diff options
| author | Cheryl Sabella <cheryl.sabella@gmail.com> | 2019-03-04 00:01:39 (GMT) |
|---|---|---|
| committer | larryhastings <larry@hastings.org> | 2019-03-04 00:01:39 (GMT) |
| commit | 2226139aa2b69047cb54dbcfd79f5c2e36f98653 (patch) | |
| tree | 2f223711a6cd167cc426007352bb702460304067 /Python/dynload_stub.c | |
| parent | 765d333512e9b58da4a4431595a0e81517ef0443 (diff) | |
| download | cpython-2226139aa2b69047cb54dbcfd79f5c2e36f98653.zip cpython-2226139aa2b69047cb54dbcfd79f5c2e36f98653.tar.gz cpython-2226139aa2b69047cb54dbcfd79f5c2e36f98653.tar.bz2 | |
[3.4] bpo-33329: Fix multiprocessing regression on newer glibcs (GH-6575) (#12145)
Starting with glibc 2.27.9000-xxx, sigaddset() can return EINVAL for some
reserved signal numbers between 1 and NSIG. The `range(1, NSIG)` idiom
is commonly used to select all signals for blocking with `pthread_sigmask`.
So we ignore the sigaddset() return value until we expose sigfillset()
to provide a better idiom.
(cherry picked from commit 25038ec)
Co-authored-by: Antoine Pitrou <pitrou@free.fr>
Diffstat (limited to 'Python/dynload_stub.c')
0 files changed, 0 insertions, 0 deletions