Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1219149 > unrolled thread

[RFC PATCH] arm64: cpuinfo: reduce cache contention on update_{feature}_support

Back to article view | Back to linux.kernel

Contents

  [RFC PATCH] arm64: cpuinfo: reduce cache contention on update_{feature}_support Yury Norov <ynorov@caviumnetworks.com> - 2015-09-04 18:10 +0200
    Re: [RFC PATCH] arm64: cpuinfo: reduce cache contention on update_{feature}_support David Daney <ddaney.cavm@gmail.com> - 2015-09-04 18:40 +0200
    Re: [RFC PATCH] arm64: cpuinfo: reduce cache contention on update_{feature}_support "Suzuki K. Poulose" <Suzuki.Poulose@arm.com> - 2015-09-04 18:50 +0200
      Re: [RFC PATCH] arm64: cpuinfo: reduce cache contention on  update_{feature}_support Yury Norov <ynorov@caviumnetworks.com> - 2015-09-04 22:00 +0200

#1219149 — [RFC PATCH] arm64: cpuinfo: reduce cache contention on update_{feature}_support

This patch is on top of https://lkml.org/lkml/2015/9/2/413

In master, there's only a single function -
	update_mixed_endian_el0_support
And similar function is on review mentioned above.

The algorithm for them is like this:
 - there's system-wide boolean marker for the feature that is
   initially enabled;
 - there's also updater for the feature that may disable it
   system-widely if feature is not supported on current CPU.
 - updater is called for each CPU on bootup.

The problem is the way updater does its work. On each CPU, it
unconditionally updates system-wide marker. For multi-core
system it makes CPU issue invalidate message for a cache
line containing marker. This invalidate increases cache
contention for nothing, because there's a single marker reset
that is really needed, and the others are useless.

If the number of system-wide markers of this sort will grow,
it may become a trouble on large-scale SOCs. The fix is trivial,
though: do system-wide marker update conditionally, and preserve
corresponding cache line in shared state for all update() calls,
except, probably, one.

Signed-off-by: Yury Norov <ynorov@caviumnetworks.com>
---
 arch/arm64/kernel/cpuinfo.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/arm64/kernel/cpuinfo.c b/arch/arm64/kernel/cpuinfo.c
index 4a6ae31..9972c1e 100644
--- a/arch/arm64/kernel/cpuinfo.c
+++ b/arch/arm64/kernel/cpuinfo.c
@@ -87,12 +87,14 @@ bool system_supports_aarch32_el0(void)
 
 static void update_mixed_endian_el0_support(struct cpuinfo_arm64 *info)
 {
-	mixed_endian_el0 &= id_aa64mmfr0_mixed_endian_el0(info->reg_id_aa64mmfr0);
+	if (mixed_endian_el0 && !id_aa64mmfr0_mixed_endian_el0(info->reg_id_aa64mmfr0))
+		mixed_endian_el0 = false;
 }
 
 static void update_aarch32_el0_support(struct cpuinfo_arm64 *info)
 {
-	aarch32_el0 &= id_aa64pfr0_aarch32_el0(info->reg_id_aa64pfr0);
+	if (aarch32_el0 && !id_aa64pfr0_aarch32_el0(info->reg_id_aa64pfr0))
+		aarch32_el0 = false;
 }
 
 static void update_cpu_features(struct cpuinfo_arm64 *info)
-- 
2.1.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1219155

