue_event(new_dir, IN_MOVED_TO|isdir, cookie, new_name); + + if (target) { + inotify_inode_queue_event(target, IN_DELETE_SELF, 0, NULL); + inotify_inode_is_dead(target); + } } /* -- cgit v1.2.1 From 0072b1389c25355ccc01048114adb9652c13fd9f Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Mon, 1 Aug 2005 21:11:39 -0700 Subject: [PATCH] include/linux/dcookies.h: dummy functions must be "static inline" We don't want these to be global functions. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/dcookies.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/dcookies.h b/include/linux/dcookies.h index c28050136164..1d68428c925d 100644 --- a/include/linux/dcookies.h +++ b/include/linux/dcookies.h @@ -48,12 +48,12 @@ int get_dcookie(struct dentry * dentry, struct vfsmount * vfsmnt, #else -struct dcookie_user * dcookie_register(void) +static inline struct dcookie_user * dcookie_register(void) { return NULL; } -void dcookie_unregister(struct dcookie_user * user) +static inline void dcookie_unregister(struct dcookie_user * user) { return; } -- cgit v1.2.1 From f33ea7f404e592e4563b12101b7a4d17da6558d7 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 3 Aug 2005 20:24:01 +1000 Subject: [PATCH] fix get_user_pages bug Checking pte_dirty instead of pte_write in __follow_page is problematic for s390, and for copy_one_pte which leaves dirty when clearing write. So revert __follow_page to check pte_write as before, and make do_wp_page pass back a special extra VM_FAULT_WRITE bit to say it has done its full job: once get_user_pages receives this value, it no longer requires pte_write in __follow_page. But most callers of handle_mm_fault, in the various architectures, have switch statements which do not expect this new case. To avoid changing them all in a hurry, make an inline wrapper function (using the old name) that masks off the new bit, and use the extended interface with double underscores. Yes, we do have a call to do_wp_page from do_swap_page, but no need to change that: in rare case it's needed, another do_wp_page will follow. Signed-off-by: Hugh Dickins [ Cleanups by Nick Piggin ] Signed-off-by: Linus Torvalds --- include/linux/mm.h | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mm.h b/include/linux/mm.h index 6eb7f48317f8..82d7024f0765 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -625,10 +625,16 @@ static inline int page_mapped(struct page *page) * Used to decide whether a process gets delivered SIGBUS or * just gets major/minor fault counters bumped up. */ -#define VM_FAULT_OOM (-1) -#define VM_FAULT_SIGBUS 0 -#define VM_FAULT_MINOR 1 -#define VM_FAULT_MAJOR 2 +#define VM_FAULT_OOM 0x00 +#define VM_FAULT_SIGBUS 0x01 +#define VM_FAULT_MINOR 0x02 +#define VM_FAULT_MAJOR 0x03 + +/* + * Special case for get_user_pages. + * Must be in a distinct bit from the above VM_FAULT_ flags. + */ +#define VM_FAULT_WRITE 0x10 #define offset_in_page(p) ((unsigned long)(p) & ~PAGE_MASK) @@ -704,7 +710,13 @@ extern pte_t *FASTCALL(pte_alloc_kernel(struct mm_struct *mm, pmd_t *pmd, unsign extern pte_t *FASTCALL(pte_alloc_map(struct mm_struct *mm, pmd_t *pmd, unsigned long address)); extern int install_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, struct page *page, pgprot_t prot); extern int install_file_pte(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, unsigned long pgoff, pgprot_t prot); -extern int handle_mm_fault(struct mm_struct *mm,struct vm_area_struct *vma, unsigned long address, int write_access); +extern int __handle_mm_fault(struct mm_struct *mm,struct vm_area_struct *vma, unsigned long address, int write_access); + +static inline int handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long address, int write_access) +{ + return __handle_mm_fault(mm, vma, address, write_access) & (~VM_FAULT_WRITE); +} + extern int make_pages_present(unsigned long addr, unsigned long end); extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write); void install_arg_page(struct vm_area_struct *, struct page *, unsigned long); -- cgit v1.2.1 From 6b8b3e8a8b3e62b4209eaa36697e3c9df457e196 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 4 Aug 2005 12:53:35 -0700 Subject: [PATCH] md: make sure md bitmap updates are flushed when array is stopped. The recent change to never ignore the bitmap, revealed that the bitmap isn't begin flushed properly when an array is stopped. We call bitmap_daemon_work three times as there is a three-stage pipeline for flushing updates to the bitmap file. Signed-off-by: Neil Brown Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/raid/bitmap.