So it's probably a heissenbug, depending on the timing between two events. Options: - show us the bad backtrace in GDB, in case it rings a bell for someone - hang the -O1 code, send it SIGABRT to make it dump core, backtrace the core, and see if it's any better - someone who knows the code to guess what needs to be slowed down, to make it reliably reproduceable On 9/11/14, Sebastian Geerken <sgeerken at dillo.org> wrote:

On Mi, Sep 10, 2014, eocene at gmx.com wrote:

...

Sebastian wrote:

...

I noticed that the current tip of dillo hangs in most cases when displaying <http://www.dillo.org/>. I succeeded to reproduce this bug in gdb, but get many "<optimized out>" values. So I tried to reduce optimization, but unfortunately had problems reproducing this bug:

- "-O2" (standard): reproducable both stand-alone and in gdb; - "-O1": reproducable stand-alone, but not in gdb; - "-O0": not reproducable at all.

Versions envolved:

$ gcc --version gcc (Debian 4.7.2-5) 4.7.2 $ gdb --version GNU gdb (GDB) 7.4.1-debian

Any ideas?

I have the same gcc and gdb, and dillo doesn't seem to want to hang for me. Does it depend on your dillorc options?

Removing ~/.dillo makes no difference.

Neither does switching to other versions (gcc and g++ 4.4).

Sebastian