* [PATCH] x86, tlb: small cleanup for native_flush_tlb_others
@ 2011-03-14 6:55 Xiao Guangrong
2011-03-14 12:33 ` Ingo Molnar
0 siblings, 1 reply; 5+ messages in thread
From: Xiao Guangrong @ 2011-03-14 6:55 UTC (permalink / raw)
To: Ingo Molnar; +Cc: hpa, LKML
Use smp_processor_id instead of get_cpu and put_cpu since
the caller has already disabled preemption
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
---
arch/x86/mm/tlb.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 6acc724..92849d0 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -211,11 +211,10 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
if (is_uv_system()) {
unsigned int cpu;
- cpu = get_cpu();
+ cpu = smp_processor_id();
cpumask = uv_flush_tlb_others(cpumask, mm, va, cpu);
if (cpumask)
flush_tlb_others_ipi(cpumask, mm, va);
- put_cpu();
return;
}
flush_tlb_others_ipi(cpumask, mm, va);
--
1.7.4
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] x86, tlb: small cleanup for native_flush_tlb_others
2011-03-14 6:55 [PATCH] x86, tlb: small cleanup for native_flush_tlb_others Xiao Guangrong
@ 2011-03-14 12:33 ` Ingo Molnar
2011-03-14 12:39 ` Ingo Molnar
2011-03-15 1:57 ` Xiao Guangrong
0 siblings, 2 replies; 5+ messages in thread
From: Ingo Molnar @ 2011-03-14 12:33 UTC (permalink / raw)
To: Xiao Guangrong; +Cc: hpa, LKML
* Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> wrote:
> Use smp_processor_id instead of get_cpu and put_cpu since
> the caller has already disabled preemption
>
> Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
> ---
> arch/x86/mm/tlb.c | 3 +--
> 1 files changed, 1 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
> index 6acc724..92849d0 100644
> --- a/arch/x86/mm/tlb.c
> +++ b/arch/x86/mm/tlb.c
> @@ -211,11 +211,10 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
> if (is_uv_system()) {
> unsigned int cpu;
>
> - cpu = get_cpu();
> + cpu = smp_processor_id();
> cpumask = uv_flush_tlb_others(cpumask, mm, va, cpu);
> if (cpumask)
> flush_tlb_others_ipi(cpumask, mm, va);
> - put_cpu();
> return;
There's multiple callers. Could you please update the changelog with the list of
callers and a statement that all of them disable preemption already?
Thanks,
Ingo
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] x86, tlb: small cleanup for native_flush_tlb_others
2011-03-14 12:33 ` Ingo Molnar
@ 2011-03-14 12:39 ` Ingo Molnar
2011-03-15 1:57 ` Xiao Guangrong
1 sibling, 0 replies; 5+ messages in thread
From: Ingo Molnar @ 2011-03-14 12:39 UTC (permalink / raw)
To: Xiao Guangrong; +Cc: hpa, LKML
Also, the title of the patch should be something like 'small micro-optimization' -
as it's not just a cleanup, it has side-effects on the code.
Thanks,
Ingo
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] x86, tlb: small cleanup for native_flush_tlb_others
2011-03-14 12:33 ` Ingo Molnar
2011-03-14 12:39 ` Ingo Molnar
@ 2011-03-15 1:57 ` Xiao Guangrong
2011-03-15 7:49 ` [tip:x86/mm] x86, tlb, UV: Do small micro-optimization for native_flush_tlb_others() tip-bot for Xiao Guangrong
1 sibling, 1 reply; 5+ messages in thread
From: Xiao Guangrong @ 2011-03-15 1:57 UTC (permalink / raw)
To: Ingo Molnar; +Cc: hpa, LKML
On 03/14/2011 08:33 PM, Ingo Molnar wrote:
>
> * Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> wrote:
>
>> Use smp_processor_id instead of get_cpu and put_cpu since
>> the caller has already disabled preemption
>>
>> Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
>> ---
>> arch/x86/mm/tlb.c | 3 +--
>> 1 files changed, 1 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
>> index 6acc724..92849d0 100644
>> --- a/arch/x86/mm/tlb.c
>> +++ b/arch/x86/mm/tlb.c
>> @@ -211,11 +211,10 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
>> if (is_uv_system()) {
>> unsigned int cpu;
>>
>> - cpu = get_cpu();
>> + cpu = smp_processor_id();
>> cpumask = uv_flush_tlb_others(cpumask, mm, va, cpu);
>> if (cpumask)
>> flush_tlb_others_ipi(cpumask, mm, va);
>> - put_cpu();
>> return;
>
> There's multiple callers. Could you please update the changelog with the list of
> callers and a statement that all of them disable preemption already?
