* Re: [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic
2017-04-05 12:38 [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic Wilco Dijkstra
@ 2017-04-05 16:17 ` Andrew Pinski
2017-04-05 16:44 ` Jim Wilson
` (2 subsequent siblings)
3 siblings, 0 replies; 5+ messages in thread
From: Andrew Pinski @ 2017-04-05 16:17 UTC (permalink / raw)
To: Wilco Dijkstra
Cc: GCC Patches, nd, James Greenhalgh, Andrew.pinski,
Evandro Menezes, jim.wilson
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. */
> };
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic
2017-04-05 12:38 [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic Wilco Dijkstra
2017-04-05 16:17 ` Andrew Pinski
@ 2017-04-05 16:44 ` Jim Wilson
2017-04-20 15:53 ` Wilco Dijkstra
2017-05-04 10:29 ` Richard Earnshaw (lists)
3 siblings, 0 replies; 5+ messages in thread
From: Jim Wilson @ 2017-04-05 16:44 UTC (permalink / raw)
To: Wilco Dijkstra
Cc: GCC Patches, nd, James Greenhalgh, Andrew.pinski, Evandro Menezes
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
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic
2017-04-05 12:38 [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic Wilco Dijkstra
2017-04-05 16:17 ` Andrew Pinski
2017-04-05 16:44 ` Jim Wilson
@ 2017-04-20 15:53 ` Wilco Dijkstra
2017-05-04 10:29 ` Richard Earnshaw (lists)
3 siblings, 0 replies; 5+ messages in thread
From: Wilco Dijkstra @ 2017-04-20 15:53 UTC (permalink / raw)
To: GCC Patches, James Greenhalgh; +Cc: nd
ping
From: Wilco Dijkstra
Sent: 05 April 2017 13:38
To: GCC Patches
Cc: nd; James Greenhalgh; Andrew.pinski@cavium.com; Evandro Menezes; jim.wilson@linaro.org
Subject: [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic
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.
--
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. */
};
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic
2017-04-05 12:38 [PATCH][AArch64] Enable AUTOPREFETCHER_WEAK with -mcpu=generic Wilco Dijkstra
` (2 preceding siblings ...)
2017-04-20 15:53 ` Wilco Dijkstra
@ 2017-05-04 10:29 ` Richard Earnshaw (lists)
3 siblings, 0 replies; 5+ messages in thread
From: Richard Earnshaw (lists) @ 2017-05-04 10:29 UTC (permalink / raw)
To: Wilco Dijkstra, GCC Patches
Cc: nd, James Greenhalgh, Andrew.pinski, Evandro Menezes, jim.wilson
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. */
> };
>
^ permalink raw reply [flat|nested] 5+ messages in thread