diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2011-05-22 14:30:36 -0700 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2011-05-22 14:30:36 -0700 |
| commit | 281dc5c5ec0fb299514567cbc358562649c1af95 (patch) | |
| tree | 2ecb019795e99147ce8d90b5e97c2bc3051f7855 | |
| parent | 71a8638480eb8fb6cfabe2ee9ca3fbc6e3453a14 (diff) | |
| download | linux-281dc5c5ec0fb299514567cbc358562649c1af95.tar.gz linux-281dc5c5ec0fb299514567cbc358562649c1af95.tar.bz2 linux-281dc5c5ec0fb299514567cbc358562649c1af95.zip | |
Give up on pushing CC_OPTIMIZE_FOR_SIZE
I still happen to believe that I$ miss costs are a major thing, but
sadly, -Os doesn't seem to be the solution. With or without it, gcc
will miss some obvious code size improvements, and with it enabled gcc
will sometimes make choices that aren't good even with high I$ miss
ratios.
For example, with -Os, gcc on x86 will turn a 20-byte constant memcpy
into a "rep movsl". While I sincerely hope that x86 CPU's will some day
do a good job at that, they certainly don't do it yet, and the cost is
higher than a L1 I$ miss would be.
Some day I hope we can re-enable this.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
| -rw-r--r-- | init/Kconfig | 1 |
1 files changed, 0 insertions, 1 deletions
diff --git a/init/Kconfig b/init/Kconfig index 4986ecc49e65..ffcdad793f14 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -908,7 +908,6 @@ endif config CC_OPTIMIZE_FOR_SIZE bool "Optimize for size" - default y help Enabling this option will pass "-Os" instead of "-O2" to gcc resulting in a smaller kernel. |