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/raid/bitmap.h b/include/linux/raid/bitmap.h index 6213e976eade..4bf1659f8aa8 100644 --- a/include/linux/raid/bitmap.h +++ b/include/linux/raid/bitmap.h @@ -248,6 +248,7 @@ struct bitmap { /* these are used only by md/bitmap */ int bitmap_create(mddev_t *mddev); +void bitmap_flush(mddev_t *mddev); void bitmap_destroy(mddev_t *mddev); int bitmap_active(struct bitmap *bitmap); -- cgit v1.2.1 From fec59a711eef002d4ef9eb8de09dd0a26986eb77 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Thu, 4 Aug 2005 18:06:10 -0700 Subject: [PATCH] PCI: restore BAR values after D3hot->D0 for devices that need it Some PCI devices (e.g. 3c905B, 3c556B) lose all configuration (including BARs) when transitioning from D3hot->D0. This leaves such a device in an inaccessible state. The patch below causes the BARs to be restored when enabling such a device, so that its driver will be able to access it. The patch also adds pci_restore_bars as a new global symbol, and adds a correpsonding EXPORT_SYMBOL_GPL for that. Some firmware (e.g. Thinkpad T21) leaves devices in D3hot after a (re)boot. Most drivers call pci_enable_device very early, so devices left in D3hot that lose configuration during the D3hot->D0 transition will be inaccessible to their drivers. Drivers could be modified to account for this, but it would be difficult to know which drivers need modification. This is especially true since often many devices are covered by the same driver. It likely would be necessary to replicate code across dozens of drivers. The patch below should trigger only when transitioning from D3hot->D0 (or at boot), and only for devices that have the "no soft reset" bit cleared in the PM control register. I believe it is safe to include this patch as part of the PCI infrastructure. The cleanest implementation of pci_restore_bars was to call pci_update_resource. Unfortunately, that does not currently exist for the sparc64 architecture. The patch below includes a null implemenation of pci_update_resource for sparc64. Some have expressed interest in making general use of the the pci_restore_bars function, so that has been exported to GPL licensed modules. Signed-off-by: John W. Linville Signed-off-by: Greg Kroah-Hartman Signed-off-by: Linus Torvalds --- include/linux/pci.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 8621cf42b46f..98bdd95fcee9 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -225,6 +225,7 @@ #define PCI_PM_CAP_PME_D3cold 0x8000 /* PME# from D3 (cold) */ #define PCI_PM_CTRL 4 /* PM control and status register */ #define PCI_PM_CTRL_STATE_MASK 0x0003 /* Current power state (D0 to D3) */ +#define PCI_PM_CTRL_NO_SOFT_RESET 0x0004 /* No reset for D3hot->D0 */ #define PCI_PM_CTRL_PME_ENABLE 0x0100 /* PME pin enable */ #define PCI_PM_CTRL_DATA_SEL_MASK 0x1e00 /* Data select (??) */ #define PCI_PM_CTRL_DATA_SCALE_MASK 0x6000 /* Data scale (??) */ @@ -816,7 +817,9 @@ int pci_set_mwi(struct pci_dev *dev); void pci_clear_mwi(struct pci_dev *dev); int pci_set_dma_mask(struct pci_dev *dev, u64 mask); int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask); +void pci_update_resource(struct pci_dev *dev, struct resource *res, int resno); int pci_assign_resource(struct pci_dev *dev, int i); +void pci_restore_bars(struct pci_dev *dev); /* ROM control related routines */ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size); -- cgit v1.2.1 From 0c3dba1534569734ba353afdf3f11def497ff2ac Mon Sep 17 00:00:00 2001 From: John McCutchan Date: Thu, 4 Aug 2005 21:12:54 -0400 Subject: [PATCH] Clean up inotify delete race fix This avoids the whole #ifdef mess by just getting a copy of dentry->d_inode before d_delete is called - that makes the codepaths the same for the INOTIFY/DNOTIFY cases as for the regular no-notify case. I've been running this under a Gnome session for the last 10 minutes. Inotify is being used extensively. Signed-off-by: John McCutchan Signed-off-by: Linus Torvalds --- include/linux/fsnotify.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index e96a4306ab3b..1cb4935348d8 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -46,10 +46,8 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, /* * fsnotify_unlink - file was unlinked */ -static inline void fsnotify_unlink(struct dentry *dentry, struct inode *dir) +static inline void fsnotify_unlink(struct dentry *dentry, struct inode *inode, struct inode *dir) { - struct inode *inode = dentry->d_inode; - inode_dir_notify(dir, DN_DELETE); inotify_inode_queue_event(dir, IN_DELETE, 0, dentry->d_name.name); inotify_inode_queue_event(inode, IN_DELETE_SELF, 0, NULL); -- cgit v1.2.1 From ba02508248e90a9d696aebd18b48a3290235b53c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 5 Aug 2005 13:28:11 -0700 Subject: [PATCH] blk: fix tag shrinking (revive real_max_size) My patch in commit fa72b903f75e4f0f0b2c2feed093005167da4023 incorrectly removed blk_queue_tag->real_max_depth. The original resize implementation was incorrect in the following points. * actual allocation size of tag_index was shorter than real_max_size, but assumed to be of the same size, possibly causing memory access beyond the allocated area. * bits in tag_map between max_deptn and real_max_depth were initialized to 1's, making the tags permanently reserved. In an attempt to fix above two bugs, I had removed allocation optimization in init_tag_map and real_max_size. Tag map/index were allocated and freed immediately during resize. Unfortunately, I wasn't considering that tag map/index can be resized dynamically with tags beyond new_depth active. This led to accessing freed area after shrinking tags and led to the following bug reporting thread on linux-scsi. http://marc.theaimsgroup.com/?l=linux-scsi&m=112319898111885&w=2 To fix the problem, I've revived real_max_depth without allocation optimization in init_tag_map, and Andrew Vasquez confirmed that the problem was fixed. As Jens is not going to be available for a week, he asked me to make sure that this patch reaches you. http://marc.theaimsgroup.com/?l=linux-scsi&m=112325778530886&w=2 Also, a comment was added to make sure that real_max_size is needed for dynamic shrinking. Signed-off-by: Tejun Heo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/blkdev.