This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[tree-ssa mudflap] per-thread items


I am spending this weekend testing the multithreading side of
mudflap. SO far it looks good, however I am getting some
read violations which I think are related to 'errno'. I have
to do more debugging.

Nevertheless, and as I should hit the bed now, is it possible
that 'errno' is allocated for each thread and as such it cannot
be registered once only at the start of the run? If so, does
the current code handle this?

I think the same question may be valid for other static memory
returned by libc (ctime, gethostbyname etc.).

What is the situation with these?

--
Eyal Lebedinsky (eyal@eyal.emu.id.au) <http://samba.org/eyal/>


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]