Re: Many pthread failures in the test suite, one setgroup failure

[Christopher Faylor <cgf at redhat dot com>]

On Sat, Sep 28, 2002 at 01:46:22AM +1000, Robert Collins wrote:
>Gah! Confused myself but good.
>
>The pthread::create routine *is* a int returning static method. Sigh.
>Ok, from the top again..
>
>pthread_create - won't error itself
>static pthread::create - will EAGAIN if pthread creation fails, and ONLY
>if pthread creation fails. This uses
>new pthread()
>and pthread->create (create params).
>and a check on object validity after the creation.
>
>So my prior email about the logic *is* correct, just not the bit about
>there being a mistake :].
>
>So, you need to strace the api test to see what's failing.

Here's one program that's failing.

Hint:  You have to do a

  handle SIGSEGV nostop noprint

in gdb if you want debug this.

You have to copy this to a directory where a new-cygwin1.dll lives.

cgf

Attachment: pthread-condvar2_1.exe.bz2
Description: BZip2 compressed data