Re: [qtile-dev] Python 2 cffi crash

2014-07-31 Thread Tycho Andersen
On Thu, Jul 31, 2014 at 05:31:42PM -0500, Sean Vig wrote: > Well, so I didn't know how to use gdb properly but I found > https://wiki.python.org/moin/DebuggingWithGdb Ah yeah, I probably had the debug symbols laying around from some other thing I was doing. > I added the backtrace, py-bt, and

Re: [qtile-dev] Python 2 cffi crash

2014-07-31 Thread Sean Vig
Well, so I didn't know how to use gdb properly but I found https://wiki.python.org/moin/DebuggingWithGdb I added the backtrace, py-bt, and py-list output to the gist: https://gist.github.com/flacjacket/118e3b3d69e076e26c11 On Thu, Jul 31, 2014 at 2:14 PM, Tycho Andersen wrote: > On Thu, J

Re: [qtile-dev] Python 2 cffi crash

2014-07-31 Thread Tycho Andersen
On Thu, Jul 31, 2014 at 12:16:14PM -0500, Sean Vig wrote: > Here's what I get from the backtrace: > > 2014-07-31 12:05:40,601 qtile call:91 Command: move_floating((10, 20, 42, > 42), {}) > > Program received signal SIGSEGV, Segmentation fault. > tupledealloc (op=0x7fffeabfd7e0) at > /var/tmp/por

Re: [qtile-dev] Python 2 cffi crash

2014-07-31 Thread Sean Vig
Here's what I get from the backtrace: 2014-07-31 12:05:40,601 qtile call:91 Command: move_floating((10, 20, 42, 42), {}) Program received signal SIGSEGV, Segmentation fault. tupledealloc (op=0x7fffeabfd7e0) at /var/tmp/portage/dev-lang/python-2.7.7/work/Python-2.7.7/Objects/tupleobject.c:222 222

Re: [qtile-dev] Python 2 cffi crash

2014-07-31 Thread Tycho Andersen
Hi Sean, On Thu, Jul 31, 2014 at 11:35:57AM -0500, Sean Vig wrote: > Hey, > > In working on [1], I ran into a weird test failure in test_manager.py: > test_default_float that happens in Python 2 [2] but not Python 3 [3]. I > thought this was related to the changes in that PR, since Python 2 tests

[qtile-dev] Python 2 cffi crash

2014-07-31 Thread Sean Vig
Hey, In working on [1], I ran into a weird test failure in test_manager.py: test_default_float that happens in Python 2 [2] but not Python 3 [3]. I thought this was related to the changes in that PR, since Python 2 tests had run successfully before for the cffi branch [4]. However, I found that I