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 0881b5cdee3d..19bd8e7e11bf 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -301,6 +301,7 @@ struct blk_queue_tag { struct list_head busy_list; /* fifo list of busy tags */ int busy; /* current depth */ int max_depth; /* what we will send to device */ + int real_max_depth; /* what the array can hold */ atomic_t refcnt; /* map can be shared */ }; -- cgit v1.2.1 From 243393c90f2b7cb781fd794e22786e9c8547901a Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 6 Aug 2005 09:39:57 -0700 Subject: Add fakey 'deflateBound()' function to the in-kernel zlib routines It's not the real deflateBound() in newer zlib libraries, partly because the upcoming usage of it won't have the "stream" available, so we can't have the same interfaces anyway. --- include/linux/zlib.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/zlib.h b/include/linux/zlib.h index 850076ea14d3..74f7b78c22d2 100644 --- a/include/linux/zlib.h +++ b/include/linux/zlib.h @@ -506,6 +506,11 @@ extern int zlib_deflateReset (z_streamp strm); stream state was inconsistent (such as zalloc or state being NULL). */ +static inline unsigned long deflateBound(unsigned long s) +{ + return s + ((s + 7) >> 3) + ((s + 63) >> 6) + 11; +} + extern int zlib_deflateParams (z_streamp strm, int level, int strategy); /* Dynamically update the compression level and compression strategy. The -- cgit v1.2.1 From 9ae5b3c703cce89a7d8ccf25fe16955ec6f016c0 Mon Sep 17 00:00:00 2001 From: Olaf Hering Date: Sun, 7 Aug 2005 09:42:24 -0700 Subject: [PATCH] remove linux/pagemap.h from linux/swap.h sparc can not include linux/pagemap.h because of the following circular dependency: asm-sparc/pgtable include linux/swap.h linux/swap.h include now linux/pagemap.h linux/pagemap.h include linux/mm.h linux/mm.h include asm/pgtable.h It needs to have the swp_entry_t type fully visible in pgtable.h, we can't work around this using macros. Signed-off-by: Olaf Hering Cc: William Lee Irwin III Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/swap.h b/include/linux/swap.h index 239f520cc49e..bfe3e763ccf2 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -7,7 +7,6 @@ #include #include #include -#include #include #include @@ -255,6 +254,8 @@ static inline void put_swap_token(struct mm_struct *mm) #define si_swapinfo(val) \ do { (val)->freeswap = (val)->totalswap = 0; } while (0) +/* only sparc can not include linux/pagemap.h in this file + * so leave page_cache_release and release_pages undeclared... */ #define free_page_and_swap_cache(page) \ page_cache_release(page) #define free_pages_and_swap_cache(pages, nr) \ -- cgit v1.2.1 From 7a91bf7f5c22c8407a9991cbd9ce5bb87caa6b4a Mon Sep 17 00:00:00 2001 From: John McCutchan Date: Mon, 8 Aug 2005 13:52:16 -0400 Subject: [PATCH] fsnotify_name/inoderemove The patch below unhooks fsnotify from vfs_unlink & vfs_rmdir. It introduces two new fsnotify calls, that are hooked in at the dcache level. This not only more closely matches how the VFS layer works, it also avoids the problem with locking and inode lifetimes. The two functions are - fsnotify_nameremove -- called when a directory entry is going away. It notifies the PARENT of the deletion. This is called from d_delete(). - inoderemove -- called when the files inode itself is going away. It notifies the inode that is being deleted. This is called from dentry_iput(). Signed-off-by: John McCutchan Signed-off-by: Linus Torvalds --- include/linux/fsnotify.h | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) (limited to 'include/linux') diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index 1cb4935348d8..9db31d251c20 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -67,6 +67,26 @@ static inline void fsnotify_rmdir(struct dentry *dentry, struct inode *inode, inotify_inode_is_dead(inode); } +/* + * fsnotify_nameremove - a filename was removed from a directory + */ +static inline void fsnotify_nameremove(struct dentry *dentry, int isdir) +{ + if (isdir) + isdir = IN_ISDIR; + dnotify_parent(dentry, DN_DELETE); + inotify_dentry_parent_queue_event(dentry, IN_DELETE|isdir, 0, dentry->d_name.name); +} + +/* + * fsnotify_inoderemove - an inode is going away + */ +static inline void fsnotify_inoderemove(struct inode *inode) +{ + inotify_inode_queue_event(inode, IN_DELETE_SELF, 0, NULL); + inotify_inode_is_dead(inode); +} + /* * fsnotify_create - 'name' was linked in */ -- cgit v1.2.1 From 4d479e40e1748a877a24015fc6727b27b77110cd Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 8 Aug 2005 13:48:02 -0700 Subject: [NETLINK]: Allocate and kill some netlink numbers. NETLINK_ARPD is unused, allocate it to the Open-iSCSI folks. NETLINK_ROUTE6 and NETLINK_TAPBASE are no longer used, delete them. Signed-off-by: David S. Miller --- include/linux/netlink.