On 3/6/18 12:00 PM, Sitsofe Wheeler wrote:
> On 6 March 2018 at 18:53, Jens Axboe <axboe@xxxxxxxxx> wrote:
>> On 3/6/18 11:45 AM, Sitsofe Wheeler wrote:
>>> [...]
>>>> +##########################################
>>>> +# sse probe
>>>> +sse="no"
>>>> +cat > $TMPC << EOF
>>>> +#include <xmmintrin.h>
>>>> +#include <immintrin.h>
>>>> +int main(int argc, char **argv)
>>>> +{
>>>> + __m128 val;
>>>> + float const *src = NULL;
>>>> + float *dst = NULL;
>>>> + val = _mm_load_ps(src);
>>>> + _mm_store_ps(dst, val);
>>>> + return 0;
>>>> +}
>>>
>>> We could try and use __builtin_ia32_addps (from
>>> https://gcc.gnu.org/onlinedocs/gcc/x86-Built-in-Functions.html#x86-Built-in-Functions
>>
>> Sure
>>
>>>> +EOF
>>>> +if test "$enable_sse" = "yes" && compile_prog "-msse" "" "sse"; then
>>>> + sse="yes"
>>>> +fi
>>>> +print_config "SSE (compiler)" "$sse"
>>>
>>> Here we wouldn't explicitly add -msse because we would have already
>>> added it to the global CFLAGS if forced or depended on the user's
>>> particular compiler flags if not.
>>
>> It's only enabled IFF the global flag (--enable-sse) is set AND the
>> configure test passes.
>
> Right and I'm saying change the how we add the flag and the configure
> test. If the user explicitly sets sse on the configure line then
> assume the user knows what they're doing and add it to the CFLAGS and
> but always detect (via comping a program that would only work if the
> user really had told the compiler to use SSE) if its already enabled
> later (rather than only if the user turned it on). My thinking is that
> platforms like x86_64 will nearly always have SSE so it would be nice
> to detect if the compiler somehow already has it on.
Ah I see, yes that might make sense. You could already just do that with
configure --extra-cflags="-msse", though.
--
Jens Axboe
--
To unsubscribe from this list: send the line "unsubscribe fio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
