[Fontconfig] fontconfig: Branch 'master' - 23 commits

[Behdad Esfahbod]

Oh, my bad.  That was against an old fontconfig.  With new one I see malloc
failures, perhaps because of a memory corruption...  Investigating.

On 13-01-07 08:23 PM, Behdad Esfahbod wrote:
> Ok, with your test I see consistent crashes with cache cleanup.  Working on
> it.  Is that the one you are seeing?
> 
> On 13-01-07 08:44 AM, Raimund Steger wrote:
>> On Mon, January 7, 2013 12:37, Raimund Steger wrote:
>>>
>>> On Mon, January 7, 2013 04:43, Behdad Esfahbod wrote:
>>>> On 13-01-06 09:15 AM, Raimund Steger wrote:
>>>>> Behdad Esfahbod wrote:
>>>>>> [...]
>>>>>
>>>>>> I'm not sure whether this is a problem for the way fontconfig uses
>>>>>> these
>>>>> operations. So far I've not been able to attribute any coredumps to
>>>>> _this_
>>>>> circumstance. This would need attention from an expert on the subject
>>>>> (which
>>>>> I'm not).
>>>>
>>>> So, did the pthread fixes make the crashes to go away or are you still
>>>> seeing
>>>> some?
>>>
>>> Unfortunately, I still see crashes. While I think the aforementioned
>>> compiler flags for Sun Studio are still necessary to enable threadsafe
>>> behavior of the standard library (well but maybe Akira can have a word
>>> where to put them...), this might not have been the main problem after
>>> all. I'll see if I can implement a shorter test snippet and send it in,
>>> maybe it's just my buggy test program :-)
>>
>> I've attached a program now which is rather simple but produces pretty
>> reproducible coredumps on a test machine here (Sun M5000 w/ 12 CPUs).
>>
>> Maybe you could first check whether the test program isn't complete
>> garbage...
>>
>> Raimund
>>
>>
>>
>>
>>
> 

-- 
behdad
http://behdad.org/