ARGH pthreads tests failing?

Thomas Pfaff
Wed Jan 22 09:54:00 GMT 2003

On Tue, 21 Jan 2003, Christopher Faylor wrote:

> FAIL: pthread/mutex5.c (execute)
> FAIL: pthread/mutex6e.c (execute)
> FAIL: pthread/mutex7e.c (execute)
> FAIL: pthread/mutex6r.c (execute)
> FAIL: pthread/mutex7r.c (execute)
> NOTE: error deleting pthread-cancel9.exe: error deleting "pthread-cancel9.exe": permission denied
> Can anyone else duplicate this?  Should I send the detailed logs?
> cgf
> (trying to release 1.3.19)

Here are my test results with the current cvs.

Test Run By pfaff on Wed Jan 22 10:05:39 2003
Native configuration is i686-pc-cygwin

		=== winsup tests ===
Schedule of variations:

Running target unix
Running /usr/src/cygwin/cvs/src/winsup/testsuite/winsup.api/winsup.exp ...
PASS: pthread/cancel9.c
PASS: pthread/mutex1.c
PASS: pthread/mutex1d.c
PASS: pthread/mutex1e.c
PASS: pthread/mutex1r.c
PASS: pthread/mutex2.c
PASS: pthread/mutex3.c
PASS: pthread/mutex4.c
PASS: pthread/mutex5.c
PASS: pthread/mutex6d.c
PASS: pthread/mutex6e.c
PASS: pthread/mutex6r.c
PASS: pthread/mutex7.c
PASS: pthread/mutex7d.c
PASS: pthread/mutex7e.c
PASS: pthread/mutex7r.c

		=== winsup Summary ===

# of expected passes 238
# of unexpected failures 1
# of unexpected successes 1
# of expected failures 20

Binary file

CYGWIN=nowinsymlinks notitle ntsec ntea nosmbntsec notty binmode

I can only guess here but it might be that the wrong pthread.h is
included in your tests. I remember some mails about -include does not
behave properly with gcc-3, it will always take the headers from
/usr/include first.

Since the default pthread mutex type has changed in pthread.h this might
be the reason why the mutex tests fail on your box.


More information about the Cygwin-developers mailing list