diff options
| author | Bob Duff <duff@adacore.com> | 2014-01-31 15:43:41 +0000 |
|---|---|---|
| committer | Arnaud Charlet <charlet@gcc.gnu.org> | 2014-01-31 16:43:41 +0100 |
| commit | 5b0e6852b1c277ca2e74ba50de7704fb740a2e5e (patch) | |
| tree | c02aaddd6a80c0a096bc5d45fcc49a67da8d7606 /gcc/tree-inline.c | |
| parent | a54ffd6cb9056b98ec4f188122b370e29ddd30d0 (diff) | |
| download | gcc-5b0e6852b1c277ca2e74ba50de7704fb740a2e5e.zip gcc-5b0e6852b1c277ca2e74ba50de7704fb740a2e5e.tar.gz gcc-5b0e6852b1c277ca2e74ba50de7704fb740a2e5e.tar.bz2 | |
s-taskin.ads: Minor comment fix.
2014-01-31 Bob Duff <duff@adacore.com>
* s-taskin.ads: Minor comment fix.
* s-tassta.adb (Abort_Dependents): Don't abort all dependents;
just direct dependents. If this is actually an abort, each task
will take care of aborting its dependents, so all dependents will
get aborted, as before. However, when this is called the second
time from Vulnerable_Complete_Master "for convenience" (i.e. to
kill off tasks waiting at terminate alternatives), aborting
indirect dependents is wrong, because it causes some unrelated
tasks to get aborted.
From-SVN: r207350
Diffstat (limited to 'gcc/tree-inline.c')
0 files changed, 0 insertions, 0 deletions