From e4afd2e5567fc5d59988025f7528f9b4794d86a5 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 24 Feb 2017 15:00:58 -0800 Subject: [PATCH] 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 Acked-by: Yury Norov Acked-by: Rasmus Villemoes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- lib/find_bit.c | 4 ++-- tools/lib/find_bit.c | 2 +- 2 files changed, 3 insertions(+), 3 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; diff --git a/tools/lib/find_bit.c b/tools/lib/find_bit.c index 6d8b8f22cf55..42c15f906aac 100644 --- a/tools/lib/find_bit.c +++ b/tools/lib/find_bit.c @@ -34,7 +34,7 @@ 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;