Merge lp:~aacid/mir/eintr_dispatch_loop into lp:mir
| Status: | Merged |
|---|---|
| Merged at revision: | 3120 |
| Proposed branch: | lp:~aacid/mir/eintr_dispatch_loop |
| Merge into: | lp:mir |
| Diff against target: |
12 lines (+2/-0) 1 file modified
src/common/dispatch/threaded_dispatcher.cpp (+2/-0) |
| To merge this branch: | bzr merge lp:~aacid/mir/eintr_dispatch_loop |
| Related bugs: |
| Reviewer | Review Type | Date Requested | Status |
|---|---|---|---|
| PS Jenkins bot | continuous-integration | Needs Fixing on 2015-11-20 | |
| Alan Griffiths | Approve on 2015-11-19 | ||
| Chris Halse Rogers | Approve on 2015-11-18 | ||
| Kevin DuBois (community) | Approve on 2015-11-18 | ||
| Alberto Aguirre | 2015-11-18 | Approve on 2015-11-18 | |
|
Review via email:
|
|||
Commit Message
Do not throw exception on EINTR when poll()ing
| PS Jenkins bot (ps-jenkins) wrote : | # |
FAILED: Continuous integration, rev:3104
http://
Executed test runs:
SUCCESS: http://
FAILURE: http://
FAILURE: http://
FAILURE: http://
SUCCESS: http://
deb: http://
None: http://
SUCCESS: http://
deb: http://
FAILURE: http://
SUCCESS: http://
FAILURE: http://
Click here to trigger a rebuild:
http://
| Kevin DuBois (kdub) wrote : | # |
> Oh yeah we used to have this with the single thread dispatcher.
>
> And we have a test for this but it's disabled.
is there a bug about the disabled test? lgtm otherwise
| Chris Halse Rogers (raof) wrote : | # |
Wait - how did you manage to get an EINTR in the dispatch thread?
We *should* be blocking all signals to that thread. SIGSTOP and SIGKILL can't be blocked, but SIGKILL will immediately end the process so that's not the problem.
We *do* send SIGSTOP to various things, but poll should *not* return EINTR after a SIGSTOP/SIGCONT pair. epoll() (erroneously) does, but poll() is not documented as suffering from that bug.
Oooh! Unless something is registering a SIGCONT handler. Stupid signal behaviour!
Quoth the signal (7) manpage:
Interruption of system calls and library functions by stop signals
On Linux, even in the absence of signal handlers, certain blocking interfaces can fail with the
error EINTR after the process is stopped by one of the stop signals and then resumed via SIG‐
CONT. This behavior is not sanctioned by POSIX.1, and doesn't occur on other systems.
The Linux interfaces that display this behavior are:
* "Input" socket interfaces, when a timeout (SO_RCVTIMEO) has been set on the socket using
* "Output" socket interfaces, when a timeout (SO_RCVTIMEO) has been set on the socket using
TIMEO) has been set.
* epoll_wait(2), epoll_pwait(2).
* semop(2), semtimedop(2).
* sigtimedwait(2), sigwaitinfo(2).
* read(2) from an inotify(7) file descriptor.
* Linux 2.6.21 and earlier: futex(2) FUTEX_WAIT, sem_timedwait(3), sem_wait(3).
* Linux 2.6.8 and earlier: msgrcv(2), msgsnd(2).
* Linux 2.4 and earlier: nanosleep(2).
| PS Jenkins bot (ps-jenkins) wrote : | # |
FAILED: Autolanding.
More details in the following jenkins job:
http://
Executed test runs:
SUCCESS: http://
FAILURE: http://
SUCCESS: http://
FAILURE: http://
SUCCESS: http://
deb: http://
SUCCESS: http://
deb: http://
SUCCESS: http://
deb: http://
SUCCESS: http://
SUCCESS: http://
FAILURE: http://
| PS Jenkins bot (ps-jenkins) wrote : | # |
FAILED: Autolanding.
More details in the following jenkins job:
http://
Executed test runs:
FAILURE: http://
SUCCESS: http://
SUCCESS: http://
FAILURE: http://
SUCCESS: http://
deb: http://
None: http://
SUCCESS: http://
deb: http://
SUCCESS: http://
SUCCESS: http://
FAILURE: http://
Oh yeah we used to have this with the single thread dispatcher.
And we have a test for this but it's disabled.