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netlink.h b/include/linux/netlink.h index 70c2a9dc4b2b..6552b71bfa73 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -12,15 +12,13 @@ #define NETLINK_NFLOG 5 /* netfilter/iptables ULOG */ #define NETLINK_XFRM 6 /* ipsec */ #define NETLINK_SELINUX 7 /* SELinux event notifications */ -#define NETLINK_ARPD 8 +#define NETLINK_ISCSI 8 /* Open-iSCSI */ #define NETLINK_AUDIT 9 /* auditing */ #define NETLINK_FIB_LOOKUP 10 -#define NETLINK_ROUTE6 11 /* af_inet6 route comm channel */ #define NETLINK_NETFILTER 12 /* netfilter subsystem */ #define NETLINK_IP6_FW 13 #define NETLINK_DNRTMSG 14 /* DECnet routing messages */ #define NETLINK_KOBJECT_UEVENT 15 /* Kernel messages to userspace */ -#define NETLINK_TAPBASE 16 /* 16 to 31 are ethertap */ #define MAX_LINKS 32 -- cgit v1.2.1 From dc836b5b6fcde95f750a4790d8200fabaf563dc9 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 8 Aug 2005 18:46:09 -0700 Subject: Revert "[PATCH] PCI: restore BAR values..." Revert commit fec59a711eef002d4ef9eb8de09dd0a26986eb77, which is breaking sparc64 that doesn't have a working pci_update_resource. We'll re-do this after 2.6.13 when we'll do it all properly. --- include/linux/pci.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 98bdd95fcee9..8621cf42b46f 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -225,7 +225,6 @@ #define PCI_PM_CAP_PME_D3cold 0x8000 /* PME# from D3 (cold) */ #define PCI_PM_CTRL 4 /* PM control and status register */ #define PCI_PM_CTRL_STATE_MASK 0x0003 /* Current power state (D0 to D3) */ -#define PCI_PM_CTRL_NO_SOFT_RESET 0x0004 /* No reset for D3hot->D0 */ #define PCI_PM_CTRL_PME_ENABLE 0x0100 /* PME pin enable */ #define PCI_PM_CTRL_DATA_SEL_MASK 0x1e00 /* Data select (??) */ #define PCI_PM_CTRL_DATA_SCALE_MASK 0x6000 /* Data scale (??) */ @@ -817,9 +816,7 @@ int pci_set_mwi(struct pci_dev *dev); void pci_clear_mwi(struct pci_dev *dev); int pci_set_dma_mask(struct pci_dev *dev, u64 mask); int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask); -void pci_update_resource(struct pci_dev *dev, struct resource *res, int resno); int pci_assign_resource(struct pci_dev *dev, int i); -void pci_restore_bars(struct pci_dev *dev); /* ROM control related routines */ void __iomem *pci_map_rom(struct pci_dev *pdev, size_t *size); -- cgit v1.2.1 From 00dd1e433967872f3997a45d5adf35056fdf2f56 Mon Sep 17 00:00:00 2001 From: John McCutchan Date: Mon, 8 Aug 2005 22:13:05 -0400 Subject: [PATCH] fsnotify-cleanups This removes the now unused fsnotify_unlink & fsnotify_rmdir code. Compile tested. Signed-off-by: John McCutchan Signed-off-by: Linus Torvalds --- include/linux/fsnotify.h | 24 ------------------------ 1 file changed, 24 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index 9db31d251c20..602c305c8585 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -43,30 +43,6 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, } } -/* - * fsnotify_unlink - file was unlinked - */ -static inline void fsnotify_unlink(struct dentry *dentry, struct inode *inode, struct inode *dir) -{ - inode_dir_notify(dir, DN_DELETE); - inotify_inode_queue_event(dir, IN_DELETE, 0, dentry->d_name.name); - inotify_inode_queue_event(inode, IN_DELETE_SELF, 0, NULL); - - inotify_inode_is_dead(inode); -} - -/* - * fsnotify_rmdir - directory was removed - */ -static inline void fsnotify_rmdir(struct dentry *dentry, struct inode *inode, - struct inode *dir) -{ - inode_dir_notify(dir, DN_DELETE); - inotify_inode_queue_event(dir,IN_DELETE|IN_ISDIR,0,dentry->d_name.name); - inotify_inode_queue_event(inode, IN_DELETE_SELF | IN_ISDIR, 0, NULL); - inotify_inode_is_dead(inode); -} - /* * fsnotify_nameremove - a filename was removed from a directory */ -- cgit v1.2.1 From 86b3786078d63242d3194ffc58ae8dae1d1bbef3 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Tue, 9 Aug 2005 19:59:21 -0700 Subject: [PATCH] Fix ide-disk.c oops caused by hwif == NULL 1. Move hwif_to_node to ide.h 2. Use hwif_to_node in ide-disk.c Signed-off-by: Christoph Lameter Signed-off-by: Linus Torvalds --- include/linux/ide.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include/linux') diff --git a/include/linux/ide.h b/include/linux/ide.h index 92129078d4f3..a6dbb51ecd7b 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -1501,4 +1501,10 @@ extern struct bus_type ide_bus_type; #define ide_id_has_flush_cache_ext(id) \ (((id)->cfs_enable_2 & 0x2400) == 0x2400) +static inline int hwif_to_node(ide_hwif_t *hwif) +{ + struct pci_dev *dev = hwif->pci_dev; + return dev ? pcibus_to_node(dev->bus) : -1; +} + #endif /* _IDE_H */ -- cgit v1.2.1 From a0d3bea3cf6c7c1b53a46432bd490b5dc784ca42 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 11 Aug 2005 16:05:50 -0700 Subject: [NET]: Make skb->protocol __be16 There are many instances of skb->protocol = htons(ETH_P_*); skb->protocol = __constant_htons(ETH_P_*); and skb->protocol = *_type_trans(...); Most of *_type_trans() are already endian-annotated, so, let's shift attention on other warnings. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- include/linux/skbuff.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 0061c9470482..948527e42a60 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -255,7 +255,7 @@ struct sk_buff { nohdr:1; /* 3 bits spare */ __u8 pkt_type; - __u16 protocol; + __be16 protocol; void (*destructor)(struct sk_buff *skb); #ifdef CONFIG_NETFILTER -- cgit v1.2.1 From 0db1d6fc1ea051af49ebe03c503d23996a7c5bbb Mon Sep 17 00:00:00 2001 From: Matt Mackall Date: Thu, 11 Aug 2005 19:25:54 -0700 Subject: [NETPOLL]: add retry timeout Add limited retry logic to netpoll_send_skb Each time we attempt to send, decrement our per-device retry counter. On every successful send, we reset the counter. We delay 50us between attempts with up to 20000 retries for a total of 1 second. After we've exhausted our retries, subsequent failed attempts will try only once until reset by success. Signed-off-by: Matt Mackall Signed-off-by: David S. Miller --- include/linux/netpoll.