Message ID | AM5PR0802MB26108A29134FE37A5335A38F830A0@AM5PR0802MB2610.eurprd08.prod.outlook.com |
---|---|
State | New |
Headers | show |
On Wed, Apr 5, 2017 at 5:38 AM, Wilco Dijkstra <Wilco.Dijkstra@arm.com> wrote: > Many supported cores use the AUTOPREFETCHER_WEAK setting which tries > to order loads and stores to improve streaming performance. Since significant > gains were reported in http://patchwork.ozlabs.org/patch/534469/ it seems > like a good idea to enable this setting too for -mcpu=generic. Since the > weak model only keeps the order if it doesn't make the schedule worse, it > should not impact performance adversely on cores that don't show a gain. > Any objections? This is ok with me. Thanks, Andrew > > ChangeLog: > 2017-04-05 Wilco Dijkstra <wdijkstr@arm.com> > > * gcc/config/aarch64/aarch64.c (generic_tunings): Update prefetch model. > > -- > diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c > index 8b729b1b1f87316e940d7fc657f235a935ffa93e..b249ce2b310707c7ded2827d505ce2ddfcfbf976 100644 > --- a/gcc/config/aarch64/aarch64.c > +++ b/gcc/config/aarch64/aarch64.c > @@ -547,7 +547,7 @@ static const struct tune_params generic_tunings = > 2, /* min_div_recip_mul_df. */ > 0, /* max_case_values. */ > 0, /* cache_line_size. */ > - tune_params::AUTOPREFETCHER_OFF, /* autoprefetcher_model. */ > + tune_params::AUTOPREFETCHER_WEAK, /* autoprefetcher_model. */ > (AARCH64_EXTRA_TUNE_NONE) /* tune_flags. */ > };
On Wed, Apr 5, 2017 at 5:38 AM, Wilco Dijkstra <Wilco.Dijkstra@arm.com> wrote: > Many supported cores use the AUTOPREFETCHER_WEAK setting which tries > to order loads and stores to improve streaming performance. Since significant > gains were reported in http://patchwork.ozlabs.org/patch/534469/ it seems > like a good idea to enable this setting too for -mcpu=generic. Since the > weak model only keeps the order if it doesn't make the schedule worse, it > should not impact performance adversely on cores that don't show a gain. > Any objections? This is fine with me also. Jim
On 05/04/17 13:38, Wilco Dijkstra wrote: > Many supported cores use the AUTOPREFETCHER_WEAK setting which tries > to order loads and stores to improve streaming performance. Since significant > gains were reported in http://patchwork.ozlabs.org/patch/534469/ it seems > like a good idea to enable this setting too for -mcpu=generic. Since the > weak model only keeps the order if it doesn't make the schedule worse, it > should not impact performance adversely on cores that don't show a gain. > Any objections? > > ChangeLog: > 2017-04-05 Wilco Dijkstra <wdijkstr@arm.com> > > * gcc/config/aarch64/aarch64.c (generic_tunings): Update prefetch model. > OK. The consensus seems to be in favour of this. R. > -- > diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c > index 8b729b1b1f87316e940d7fc657f235a935ffa93e..b249ce2b310707c7ded2827d505ce2ddfcfbf976 100644 > --- a/gcc/config/aarch64/aarch64.c > +++ b/gcc/config/aarch64/aarch64.c > @@ -547,7 +547,7 @@ static const struct tune_params generic_tunings = > 2, /* min_div_recip_mul_df. */ > 0, /* max_case_values. */ > 0, /* cache_line_size. */ > - tune_params::AUTOPREFETCHER_OFF, /* autoprefetcher_model. */ > + tune_params::AUTOPREFETCHER_WEAK, /* autoprefetcher_model. */ > (AARCH64_EXTRA_TUNE_NONE) /* tune_flags. */ > }; >
diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c index 8b729b1b1f87316e940d7fc657f235a935ffa93e..b249ce2b310707c7ded2827d505ce2ddfcfbf976 100644 --- a/gcc/config/aarch64/aarch64.c +++ b/gcc/config/aarch64/aarch64.c @@ -547,7 +547,7 @@ static const struct tune_params generic_tunings = 2, /* min_div_recip_mul_df. */ 0, /* max_case_values. */ 0, /* cache_line_size. */ - tune_params::AUTOPREFETCHER_OFF, /* autoprefetcher_model. */ + tune_params::AUTOPREFETCHER_WEAK, /* autoprefetcher_model. */ (AARCH64_EXTRA_TUNE_NONE) /* tune_flags. */ };