summaryrefslogtreecommitdiff
path: root/arch/ppc64
diff options
context:
space:
mode:
Diffstat (limited to 'arch/ppc64')
-rw-r--r--arch/ppc64/Kconfig13
-rw-r--r--arch/ppc64/boot/main.c11
-rw-r--r--arch/ppc64/kernel/asm-offsets.c3
-rw-r--r--arch/ppc64/kernel/head.S300
-rw-r--r--arch/ppc64/kernel/kprobes.c138
-rw-r--r--arch/ppc64/kernel/lparcfg.c4
-rw-r--r--arch/ppc64/kernel/pacaData.c2
-rw-r--r--arch/ppc64/kernel/prom.c94
-rw-r--r--arch/ppc64/kernel/scanlog.c3
-rw-r--r--arch/ppc64/kernel/sysfs.c1
10 files changed, 358 insertions, 211 deletions
diff --git a/arch/ppc64/Kconfig b/arch/ppc64/Kconfig
index b987164fca4c..2130cc315957 100644
--- a/arch/ppc64/Kconfig
+++ b/arch/ppc64/Kconfig
@@ -47,6 +47,10 @@ config ARCH_MAY_HAVE_PC_FDC
bool
default y
+config PPC_STD_MMU
+ bool
+ default y
+
# We optimistically allocate largepages from the VM, so make the limit
# large enough (16MB). This badly named config option is actually
# max order + 1
@@ -294,6 +298,15 @@ config NODES_SPAN_OTHER_NODES
def_bool y
depends on NEED_MULTIPLE_NODES
+config PPC_64K_PAGES
+ bool "64k page size"
+ help
+ This option changes the kernel logical page size to 64k. On machines
+ without processor support for 64k pages, the kernel will simulate
+ them by loading each individual 4k page on demand transparently,
+ while on hardware with such support, it will be used to map
+ normal application pages.
+
config SCHED_SMT
bool "SMT (Hyperthreading) scheduler support"
depends on SMP
diff --git a/arch/ppc64/boot/main.c b/arch/ppc64/boot/main.c
index c1dc876bccab..e0dde24a72ce 100644
--- a/arch/ppc64/boot/main.c
+++ b/arch/ppc64/boot/main.c
@@ -203,8 +203,15 @@ void start(unsigned long a1, unsigned long a2, void *promptr, void *sp)
if (elf64ph->p_type == PT_LOAD && elf64ph->p_offset != 0)
break;
}
- vmlinux.size = (unsigned long)elf64ph->p_filesz;
- vmlinux.memsize = (unsigned long)elf64ph->p_memsz;
+ vmlinux.size = (unsigned long)elf64ph->p_filesz +
+ (unsigned long)elf64ph->p_offset;
+ /* We need to claim the memsize plus the file offset since gzip
+ * will expand the header (file offset), then the kernel, then
+ * possible rubbish we don't care about. But the kernel bss must
+ * be claimed (it will be zero'd by the kernel itself)
+ */
+ vmlinux.memsize = (unsigned long)elf64ph->p_memsz +
+ (unsigned long)elf64ph->p_offset;
printf("Allocating 0x%lx bytes for kernel ...\n\r", vmlinux.memsize);
vmlinux.addr = try_claim(vmlinux.memsize);
if (vmlinux.addr == 0) {
diff --git a/arch/ppc64/kernel/asm-offsets.c b/arch/ppc64/kernel/asm-offsets.c
index 504dee836d29..bce9065da6cb 100644
--- a/arch/ppc64/kernel/asm-offsets.c
+++ b/arch/ppc64/kernel/asm-offsets.c
@@ -93,6 +93,9 @@ int main(void)
DEFINE(PACASLBCACHE, offsetof(struct paca_struct, slb_cache));
DEFINE(PACASLBCACHEPTR, offsetof(struct paca_struct, slb_cache_ptr));
DEFINE(PACACONTEXTID, offsetof(struct paca_struct, context.id));
+#ifdef CONFIG_PPC_64K_PAGES
+ DEFINE(PACAPGDIR, offsetof(struct paca_struct, pgdir));
+#endif
#ifdef CONFIG_HUGETLB_PAGE
DEFINE(PACALOWHTLBAREAS, offsetof(struct paca_struct, context.low_htlb_areas));
DEFINE(PACAHIGHHTLBAREAS, offsetof(struct paca_struct, context.high_htlb_areas));
diff --git a/arch/ppc64/kernel/head.S b/arch/ppc64/kernel/head.S
index db1cf397be2d..9e8050ea1225 100644
--- a/arch/ppc64/kernel/head.S
+++ b/arch/ppc64/kernel/head.S
@@ -195,11 +195,11 @@ exception_marker:
#define EX_R12 24
#define EX_R13 32
#define EX_SRR0 40
-#define EX_R3 40 /* SLB miss saves R3, but not SRR0 */
#define EX_DAR 48
-#define EX_LR 48 /* SLB miss saves LR, but not DAR */
#define EX_DSISR 56
#define EX_CCR 60
+#define EX_R3 64
+#define EX_LR 72
#define EXCEPTION_PROLOG_PSERIES(area, label) \
mfspr r13,SPRN_SPRG3; /* get paca address into r13 */ \
@@ -419,17 +419,22 @@ data_access_slb_pSeries:
mtspr SPRN_SPRG1,r13
RUNLATCH_ON(r13)
mfspr r13,SPRN_SPRG3 /* get paca address into r13 */
+ std r3,PACA_EXSLB+EX_R3(r13)
+ mfspr r3,SPRN_DAR
std r9,PACA_EXSLB+EX_R9(r13) /* save r9 - r12 */
+ mfcr r9
+#ifdef __DISABLED__
+ /* Keep that around for when we re-implement dynamic VSIDs */
+ cmpdi r3,0
+ bge slb_miss_user_pseries
+#endif /* __DISABLED__ */
std r10,PACA_EXSLB+EX_R10(r13)
std r11,PACA_EXSLB+EX_R11(r13)
std r12,PACA_EXSLB+EX_R12(r13)
- std r3,PACA_EXSLB+EX_R3(r13)
- mfspr r9,SPRN_SPRG1
- std r9,PACA_EXSLB+EX_R13(r13)
- mfcr r9
+ mfspr r10,SPRN_SPRG1
+ std r10,PACA_EXSLB+EX_R13(r13)
mfspr r12,SPRN_SRR1 /* and SRR1 */
- mfspr r3,SPRN_DAR
- b .do_slb_miss /* Rel. branch works in real mode */
+ b .slb_miss_realmode /* Rel. branch works in real mode */
STD_EXCEPTION_PSERIES(0x400, instruction_access)
@@ -440,17 +445,22 @@ instruction_access_slb_pSeries:
mtspr SPRN_SPRG1,r13
RUNLATCH_ON(r13)
mfspr r13,SPRN_SPRG3 /* get paca address into r13 */
+ std r3,PACA_EXSLB+EX_R3(r13)
+ mfspr r3,SPRN_SRR0 /* SRR0 is faulting address */
std r9,PACA_EXSLB+EX_R9(r13) /* save r9 - r12 */
+ mfcr r9
+#ifdef __DISABLED__
+ /* Keep that around for when we re-implement dynamic VSIDs */
+ cmpdi r3,0
+ bge slb_miss_user_pseries
+#endif /* __DISABLED__ */
std r10,PACA_EXSLB+EX_R10(r13)
std r11,PACA_EXSLB+EX_R11(r13)
std r12,PACA_EXSLB+EX_R12(r13)
- std r3,PACA_EXSLB+EX_R3(r13)
- mfspr r9,SPRN_SPRG1
- std r9,PACA_EXSLB+EX_R13(r13)
- mfcr r9
+ mfspr r10,SPRN_SPRG1
+ std r10,PACA_EXSLB+EX_R13(r13)
mfspr r12,SPRN_SRR1 /* and SRR1 */
- mfspr r3,SPRN_SRR0 /* SRR0 is faulting address */
- b .do_slb_miss /* Rel. branch works in real mode */
+ b .slb_miss_realmode /* Rel. branch works in real mode */
STD_EXCEPTION_PSERIES(0x500, hardware_interrupt)
STD_EXCEPTION_PSERIES(0x600, alignment)
@@ -509,6 +519,38 @@ _GLOBAL(do_stab_bolted_pSeries)
EXCEPTION_PROLOG_PSERIES(PACA_EXSLB, .do_stab_bolted)
/*
+ * We have some room here we use that to put
+ * the peries slb miss user trampoline code so it's reasonably
+ * away from slb_miss_user_common to avoid problems with rfid
+ *
+ * This is used for when the SLB miss handler has to go virtual,
+ * which doesn't happen for now anymore but will once we re-implement
+ * dynamic VSIDs for shared page tables
+ */
+#ifdef __DISABLED__
+slb_miss_user_pseries:
+ std r10,PACA_EXGEN+EX_R10(r13)
+ std r11,PACA_EXGEN+EX_R11(r13)
+ std r12,PACA_EXGEN+EX_R12(r13)
+ mfspr r10,SPRG1
+ ld r11,PACA_EXSLB+EX_R9(r13)
+ ld r12,PACA_EXSLB+EX_R3(r13)
+ std r10,PACA_EXGEN+EX_R13(r13)
+ std r11,PACA_EXGEN+EX_R9(r13)
+ std r12,PACA_EXGEN+EX_R3(r13)
+ clrrdi r12,r13,32
+ mfmsr r10
+ mfspr r11,SRR0 /* save SRR0 */
+ ori r12,r12,slb_miss_user_common@l /* virt addr of handler */
+ ori r10,r10,MSR_IR|MSR_DR|MSR_RI
+ mtspr SRR0,r12
+ mfspr r12,SRR1 /* and SRR1 */
+ mtspr SRR1,r10
+ rfid
+ b . /* prevent spec. execution */
+#endif /* __DISABLED__ */
+
+/*
* Vectors for the FWNMI option. Share common code.