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index bcd0ac33f592..be68d94b03d5 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -26,6 +26,7 @@ struct netpoll { struct netpoll_info { spinlock_t poll_lock; int poll_owner; + int tries; int rx_flags; spinlock_t rx_lock; struct netpoll *rx_np; /* netpoll that registered an rx_hook */ -- cgit v1.2.1 From 53fb95d3c14290fd6ee808b221e35493f096246f Mon Sep 17 00:00:00 2001 From: Matt Mackall Date: Thu, 11 Aug 2005 19:27:43 -0700 Subject: [NETPOLL]: fix initialization/NAPI race This fixes a race during initialization with the NAPI softirq processing by using an RCU approach. This race was discovered when refill_skbs() was added to the setup code. Signed-off-by: Matt Mackall Signed-off-by: David S. Miller --- include/linux/netpoll.h | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h index be68d94b03d5..5ade54a78dbb 100644 --- a/include/linux/netpoll.h +++ b/include/linux/netpoll.h @@ -9,6 +9,7 @@ #include #include +#include #include struct netpoll; @@ -61,25 +62,31 @@ static inline int netpoll_rx(struct sk_buff *skb) return ret; } -static inline void netpoll_poll_lock(struct net_device *dev) +static inline void *netpoll_poll_lock(struct net_device *dev) { + rcu_read_lock(); /* deal with race on ->npinfo */ if (dev->npinfo) { spin_lock(&dev->npinfo->poll_lock); dev->npinfo->poll_owner = smp_processor_id(); + return dev->npinfo; } + return NULL; } -static inline void netpoll_poll_unlock(struct net_device *dev) +static inline void netpoll_poll_unlock(void *have) { - if (dev->npinfo) { - dev->npinfo->poll_owner = -1; - spin_unlock(&dev->npinfo->poll_lock); + struct netpoll_info *npi = have; + + if (npi) { + npi->poll_owner = -1; + spin_unlock(&npi->poll_lock); } + rcu_read_unlock(); } #else #define netpoll_rx(a) 0 -#define netpoll_poll_lock(a) +#define netpoll_poll_lock(a) 0 #define netpoll_poll_unlock(a) #endif -- cgit v1.2.1 From 89204c40a03346cd951e698d854105db4cfedc28 Mon Sep 17 00:00:00 2001 From: John McCutchan Date: Mon, 15 Aug 2005 12:13:28 -0400 Subject: [PATCH] inotify: add MOVE_SELF event This adds a MOVE_SELF event to inotify. It is sent whenever the inode you are watching is moved. We need this event so that we can catch something like this: - app1: watch /etc/mtab - app2: cp /etc/mtab /tmp/mtab-work mv /etc/mtab /etc/mtab~ mv /tmp/mtab-work /etc/mtab app1 still thinks it's watching /etc/mtab but it's actually watching /etc/mtab~. Signed-off-by: John McCutchan Signed-off-by: Robert Love Signed-off-by: Linus Torvalds --- include/linux/fsnotify.h | 6 +++++- include/linux/inotify.h | 4 +++- 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index 602c305c8585..03b8e7932b83 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -21,7 +21,7 @@ */ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, const char *old_name, const char *new_name, - int isdir, struct inode *target) + int isdir, struct inode *target, struct inode *source) { u32 cookie = inotify_get_cookie(); @@ -41,6 +41,10 @@ static inline void fsnotify_move(struct inode *old_dir, struct inode *new_dir, inotify_inode_queue_event(target, IN_DELETE_SELF, 0, NULL); inotify_inode_is_dead(target); } + + if (source) { + inotify_inode_queue_event(source, IN_MOVE_SELF, 0, NULL); + } } /* diff --git a/include/linux/inotify.h b/include/linux/inotify.h index a40c2bf0408e..93bb3afe646b 100644 --- a/include/linux/inotify.h +++ b/include/linux/inotify.h @@ -35,6 +35,7 @@ struct inotify_event { #define IN_CREATE 0x00000100 /* Subfile was created */ #define IN_DELETE 0x00000200 /* Subfile was deleted */ #define IN_DELETE_SELF 0x00000400 /* Self was deleted */ +#define IN_MOVE_SELF 0x00000800 /* Self was moved */ /* the following are legal events. they are sent as needed to any watch */ #define IN_UNMOUNT 0x00002000 /* Backing fs was unmounted */ @@ -56,7 +57,8 @@ struct inotify_event { */ #define IN_ALL_EVENTS (IN_ACCESS | IN_MODIFY | IN_ATTRIB | IN_CLOSE_WRITE | \ IN_CLOSE_NOWRITE | IN_OPEN | IN_MOVED_FROM | \ - IN_MOVED_TO | IN_DELETE | IN_CREATE | IN_DELETE_SELF) + IN_MOVED_TO | IN_DELETE | IN_CREATE | IN_DELETE_SELF | \ + IN_MOVE_SELF) #ifdef __KERNEL__ -- cgit v1.2.1 From 58fcb8df0bf663bb6b8f46cd3010bfe8d13d97cf Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Wed, 10 Aug 2005 18:15:12 -0400 Subject: [PATCH] NFS: Ensure ACL xdr code doesn't overflow. Signed-off-by: Trond Myklebust Signed-off-by: Linus Torvalds --- include/linux/sunrpc/xdr.