Fix attribute reference in __wrap_pthread_create
PATCH 1/1 Fix attribute reference in __wrap_pthread_create
Hi Gilles,
my rt-cap changes to the posix lib caused another regression. I don't understand what my intention of changing the attribute reference once was (probably an intermediate change), but it was wrong. The attached patch now makes all programs happy that create pthreads without attributes (clocktest e.g.).
Patch against 2.3.x, should be applied to trunk as well.
Jan
Signature · 2 lines
Siemens AG, Corporate Technology, CT SE 2 Corporate Competence Center Embedded Linux
--- ChangeLog | 5 +++++ src/skins/posix/thread.c | 2 +- 2 files changed, 6 insertions(+), 1 deletion(-)
ChangeLog +5 -0
@@ -1,3 +1,8 @@ +2007-09-26 Jan Kiszka <[email protected]> + + * src/skins/posix/thread.c (__wrap_pthread_create): Use the correct + attribute reference when creating the real pthread. + 2007-09-16 Philippe Gerum <[email protected]> * RELEASE: Xenomai 2.3.4 (Cool #9)
src/skins/posix/thread.c +1 -1
@@ -170,7 +170,7 @@ int __wrap_pthread_create(pthread_t *tid iargs.ret =EAGAIN ;__real_sem_init (&iargs.sync,0 ,0 ); - err =__real_pthread_create (<id, &iattr, &__pthread_trampoline, &iargs); + err =__real_pthread_create (<id, attr, &__pthread_trampoline, &iargs);if (!err)while (__real_sem_wait (&iargs.sync) && errno ==EINTR ) ;__real_sem_destroy (&iargs.sync);
Quoted 9 lines · Jan Kiszka
Hi Gilles,
my rt-cap changes to the posix lib caused another regression. I don't understand what my intention of changing the attribute reference once was (probably an intermediate change), but it was wrong. The attached patch now makes all programs happy that create pthreads without attributes (clocktest e.g.).
Patch against 2.3.x, should be applied to trunk as well.
Applied on both branches, thanks.
Signature · 1 line
Gilles Chanteperdrix