http://bugs.openembedded.org/show_bug.cgi?id=877





------- Comment #1 from [EMAIL PROTECTED]  2006-08-03 05:50 -------
I can confirm that qplot has similar problems within OZ 3.5.4.1 on an SL-C3200
(terrier). However, I think the "broken pipe" is related to the bc app which
qplot exec()'s and which isn't with the qplot .ipk. Even worse, qplot expects
the bc executable to be in /opt/Qtopia/bin, but the feed's bc .ipk puts the
executable in /usr/bin.

Sadly, even when these things are fixed, qplot still dies with either a SIGSEGV
or SIGABRT, depending on who you talk to. For me, the failure is receipt of a
SIGABRT after glibc complains about a bad free() call. I've run the qplot
executable, with debug symbols compiled in, in gdb in an attempt to find trhe
problem but have gotten inconsistent results.

As an aside, running qplot with 'strace -fF' never shows the child process
being spawned. Just to make sure OZ's strace wasn't broken, I wrote a simple
app which performed the same pipe() and fork() steps as qplot, and confirmed
that strace properly reports the child process spawn.

Upon request I will attach here the strace output and backtrace from qplot;
they're rather long.


-- 
Configure bugmail: http://bugs.openembedded.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
_______________________________________________
Oe mailing list
[email protected]
https://www.handhelds.org/mailman/listinfo/oe

Reply via email to