*/
.globl system_reset_fwnmi
@@ -559,22 +601,59 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SLB)
.globl data_access_slb_iSeries
data_access_slb_iSeries:
mtspr SPRN_SPRG1,r13 /* save r13 */
- EXCEPTION_PROLOG_ISERIES_1(PACA_EXSLB)
+ mfspr r13,SPRN_SPRG3 /* get paca address into r13 */
std r3,PACA_EXSLB+EX_R3(r13)
- ld r12,PACALPPACA+LPPACASRR1(r13)
mfspr r3,SPRN_DAR
- b .do_slb_miss
+ std r9,PACA_EXSLB+EX_R9(r13)
+ mfcr r9
+#ifdef __DISABLED__
+ cmpdi r3,0
+ bge slb_miss_user_iseries
+#endif
+ std r10,PACA_EXSLB+EX_R10(r13)
+ std r11,PACA_EXSLB+EX_R11(r13)
+ std r12,PACA_EXSLB+EX_R12(r13)
+ mfspr r10,SPRN_SPRG1
+ std r10,PACA_EXSLB+EX_R13(r13)
+ ld r12,PACALPPACA+LPPACASRR1(r13);
+ b .slb_miss_realmode
STD_EXCEPTION_ISERIES(0x400, instruction_access, PACA_EXGEN)
.globl instruction_access_slb_iSeries
instruction_access_slb_iSeries:
mtspr SPRN_SPRG1,r13 /* save r13 */
- EXCEPTION_PROLOG_ISERIES_1(PACA_EXSLB)
+ mfspr r13,SPRN_SPRG3 /* get paca address into r13 */
std r3,PACA_EXSLB+EX_R3(r13)
- ld r12,PACALPPACA+LPPACASRR1(r13)
- ld r3,PACALPPACA+LPPACASRR0(r13)
- b .do_slb_miss
+ ld r3,PACALPPACA+LPPACASRR0(r13) /* get SRR0 value */
+ std r9,PACA_EXSLB+EX_R9(r13)
+ mfcr r9
+#ifdef __DISABLED__
+ cmpdi r3,0
+ bge .slb_miss_user_iseries
+#endif
+ std r10,PACA_EXSLB+EX_R10(r13)
+ std r11,PACA_EXSLB+EX_R11(r13)
+ std r12,PACA_EXSLB+EX_R12(r13)
+ mfspr r10,SPRN_SPRG1
+ std r10,PACA_EXSLB+EX_R13(r13)
+ ld r12,PACALPPACA+LPPACASRR1(r13);
+ b .slb_miss_realmode
+
+#ifdef __DISABLED__
+slb_miss_user_iseries:
+ std r10,PACA_EXGEN+EX_R10(r13)
+ std r11,PACA_EXGEN+EX_R11(r13)
+ std r12,PACA_EXGEN+EX_R12(r13)
+ mfspr r10,SPRG1
+ ld r11,PACA_EXSLB+EX_R9(r13)
+ ld r12,PACA_EXSLB+EX_R3(r13)
+ std r10,PACA_EXGEN+EX_R13(r13)
+ std r11,PACA_EXGEN+EX_R9(r13)
+ std r12,PACA_EXGEN+EX_R3(r13)
+ EXCEPTION_PROLOG_ISERIES_2
+ b slb_miss_user_common
+#endif
MASKABLE_EXCEPTION_ISERIES(0x500, hardware_interrupt)
STD_EXCEPTION_ISERIES(0x600, alignment, PACA_EXGEN)
@@ -809,6 +888,126 @@ instruction_access_common:
li r5,0x400
b .do_hash_page /* Try to handle as hpte fault */
+/*
+ * Here is the common SLB miss user that is used when going to virtual
+ * mode for SLB misses, that is currently not used
+ */
+#ifdef __DISABLED__
+ .align 7
+ .globl slb_miss_user_common
+slb_miss_user_common:
+ mflr r10
+ std r3,PACA_EXGEN+EX_DAR(r13)
+ stw r9,PACA_EXGEN+EX_CCR(r13)
+ std r10,PACA_EXGEN+EX_LR(r13)
+ std r11,PACA_EXGEN+EX_SRR0(r13)
+ bl .slb_allocate_user
+
+ ld r10,PACA_EXGEN+EX_LR(r13)
+ ld r3,PACA_EXGEN+EX_R3(r13)
+ lwz r9,PACA_EXGEN+EX_CCR(r13)
+ ld r11,PACA_EXGEN+EX_SRR0(r13)
+ mtlr r10
+ beq- slb_miss_fault
+
+ andi. r10,r12,MSR_RI /* check for unrecoverable exception */
+ beq- unrecov_user_slb
+ mfmsr r10
+
+.machine push
+.machine "power4"
+ mtcrf 0x80,r9
+.machine pop
+
+ clrrdi r10,r10,2 /* clear RI before setting SRR0/1 */
+ mtmsrd r10,1
+
+ mtspr SRR0,r11
+ mtspr SRR1,r12
+
+ ld r9,PACA_EXGEN+EX_R9(r13)
+ ld r10,PACA_EXGEN+EX_R10(r13)
+ ld r11,PACA_EXGEN+EX_R11(r13)
+ ld r12,PACA_EXGEN+EX_R12(r13)
+ ld r13,PACA_EXGEN+EX_R13(r13)
+ rfid
+ b .
+
+slb_miss_fault:
+ EXCEPTION_PROLOG_COMMON(0x380, PACA_EXGEN)
+ ld r4,PACA_EXGEN+EX_DAR(r13)
+ li r5,0
+ std r4,_DAR(r1)
+ std r5,_DSISR(r1)
+ b .handle_page_fault
+
+unrecov_user_slb:
+ EXCEPTION_PROLOG_COMMON(0x4200, PACA_EXGEN)
+ DISABLE_INTS
+ bl .save_nvgprs
+1: addi r3,r1,STACK_FRAME_OVERHEAD
+ bl .unrecoverable_exception
+ b 1b
+
+#endif /* __DISABLED__ */
+
+
+/*
+ * r13 points to the PACA, r9 contains the saved CR,
+ * r12 contain the saved SRR1, SRR0 is still ready for return
+ * r3 has the faulting address
+ * r9 - r13 are saved in paca->exslb.
+ * r3 is saved in paca->slb_r3
+ * We assume we aren't going to take any exceptions during this procedure.
