Summary: | New check added in pthread_attr_setschedparam() causes hard-to-explain behavior | ||
---|---|---|---|
Product: | glibc | Reporter: | Jiri Dluhos <jdluhos> |
Component: | nptl | Assignee: | Ulrich Drepper <drepper.fsp> |
Status: | RESOLVED DUPLICATE | ||
Severity: | normal | CC: | glibc-bugs |
Priority: | P2 | Flags: | fweimer:
security-
|
Version: | 2.10 | ||
Target Milestone: | --- | ||
Host: | Target: | ||
Build: | Last reconfirmed: |
Description
Jiri Dluhos
2009-10-22 13:40:47 UTC
File a bug with the LSB, as usual their code is broken. I agree, LSB tests are often broken. But in this specific case, I'd say the programmer did nothing wrong. POSIX does not specify that the pthread_attr structure should be filled in a specific order, and the EINVAL error in the case of wrong order is hard to explain (the incorrect priority is not an error *yet*; it would become an error in the moment of thread creation, but before this, the values in the structure can still be changed). Could you please consider this problem once more? If you still have the same opinion, I will open the bug with LSB - but I'm afraid the same problem can occur in another codes, as well. Thank you for your patience. Problem also reported in LSB bugzilla: http://bugs.linuxbase.org/show_bug.cgi?id=2784 (which seems equivalent to http://bugs.linuxbase.org/show_bug.cgi?id=2663) |