Re: [ecasound] bad_alloc exception in ecasound 2.5.2, 2.6.0 and -git

From: Krzysztof Foltman <wdev@email-addr-hidden>
Date: Sun Feb 22 2009 - 15:55:23 EET

Krzysztof Foltman wrote:
> This is happening with JACK2 (jackdmp) - didn't try with JACK1 yet.
Happens with JACK1 and with ALSA too (after replacing jack_alsa
inputs/ouputs with the ALSA counterparts). The problem seems to be
related to using loopback and splitting the signal, because if I replace
that part with simple output, it works.

There are warnings about uninitialised values, but they might be related
to valgrind not being aware about ALSA ioctl operation, not real problems.

The same crash also happens with a much simpler setup:

# general

# audio inputs
-a:1,2 -i:jack_alsa

# audio outputs
-a:1 -o:vocal.wav,
-a:2 -o:jack_alsa

It runs perfectly fine with heap checks disabled (MALLOC_CHECK_=1) but
that's hardly a solution.

Another user got the same kind of crash (glibc finding a heap corruption
which is not detectable when running under valgrind) with my own
software. Maybe it's a bug in some other library that's used by both
calf and ecasound, or over-zealous heap checks detecting a problem where
there's none. Haven't found any obvious candidates though.

Kernel version, glibc 2.7-1, gcc 4.3.3.


Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
-Strategies to boost innovation and cut costs with open source participation
-Receive a $600 discount off the registration fee with the source code: SFAD
Ecasound-list mailing list
Received on Sun Feb 22 16:15:02 2009

This archive was generated by hypermail 2.1.8 : Sun Feb 22 2009 - 16:15:02 EET