[PATCH, i386] Introduce switch for Skylake Server CPU.

Uros Bizjak ubizjak@gmail.com
Mon Sep 21 15:41:00 GMT 2015


On Mon, Sep 21, 2015 at 4:27 PM, Jakub Jelinek <jakub@redhat.com> wrote:
> On Mon, Sep 21, 2015 at 05:14:45PM +0300, Kirill Yukhin wrote:
>> Hello,
>> This patch introduces switches necessary for new Intel Server CPU
>> (code-named Skylake).
>>
>> Bootstrapped & regtested.
>>
>> Is it ok for trunk?
>>
>> gcc/
>>       * config.gcc: Support "skx".
>>         * config/i386/i386-c.c (ix86_target_macros_internal): Handle
>>         PROCESSOR_SKX.
>>         * config/i386/i386.c (m_SKX): Define.
>>         (processor_target_table): Add "skx".
>>         (PTA_SKX): Define.
>>       (ix86_option_override_internal): Add "skx".
>>         (fold_builtin_cpu): Handle "skx".
>>         * config/i386/i386.h (TARGET_SKX): Define.
>>         (processor_type): Add PROCESSOR_SKX.
>>         * config/i386/i386.md (attr "cpu"): Add knl.
>>         * config/i386/x86-tune.def: Add m_KNL.
>>
>> gcc/testsuite/
>>         * gcc.target/i386/funcspec-5.c: Test avx512vl, avx512bw,
>>       avx512dq, avx512cd, avx512er, avx512pf and skx.
>
> Is it a good idea to introduce further abbrevs like this?
> I thought we went away from e.g. slm to silvermont, we didn't
> introduce hsw but haswell, etc.
> So, wouldn't it be better to add skylake-xeon instead of skx?

Those "short names" were invented due to some
restrictions/recommendations on the usage of internal project names.
These restrictions were lifted years ago, so we are free to add more
descriptive names. As Jakub suggested, "skylake-xeon" is IMO much
better.

Uros.



More information about the Gcc-patches mailing list