ffado-test-streaming crashed with SIGSEGV in __GI___libc_free()

[David Kastrup]

Just called ffado-test-streaming on the current setup (one builtin Ricoh Firewire controller without attached hardware, one Expresscard with TI chipset and an Alesis io|14 attached) and it crashed after some test duration, apparently before wanting to exit.

ProblemType: Crash
DistroRelease: Ubuntu 17.10
Package: ffado-tools 2.3.0-2
ProcVersionSignature: Ubuntu 4.12.0-11.12-lowlatency 4.12.5
Uname: Linux 4.12.0-11-lowlatency x86_64
ApportVersion: 2.20.7-0ubuntu1
Architecture: amd64
CrashCounter: 1
Date: Thu Aug 31 15:28:05 2017
ExecutablePath: /usr/bin/ffado-test-streaming
InstallationDate: Installed on 2011-10-14 (2148 days ago)
InstallationMedia: Ubuntu 11.10 "Oneiric Ocelot" - Release i386 (20111011)
ProcCmdline: ffado-test-streaming
SegvAnalysis:
 Segfault happened at: 0x7f25bb6b9d32 <__GI___libc_free+34>: mov -0x8(%rdi),%rax
 PC (0x7f25bb6b9d32) ok
 source "-0x8(%rdi)" (0x000003f9) not located in a known VMA region (needed readable region)!
 destination "%rax" ok
SegvReason: reading NULL VMA
Signal: 11
SourcePackage: libffado
StacktraceTop:
 __GI___libc_free (mem=0x401) at malloc.c:2966
 ()
 __libc_start_main (main=0xf6ba13cd97, argc=1, argv=0x7ffc0fddf5f8, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7ffc0fddf5e8) at ../csu/libc-start.c:291
 ()
Title: ffado-test-streaming crashed with SIGSEGV in __GI___libc_free()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups:

[Apport retracing service (apport)]

StacktraceTop:
 __GI___libc_free (mem=0x401) at malloc.c:2966
 ?? ()
 __libc_start_main (main=0xf6ba13cd97, argc=1, argv=0x7ffc0fddf5f8, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7ffc0fddf5e8) at ../csu/libc-start.c:291
 ?? ()

[Pander (pander)]

From FFADO maintainer:

   Bug: segfault in "ffado-test Discover"

   Suggestion:
       The reporter acknowledges that prior to running ffado-test "there
       already were problems with the Firewire connection". The crash is
       associated with a bus reset. While bus resets should be handled
       cleanly by FFADO, it is not always the case. In a properly
       functioning system there should be no sporadic bus resets. Bus
       resets normally occur when the bus topography changes (a device is
       added or removed for example).

       It seems that the system is experiencing frequent bus resets. This
       most likely indicates a hardware fault. This could be in the audio
       device, the firewire cable or the firewire card in the computer.
       The first thing to try is a different firewire cable from a reputable
       manufacturer. There were many dodgy firewire cables made, and bad
       cables can give rise to all kinds of strange problems.

       Another option is a firewire card which isn't up to the task of
       managing the datastream associated with an audio interface. This is
       unlikely to be the cause here though, since the fault is occurring
       during device probing which isn't particularly taxing on the firewire
       bus.