summaryrefslogtreecommitdiffstats
path: root/lib
diff options
context:
space:
mode:
authorMatthew Wilcox <mawilcox@microsoft.com>2017-02-24 15:00:58 -0800
committerLinus Torvalds <torvalds@linux-foundation.org>2017-02-24 17:46:57 -0800
commite4afd2e5567fc5d59988025f7528f9b4794d86a5 (patch)
treed00cf7c2409d3fdeb4dbb81ec04d6cb9c0e7965d /lib
parent55ded9551f9a64f2872df77a954d4c30f8958e82 (diff)
downloadlinux-e4afd2e5567fc5d59988025f7528f9b4794d86a5.tar.gz
linux-e4afd2e5567fc5d59988025f7528f9b4794d86a5.tar.xz
lib/find_bit.c: micro-optimise find_next_*_bit
This saves 32 bytes on my x86-64 build, mostly due to alignment considerations and sharing more code between find_next_bit and find_next_zero_bit, but it does save a couple of instructions. There's really two parts to this commit: - First, the first half of the test: (!nbits || start >= nbits) is trivially a subset of the second half, since nbits and start are both unsigned - Second, while looking at the disassembly, I noticed that GCC was predicting the branch taken. Since this is a failure case, it's clearly the less likely of the two branches, so add an unlikely() to override GCC's heuristics. [mawilcox@microsoft.com: v2] Link: http://lkml.kernel.org/r/1483709016-1834-1-git-send-email-mawilcox@linuxonhyperv.com Link: http://lkml.kernel.org/r/1483709016-1834-1-git-send-email-mawilcox@linuxonhyperv.com Signed-off-by: Matthew Wilcox <mawilcox@microsoft.com> Acked-by: Yury Norov <ynorov@caviumnetworks.com> Acked-by: Rasmus Villemoes <linux@rasmusvillemoes.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'lib')
-rw-r--r--lib/find_bit.c4
1 files changed, 2 insertions, 2 deletions
diff --git a/lib/find_bit.c b/lib/find_bit.c
index 18072ea9c20e..6ed74f78380c 100644
--- a/lib/find_bit.c
+++ b/lib/find_bit.c
@@ -33,7 +33,7 @@ static unsigned long _find_next_bit(const unsigned long *addr,
{
unsigned long tmp;
- if (!nbits || start >= nbits)
+ if (unlikely(start >= nbits))
return nbits;
tmp = addr[start / BITS_PER_LONG] ^ invert;
@@ -151,7 +151,7 @@ static unsigned long _find_next_bit_le(const unsigned long *addr,
{
unsigned long tmp;
- if (!nbits || start >= nbits)
+ if (unlikely(start >= nbits))
return nbits;
tmp = addr[start / BITS_PER_LONG] ^ invert;