>
Thanks Ingo, there is the new version.
From: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Subject: [PATCH] x86, tlb: small micro-optimization for native_flush_tlb_others
native_flush_tlb_others is called from:
flush_tlb_current_task
flush_tlb_mm
flush_tlb_page
All these functions disable preemption explicitly, so we can use smp_processor_id
instead of get_cpu and put_cpu
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
---
arch/x86/mm/tlb.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 6acc724..92849d0 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -211,11 +211,10 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
if (is_uv_system()) {
unsigned int cpu;
- cpu = get_cpu();
+ cpu = smp_processor_id();
cpumask = uv_flush_tlb_others(cpumask, mm, va, cpu);
if (cpumask)
flush_tlb_others_ipi(cpumask, mm, va);
- put_cpu();
return;
}
flush_tlb_others_ipi(cpumask, mm, va);
--
1.7.4
^ permalink raw reply related [flat|nested] 5+ messages in thread
* [tip:x86/mm] x86, tlb, UV: Do small micro-optimization for native_flush_tlb_others()
2011-03-15 1:57 ` Xiao Guangrong
@ 2011-03-15 7:49 ` tip-bot for Xiao Guangrong
0 siblings, 0 replies; 5+ messages in thread
From: tip-bot for Xiao Guangrong @ 2011-03-15 7:49 UTC (permalink / raw)
To: linux-tip-commits
Cc: linux-kernel, hpa, mingo, cpw, xiaoguangrong, tglx, mingo
Commit-ID: 25542c646afbf14c43fa7d2b443055cadb73b07a
Gitweb: http://git.kernel.org/tip/25542c646afbf14c43fa7d2b443055cadb73b07a
Author: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
AuthorDate: Tue, 15 Mar 2011 09:57:37 +0800
Committer: Ingo Molnar <mingo@elte.hu>
CommitDate: Tue, 15 Mar 2011 08:30:34 +0100
x86, tlb, UV: Do small micro-optimization for native_flush_tlb_others()
native_flush_tlb_others() is called from:
flush_tlb_current_task()
flush_tlb_mm()
flush_tlb_page()
All these functions disable preemption explicitly, so we can use
smp_processor_id() instead of get_cpu() and put_cpu().
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Cc: Cliff Wickman <cpw@sgi.com>
LKML-Reference: <4D7EC791.4040003@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
arch/x86/mm/tlb.c | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 55272d7..d6c0418 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -208,11 +208,10 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
if (is_uv_system()) {
unsigned int cpu;
- cpu = get_cpu();
+ cpu = smp_processor_id();
cpumask = uv_flush_tlb_others(cpumask, mm, va, cpu);
if (cpumask)
flush_tlb_others_ipi(cpumask, mm, va);
- put_cpu();
return;
}
flush_tlb_others_ipi(cpumask, mm, va);
^ permalink raw reply related [flat|nested] 5+ messages in thread
end of thread, other threads:[~2011-03-15 7:49 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-03-14 6:55 [PATCH] x86, tlb: small cleanup for native_flush_tlb_others Xiao Guangrong
2011-03-14 12:33 ` Ingo Molnar
2011-03-14 12:39 ` Ingo Molnar
2011-03-15 1:57 ` Xiao Guangrong
2011-03-15 7:49 ` [tip:x86/mm] x86, tlb, UV: Do small micro-optimization for native_flush_tlb_others() tip-bot for Xiao Guangrong
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox