x86_64 provides an optimized way to determine the local per cpu area
offset through the pda and determines the base by accessing a remote
pda.
Cc: Rusty Russell <[EMAIL PROTECTED]>
Cc: Andi Kleen <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Signed-off-by: Christoph Lameter <[EMA
x86_64 provides an optimized way to determine the local per cpu area
offset through the pda and determines the base by accessing a remote
pda.
Cc: Rusty Russell <[EMAIL PROTECTED]>
Cc: Andi Kleen <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Signed-off-by: Christoph Lameter <[EMA
* Christoph Lameter <[EMAIL PROTECTED]> wrote:
> On Tue, 1 Jan 2008, Ingo Molnar wrote:
>
> > FYI, i tried your patchset on 2.6.24-rc6+x86.git, and randconfig testing
> > found a faulty 32-bit config below - the bootup would spontaneously
> > reboot shortly after hitting user-space. (which sug
On Tue, 1 Jan 2008, Ingo Molnar wrote:
> FYI, i tried your patchset on 2.6.24-rc6+x86.git, and randconfig testing
> found a faulty 32-bit config below - the bootup would spontaneously
> reboot shortly after hitting user-space. (which suggests a triple fault)
> No log messages on the serial cons
On Sun, 30 Dec 2007, Ingo Molnar wrote:
> to get more test feedback: what would be the best way to get this tested
> in x86.git in a standalone way? Can i just pick up these 10 patches and
> remove all the non-x86 arch changes, and expect it to work - or are the
> other percpu preparatory/clean
On Sat, 29 Dec 2007, Andi Kleen wrote:
> Anyways the difference between the x86 percpu.h and the generic one is
> that x86-64 uses a short cut through the PDA to get the current cpu
> offset for the current CPU case. The generic one goes through
> smp_processor_id()->array reference instead.
No
On Fri, 28 Dec 2007, Andi Kleen wrote:
> On Friday 28 December 2007 01:10:51 [EMAIL PROTECTED] wrote:
> > x86_64 provides an optimized way to determine the local per cpu area
> > offset through the pda and determines the base by accessing a remote
> > pda.
>
> And? The rationale for this patch se
* Mike Travis <[EMAIL PROTECTED]> wrote:
> > I've tested some x86_64 configs but the UP model is currently broken
> > so I haven't been able to test that. (the "fs/nfs/super.c" build
> > problem with TASK_NORMAL and TASK_UNINTERRUPTIBLE undefined.)
>
> Sorry, I guess I didn't exactly answer th
Mike Travis wrote:
> Ingo Molnar wrote:
>> * Mike Travis <[EMAIL PROTECTED]> wrote:
>>
Also for such changes .text size comparisons before/after are a good
idea.
>>> x86_64-defconfig:
>>>
>>> pre-percpu post-percpu
>>> 159373 .init.text
Andi Kleen wrote:
> On Friday 28 December 2007 23:05:05 Mike Travis wrote:
>> Andi Kleen wrote:
>>> On Friday 28 December 2007 01:10:51 [EMAIL PROTECTED] wrote:
x86_64 provides an optimized way to determine the local per cpu area
offset through the pda and determines the base by accessing
Ingo Molnar wrote:
> * Mike Travis <[EMAIL PROTECTED]> wrote:
>
>>> Also for such changes .text size comparisons before/after are a good
>>> idea.
>> x86_64-defconfig:
>>
>> pre-percpu post-percpu
>> 159373 .init.text +3 .init.text
>> 1411
* Mike Travis <[EMAIL PROTECTED]> wrote:
> > Also for such changes .text size comparisons before/after are a good
> > idea.
>
> x86_64-defconfig:
>
> pre-percpu post-percpu
> 159373 .init.text +3 .init.text
> 1411137 .rodata
On Friday 28 December 2007 23:05:05 Mike Travis wrote:
> Andi Kleen wrote:
> > On Friday 28 December 2007 01:10:51 [EMAIL PROTECTED] wrote:
> >> x86_64 provides an optimized way to determine the local per cpu area
> >> offset through the pda and determines the base by accessing a remote
> >> pda.
>
Andi Kleen wrote:
> On Friday 28 December 2007 01:10:51 [EMAIL PROTECTED] wrote:
>> x86_64 provides an optimized way to determine the local per cpu area
>> offset through the pda and determines the base by accessing a remote
>> pda.
>
> And? The rationale for this patch seems to be incomplete.
>
>
On Friday 28 December 2007 01:10:51 [EMAIL PROTECTED] wrote:
> x86_64 provides an optimized way to determine the local per cpu area
> offset through the pda and determines the base by accessing a remote
> pda.
And? The rationale for this patch seems to be incomplete.
As far as I can figure out yo
x86_64 provides an optimized way to determine the local per cpu area
offset through the pda and determines the base by accessing a remote
pda.
Cc: Rusty Russell <[EMAIL PROTECTED]>
Cc: Andi Kleen <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Signed-off-by: Christoph Lameter <[EMA
x86_64 provides an optimized way to determine the local per cpu area
offset through the pda and determines the base by accessing a remote
pda.
Cc: Rusty Russell <[EMAIL PROTECTED]>
Cc: Andi Kleen <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Signed-off-by: Christoph Lameter <[EMA
had rejects - hand-merged to x86.git - please check.
Ingo
>
Subject: x86_64: Use generic percpu
From: Christoph Lameter <[EMAIL PROTECTED]>
x86_64 provides an optimized way to determine the local per cpu area
offset through the pda and determines the base by accessing a
x86_64 provides an optimized way to determine the local per cpu area
offset through the pda and determines the base by accessing a remote
pda.
Cc: Rusty Russell <[EMAIL PROTECTED]>
Cc: Andi Kleen <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Signed-off-by: Christoph Lameter <[EMA
19 matches
Mail list logo