* + zlib-optimize-inffast-when-copying-direct-from-output.patch added to -mm tree
@ 2009-11-11 0:43 akpm
0 siblings, 0 replies; 2+ messages in thread
From: akpm @ 2009-11-11 0:43 UTC (permalink / raw)
To: mm-commits; +Cc: Joakim.Tjernlund, alain, dwmw2, hpa, rpurdie
The patch titled
zlib: optimize inffast when copying direct from output
has been added to the -mm tree. Its filename is
zlib-optimize-inffast-when-copying-direct-from-output.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find
out what to do about this
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: zlib: optimize inffast when copying direct from output
From: Joakim Tjernlund <Joakim.Tjernlund@transmode.se>
JFFS2 uses lesser compression ratio and inflate always ends up in "copy
direct from output" case.
This patch tries to optimize the copy procedure for arch's that have
CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS defined. On my MPC8321 this is
about 14% faster on my JFFS2 root FS than the original.
Signed-off-by: Joakim Tjernlund <Joakim.Tjernlund@transmode.se>
Cc: Alain Knaff <alain@knaff.lu>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Richard Purdie <rpurdie@rpsys.net>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
diff -puN lib/zlib_inflate/inffast.c~zlib-optimize-inffast-when-copying-direct-from-output lib/zlib_inflate/inffast.c
--- a/lib/zlib_inflate/inffast.c~zlib-optimize-inffast-when-copying-direct-from-output
+++ a/lib/zlib_inflate/inffast.c
@@ -240,6 +240,40 @@ void inflate_fast(z_streamp strm, unsign
}
else {
from = out - dist; /* copy direct from output */
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+ /* minimum length is three */
+ if (dist > 2 ) {
+ unsigned short *sout = (unsigned short *)(out - OFF);
+ unsigned short *sfrom = (unsigned short *)(from - OFF);
+ unsigned long loops = len >> 1;
+
+ do
+ PUP(sout) = PUP(sfrom);
+ while (--loops);
+ out = (unsigned char *)sout + OFF;
+ from = (unsigned char *)sfrom + OFF;
+ if (len & 1)
+ PUP(out) = PUP(from);
+ } else if (dist == 2) {
+ unsigned short *sout = (unsigned short *)(out - OFF);
+ unsigned short pat16;
+ unsigned long loops = len >> 1;
+
+ pat16 = *(sout-2+2*OFF);
+ do
+ PUP(sout) = pat16;
+ while (--loops);
+ out = (unsigned char *)sout + OFF;
+ if (len & 1)
+ PUP(out) = PUP(from);
+ } else {
+ unsigned char pat8 = *(out - 1 + OFF);
+
+ do {
+ PUP(out) = pat8;
+ } while (--len);
+ }
+#else
do { /* minimum length is three */
PUP(out) = PUP(from);
PUP(out) = PUP(from);
@@ -251,6 +285,7 @@ void inflate_fast(z_streamp strm, unsign
if (len > 1)
PUP(out) = PUP(from);
}
+#endif
}
}
else if ((op & 64) == 0) { /* 2nd level distance code */
_
Patches currently in -mm which might be from Joakim.Tjernlund@transmode.se are
zlib-optimize-inffast-when-copying-direct-from-output.patch
^ permalink raw reply [flat|nested] 2+ messages in thread
* + zlib-optimize-inffast-when-copying-direct-from-output.patch added to -mm tree
@ 2009-11-13 22:47 akpm
0 siblings, 0 replies; 2+ messages in thread
From: akpm @ 2009-11-13 22:47 UTC (permalink / raw)
To: mm-commits; +Cc: Joakim.Tjernlund, dwmw2, roel.kluin, rpurdie
The patch titled
zlib: optimize inffast when copying direct from output
has been added to the -mm tree. Its filename is
zlib-optimize-inffast-when-copying-direct-from-output.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find
out what to do about this
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: zlib: optimize inffast when copying direct from output
From: Joakim Tjernlund <Joakim.Tjernlund@transmode.se>
JFFS2 uses lesser compression ratio and inflate always ends up in "copy
direct from output" case.
This patch tries to optimize the direct copy procedure. Uses
get_unaligned() but only in one place.
The copy loop just above this one can also use this optimization, but I
havn't done so as I have not tested if it is a win there too.
On my MPC8321 this is about 17% faster on my JFFS2 root FS than the
original.