+ */
+_GLOBAL(slb_miss_realmode)
+ mflr r10
+
+ stw r9,PACA_EXSLB+EX_CCR(r13) /* save CR in exc. frame */
+ std r10,PACA_EXSLB+EX_LR(r13) /* save LR */
+
+ bl .slb_allocate_realmode
+
+ /* All done -- return from exception. */
+
+ ld r10,PACA_EXSLB+EX_LR(r13)
+ ld r3,PACA_EXSLB+EX_R3(r13)
+ lwz r9,PACA_EXSLB+EX_CCR(r13) /* get saved CR */
+#ifdef CONFIG_PPC_ISERIES
+ ld r11,PACALPPACA+LPPACASRR0(r13) /* get SRR0 value */
+#endif /* CONFIG_PPC_ISERIES */
+
+ mtlr r10
+
+ andi. r10,r12,MSR_RI /* check for unrecoverable exception */
+ beq- unrecov_slb
+
+.machine push
+.machine "power4"
+ mtcrf 0x80,r9
+ mtcrf 0x01,r9 /* slb_allocate uses cr0 and cr7 */
+.machine pop
+
+#ifdef CONFIG_PPC_ISERIES
+ mtspr SPRN_SRR0,r11
+ mtspr SPRN_SRR1,r12
+#endif /* CONFIG_PPC_ISERIES */
+ ld r9,PACA_EXSLB+EX_R9(r13)
+ ld r10,PACA_EXSLB+EX_R10(r13)
+ ld r11,PACA_EXSLB+EX_R11(r13)
+ ld r12,PACA_EXSLB+EX_R12(r13)
+ ld r13,PACA_EXSLB+EX_R13(r13)
+ rfid
+ b . /* prevent speculative execution */
+
+unrecov_slb:
+ EXCEPTION_PROLOG_COMMON(0x4100, PACA_EXSLB)
+ DISABLE_INTS
+ bl .save_nvgprs
+1: addi r3,r1,STACK_FRAME_OVERHEAD
+ bl .unrecoverable_exception
+ b 1b
+
.align 7
.globl hardware_interrupt_common
.globl hardware_interrupt_entry
@@ -1139,62 +1338,6 @@ _GLOBAL(do_stab_bolted)
b . /* prevent speculative execution */
/*
- * r13 points to the PACA, r9 contains the saved CR,
- * r11 and r12 contain the saved SRR0 and SRR1.
- * r3 has the faulting address
- * r9 - r13 are saved in paca->exslb.
- * r3 is saved in paca->slb_r3
- * We assume we aren't going to take any exceptions during this procedure.
- */
-_GLOBAL(do_slb_miss)
- mflr r10
-
- stw r9,PACA_EXSLB+EX_CCR(r13) /* save CR in exc. frame */
- std r10,PACA_EXSLB+EX_LR(r13) /* save LR */
-
- bl .slb_allocate /* handle it */
-
- /* All done -- return from exception. */
-
- ld r10,PACA_EXSLB+EX_LR(r13)
- ld r3,PACA_EXSLB+EX_R3(r13)
- lwz r9,PACA_EXSLB+EX_CCR(r13) /* get saved CR */
-#ifdef CONFIG_PPC_ISERIES
- ld r11,PACALPPACA+LPPACASRR0(r13) /* get SRR0 value */
-#endif /* CONFIG_PPC_ISERIES */
-
- mtlr r10
-
- andi. r10,r12,MSR_RI /* check for unrecoverable exception */
- beq- unrecov_slb
-
-.machine push
-.machine "power4"
- mtcrf 0x80,r9
- mtcrf 0x01,r9 /* slb_allocate uses cr0 and cr7 */
-.machine pop
-
-#ifdef CONFIG_PPC_ISERIES
- mtspr SPRN_SRR0,r11
- mtspr SPRN_SRR1,r12
-#endif /* CONFIG_PPC_ISERIES */
- ld r9,PACA_EXSLB+EX_R9(r13)
- ld r10,PACA_EXSLB+EX_R10(r13)
- ld r11,PACA_EXSLB+EX_R11(r13)
- ld r12,PACA_EXSLB+EX_R12(r13)
- ld r13,PACA_EXSLB+EX_R13(r13)
- rfid
- b . /* prevent speculative execution */
-
-unrecov_slb:
- EXCEPTION_PROLOG_COMMON(0x4100, PACA_EXSLB)
- DISABLE_INTS
- bl .save_nvgprs
-1: addi r3,r1,STACK_FRAME_OVERHEAD
- bl .unrecoverable_exception
- b 1b
-
-/*
* Space for CPU0's segment table.
*
* On iSeries, the hypervisor must fill in at least one entry before
@@ -1569,7 +1712,10 @@ _GLOBAL(__secondary_start)
#endif
/* Initialize the first segment table (or SLB) entry */
ld r3,PACASTABVIRT(r13) /* get addr of segment table */
+BEGIN_FTR_SECTION
bl .stab_initialize
+END_FTR_SECTION_IFCLR(CPU_FTR_SLB)
+ bl .slb_initialize
/* Initialize the kernel stack. Just a repeat for iSeries. */
LOADADDR(r3,current_set)
diff --git a/arch/ppc64/kernel/kprobes.c b/arch/ppc64/kernel/kprobes.c
index ed876a5178ae..511af54e6230 100644
--- a/arch/ppc64/kernel/kprobes.c
+++ b/arch/ppc64/kernel/kprobes.c
@@ -30,19 +30,14 @@
#include <linux/config.h>
#include <linux/kprobes.h>
#include <linux/ptrace.h>
-#include <linux/spinlock.h>
#include <linux/preempt.h>
#include <asm/cacheflush.h>
#include <asm/kdebug.h>
#include <asm/sstep.h>
static DECLARE_MUTEX(kprobe_mutex);
-
-static struct kprobe *current_kprobe;
-static unsigned long kprobe_status, kprobe_saved_msr;
-static struct kprobe *kprobe_prev;
-static unsigned long kprobe_status_prev, kprobe_saved_msr_prev;
-static struct pt_regs jprobe_saved_regs;
+DEFINE_PER_CPU(struct kprobe *, current_kprobe) = NULL;
+DEFINE_PER_CPU(struct kprobe_ctlblk, kprobe_ctlblk);
int __kprobes arch_prepare_kprobe(struct kprobe *p)
{
@@ -108,20 +103,28 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
regs->nip = (unsigned long)p->ainsn.insn;
}
-static inline void save_previous_kprobe(void)
