anyone familiar with the nuances of clone(2) care to comment? I am running the attached program on a quad machine with a inux version 2.6.11-1.1369_FC4smp (bhcompile@decompose.build.redhat.com) and gcc version 4.0.2 20051125 (Red Hat 4.0.2-8)
and I do not get a single counter increased by multiple threads, which is what I want.
I don't even seem to get four different counters incremented by multiple threads that are on the same CPU. It appears that all the threads are running their own counter, and all starting from 0.
Advice?
Ugh. Something is seriously broken somewhere.
---------------------
[hald@iolo ~]$ strace -o /dev/null ./clone 0 [hald@iolo ~]$ strace -f -o /dev/null ./clone 4
kernel: 2.6.12-2.3.legacy_FC3 gcc: 3.4.4 20050721 (Red Hat 3.4.4-2) libc: 2.3.6-0.fc3.1
---------------------
[hald@zeus ~]$ strace -o /dev/null ./clone 0 [hald@zeus ~]$ strace -o /dev/null -f ./clone 4
kernel: 2.6.17-1.2157_FC5smp gcc: 4.1.1 20060525 (Red Hat 4.1.1-1) libc: 2.4.8
On Thu, Aug 24, 2006 at 03:58:49PM -0500, David Nicol wrote:
anyone familiar with the nuances of clone(2) care to comment? I am running the attached program on a quad machine with a Linux version 2.6.11-1.1369_FC4smp (bhcompile@decompose.build.redhat.com) and gcc version 4.0.2 20051125 (Red Hat 4.0.2-8)
and I do not get a single counter increased by multiple threads, which is what I want.
I don't even seem to get four different counters incremented by multiple threads that are on the same CPU. It appears that all the threads are running their own counter, and all starting from 0.
Advice?
-- David L Nicol Dickenson on the flag http://cronos.advenge.com/pc/EmilyDickenson/SecondBook/p39.html _______________________________________________ Kclug mailing list Kclug@kclug.org http://kclug.org/mailman/listinfo/kclug