Signed-off-by: Joakim Tjernlund <Joakim.Tjernlund@transmode.se>
Cc: Roel Kluin <roel.kluin@gmail.com>
Cc: Richard Purdie <rpurdie@rpsys.net>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
arch/powerpc/boot/Makefile | 4 +-
lib/zlib_inflate/inffast.c | 55 ++++++++++++++++++++++++++++-------
2 files changed, 47 insertions(+), 12 deletions(-)
diff -puN arch/powerpc/boot/Makefile~zlib-optimize-inffast-when-copying-direct-from-output arch/powerpc/boot/Makefile
--- a/arch/powerpc/boot/Makefile~zlib-optimize-inffast-when-copying-direct-from-output
+++ a/arch/powerpc/boot/Makefile
@@ -20,7 +20,7 @@
all: $(obj)/zImage
BOOTCFLAGS := -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs \
- -fno-strict-aliasing -Os -msoft-float -pipe \
+ -fno-strict-aliasing -Os -msoft-float -pipe -D__KERNEL__\
-fomit-frame-pointer -fno-builtin -fPIC -nostdinc \
-isystem $(shell $(CROSS32CC) -print-file-name=include)
BOOTAFLAGS := -D__ASSEMBLY__ $(BOOTCFLAGS) -traditional -nostdinc
@@ -34,6 +34,8 @@ BOOTCFLAGS += -fno-stack-protector
endif
BOOTCFLAGS += -I$(obj) -I$(srctree)/$(obj)
+BOOTCFLAGS += -include include/linux/autoconf.h -Iarch/powerpc/include
+BOOTCFLAGS += -Iinclude
DTS_FLAGS ?= -p 1024
diff -puN lib/zlib_inflate/inffast.c~zlib-optimize-inffast-when-copying-direct-from-output lib/zlib_inflate/inffast.c
--- a/lib/zlib_inflate/inffast.c~zlib-optimize-inffast-when-copying-direct-from-output
+++ a/lib/zlib_inflate/inffast.c
@@ -4,6 +4,8 @@
*/
#include <linux/zutil.h>
+#include <asm/unaligned.h>
+#include <asm/byteorder.h>
#include "inftrees.h"
#include "inflate.h"
#include "inffast.h"
@@ -24,9 +26,11 @@
#ifdef POSTINC
# define OFF 0
# define PUP(a) *(a)++
+# define UP_UNALIGNED(a) get_unaligned((a)++)
#else
# define OFF 1
# define PUP(a) *++(a)
+# define UP_UNALIGNED(a) get_unaligned(++(a))
#endif
/*
@@ -239,18 +243,47 @@ void inflate_fast(z_streamp strm, unsign
}
}
else {
+ unsigned short *sout;
+ unsigned long loops;
+
from = out - dist; /* copy direct from output */
- do { /* minimum length is three */
- PUP(out) = PUP(from);
- PUP(out) = PUP(from);
- PUP(out) = PUP(from);
- len -= 3;
- } while (len > 2);
- if (len) {
- PUP(out) = PUP(from);
- if (len > 1)
- PUP(out) = PUP(from);
- }
+ /* minimum length is three */
+ /* Align out addr */
+ if (!((long)(out - 1 + OFF) & 1)) {
+ PUP(out) = PUP(from);
+ len--;
+ }
+ sout = (unsigned short *)(out - OFF);
+ if (dist > 2 ) {
+ unsigned short *sfrom;
+
+ sfrom = (unsigned short *)(from - OFF);
+ loops = len >> 1;
+ do
+ PUP(sout) = UP_UNALIGNED(sfrom);
+ while (--loops);
+ out = (unsigned char *)sout + OFF;
+ from = (unsigned char *)sfrom + OFF;
+ } else { /* dist == 1 or dist == 2 */
+ unsigned short pat16;
+
+ pat16 = *(sout-2+2*OFF);
+ if (dist == 1)
+#if defined(__BIG_ENDIAN)
+ pat16 = (pat16 & 0xff) | ((pat16 & 0xff ) << 8);
+#elif defined(__LITTLE_ENDIAN)
+ pat16 = (pat16 & 0xff00) | ((pat16 & 0xff00 ) >> 8);
+#else
+#error __BIG_ENDIAN nor __LITTLE_ENDIAN is defined
+#endif
+ loops = len >> 1;
+ do
+ PUP(sout) = pat16;
+ while (--loops);
+ out = (unsigned char *)sout + OFF;
+ }
+ if (len & 1)
+ PUP(out) = PUP(from);
}
}
else if ((op & 64) == 0) { /* 2nd level distance code */
_
Patches currently in -mm which might be from Joakim.Tjernlund@transmode.se are
crc32-minor-optimizations-and-cleanup.patch
crc32-minor-optimizations-and-cleanup-checkpatch-fixes.patch
zlib-optimize-inffast-when-copying-direct-from-output.patch
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2009-11-13 22:47 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2009-11-11 0:43 + zlib-optimize-inffast-when-copying-direct-from-output.patch added to -mm tree akpm
2009-11-13 22:47 akpm
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.