windows: support for large files

note that this change will break binary compatibility with the
default 5.8.0 build options; nevertheless I think it is worth
having in 5.8.1 (people who want the compatibility can disable
the option in the makefile)

p4raw-id: //depot/perl@18327

14 files changed, 103 insertions, 28 deletions

diff --git a/win32/Makefile b/win32/Makefile
index 901ff0cda2..bba77a92cf 100644
--- a/win32/Makefile
+++ b/win32/Makefile
@@ -74,11 +74,17 @@ USE_IMP_SYS = define
 USE_PERLIO	= define
 
 #
+# Comment this out if you don't want to enable large file support for
+# some reason.  Should normally only be changed to maintain compatibility
+# with an older release of perl.
+USE_LARGE_FILES = define
+
+#
 # uncomment one of the following lines if you are using either
 # Visual C++ 2.x or Visual C++ 6.x (aka Visual Studio 98)
 #
 #CCTYPE		= MSVC20
-#CCTYPE		= MSVC60
+CCTYPE		= MSVC60
 
 #
 # uncomment next line if you want debug version of perl (big,slow)
@@ -216,6 +222,10 @@ USE_IMP_SYS	= undef
 USE_PERLIO	= undef
 !ENDIF
 
+!IF "$(USE_LARGE_FILES)" == ""
+USE_LARGE_FILES	= undef
+!ENDIF
+
 !IF "$(USE_PERLCRT)" == ""
 USE_PERLCRT	= undef
 !ENDIF
@@ -767,6 +777,7 @@ CFG_VARS	=					\
 		"useithreads=$(USE_ITHREADS)"		\
 		"usemultiplicity=$(USE_MULTI)"		\
 		"useperlio=$(USE_PERLIO)"		\
+		"uselargefiles=$(USE_LARGE_FILES)"	\
 		"LINK_FLAGS=$(LINK_FLAGS:"=\")"		\
 		"optimize=$(OPTIMIZE:"=\")"
 
diff --git a/win32/config.bc b/win32/config.bc
index 7d669fffa8..d8ac588975 100644
--- a/win32/config.bc
+++ b/win32/config.bc
@@ -697,8 +697,8 @@ longsize='4'
 lp=''
 lpr=''
 ls='dir'
-lseeksize='4'
-lseektype='off_t'
+lseeksize='8'
+lseektype='__int64'
 mail=''
 mailx=''
 make='dmake'
@@ -904,7 +904,7 @@ use64bitint='undef'
 usecrosscompile='undef'
 usedl='define'
 useithreads='undef'
-uselargefiles='undef'
+uselargefiles='define'
 uselongdouble='undef'
 usemorebits='undef'
 usemultiplicity='undef'
diff --git a/win32/config.gc b/win32/config.gc
index 141fb542ee..f603bee878 100644
--- a/win32/config.gc
+++ b/win32/config.gc
@@ -697,8 +697,8 @@ longsize='4'
 lp=''
 lpr=''
 ls='dir'
-lseeksize='4'
-lseektype='off_t'
+lseeksize='8'
+lseektype='long long'
 mail=''
 mailx=''
 make='dmake'
@@ -904,7 +904,7 @@ use64bitint='undef'
 usecrosscompile='undef'
 usedl='define'
 useithreads='undef'
-uselargefiles='undef'
+uselargefiles='define'
 uselongdouble='undef'
 usemorebits='undef'
 usemultiplicity='undef'
diff --git a/win32/config.vc b/win32/config.vc
index 89192be0a0..3400eff7d3 100644
--- a/win32/config.vc
+++ b/win32/config.vc
@@ -697,8 +697,8 @@ longsize='4'
 lp=''
 lpr=''
 ls='dir'
-lseeksize='4'
-lseektype='off_t'
+lseeksize='8'
+lseektype='__int64'
 mail=''
 mailx=''
 make='nmake'
@@ -904,7 +904,7 @@ use64bitint='undef'
 usecrosscompile='undef'
 usedl='define'
 useithreads='undef'
-uselargefiles='undef'
+uselargefiles='define'
 uselongdouble='undef'
 usemorebits='undef'
 usemultiplicity='undef'
diff --git a/win32/config.vc64 b/win32/config.vc64
index bc969884bb..ca88714ce3 100644
--- a/win32/config.vc64
+++ b/win32/config.vc64
@@ -904,7 +904,7 @@ use64bitint='define'
 usecrosscompile='undef'
 usedl='define'
 useithreads='undef'
-uselargefiles='undef'
+uselargefiles='define'
 uselongdouble='undef'
 usemorebits='undef'
 usemultiplicity='undef'
diff --git a/win32/config_H.bc b/win32/config_H.bc
index f4fec85114..8bcd5fbf47 100644
--- a/win32/config_H.bc
+++ b/win32/config_H.bc
@@ -2950,9 +2950,9 @@
 /* Off_t_size:
  *	This symbol holds the number of bytes used by the Off_t.
  */
-#define Off_t off_t		/* <offset> type */
-#define LSEEKSIZE 4		/* <offset> size */
-#define Off_t_size 4	/* <offset> size */
+#define Off_t __int64		/* <offset> type */
+#define LSEEKSIZE 8		/* <offset> size */
+#define Off_t_size 8	/* <offset> size */
 
