*** This bug is a duplicate of bug 797078 ***
https://bugs.launchpad.net/bugs/797078
@C de-Avillez
"Since the OP action was voluntary, we should not have re-subscribed"
I'll be mindful of that in the future.
It's not important that the OP receive this message, but I cannot myself
unsubscribe him.
"it is difficult to say if it is the same bug or not."
To make my reasoning more explicit, removing either #3 or #4 in
https://launchpadlibrarian.net/73496243/metacity.gdb (bug 797098) seems to make
the symbolic function names line up with
https://launchpadlibrarian.net/72993752/Stacktrace.txt (this bug) where they
exist, and line up with the library binaries that contain them everywhere else.
Specifically, it looks like in both bugs, whatever is going wrong seems to
occur (or crystallize into full fruition) in _XReadEvents (it's from there that
error handling functions seem to begin to be called). It seems to me that the
x_error_handler self-call (from #4 to #3) might be optimized out for some cases
(that is, the library might be built with that function implemented as two or
more separate code paths as an optimization), producing two steps in that
function in some traces and just one in others, though I have not examined the
source to determine how plausible this is. However, I understand that,
technically, there's insufficient information to determine if they're the same
bug; in spite of the notable _XReadEvents similarity, this could be a different
bug that looks similar.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/792874
Title:
metacity crashed with SIGABRT in raise()
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/metacity/+bug/792874/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