On 09/04/2015 09:04 AM, Yury Norov wrote:
> This patch is on top of https://lkml.org/lkml/2015/9/2/413
>
> In master, there's only a single function -
> 	update_mixed_endian_el0_support
> And similar function is on review mentioned above.
>
> The algorithm for them is like this:
>   - there's system-wide boolean marker for the feature that is
>     initially enabled;
>   - there's also updater for the feature that may disable it
>     system-widely if feature is not supported on current CPU.
>   - updater is called for each CPU on bootup.
>
> The problem is the way updater does its work. On each CPU, it
> unconditionally updates system-wide marker. For multi-core
> system it makes CPU issue invalidate message for a cache
> line containing marker. This invalidate increases cache
> contention for nothing, because there's a single marker reset
> that is really needed, and the others are useless.
>
> If the number of system-wide markers of this sort will grow,
> it may become a trouble on large-scale SOCs. The fix is trivial,
> though: do system-wide marker update conditionally, and preserve
> corresponding cache line in shared state for all update() calls,
> except, probably, one.
>
> Signed-off-by: Yury Norov <ynorov@caviumnetworks.com>
> ---
>   arch/arm64/kernel/cpuinfo.c | 6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/arch/arm64/kernel/cpuinfo.c b/arch/arm64/kernel/cpuinfo.c
> index 4a6ae31..9972c1e 100644
> --- a/arch/arm64/kernel/cpuinfo.c
> +++ b/arch/arm64/kernel/cpuinfo.c
> @@ -87,12 +87,14 @@ bool system_supports_aarch32_el0(void)
>
>   static void update_mixed_endian_el0_support(struct cpuinfo_arm64 *info)
>   {
> -	mixed_endian_el0 &= id_aa64mmfr0_mixed_endian_el0(info->reg_id_aa64mmfr0);
> +	if (mixed_endian_el0 && !id_aa64mmfr0_mixed_endian_el0(info->reg_id_aa64mmfr0))
> +		mixed_endian_el0 = false;
>   }
>
>   static void update_aarch32_el0_support(struct cpuinfo_arm64 *info)
>   {
> -	aarch32_el0 &= id_aa64pfr0_aarch32_el0(info->reg_id_aa64pfr0);
> +	if (aarch32_el0 && !id_aa64pfr0_aarch32_el0(info->reg_id_aa64pfr0))
> +		aarch32_el0 = false;
>   }
>

How many times in the lifetime of the kernel are these functions called?

If it is just done at startup, then there is no "steady state" 
performance impact, and the burden of complicating the code may not be 
worthwhile.

David Daney


>   static void update_cpu_features(struct cpuinfo_arm64 *info)
>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1219168

On 04/09/15 17:04, Yury Norov wrote:
> This patch is on top of https://lkml.org/lkml/2015/9/2/413
>
> In master, there's only a single function -
> 	update_mixed_endian_el0_support
> And similar function is on review mentioned above.
>
> The algorithm for them is like this:
>   - there's system-wide boolean marker for the feature that is
>     initially enabled;
>   - there's also updater for the feature that may disable it
>     system-widely if feature is not supported on current CPU.
>   - updater is called for each CPU on bootup.
>
> The problem is the way updater does its work. On each CPU, it
> unconditionally updates system-wide marker. For multi-core
> system it makes CPU issue invalidate message for a cache
> line containing marker. This invalidate increases cache
> contention for nothing, because there's a single marker reset
> that is really needed, and the others are useless.
>
> If the number of system-wide markers of this sort will grow,
> it may become a trouble on large-scale SOCs. The fix is trivial,
> though: do system-wide marker update conditionally, and preserve
> corresponding cache line in shared state for all update() calls,
> except, probably, one.

As I have mentioned already, this patch (and the per feature functions)
won't be needed once we merge my series (which is waiting for the merge
window to see the public lights)

Cheers
Suzuki

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1219278 — Re: [RFC PATCH] arm64: cpuinfo: reduce cache contention on update_{feature}_support

On Fri, Sep 04, 2015 at 05:40:57PM +0100, Suzuki K. Poulose wrote:
> On 04/09/15 17:04, Yury Norov wrote:
> >This patch is on top of https://lkml.org/lkml/2015/9/2/413
> >
> >In master, there's only a single function -
> >	update_mixed_endian_el0_support
> >And similar function is on review mentioned above.
> >
> >The algorithm for them is like this:
> >  - there's system-wide boolean marker for the feature that is
> >    initially enabled;
> >  - there's also updater for the feature that may disable it
> >    system-widely if feature is not supported on current CPU.
> >  - updater is called for each CPU on bootup.
> >
> >The problem is the way updater does its work. On each CPU, it
> >unconditionally updates system-wide marker. For multi-core
> >system it makes CPU issue invalidate message for a cache
> >line containing marker. This invalidate increases cache
> >contention for nothing, because there's a single marker reset
> >that is really needed, and the others are useless.
> >
> >If the number of system-wide markers of this sort will grow,
> >it may become a trouble on large-scale SOCs. The fix is trivial,
> >though: do system-wide marker update conditionally, and preserve
> >corresponding cache line in shared state for all update() calls,
> >except, probably, one.
> 
> As I have mentioned already, this patch (and the per feature functions)
> won't be needed once we merge my series (which is waiting for the merge
> window to see the public lights)
> 

OK. Than waiting for your patchset.

BR,
Yury

> Cheers
> Suzuki
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web