 /* Free_t:
  *	This variable contains the return type of free().  It is usually
@@ -3445,7 +3445,7 @@
  *	should be used when available.
  */
 #ifndef USE_LARGE_FILES
-/*#define	USE_LARGE_FILES		/**/
+#define	USE_LARGE_FILES		/**/
 #endif
 
 /* USE_LONG_DOUBLE:
diff --git a/win32/config_H.gc b/win32/config_H.gc
index 2422466cc2..c9ce6673ac 100644
--- a/win32/config_H.gc
+++ b/win32/config_H.gc
@@ -2950,9 +2950,9 @@
 /* Off_t_size:
  *	This symbol holds the number of bytes used by the Off_t.
  */
-#define Off_t off_t		/* <offset> type */
-#define LSEEKSIZE 4		/* <offset> size */
-#define Off_t_size 4	/* <offset> size */
+#define Off_t long long		/* <offset> type */
+#define LSEEKSIZE 8		/* <offset> size */
+#define Off_t_size 8	/* <offset> size */
 
 /* Free_t:
  *	This variable contains the return type of free().  It is usually
@@ -3445,7 +3445,7 @@
  *	should be used when available.
  */
 #ifndef USE_LARGE_FILES
-/*#define	USE_LARGE_FILES		/**/
+#define	USE_LARGE_FILES		/**/
 #endif
 
 /* USE_LONG_DOUBLE:
diff --git a/win32/config_H.vc b/win32/config_H.vc
index 8fcfb0fbfb..c4ba4a51cc 100644
--- a/win32/config_H.vc
+++ b/win32/config_H.vc
@@ -2950,9 +2950,9 @@
 /* Off_t_size:
  *	This symbol holds the number of bytes used by the Off_t.
  */
-#define Off_t off_t		/* <offset> type */
-#define LSEEKSIZE 4		/* <offset> size */
-#define Off_t_size 4	/* <offset> size */
+#define Off_t __int64		/* <offset> type */
+#define LSEEKSIZE 8		/* <offset> size */
+#define Off_t_size 8	/* <offset> size */
 
 /* Free_t:
  *	This variable contains the return type of free().  It is usually
@@ -3445,7 +3445,7 @@
  *	should be used when available.
  */
 #ifndef USE_LARGE_FILES
-/*#define	USE_LARGE_FILES		/**/
+#define	USE_LARGE_FILES		/**/
 #endif
 
 /* USE_LONG_DOUBLE:
diff --git a/win32/config_H.vc64 b/win32/config_H.vc64
index 5f1ff0fc8d..5572c2613c 100644
--- a/win32/config_H.vc64
+++ b/win32/config_H.vc64
@@ -3445,7 +3445,7 @@
  *	should be used when available.
  */
 #ifndef USE_LARGE_FILES
-/*#define	USE_LARGE_FILES		/**/
+#define	USE_LARGE_FILES		/**/
 #endif
 
 /* USE_LONG_DOUBLE:
diff --git a/win32/config_sh.PL b/win32/config_sh.PL
index b31415d88a..3b76cd04ea 100644
--- a/win32/config_sh.PL
+++ b/win32/config_sh.PL
@@ -90,6 +90,11 @@ if (defined(&Win32::IsWin95) && Win32::IsWin95()) {
     $opt{d_link} = 'undef';
 }
 
+if ($opt{uselargefiles} ne 'define') {
+    $opt{lseeksize} = 4;
+    $opt{lseektype} = 'off_t';
+}
+
 while (<>) {
     s/~([\w_]+)~/$opt{$1}/g;
     if (/^([\w_]+)=(.*)$/) {
diff --git a/win32/makefile.mk b/win32/makefile.mk
index 683cb420b8..723d5b2815 100644
--- a/win32/makefile.mk
+++ b/win32/makefile.mk
@@ -73,7 +73,13 @@ USE_IMP_SYS	*= define
 # then get a number of fails from make test i.e. bugs - complain to them not us ;-). 
 # You will also be unable to take full advantage of perl5.8's support for multiple 
 # encodings and may see lower IO performance. You have been warned.
-USE_PERLIO	= define
+USE_PERLIO	*= define
+
+#
+# Comment this out if you don't want to enable large file support for
+# some reason.  Should normally only be changed to maintain compatibility
+# with an older release of perl.
+USE_LARGE_FILES *= define
 
