diff options
author | Anton Blanchard <anton@samba.org> | 2012-05-27 19:54:03 +0000 |
---|---|---|
committer | Benjamin Herrenschmidt <benh@kernel.crashing.org> | 2012-07-03 14:14:41 +1000 |
commit | 17968fbbd19f1bb281ee4eb2548764ac5664c4ec (patch) | |
tree | c6b7a68ea7897e6bf213bffa4795fe209609cba9 /arch/powerpc/lib/string.S | |
parent | d136e27326a3bd50d7929a43c018abf13e426b7e (diff) |
powerpc: 64bit optimised __clear_user
I noticed __clear_user high up in a profile of one of my RAID stress
tests. The testcase was doing a dd from /dev/zero which ends up
calling __clear_user.
__clear_user is basically a loop with a single 4 byte store which
is horribly slow. We can do much better by aligning the desination
and doing 32 bytes of 8 byte stores in a loop.
The following testcase was used to verify the patch:
http://ozlabs.org/~anton/junkcode/stress_clear_user.c
To show the improvement in performance I ran a dd from /dev/zero
to /dev/null on a POWER7 box:
Before:
# dd if=/dev/zero of=/dev/null bs=1M count=10000
10485760000 bytes (10 GB) copied, 3.72379 s, 2.8 GB/s
After:
# time dd if=/dev/zero of=/dev/null bs=1M count=10000
10485760000 bytes (10 GB) copied, 0.728318 s, 14.4 GB/s
Over 5x faster.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Diffstat (limited to 'arch/powerpc/lib/string.S')
-rw-r--r-- | arch/powerpc/lib/string.S | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/arch/powerpc/lib/string.S b/arch/powerpc/lib/string.S index 093d6316435c..1b5a0a09d609 100644 --- a/arch/powerpc/lib/string.S +++ b/arch/powerpc/lib/string.S @@ -119,6 +119,7 @@ _GLOBAL(memchr) 2: li r3,0 blr +#ifdef CONFIG_PPC32 _GLOBAL(__clear_user) addi r6,r3,-4 li r3,0 @@ -160,3 +161,4 @@ _GLOBAL(__clear_user) PPC_LONG 1b,91b PPC_LONG 8b,92b .text +#endif |