Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1318095 > unrolled thread
| Started by | Waiman Long <Waiman.Long@hpe.com> |
|---|---|
| First post | 2016-01-26 17:10 +0100 |
| Last post | 2016-01-27 21:40 +0100 |
| Articles | 4 — 4 participants |
Back to article view | Back to linux.kernel
This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by
below is the oldest one visible, not the original post.
[RFC PATCH 2/3] lib/list_batch, x86: Enable list insertion/deletion batching in x86-64 Waiman Long <Waiman.Long@hpe.com> - 2016-01-26 17:10 +0100
Re: [RFC PATCH 2/3] lib/list_batch, x86: Enable list insertion/deletion batching in x86-64 Andi Kleen <andi@firstfloor.org> - 2016-01-26 22:50 +0100
Re: [RFC PATCH 2/3] lib/list_batch, x86: Enable list insertion/deletion batching in x86-64 Peter Zijlstra <peterz@infradead.org> - 2016-01-27 17:40 +0100
Re: [RFC PATCH 2/3] lib/list_batch, x86: Enable list insertion/deletion batching in x86-64 Waiman Long <waiman.long@hpe.com> - 2016-01-27 21:40 +0100
| From | Waiman Long <Waiman.Long@hpe.com> |
|---|---|
| Date | 2016-01-26 17:10 +0100 |
| Subject | [RFC PATCH 2/3] lib/list_batch, x86: Enable list insertion/deletion batching in x86-64 |
| Message-ID | <qVelP-6xm-17@gated-at.bofh.it> |
System with a large number of CPUs will benefit from the list batching facility. This patch enables it for x86-64 as 32-bit i386 is not likely to get much gain from it. Signed-off-by: Waiman Long <Waiman.Long@hpe.com> --- arch/x86/Kconfig | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 330e738..443e41d 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -42,6 +42,7 @@ config X86 select ARCH_SUPPORTS_NUMA_BALANCING if X86_64 select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if X86_64 + select ARCH_USE_LIST_BATCHING if X86_64 select ARCH_USE_QUEUED_RWLOCKS select ARCH_USE_QUEUED_SPINLOCKS select ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH if SMP -- 1.7.1
[toc] | [next] | [standalone]
| From | Andi Kleen <andi@firstfloor.org> |
|---|---|
| Date | 2016-01-26 22:50 +0100 |
| Message-ID | <qVjES-1TK-19@gated-at.bofh.it> |
| In reply to | #1318095 |
Waiman Long <Waiman.Long@hpe.com> writes: > > diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig > index 330e738..443e41d 100644 > --- a/arch/x86/Kconfig > +++ b/arch/x86/Kconfig > @@ -42,6 +42,7 @@ config X86 > select ARCH_SUPPORTS_NUMA_BALANCING if X86_64 > select ARCH_USE_BUILTIN_BSWAP > select ARCH_USE_CMPXCHG_LOCKREF if X86_64 > + select ARCH_USE_LIST_BATCHING if X86_64 I would make it unconditional. The code is simple enough and shouldn't have drawbacks on smaller systems. -Andi
[toc] | [prev] | [next] | [standalone]
| From | Peter Zijlstra <peterz@infradead.org> |
|---|---|
| Date | 2016-01-27 17:40 +0100 |
| Subject | Re: [RFC PATCH 2/3] lib/list_batch, x86: Enable list insertion/deletion batching in x86-64 |
| Message-ID | <qVBir-6lh-23@gated-at.bofh.it> |
| In reply to | #1318420 |
On Tue, Jan 26, 2016 at 01:44:13PM -0800, Andi Kleen wrote: > Waiman Long <Waiman.Long@hpe.com> writes: > > > > diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig > > index 330e738..443e41d 100644 > > --- a/arch/x86/Kconfig > > +++ b/arch/x86/Kconfig > > @@ -42,6 +42,7 @@ config X86 > > select ARCH_SUPPORTS_NUMA_BALANCING if X86_64 > > select ARCH_USE_BUILTIN_BSWAP > > select ARCH_USE_CMPXCHG_LOCKREF if X86_64 > > + select ARCH_USE_LIST_BATCHING if X86_64 > > I would make it unconditional. The code is simple enough > and shouldn't have drawbacks on smaller systems. I agree with the sentiment but would like to see a benchmark done on a small system to verify all the same.
[toc] | [prev] | [next] | [standalone]
| From | Waiman Long <waiman.long@hpe.com> |
|---|---|
| Date | 2016-01-27 21:40 +0100 |
| Subject | Re: [RFC PATCH 2/3] lib/list_batch, x86: Enable list insertion/deletion batching in x86-64 |
| Message-ID | <qVF2G-Kd-13@gated-at.bofh.it> |
| In reply to | #1318420 |
On 01/26/2016 04:44 PM, Andi Kleen wrote: > Waiman Long<Waiman.Long@hpe.com> writes: >> diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig >> index 330e738..443e41d 100644 >> --- a/arch/x86/Kconfig >> +++ b/arch/x86/Kconfig >> @@ -42,6 +42,7 @@ config X86 >> select ARCH_SUPPORTS_NUMA_BALANCING if X86_64 >> select ARCH_USE_BUILTIN_BSWAP >> select ARCH_USE_CMPXCHG_LOCKREF if X86_64 >> + select ARCH_USE_LIST_BATCHING if X86_64 > I would make it unconditional. The code is simple enough > and shouldn't have drawbacks on smaller systems. > > -Andi You are probably right. I will look into that. Cheers, Longman
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web