+static inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
+{
+ kcb->prev_kprobe.kp = kprobe_running();
+ kcb->prev_kprobe.status = kcb->kprobe_status;
+ kcb->prev_kprobe.saved_msr = kcb->kprobe_saved_msr;
+}
+
+static inline void restore_previous_kprobe(struct kprobe_ctlblk *kcb)
{
- kprobe_prev = current_kprobe;
- kprobe_status_prev = kprobe_status;
- kprobe_saved_msr_prev = kprobe_saved_msr;
+ __get_cpu_var(current_kprobe) = kcb->prev_kprobe.kp;
+ kcb->kprobe_status = kcb->prev_kprobe.status;
+ kcb->kprobe_saved_msr = kcb->prev_kprobe.saved_msr;
}
-static inline void restore_previous_kprobe(void)
+static inline void set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
+ struct kprobe_ctlblk *kcb)
{
- current_kprobe = kprobe_prev;
- kprobe_status = kprobe_status_prev;
- kprobe_saved_msr = kprobe_saved_msr_prev;
+ __get_cpu_var(current_kprobe) = p;
+ kcb->kprobe_saved_msr = regs->msr;
}
+/* Called with kretprobe_lock held */
void __kprobes arch_prepare_kretprobe(struct kretprobe *rp,
struct pt_regs *regs)
{
@@ -145,19 +148,24 @@ static inline int kprobe_handler(struct pt_regs *regs)
struct kprobe *p;
int ret = 0;
unsigned int *addr = (unsigned int *)regs->nip;
+ struct kprobe_ctlblk *kcb;
+
+ /*
+ * We don't want to be preempted for the entire
+ * duration of kprobe processing
+ */
+ preempt_disable();
+ kcb = get_kprobe_ctlblk();
/* Check we're not actually recursing */
if (kprobe_running()) {
- /* We *are* holding lock here, so this is safe.
- Disarm the probe we just hit, and ignore it. */
p = get_kprobe(addr);
if (p) {
kprobe_opcode_t insn = *p->ainsn.insn;
- if (kprobe_status == KPROBE_HIT_SS &&
+ if (kcb->kprobe_status == KPROBE_HIT_SS &&
is_trap(insn)) {
regs->msr &= ~MSR_SE;
- regs->msr |= kprobe_saved_msr;
- unlock_kprobes();
+ regs->msr |= kcb->kprobe_saved_msr;
goto no_kprobe;
}
/* We have reentered the kprobe_handler(), since
@@ -166,27 +174,24 @@ static inline int kprobe_handler(struct pt_regs *regs)
* just single step on the instruction of the new probe
* without calling any user handlers.
*/
- save_previous_kprobe();
- current_kprobe = p;
- kprobe_saved_msr = regs->msr;
+ save_previous_kprobe(kcb);
+ set_current_kprobe(p, regs, kcb);
+ kcb->kprobe_saved_msr = regs->msr;
p->nmissed++;
prepare_singlestep(p, regs);
- kprobe_status = KPROBE_REENTER;
+ kcb->kprobe_status = KPROBE_REENTER;
return 1;
} else {
- p = current_kprobe;
+ p = __get_cpu_var(current_kprobe);
if (p->break_handler && p->break_handler(p, regs)) {
goto ss_probe;
}
}
- /* If it's not ours, can't be delete race, (we hold lock). */
goto no_kprobe;
}
- lock_kprobes();
p = get_kprobe(addr);
if (!p) {
- unlock_kprobes();
if (*addr != BREAKPOINT_INSTRUCTION) {
/*
* PowerPC has multiple variants of the "trap"
@@ -209,24 +214,19 @@ static inline int kprobe_handler(struct pt_regs *regs)
goto no_kprobe;
}
- kprobe_status = KPROBE_HIT_ACTIVE;
- current_kprobe = p;
- kprobe_saved_msr = regs->msr;
+ kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+ set_current_kprobe(p, regs, kcb);
if (p->pre_handler && p->pre_handler(p, regs))
/* handler has already set things up, so skip ss setup */
return 1;
ss_probe:
prepare_singlestep(p, regs);
- kprobe_status = KPROBE_HIT_SS;
- /*
- * This preempt_disable() matches the preempt_enable_no_resched()
- * in post_kprobe_handler().
- */
- preempt_disable();
+ kcb->kprobe_status = KPROBE_HIT_SS;
return 1;
no_kprobe:
+ preempt_enable_no_resched();
return ret;
}
@@ -251,9 +251,10 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
struct kretprobe_instance *ri = NULL;
struct hlist_head *head;
struct hlist_node *node, *tmp;
- unsigned long orig_ret_address = 0;
+ unsigned long flags, orig_ret_address = 0;
unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
+ spin_lock_irqsave(&kretprobe_lock, flags);
head = kretprobe_inst_table_head(current);
/*
@@ -292,12 +293,14 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
BUG_ON(!orig_ret_address || (orig_ret_address == trampoline_address));
regs->nip = orig_ret_address;
- unlock_kprobes();
+ reset_current_kprobe();
+ spin_unlock_irqrestore(&kretprobe_lock, flags);
+ preempt_enable_no_resched();
/*
* By returning a non-zero value, we are telling
- * kprobe_handler() that we have handled unlocking
- * and re-enabling preemption.
