This is the mail archive of the
mailing list for the pthreas-win32 project.
RE: Pthreads-w32 on Win2k+Cygwin env.
- To: 'Ildar Mulyukov' <ildar at sendmail dot ru>, pthreads-win32 at sources dot redhat dot com
- Subject: RE: Pthreads-w32 on Win2k+Cygwin env.
- From: "Bossom, John" <John dot Bossom at Cognos dot COM>
- Date: Mon, 4 Jun 2001 10:58:37 -0400
I am surprised to see a call to pthread_mutex_trylock from within
I am also surprised to see a call to pthread_self from
In the end, you will notice that the key passed into pthread_getspecific is
0 (zero) indicating
that it either hasn't been initialized yet or has already been destroyed.
You did not indicate if you are using the static library or the DLL. I am
betting you are
linking against the static library which, I believe, requires an
initialization and terminate call.
The DLL version is self boot-strapping...
From: Ildar Mulyukov [mailto:email@example.com]
Sent: June 4, 2001 3:28 AM
Subject: Pthreads-w32 on Win2k+Cygwin env.
The thing is I've tried this pthreads-snap-2000-12-29 on my environment.
I have Win2k localized(russian) + Cygwin crosscompiler (to MinGW).
This pthreads doesn't work for me :-(
Libs are built after I've added -mno-cygwin in makefiles (GNUmakefile).
But test mutex1 fails on pthread_mutex_destroy. Gdb backtrace:
#0 0x00403194 in pthread_getspecific (key=0x0) at tsd.c:325
#1 0x00401d8d in pthread_self () at misc.c:161
#2 0x00401cc3 in pthread_mutex_trylock (mutex=0x247ff38) at mutex.c:686
#3 0x004017f8 in pthread_mutex_destroy (mutex=0x4080b4) at mutex.c:216
#4 0x00401535 in main () at mutex1.c:27
Could anyone explain me this behaviour?
Thank you in advance, Ildar.