This is the mail archive of the
mailing list for the glibc project.
Re: linuxthreads tst-cancel5 failure
- From: Andrew Walrond <andrew at walrond dot org>
- To: Ulrich Drepper <drepper at redhat dot com>
- Cc: libc-alpha at sources dot redhat dot com
- Date: Thu, 06 Mar 2003 23:57:45 +0000
- Subject: Re: linuxthreads tst-cancel5 failure
- References: <3E67BA87.email@example.com> <3E67C0CC.firstname.lastname@example.org>
Ulrich Drepper wrote:
I fail to say which kernel version you use. That's the factor here.
The 2.5 kernels have an unfortunate limitation on sleep intervals these
days. Too long intervals are flagged as errors which isn't a valid
response from what I can see in the moment. Nevertheless, I changed the
NPTL version of the test some time ago. Now I made the samme change to
the linuxthreads test. Update your sources from CVS and run the test
again. It should work now.
You are quite right; I am using 2.5-bk, and your change worked. make
check now gets further, but I see some further failures:
make: *** [/home/ftl/wrk/elf/reldep3.out] Error 1
make: *** Waiting for unfinished jobs....
make: *** [/home/ftl/wrk/elf/nodelete.out] Error 1
make: *** [elf/tests] Error 2
make: Leaving directory `/home/ftl/glibc-HEAD'
make: *** [check] Error 2
ftl at hal3 wrk $ cat elf/reldep3.out
variable "some_var" not reset
ftl at hal3 wrk $ cat elf/nodelete.out
succeeded loading "nodelmod1.so"
-z nodelete test succeeded
succeeded loading "nodelmod2.so"
RTLD_NODELETE test succeeded
succeeded loading "nodelmod3.so"
destructor of "nodelmod4.so" ran
ftl at hal3 wrk $
Any thoughts on these? (I am trying to make sure my environment is
completely sane before rebuilding/debugging the nptl problem reported