+ * kprobe_handler() that we don't want the post_handler
+ * to run (and have re-enabled preemption)
*/
return 1;
}
@@ -323,23 +326,26 @@ static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
static inline int post_kprobe_handler(struct pt_regs *regs)
{
- if (!kprobe_running())
+ struct kprobe *cur = kprobe_running();
+ struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+ if (!cur)
return 0;
- if ((kprobe_status != KPROBE_REENTER) && current_kprobe->post_handler) {
- kprobe_status = KPROBE_HIT_SSDONE;
- current_kprobe->post_handler(current_kprobe, regs, 0);
+ if ((kcb->kprobe_status != KPROBE_REENTER) && cur->post_handler) {
+ kcb->kprobe_status = KPROBE_HIT_SSDONE;
+ cur->post_handler(cur, regs, 0);
}
- resume_execution(current_kprobe, regs);
- regs->msr |= kprobe_saved_msr;
+ resume_execution(cur, regs);
+ regs->msr |= kcb->kprobe_saved_msr;
/*Restore back the original saved kprobes variables and continue. */
- if (kprobe_status == KPROBE_REENTER) {
- restore_previous_kprobe();
+ if (kcb->kprobe_status == KPROBE_REENTER) {
+ restore_previous_kprobe(kcb);
goto out;
}
- unlock_kprobes();
+ reset_current_kprobe();
out:
preempt_enable_no_resched();
@@ -354,19 +360,20 @@ out:
return 1;
}
-/* Interrupts disabled, kprobe_lock held. */
static inline int kprobe_fault_handler(struct pt_regs *regs, int trapnr)
{
- if (current_kprobe->fault_handler
- && current_kprobe->fault_handler(current_kprobe, regs, trapnr))
+ struct kprobe *cur = kprobe_running();
+ struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
+ if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
return 1;
- if (kprobe_status & KPROBE_HIT_SS) {
- resume_execution(current_kprobe, regs);
+ if (kcb->kprobe_status & KPROBE_HIT_SS) {
+ resume_execution(cur, regs);
regs->msr &= ~MSR_SE;
- regs->msr |= kprobe_saved_msr;
+ regs->msr |= kcb->kprobe_saved_msr;
- unlock_kprobes();
+ reset_current_kprobe();
preempt_enable_no_resched();
}
return 0;
@@ -381,11 +388,6 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
struct die_args *args = (struct die_args *)data;
int ret = NOTIFY_DONE;
- /*
- * Interrupts are not disabled here. We need to disable
- * preemption, because kprobe_running() uses smp_processor_id().
- */
- preempt_disable();
switch (val) {
case DIE_BPT:
if (kprobe_handler(args->regs))
@@ -396,22 +398,25 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
ret = NOTIFY_STOP;
break;
case DIE_PAGE_FAULT:
+ /* kprobe_running() needs smp_processor_id() */
+ preempt_disable();
if (kprobe_running() &&
kprobe_fault_handler(args->regs, args->trapnr))
ret = NOTIFY_STOP;
+ preempt_enable();
break;
default:
break;
}
- preempt_enable_no_resched();
return ret;
}
int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
{
struct jprobe *jp = container_of(p, struct jprobe, kp);
+ struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
- memcpy(&jprobe_saved_regs, regs, sizeof(struct pt_regs));
+ memcpy(&kcb->jprobe_saved_regs, regs, sizeof(struct pt_regs));
/* setup return addr to the jprobe handler routine */
regs->nip = (unsigned long)(((func_descr_t *)jp->entry)->entry);
@@ -431,12 +436,15 @@ void __kprobes jprobe_return_end(void)
int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
{
+ struct kprobe_ctlblk *kcb = get_kprobe_ctlblk();
+
/*
* FIXME - we should ideally be validating that we got here 'cos
* of the "trap" in jprobe_return() above, before restoring the
* saved regs...
*/
- memcpy(regs, &jprobe_saved_regs, sizeof(struct pt_regs));
+ memcpy(regs, &kcb->jprobe_saved_regs, sizeof(struct pt_regs));
+ preempt_enable_no_resched();
return 1;
}
diff --git a/arch/ppc64/kernel/lparcfg.c b/arch/ppc64/kernel/lparcfg.c
index e86155770bbc..3e7b2f28ec83 100644
--- a/arch/ppc64/kernel/lparcfg.c
+++ b/arch/ppc64/kernel/lparcfg.c
@@ -599,9 +599,7 @@ int __init lparcfg_init(void)
void __exit lparcfg_cleanup(void)
{
if (proc_ppc64_lparcfg) {
- if (proc_ppc64_lparcfg->data) {
- kfree(proc_ppc64_lparcfg->data);
- }
+ kfree(proc_ppc64_lparcfg->data);
remove_proc_entry("lparcfg", proc_ppc64_lparcfg->parent);
}
}
diff --git a/arch/ppc64/kernel/pacaData.c b/arch/ppc64/kernel/pacaData.c
index 5e27e5a6a35d..3133c72b28ec 100644
--- a/arch/ppc64/kernel/pacaData.c
+++ b/arch/ppc64/kernel/pacaData.c
@@ -23,7 +23,7 @@
static union {
struct systemcfg data;
u8 page[PAGE_SIZE];
-} systemcfg_store __page_aligned;
+} systemcfg_store __attribute__((__section__(".data.page.aligned")));
struct systemcfg *systemcfg = &systemcfg_store.data;
EXPORT_SYMBOL(systemcfg);
diff --git a/arch/ppc64/kernel/prom.c b/arch/ppc64/kernel/prom.c