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/sunrpc/xdr.h b/include/linux/sunrpc/xdr.h index 34ec3e8d99b3..23448d0fb5bc 100644 --- a/include/linux/sunrpc/xdr.h +++ b/include/linux/sunrpc/xdr.h @@ -177,6 +177,7 @@ typedef int (*xdr_xcode_elem_t)(struct xdr_array2_desc *desc, void *elem); struct xdr_array2_desc { unsigned int elem_size; unsigned int array_len; + unsigned int array_maxlen; xdr_xcode_elem_t xcode; }; -- cgit v1.2.1 From 65e4308d2500e7daf60c3dccc202c61ffb066c63 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Tue, 16 Aug 2005 11:49:44 -0400 Subject: [PATCH] NFS: Ensure we always update inode->i_mode when doing O_EXCL creates When the client performs an exclusive create and opens the file for writing, a Netapp filer will first create the file using the mode 01777. It does this since an NFSv3/v4 exclusive create cannot immediately set the mode bits. The 01777 mode then gets put into the inode->i_mode. After the file creation is successful, we then do a setattr to change the mode to the correct value (as per the NFS spec). The problem is that nfs_refresh_inode() no longer updates inode->i_mode, so the latter retains the 01777 mode. A bit later, the VFS notices this, and calls remove_suid(). This of course now resets the file mode to inode->i_mode & 0777. Hey presto, the file mode on the server is now magically changed to 0777. Duh... Fixes http://bugzilla.linux-nfs.org/show_bug.cgi?id=32 Signed-off-by: Trond Myklebust Signed-off-by: Linus Torvalds --- include/linux/nfs_fs.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 8ea249110fb0..7d78a783c64a 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -292,6 +292,7 @@ extern int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode); extern int __nfs_revalidate_inode(struct nfs_server *, struct inode *); extern void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping); extern int nfs_setattr(struct dentry *, struct iattr *); +extern void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr); extern void nfs_begin_attr_update(struct inode *); extern void nfs_end_attr_update(struct inode *); extern void nfs_begin_data_update(struct inode *); -- cgit v1.2.1 From 4602b88d9743b5f20655de8078fb42e9fd25581f Mon Sep 17 00:00:00 2001 From: Kristen Accardi Date: Tue, 16 Aug 2005 15:15:58 -0700 Subject: [PATCH] PCI: 6700/6702PXH quirk On the 6700/6702 PXH part, a MSI may get corrupted if an ACPI hotplug driver and SHPC driver in MSI mode are used together. This patch will prevent MSI from being enabled for the SHPC as part of an early pci quirk, as well as on any pci device which sets the no_msi bit. Signed-off-by: Kristen Carlson Accardi Signed-off-by: Greg Kroah-Hartman Signed-off-by: Linus Torvalds --- include/linux/pci.h | 3 ++- include/linux/pci_ids.h | 5 +++++ 2 files changed, 7 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/pci.h b/include/linux/pci.h index 8621cf42b46f..bc4c40000c0d 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -556,7 +556,8 @@ struct pci_dev { /* keep track of device state */ unsigned int is_enabled:1; /* pci_enable_device has been called */ unsigned int is_busmaster:1; /* device is busmaster */ - + unsigned int no_msi:1; /* device may not use msi */ + u32 saved_config_space[16]; /* config space saved at suspend time */ struct bin_attribute *rom_attr; /* attribute descriptor for sysfs ROM entry */ int rom_attr_enabled; /* has display of the rom attribute been enabled? */ diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index bc4cc10fabe9..51e61e96051c 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2281,6 +2281,11 @@ #define PCI_VENDOR_ID_INTEL 0x8086 #define PCI_DEVICE_ID_INTEL_EESSC 0x0008 #define PCI_DEVICE_ID_INTEL_21145 0x0039 +#define PCI_DEVICE_ID_INTEL_PXHD_0 0x0320 +#define PCI_DEVICE_ID_INTEL_PXHD_1 0x0321 +#define PCI_DEVICE_ID_INTEL_PXH_0 0x0329 +#define PCI_DEVICE_ID_INTEL_PXH_1 0x032A +#define PCI_DEVICE_ID_INTEL_PXHV 0x032C #define PCI_DEVICE_ID_INTEL_82375 0x0482 #define PCI_DEVICE_ID_INTEL_82424 0x0483 #define PCI_DEVICE_ID_INTEL_82378 0x0484 -- cgit v1.2.1 From 5529680981807b44abf3be30fb6d612ff04f68ff Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 18 Aug 2005 11:24:09 -0700 Subject: [PATCH] NFS: split nfsi->flags into two fields Certain bits in nfsi->flags can be manipulated with atomic bitops, and some are better manipulated via logical bitmask operations. This patch splits the flags field into two. The next patch introduces atomic bitops for one of the fields. Test plan: Millions of fsx ops on SMP clients. Signed-off-by: Chuck Lever Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfs_fs.h | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 7d78a783c64a..229a1755842a 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -113,6 +113,7 @@ struct nfs_inode { * Various flags */ unsigned int flags; + unsigned long cache_validity; /* * read_cache_jiffies is when we started read-caching this inode, @@ -188,17 +189,21 @@ struct nfs_inode { }; /* - * Legal inode flag values + * Cache validity bit flags */ -#define NFS_INO_STALE 0x0001 /* possible stale inode */ -#define NFS_INO_ADVISE_RDPLUS 0x0002 /* advise readdirplus */ -#define NFS_INO_REVALIDATING 0x0004 /* revalidating attrs */ -#define NFS_INO_INVALID_ATTR 0x0008 /* cached attrs are invalid */ -#define NFS_INO_INVALID_DATA 0x0010 /* cached data is invalid */ -#define NFS_INO_INVALID_ATIME 0x0020 /* cached atime is invalid */ -#define NFS_INO_INVALID_ACCESS 0x0040 /* cached access cred invalid */ -#define NFS_INO_INVALID_ACL 0x0080 /* cached acls are invalid */ -#define NFS_INO_REVAL_PAGECACHE 0x1000 /* must revalidate pagecache */ +#define NFS_INO_INVALID_ATTR 0x0001 /* cached attrs are invalid */ +#define NFS_INO_INVALID_DATA 0x0002 /* cached data is invalid */ +#define NFS_INO_INVALID_ATIME 