 #
 # uncomment exactly one of the following
@@ -233,6 +239,7 @@ USE_MULTI	*= undef
 USE_ITHREADS	*= undef
 USE_IMP_SYS	*= undef
 USE_PERLIO	*= undef
+USE_LARGE_FILES	*= undef
 USE_PERLCRT	*= undef
 
 .IF "$(USE_IMP_SYS)$(USE_MULTI)" == "defineundef"
@@ -813,7 +820,8 @@ CFG_VARS	=					\
 		useithreads=$(USE_ITHREADS)	~	\
 		usemultiplicity=$(USE_MULTI)	~	\
 		useperlio=$(USE_PERLIO)		~	\
-		LINK_FLAGS=$(LINK_FLAGS:s/\/\\/)		~	\
+		uselargefiles=$(USE_LARGE_FILES) ~	\
+		LINK_FLAGS=$(LINK_FLAGS:s/\/\\/) ~	\
 		optimize=$(OPTIMIZE)
 
 #
diff --git a/win32/perlhost.h b/win32/perlhost.h
index 7926142954..b68f5c771b 100644
--- a/win32/perlhost.h
+++ b/win32/perlhost.h
@@ -939,9 +939,9 @@ PerlLIOChown(struct IPerlLIO* piPerl, const char *filename, uid_t owner, gid_t g
 }
 
 int
-PerlLIOChsize(struct IPerlLIO* piPerl, int handle, long size)
+PerlLIOChsize(struct IPerlLIO* piPerl, int handle, Off_t size)
 {
-    return chsize(handle, size);
+    return win32_chsize(handle, size);
 }
 
 int
diff --git a/win32/win32.c b/win32/win32.c
index 633dbbdbcc..556d62145c 100644
--- a/win32/win32.c
+++ b/win32/win32.c
@@ -3073,6 +3073,55 @@ win32_setmode(int fd, int mode)
     return setmode(fd, mode);
 }
 
+DllExport int
+win32_chsize(int fd, Off_t size)
+{
+#if defined(WIN64) || defined(USE_LARGE_FILES)
+    int retval = 0;
+    Off_t cur, end, extend;
+
+    cur = win32_tell(fd);
+    if (cur < 0)
+	return -1;
+    end = win32_lseek(fd, 0, SEEK_END);
+    if (end < 0)
+	return -1;
+    extend = size - end;
+    if (extend == 0) {
+	/* do nothing */
+    }
+    else if (extend > 0) {
+	/* must grow the file, padding with nulls */
+	char b[4096];
+	int oldmode = win32_setmode(fd, O_BINARY);
+	size_t count;
+	memset(b, '\0', sizeof(b));
+	do {
+	    count = extend >= sizeof(b) ? sizeof(b) : (size_t)extend;
+	    count = win32_write(fd, b, count);
+	    if (count < 0) {
+		retval = -1;
+		break;
+	    }
+	} while ((extend -= count) > 0);
+	win32_setmode(fd, oldmode);
+    }
+    else {
+	/* shrink the file */
+	win32_lseek(fd, size, SEEK_SET);
+	if (!SetEndOfFile((HANDLE)_get_osfhandle(fd))) {
+	    errno = EACCES;
+	    retval = -1;
+	}
+    }
+finish:
+    win32_lseek(fd, cur, SEEK_SET);
+    return retval;
+#else
+    return chsize(fd, size);
+#endif
+}
+
 DllExport Off_t
 win32_lseek(int fd, Off_t offset, int origin)
 {
diff --git a/win32/win32iop.h b/win32/win32iop.h
index 52c59b45b5..e835b2eea6 100644
--- a/win32/win32iop.h
+++ b/win32/win32iop.h
@@ -77,6 +77,7 @@ DllExport  PerlIO*	win32_popenlist(const char *mode, IV narg, SV **args);
 DllExport  int		win32_pclose( PerlIO *pf);
 DllExport  int		win32_rename( const char *oname, const char *newname);
 DllExport  int		win32_setmode( int fd, int mode);
+DllExport  int		win32_chsize(int fd, Off_t size);
 DllExport  Off_t	win32_lseek( int fd, Off_t offset, int origin);
 DllExport  Off_t	win32_tell( int fd);
 DllExport  int		win32_dup( int fd);
@@ -236,6 +237,7 @@ END_EXTERN_C
 #define longpath(pth)   	win32_longpath(pth)
 #define rename(old,new)		win32_rename(old,new)
 #define setmode(fd,mode)	win32_setmode(fd,mode)
+#define chsize(fd,sz)		win32_chsize(fd,sz)
 #define lseek(fd,offset,orig)	win32_lseek(fd,offset,orig)
 #define tell(fd)		win32_tell(fd)
 #define dup(fd)			win32_dup(fd)