I'm not sure what you mean you could generate a backtrace but it stopped. E should generate the backtrace itself whenever it crashes. Is E locking up?
]]>@WW; this is in a -git AUR package, not the repos so it doesn't make sense to file it in the Arch bugtracker.
@pelpix; were you ever able to generate a backtrace? I'd be interested in seeing just where it's happening, and I'm sure they'd be interested upstream as well. They're preparing to release to release EFL 1.8/E 0.18 on Dec 1 and a consistent segv is something to look into.
The fact that working E18 builds were retroactively broken recently makes me think it may be some change or error in Arch's official versions of the dependencies, though.
Also, I could generate a backtrace, but it just...stopped. No error. The process just crashed, without even signalling that the E SEGV is being called. It just freezes solid with no indication of what or where the error is.
@pelpix; were you ever able to generate a backtrace? I'd be interested in seeing just where it's happening, and I'm sure they'd be interested upstream as well. They're preparing to release to release EFL 1.8/E 0.18 on Dec 1 and a consistent segv is something to look into.
]]>Just to clarify, we're talking about Enlightenment 0.17.99 (or now 0.18-alpha1)? If you have GDB installed, it should create a backtrace at ~/.e-crashdump.txt
Yep.
I compiled it with GDB support and have GDB installed, but it hangs. ~/.e-crashdump.txt is created, but it's completely blank.
Hey, there was a commit just an hour ago:
wsod: Fixed gdb hangs when generating backtraces
Perhaps this will fix my problem with the logs? Time to test!
]]>On top of that, it doesn't dump any telling logs even in debug mode.
Ideas?
]]>