Re: [PATCH] x86, percpu: revert commit fe8e0c25

[Tejun Heo <tj@kernel.org>]

Hello,

On 10/26/2010 03:25 PM, Ingo Molnar wrote:
> 
> * Tejun Heo <tj@kernel.org> wrote:
> 
>> Commit fe8e0c25 (x86, 32-bit: Align percpu area and irq stacks to THREAD_SIZE) 
>> aligned PERCPU section to THREAD_SIZE which can be larger than PAGE_SIZE, 
>> introduced DEFINE_PER_CPU_MULTIPAGE_ALIGNED() and used it to make irq stacks 
>> aligned to THREAD_SIZE on x86_32.
>>
>> This won't work.  The PERCPU output section is used as the template to prepare the 
>> percpu area and the actual percpu area is _alwasy_ aligned to PAGE_SIZE whether 
>> the source area is aligned to larger size or not.
> 
> The problem is, this will reintroduce a nasty boot crash which commit fe8e0c25 
> fixed. Do you say that fe8e0c25 didnt have the alignment effect?

AFAICS, not in a way which is correct.  The patch probably made the
following two differences.

* The stack in the template area is THREAD_SIZE aligned.  If something
  was dereferencing it before percpu init, this could have helped.
  IIRC, x86 early init code does use the template area.

* The percpu address would be THREAD_SIZE aligned while the translated
  kernel address for each cpu wouldn't be.  For masking stack pointer
  to find out task struct, I don't think aligning the percpu address
  would have been helpful.

It's simply broken and needs to be reverted.  If the patch somehow
fixed boot crash, yeah, we probably want to put a fix for it first
tho.

Thanks.

-- 
tejun