On 03/09/2010 02:02 PM, Sven Joachim wrote:
I will do that anyway (might take some time to find a sponsor, though),
but I would also like to understand the problem. In case it isn't
clear: this assertion error is very weird and might just be a symptom of
a problem that has nothing to do with the symlinks package. Could be a
bug in libc6, the kernel or your hardware. There has been a similar
report in bug #566947¹, except that that bug occurred on rather exotic
hardware.
Come on: the symlinks program has a protected address space that is
protected so nothing is going to corrupt the malloc internal free buffer
management except code executed the program itself. Now, and I can
follow you on libc6, there are other functions that do do return
pointers (but generally its a global/static buffer not a malloced one
and you should not/never call free). But exposing a NEW bug in libc6 is
something strange.
Kernel/hardware/libc6 does not change and the problem vanish.
As I said, the problem may actually be elsewhere. Anyway, I found
indeed a bug in my patch for #61140², but I find it hard to imagine how
that could lead to a malloc assertion error.
a free of an address returned by a non reentrant function could
perfectly cause it (try the _r version with explicit buffer allocation)
-- eric
*********************************
This message and any attachments (the "message") are confidential and intended solely for the addressees.
Any unauthorised use or dissemination is prohibited.
Messages are susceptible to alteration.
France Telecom Group shall not be liable for the message if altered, changed or falsified.
If you are not the intended addressee of this message, please cancel it
immediately and inform the sender.
********************************