index 97bfceb5353b..dece31e58bc4 100644
--- a/arch/ppc64/kernel/prom.c
+++ b/arch/ppc64/kernel/prom.c
@@ -635,10 +635,10 @@ static inline char *find_flat_dt_string(u32 offset)
* used to extract the memory informations at boot before we can
* unflatten the tree
*/
-static int __init scan_flat_dt(int (*it)(unsigned long node,
- const char *uname, int depth,
- void *data),
- void *data)
+int __init of_scan_flat_dt(int (*it)(unsigned long node,
+ const char *uname, int depth,
+ void *data),
+ void *data)
{
unsigned long p = ((unsigned long)initial_boot_params) +
initial_boot_params->off_dt_struct;
@@ -695,8 +695,8 @@ static int __init scan_flat_dt(int (*it)(unsigned long node,
* This function can be used within scan_flattened_dt callback to get
* access to properties
*/
-static void* __init get_flat_dt_prop(unsigned long node, const char *name,
- unsigned long *size)
+void* __init of_get_flat_dt_prop(unsigned long node, const char *name,
+ unsigned long *size)
{
unsigned long p = node;
@@ -996,7 +996,7 @@ void __init unflatten_device_tree(void)
static int __init early_init_dt_scan_cpus(unsigned long node,
const char *uname, int depth, void *data)
{
- char *type = get_flat_dt_prop(node, "device_type", NULL);
+ char *type = of_get_flat_dt_prop(node, "device_type", NULL);
u32 *prop;
unsigned long size;
@@ -1004,17 +1004,6 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
if (type == NULL || strcmp(type, "cpu") != 0)
return 0;
- /* On LPAR, look for the first ibm,pft-size property for the hash table size
- */
- if (systemcfg->platform == PLATFORM_PSERIES_LPAR && ppc64_pft_size == 0) {
- u32 *pft_size;
- pft_size = (u32 *)get_flat_dt_prop(node, "ibm,pft-size", NULL);
- if (pft_size != NULL) {
- /* pft_size[0] is the NUMA CEC cookie */
- ppc64_pft_size = pft_size[1];
- }
- }
-
if (initial_boot_params && initial_boot_params->version >= 2) {
/* version 2 of the kexec param format adds the phys cpuid
* of booted proc.
@@ -1023,8 +1012,9 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
boot_cpuid = 0;
} else {
/* Check if it's the boot-cpu, set it's hw index in paca now */
- if (get_flat_dt_prop(node, "linux,boot-cpu", NULL) != NULL) {
- u32 *prop = get_flat_dt_prop(node, "reg", NULL);
+ if (of_get_flat_dt_prop(node, "linux,boot-cpu", NULL)
+ != NULL) {
+ u32 *prop = of_get_flat_dt_prop(node, "reg", NULL);
set_hard_smp_processor_id(0, prop == NULL ? 0 : *prop);
boot_cpuid_phys = get_hard_smp_processor_id(0);
}
@@ -1032,14 +1022,14 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
#ifdef CONFIG_ALTIVEC
/* Check if we have a VMX and eventually update CPU features */
- prop = (u32 *)get_flat_dt_prop(node, "ibm,vmx", NULL);
+ prop = (u32 *)of_get_flat_dt_prop(node, "ibm,vmx", NULL);
if (prop && (*prop) > 0) {
cur_cpu_spec->cpu_features |= CPU_FTR_ALTIVEC;
cur_cpu_spec->cpu_user_features |= PPC_FEATURE_HAS_ALTIVEC;
}
/* Same goes for Apple's "altivec" property */
- prop = (u32 *)get_flat_dt_prop(node, "altivec", NULL);
+ prop = (u32 *)of_get_flat_dt_prop(node, "altivec", NULL);
if (prop) {
cur_cpu_spec->cpu_features |= CPU_FTR_ALTIVEC;
cur_cpu_spec->cpu_user_features |= PPC_FEATURE_HAS_ALTIVEC;
@@ -1051,7 +1041,7 @@ static int __init early_init_dt_scan_cpus(unsigned long node,
* this by looking at the size of the ibm,ppc-interrupt-server#s
* property
*/
- prop = (u32 *)get_flat_dt_prop(node, "ibm,ppc-interrupt-server#s",
+ prop = (u32 *)of_get_flat_dt_prop(node, "ibm,ppc-interrupt-server#s",
&size);
cur_cpu_spec->cpu_features &= ~CPU_FTR_SMT;
if (prop && ((size / sizeof(u32)) > 1))
@@ -1072,26 +1062,26 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
return 0;
/* get platform type */
- prop = (u32 *)get_flat_dt_prop(node, "linux,platform", NULL);
+ prop = (u32 *)of_get_flat_dt_prop(node, "linux,platform", NULL);
if (prop == NULL)
return 0;
systemcfg->platform = *prop;
/* check if iommu is forced on or off */
- if (get_flat_dt_prop(node, "linux,iommu-off", NULL) != NULL)
+ if (of_get_flat_dt_prop(node, "linux,iommu-off", NULL) != NULL)
iommu_is_off = 1;
- if (get_flat_dt_prop(node, "linux,iommu-force-on", NULL) != NULL)
+ if (of_get_flat_dt_prop(node, "linux,iommu-force-on", NULL) != NULL)
iommu_force_on = 1;
- prop64 = (u64*)get_flat_dt_prop(node, "linux,memory-limit", NULL);
+ prop64 = (u64*)of_get_flat_dt_prop(node, "linux,memory-limit", NULL);
if (prop64)
memory_limit = *prop64;
- prop64 = (u64*)get_flat_dt_prop(node, "linux,tce-alloc-start", NULL);
+ prop64 = (u64*)of_get_flat_dt_prop(node, "linux,tce-alloc-start",NULL);
if (prop64)