0x0004 /* cached atime is invalid */ +#define NFS_INO_INVALID_ACCESS 0x0008 /* cached access cred invalid */ +#define NFS_INO_INVALID_ACL 0x0010 /* cached acls are invalid */ +#define NFS_INO_REVAL_PAGECACHE 0x0020 /* must revalidate pagecache */ + +/* + * Legal values of flags field + */ +#define NFS_INO_REVALIDATING 0x0001 /* revalidating attrs */ +#define NFS_INO_ADVISE_RDPLUS 0x0002 /* advise readdirplus */ +#define NFS_INO_STALE 0x0004 /* possible stale inode */ static inline struct nfs_inode *NFS_I(struct inode *inode) { @@ -237,7 +242,7 @@ static inline int nfs_caches_unstable(struct inode *inode) static inline void NFS_CACHEINV(struct inode *inode) { if (!nfs_caches_unstable(inode)) - NFS_FLAGS(inode) |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; } static inline int nfs_server_capable(struct inode *inode, int cap) -- cgit v1.2.1 From 412d582ec1dd59aab2353f8cb7e74f2c79cd20b9 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 18 Aug 2005 11:24:11 -0700 Subject: [PATCH] NFS: use atomic bitops to manipulate flags in nfsi->flags Introduce atomic bitops to manipulate the bits in the nfs_inode structure's "flags" field. Using bitops means we can use a generic wait_on_bit call instead of an ad hoc locking scheme in fs/nfs/inode.c, so we can remove the "nfs_i_wait" field from nfs_inode at the same time. The other new flags field will continue to use bitmask and logic AND and OR. This permits several flags to be set at the same time efficiently. The following patch adds a spin lock to protect these flags, and this spin lock will later cover other fields in the nfs_inode structure, amortizing the cost of using this type of serialization. Test plan: Millions of fsx ops on SMP clients. Signed-off-by: Chuck Lever Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfs_fs.h | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index 229a1755842a..deef9567788a 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -112,8 +112,8 @@ struct nfs_inode { /* * Various flags */ - unsigned int flags; - unsigned long cache_validity; + unsigned long flags; /* atomic bit ops */ + unsigned long cache_validity; /* bit mask */ /* * read_cache_jiffies is when we started read-caching this inode, @@ -175,8 +175,6 @@ struct nfs_inode { /* Open contexts for shared mmap writes */ struct list_head open_files; - wait_queue_head_t nfs_i_wait; - #ifdef CONFIG_NFS_V4 struct nfs4_cached_acl *nfs4_acl; /* NFSv4 state */ @@ -199,11 +197,11 @@ struct nfs_inode { #define NFS_INO_REVAL_PAGECACHE 0x0020 /* must revalidate pagecache */ /* - * Legal values of flags field + * Bit offsets in flags field */ -#define NFS_INO_REVALIDATING 0x0001 /* revalidating attrs */ -#define NFS_INO_ADVISE_RDPLUS 0x0002 /* advise readdirplus */ -#define NFS_INO_STALE 0x0004 /* possible stale inode */ +#define NFS_INO_REVALIDATING (0) /* revalidating attrs */ +#define NFS_INO_ADVISE_RDPLUS (1) /* advise readdirplus */ +#define NFS_INO_STALE (2) /* possible stale inode */ static inline struct nfs_inode *NFS_I(struct inode *inode) { @@ -229,8 +227,7 @@ static inline struct nfs_inode *NFS_I(struct inode *inode) #define NFS_ATTRTIMEO_UPDATE(inode) (NFS_I(inode)->attrtimeo_timestamp) #define NFS_FLAGS(inode) (NFS_I(inode)->flags) -#define NFS_REVALIDATING(inode) (NFS_FLAGS(inode) & NFS_INO_REVALIDATING) -#define NFS_STALE(inode) (NFS_FLAGS(inode) & NFS_INO_STALE) +#define NFS_STALE(inode) (test_bit(NFS_INO_STALE, &NFS_FLAGS(inode))) #define NFS_FILEID(inode) (NFS_I(inode)->fileid) @@ -252,7 +249,7 @@ static inline int nfs_server_capable(struct inode *inode, int cap) static inline int NFS_USE_READDIRPLUS(struct inode *inode) { - return NFS_FLAGS(inode) & NFS_INO_ADVISE_RDPLUS; + return test_bit(NFS_INO_ADVISE_RDPLUS, &NFS_FLAGS(inode)); } /** -- cgit v1.2.1 From dc59250c6ebed099a9bc0a11298e2281dd896657 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Thu, 18 Aug 2005 11:24:12 -0700 Subject: [PATCH] NFS: Introduce the use of inode->i_lock to protect fields in nfsi Down the road we want to eliminate the use of the global kernel lock entirely from the NFS client. To do this, we need to protect the fields in the nfs_inode structure adequately. Start by serializing updates to the "cache_validity" field. Note this change addresses an SMP hang found by njw@osdl.org, where processes deadlock because nfs_end_data_update and nfs_revalidate_mapping update the "cache_validity" field without proper serialization. Test plan: Millions of fsx ops on SMP clients. Run Nick Wilson's breaknfs program on large SMP clients. Signed-off-by: Chuck Lever Cc: Trond Myklebust Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/nfs_fs.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index deef9567788a..9a6047ff1b25 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -238,8 +238,11 @@ static inline int nfs_caches_unstable(struct inode *inode) static inline void NFS_CACHEINV(struct inode *inode) { - if (!nfs_caches_unstable(inode)) + if (!nfs_caches_unstable(inode)) { + spin_lock(&inode->i_lock); NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; + spin_unlock(&inode->i_lock); + } } static inline int nfs_server_capable(struct inode *inode, int cap) -- cgit v1.2.1 From b07e5eccaf512ae3209beae5cd2e3a27c92c300b Mon Sep 17 00:00:00 2001 From: Grant Coady Date: Thu, 18 Aug 2005 22:19:55 +0200 Subject: [PATCH] ide: fix PCI_DEVIEC_ID_APPLE_UNI_N_ATA spelling Signed-off-by: Grant Coady Signed-off-by: Bartlomiej Zolnierkiewicz --- include/linux/pci_ids.