Hi Jakub, That sounds fine, and I'll review any patch to gjs you guys submit. Just curious, have you investigated what Mozilla does for their stack traces? We can't be the only ones having this issue. A little Googling suggests there might be a way to grab a JS stack trace out of a core dump file, which might be more interesting than having some sketchy code run in the SIGSEGV handler:
http://www.visophyte.org/blog/2009/09/12/prototype-unified-javascriptc-back-traces-for-mozilla-in-archer-gdb/ http://hg.mozilla.org/users/bugmail_asutherland.org/pythongdb-gaudy/ On Wed, Mar 26, 2014 at 5:18 AM, Jakub Filak <jfi...@redhat.com> wrote: > Hello, > > I am a member of the ABRT team. We have recently found an interesting page > about > GNOME Shell debugging[1]. According to that page, it is possible to obtain > a > stack trace of executed JavaScript code in gdb. So we got an idea to > include > the JavaScript stack trace in Bugzilla bugs filed by ABRT. I tried to > obtain > the JavaScript stack trace by attaching gdb to a core dump file but it > didn't > work because we can call a method only when gdb is connected to a process. > > But we got another idea. Would it be possible to write the JavaScript stack > trace to a file (or journald) in a well known path upon receiving a UNIX > signal? This already works for JVM, which creates the hs_err_pid.log > file[2] > when a JVM process is to die and ABRT includes the hs err log file in > Bugzilla > bugs. > > > > Regards, > Jakub > > > 1: https://wiki.gnome.org/Projects/GnomeShell/Debugging > 2: http://www.oracle.com/technetwork/java/javase/felog-138657.html > _______________________________________________ > gnome-shell-list mailing list > gnome-shell-list@gnome.org > https://mail.gnome.org/mailman/listinfo/gnome-shell-list > -- Jasper
_______________________________________________ gnome-shell-list mailing list gnome-shell-list@gnome.org https://mail.gnome.org/mailman/listinfo/gnome-shell-list