tce_alloc_start = *prop64;
- prop64 = (u64*)get_flat_dt_prop(node, "linux,tce-alloc-end", NULL);
+ prop64 = (u64*)of_get_flat_dt_prop(node, "linux,tce-alloc-end", NULL);
if (prop64)
tce_alloc_end = *prop64;
@@ -1102,9 +1092,12 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
{
u64 *basep, *entryp;
- basep = (u64*)get_flat_dt_prop(node, "linux,rtas-base", NULL);
- entryp = (u64*)get_flat_dt_prop(node, "linux,rtas-entry", NULL);
- prop = (u32*)get_flat_dt_prop(node, "linux,rtas-size", NULL);
+ basep = (u64*)of_get_flat_dt_prop(node,
+ "linux,rtas-base", NULL);
+ entryp = (u64*)of_get_flat_dt_prop(node,
+ "linux,rtas-entry", NULL);
+ prop = (u32*)of_get_flat_dt_prop(node,
+ "linux,rtas-size", NULL);
if (basep && entryp && prop) {
rtas.base = *basep;
rtas.entry = *entryp;
@@ -1125,11 +1118,11 @@ static int __init early_init_dt_scan_root(unsigned long node,
if (depth != 0)
return 0;
- prop = (u32 *)get_flat_dt_prop(node, "#size-cells", NULL);
+ prop = (u32 *)of_get_flat_dt_prop(node, "#size-cells", NULL);
dt_root_size_cells = (prop == NULL) ? 1 : *prop;
DBG("dt_root_size_cells = %x\n", dt_root_size_cells);
- prop = (u32 *)get_flat_dt_prop(node, "#address-cells", NULL);
+ prop = (u32 *)of_get_flat_dt_prop(node, "#address-cells", NULL);
dt_root_addr_cells = (prop == NULL) ? 2 : *prop;
DBG("dt_root_addr_cells = %x\n", dt_root_addr_cells);
@@ -1161,7 +1154,7 @@ static unsigned long __init dt_mem_next_cell(int s, cell_t **cellp)
static int __init early_init_dt_scan_memory(unsigned long node,
const char *uname, int depth, void *data)
{
- char *type = get_flat_dt_prop(node, "device_type", NULL);
+ char *type = of_get_flat_dt_prop(node, "device_type", NULL);
cell_t *reg, *endp;
unsigned long l;
@@ -1169,7 +1162,7 @@ static int __init early_init_dt_scan_memory(unsigned long node,
if (type == NULL || strcmp(type, "memory") != 0)
return 0;
- reg = (cell_t *)get_flat_dt_prop(node, "reg", &l);
+ reg = (cell_t *)of_get_flat_dt_prop(node, "reg", &l);
if (reg == NULL)
return 0;
@@ -1225,19 +1218,16 @@ void __init early_init_devtree(void *params)
/* Setup flat device-tree pointer */
initial_boot_params = params;
- /* By default, hash size is not set */
- ppc64_pft_size = 0;
-
/* Retreive various informations from the /chosen node of the
* device-tree, including the platform type, initrd location and
* size, TCE reserve, and more ...
*/
- scan_flat_dt(early_init_dt_scan_chosen, NULL);
+ of_scan_flat_dt(early_init_dt_scan_chosen, NULL);
/* Scan memory nodes and rebuild LMBs */
lmb_init();
- scan_flat_dt(early_init_dt_scan_root, NULL);
- scan_flat_dt(early_init_dt_scan_memory, NULL);
+ of_scan_flat_dt(early_init_dt_scan_root, NULL);
+ of_scan_flat_dt(early_init_dt_scan_memory, NULL);
lmb_enforce_memory_limit(memory_limit);
lmb_analyze();
systemcfg->physicalMemorySize = lmb_phys_mem_size();
@@ -1253,26 +1243,8 @@ void __init early_init_devtree(void *params)
/* Retreive hash table size from flattened tree plus other
* CPU related informations (altivec support, boot CPU ID, ...)
*/
- scan_flat_dt(early_init_dt_scan_cpus, NULL);
-
- /* If hash size wasn't obtained above, we calculate it now based on
- * the total RAM size
- */
- if (ppc64_pft_size == 0) {
- unsigned long rnd_mem_size, pteg_count;
-
- /* round mem_size up to next power of 2 */
- rnd_mem_size = 1UL << __ilog2(systemcfg->physicalMemorySize);
- if (rnd_mem_size < systemcfg->physicalMemorySize)
- rnd_mem_size <<= 1;
-
- /* # pages / 2 */
- pteg_count = max(rnd_mem_size >> (12 + 1), 1UL << 11);
-
- ppc64_pft_size = __ilog2(pteg_count << 7);
- }
+ of_scan_flat_dt(early_init_dt_scan_cpus, NULL);
- DBG("Hash pftSize: %x\n", (int)ppc64_pft_size);
DBG(" <- early_init_devtree()\n");
}
diff --git a/arch/ppc64/kernel/scanlog.c b/arch/ppc64/kernel/scanlog.c
index 215bf8900304..2edc947f7c44 100644
--- a/arch/ppc64/kernel/scanlog.c
+++ b/arch/ppc64/kernel/scanlog.c
@@ -225,8 +225,7 @@ int __init scanlog_init(void)
void __exit scanlog_cleanup(void)
{
if (proc_ppc64_scan_log_dump) {
- if (proc_ppc64_scan_log_dump->data)
- kfree(proc_ppc64_scan_log_dump->data);
+ kfree(proc_ppc64_scan_log_dump->data);
remove_proc_entry("scan-log-dump", proc_ppc64_scan_log_dump->parent);
}
}
diff --git a/arch/ppc64/kernel/sysfs.c b/arch/ppc64/kernel/sysfs.c
index 6654b350979c..e99ec62c2c52 100644
--- a/arch/ppc64/kernel/sysfs.c
+++ b/arch/ppc64/kernel/sysfs.c
@@ -20,6 +20,7 @@
#include <asm/paca.h>
#include <asm/lppaca.h>
#include <asm/machdep.h>
+#include <asm/smp.h>
static DEFINE_PER_CPU(struct cpu, cpu_devices);
View Lorry Controller Status