Re: [git pull] core kernel fixes

["Vegard Nossum" <vegard.nossum@gmail.com>]

On Mon, Jun 30, 2008 at 9:46 PM, Thomas Gleixner <tglx@linutronix.de> wrote:
> On Mon, 30 Jun 2008, Vegard Nossum wrote:
>> On Mon, Jun 30, 2008 at 8:20 PM, Ingo Molnar <mingo@elte.hu> wrote:
>> > The patch was tested with our standard tests so it's certainly good in
>> > practice - but i havent specifically tried your testcase (maybe Thomas
>> > has). Can you see any problem with the fix?
>>
>> Well, what I can see is that the patch that was committed has some
>> missing changes. In Daniel's patch:
>>
>> -repeat:
>> -     spin_lock(&pool_lock);
>> +     spin_lock_irqsave(&pool_lock, flags);
>>       if (obj_pool.first) {
>>               obj         = hlist_entry(obj_pool.first, typeof(*obj), node);
>>
>> The patch that was committed:
>>
>> -repeat:
>>         spin_lock(&pool_lock);
>>         if (obj_pool.first) {
>>                 obj         = hlist_entry(obj_pool.first, typeof(*obj), node);
>>
>> Was it not necessary to make the pool lock irq-safe in this place?
>>
>> For reference:
>>
>> Daniel's patch: http://lkml.org/lkml/2008/6/15/27
>> Actual commit: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=70c85057e0bde35eb56352a293ecb5d1641a0334;hp=e6100f23375c0c71ce595d04551fa6553b611918
>
> alloc_object() is called with interrupts disabled from __debug_object_init()

Ok, thanks for clearing that up! :-)

(I'm just wary of patches that mutate when they have my name on the top...)


Vegard

-- 
"The animistic metaphor of the bug that maliciously sneaked in while
the programmer was not looking is intellectually dishonest as it
disguises that the error is the programmer's own creation."
	-- E. W. Dijkstra, EWD1036