Re: [PATCH,x86] Fix combine for condditional instructions.

[Yuri Rumyantsev <ysrumyan at gmail dot com>]

Hi Uros,

This fix is for all x86 platforms, we tested it on core2/corei7,
atom/atom2 and AMD and got performance improvement +6% -- +11%. So I
don' think we need to introduce additioanl tune feature.

Sorry for my typo with gcc version - I ment mainline only since 4.7
does not use LRA.

Thanks.
Yuri.



2012/12/12 Uros Bizjak <ubizjak@gmail.com>:
> On Wed, Dec 12, 2012 at 12:27 PM, Yuri Rumyantsev <ysrumyan@gmail.com> wrote:
>
>> This fix is aimed to remove performance degradation introduced by new
>> LRA phase that in fact is combining problem. Gcc combiner does
>> propagation of memory load to if-then-else gimple that was splitted
>> back by old reload phase. LRA does not perform such splitting. To
>> avoid performance slowdown on important benchmark (this is true for
>> all x86 targets) we decided to enhance 'ix86_legitimate_combined_insn'
>> with a check on such propagation and consider such conditional
>> instruction with memory operand as illegal one from performance point
>> of view.
>
> Is this true for all x86 targets? I have no objections to the
> implementation, but these fine-tunings should be declared in
> ix86_tune_features[] array, and used as conditions involving
> TARGET_xxx in the code. Please see many examples in the i386 source
> dir.
>
>> Is it OK for 4.8 and mainline?
>
> Hm, currently 4.8 _is_ mainline. Did you mean 4.7?
>
> Thanks,
> Uros.