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 51e61e96051c..953b4dc819ae 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -881,7 +881,7 @@ #define PCI_DEVICE_ID_APPLE_UNI_N_PCI15 0x002e #define PCI_DEVICE_ID_APPLE_UNI_N_FW2 0x0030 #define PCI_DEVICE_ID_APPLE_UNI_N_GMAC2 0x0032 -#define PCI_DEVIEC_ID_APPLE_UNI_N_ATA 0x0033 +#define PCI_DEVICE_ID_APPLE_UNI_N_ATA 0x0033 #define PCI_DEVICE_ID_APPLE_UNI_N_AGP2 0x0034 #define PCI_DEVICE_ID_APPLE_IPID_ATA100 0x003b #define PCI_DEVICE_ID_APPLE_KEYLARGO_I 0x003e -- cgit v1.2.1 From 2f09a7f4af131bf23c013ead89373deba1c7593c Mon Sep 17 00:00:00 2001 From: Matt Gillette Date: Thu, 18 Aug 2005 22:27:07 +0200 Subject: [PATCH] ide: add support for Netcell Revolution to pci-ide generic driver Adds support for Netcell Revolution to pci-ide generic driver by including it in the list of devices matched. Includes the Revolution in the list of simplex devices forced into DMA mode. Signed-off-by: Matt Gillette Cc: Bartlomiej Zolnierkiewicz Cc: Jeff Garzik Signed-off-by: Andrew Morton Signed-off-by: Bartlomiej Zolnierkiewicz --- include/linux/pci_ids.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 953b4dc819ae..163036867ac7 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2184,6 +2184,9 @@ #define PCI_VENDOR_ID_SIBYTE 0x166d #define PCI_DEVICE_ID_BCM1250_HT 0x0002 +#define PCI_VENDOR_ID_NETCELL 0x169c +#define PCI_DEVICE_ID_REVOLUTION 0x0044 + #define PCI_VENDOR_ID_LINKSYS 0x1737 #define PCI_DEVICE_ID_LINKSYS_EG1032 0x1032 #define PCI_DEVICE_ID_LINKSYS_EG1064 0x1064 -- cgit v1.2.1 From 84f57fbc724e3b56dc87c37dddac89f82cf75ef6 Mon Sep 17 00:00:00 2001 From: Narendra Sankar Date: Thu, 18 Aug 2005 22:30:35 +0200 Subject: [PATCH] serverworks: add support for new southbridge IDE BCM5785 (HT1000) is a Opteron Southbridge from Serverworks/Broadcom that incorporates a single channel ATA100 IDE controller that is functionally identical to the Serverworks CSB6 IDE controller. This patch adds support for the new PCI device ID and also the support for this controller. Signed-off-by: Narendra Sankar Acked-by: Jeff Garzik Signed-off-by: Bartlomiej Zolnierkiewicz --- include/linux/pci_ids.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 163036867ac7..927ed487630d 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1580,6 +1580,7 @@ #define PCI_DEVICE_ID_SERVERWORKS_OSB4IDE 0x0211 #define PCI_DEVICE_ID_SERVERWORKS_CSB5IDE 0x0212 #define PCI_DEVICE_ID_SERVERWORKS_CSB6IDE 0x0213 +#define PCI_DEVICE_ID_SERVERWORKS_HT1000IDE 0x0214 #define PCI_DEVICE_ID_SERVERWORKS_CSB6IDE2 0x0217 #define PCI_DEVICE_ID_SERVERWORKS_OSB4USB 0x0220 #define PCI_DEVICE_ID_SERVERWORKS_CSB5USB PCI_DEVICE_ID_SERVERWORKS_OSB4USB -- cgit v1.2.1 From cc314eef0128a807e50fa03baf2d0abc0647952c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Fri, 19 Aug 2005 18:02:56 -0700 Subject: Fix nasty ncpfs symlink handling bug. This bug could cause oopses and page state corruption, because ncpfs used the generic page-cache symlink handlign functions. But those functions only work if the page cache is guaranteed to be "stable", ie a page that was installed when the symlink walk was started has to still be installed in the page cache at the end of the walk. We could have fixed ncpfs to not use the generic helper routines, but it is in many ways much cleaner to instead improve on the symlink walking helper routines so that they don't require that absolute stability. We do this by allowing "follow_link()" to return a error-pointer as a cookie, which is fed back to the cleanup "put_link()" routine. This also simplifies NFS symlink handling. Signed-off-by: Linus Torvalds --- include/linux/fs.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/fs.h b/include/linux/fs.h index f9adf75fd9b4..67e6732d4fdc 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -993,8 +993,8 @@ struct inode_operations { int (*rename) (struct inode *, struct dentry *, struct inode *, struct dentry *); int (*readlink) (struct dentry *, char __user *,int); - int (*follow_link) (struct dentry *, struct nameidata *); - void (*put_link) (struct dentry *, struct nameidata *); + void * (*follow_link) (struct dentry *, struct nameidata *); + void (*put_link) (struct dentry *, struct nameidata *, void *); void (*truncate) (struct inode *); int (*permission) (struct inode *, int, struct nameidata *); int (*setattr) (struct dentry *, struct iattr *); @@ -1602,8 +1602,8 @@ extern struct file_operations generic_ro_fops; extern int vfs_readlink(struct dentry *, char __user *, int, const char *); extern int vfs_follow_link(struct nameidata *, const char *); extern int page_readlink(struct dentry *, char __user *, int); -extern int page_follow_link_light(struct dentry *, struct nameidata *); -extern void page_put_link(struct dentry *, struct nameidata *); +extern void *page_follow_link_light(struct dentry *, struct nameidata *); +extern void page_put_link(struct dentry *, struct nameidata *, void *); extern int page_symlink(struct inode *inode, const char *symname, int len); extern struct inode_operations page_symlink_inode_operations; extern int generic_readlink(struct dentry *, char __user *, int); -- cgit v1.2.1