Re: AVX-512 for LAME: code to try, if you have the hardware

Am 2026-07-26 15:49, schrieb Todd Zullinger:
> Alexander Leidinger via Lame-dev wrote:
>> Am 2026-07-26 12:44, schrieb Tobias Damisch via Lame-dev:
>>> I assume the performance comparisons for stage 2 are obsolete if
>>> "make check" shows fails...?
>> 
>> Please provide test/unit/test-suite.log to understand what exactly 
>> failed.
>> This is more out of curiosity, it will not change the outcome.
> 
> I needed to patch a few test files to include stdint.h on
> Fedora.  Otherwise make check fails.
> 
> Index: test/unit/test_parse_pathlen.c
> ===================================================================
> --- test/unit/test_parse_pathlen.c	(revision 6662)
> +++ test/unit/test_parse_pathlen.c	(working copy)
> @@ -19,6 +19,7 @@
>  #include <stddef.h>
>  #include <setjmp.h>
>  #include <stdlib.h>
> +#include <stdint.h>
>  #include <string.h>
> 
>  #include <cmocka.h>
> Index: test/unit/test_merge_argv.c
> ===================================================================
> --- test/unit/test_merge_argv.c	(revision 6662)
> +++ test/unit/test_merge_argv.c	(working copy)
> @@ -21,6 +21,7 @@
> 
>  #include <stdarg.h>
>  #include <stddef.h>
> +#include <stdint.h>
>  #include <setjmp.h>
> 
>  #include <cmocka.h>

Thanks, I have that already pending on my own...

> The timing tests looked identical to me, using hyperfine
> directly, so I didn't proceed to further tests at this
> point.  I can try to do that sometime today, if they'd be
> helpful.  Without any measuable speed up, I wasn't sure if
> it was worth the effort.

Thanks for the hyperfine run. Perfect!
Could you please do the "make check" for the stage 2? That would be 
enough to give some data for that CPU.

Bye,
Alexander.

-- 
http://www.Leidinger.net [email protected]: PGP 0x8F31830F9F2772BF
http://www.FreeBSD.org    [email protected]  : PGP 0x8F31830F9F2772BF