> > The diffrerence is that Cores understand the fact that fmadd does not need
> > all three parameters to start computation, while Zen cores doesn't.
> >
> > Since this seems noticeable win on zen and not loss on Core it seems like
> > good
> > default for generic.
> >
> > I plan to commit the patch next week if there are no compplains.
> The generic part LGTM.(It's exactly what we proposed in [1])
>
> [1] https://gcc.gnu.org/pipermail/gcc-patches/2023-November/637721.html
Thanks. I wonder if can think of other generic changes that would make
sense to do?
Concerning zen4 and FMA, it is not really win with AVX512 enabled
(which is what I was benchmarking for znver4 tuning), but indeed it is
win with AVX256 where the extra latency is not hidden by the parallelism
exposed by doing evertyhing twice.
I re-benmchmarked zen4 and it behaves similarly to zen3 with avx256, so
for x86-64-v3 this makes sense.
Honza
> >
> > Honza
> >
> > #include <stdio.h>
> > #include <time.h>
> >
> > #define SIZE 1000
> >
> > float a[SIZE][SIZE];
> > float b[SIZE][SIZE];
> > float c[SIZE][SIZE];
> >
> > void init(void)
> > {
> > int i, j, k;
> > for(i=0; i<SIZE; ++i)
> > {
> > for(j=0; j<SIZE; ++j)
> > {
> > a[i][j] = (float)i + j;
> > b[i][j] = (float)i - j;
> > c[i][j] = 0.0f;
> > }
> > }
> > }
> >
> > void mult(void)
> > {
> > int i, j, k;
> >
> > for(i=0; i<SIZE; ++i)
> > {
> > for(j=0; j<SIZE; ++j)
> > {
> > for(k=0; k<SIZE; ++k)
> > {
> > c[i][j] += a[i][k] * b[k][j];
> > }
> > }
> > }
> > }
> >
> > int main(void)
> > {
> > clock_t s, e;
> >
> > init();
> > s=clock();
> > mult();
> > e=clock();
> > printf(" mult took %10d clocks\n", (int)(e-s));
> >
> > return 0;
> >
> > }
> >
> > * confg/i386/x86-tune.def (X86_TUNE_AVOID_128FMA_CHAINS,
> > X86_TUNE_AVOID_256FMA_CHAINS)
> > Enable for znver4 and Core.
> >
> > diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def
> > index 43fa9e8fd6d..74b03cbcc60 100644
> > --- a/gcc/config/i386/x86-tune.def
> > +++ b/gcc/config/i386/x86-tune.def
> > @@ -515,13 +515,13 @@ DEF_TUNE (X86_TUNE_USE_SCATTER_8PARTS,
> > "use_scatter_8parts",
> >
> > /* X86_TUNE_AVOID_128FMA_CHAINS: Avoid creating loops with tight 128bit or
> > smaller FMA chain. */
> > -DEF_TUNE (X86_TUNE_AVOID_128FMA_CHAINS, "avoid_fma_chains", m_ZNVER1 |
> > m_ZNVER2 | m_ZNVER3
> > - | m_YONGFENG)
> > +DEF_TUNE (X86_TUNE_AVOID_128FMA_CHAINS, "avoid_fma_chains", m_ZNVER1 |
> > m_ZNVER2 | m_ZNVER3 | m_ZNVER4
> > + | m_YONGFENG | m_GENERIC)
> >
> > /* X86_TUNE_AVOID_256FMA_CHAINS: Avoid creating loops with tight 256bit or
> > smaller FMA chain. */
> > -DEF_TUNE (X86_TUNE_AVOID_256FMA_CHAINS, "avoid_fma256_chains", m_ZNVER2 |
> > m_ZNVER3
> > - | m_CORE_HYBRID | m_SAPPHIRERAPIDS | m_CORE_ATOM)
> > +DEF_TUNE (X86_TUNE_AVOID_256FMA_CHAINS, "avoid_fma256_chains", m_ZNVER2 |
> > m_ZNVER3 | m_ZNVER4
> > + | m_CORE_HYBRID | m_SAPPHIRERAPIDS | m_CORE_ATOM | m_GENERIC)
> >
> > /* X86_TUNE_AVOID_512FMA_CHAINS: Avoid creating loops with tight 512bit or
> > smaller FMA chain. */
>
>
>
> --
> BR,
> Hongtao
