Ubuntu: portmidi calls exit(-1) in error conditions

Bug #1527888 reported by Daniel Schürmann on 2015-12-19
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Mixxx
Critical
Unassigned

Bug Description

On Ubuntu Trusty, libportmidi0 1:200-0ubuntu3 is compiled with
PM_CHECK_ERRORS

which calls this is any error condition and crasches Mixxx without allowing Mixxx to gracefull handle the errors like buffer overflows which can likely happen.

static void prompt_and_exit(void)
{
    char line[STRING_MAX];
    printf("type ENTER...");
    fgets(line, STRING_MAX, stdin);
    /* this will clean up open ports: */
    exit(-1);
}

Mixxx log look like that

Debug [Controller]: PortMidiController::poll() 2
Debug [Controller]: PortMidiController::poll() 2
PortMidi call failed...
  PortMidi: `Buffer overflow'
type ENTER...

Program received signal SIGSEGV, Segmentation fault.
QBrush::QBrush (this=0x3ae5e30) at painting/qbrush.cpp:425
425 d->ref.ref();
(gdb)

Daniel Schürmann (daschuer) wrote :

We have to check if his is the case on our buildserver as well and we need to provide our own version of portmidi in our ppa and depend Mixxx on that.

@rryan can you adopt this bug?

Changed in mixxx:
importance: Undecided → Critical
milestone: none → 2.0.0

We don't do this on our windows/Mac builds. I filed a bug with Ubuntu years
ago to get this fixed and it was! Annoying that this has reemerged.

On Sat, Dec 19, 2015, 9:40 AM Daniel Schürmann <email address hidden>
wrote:

> We have to check if his is the case on our buildserver as well and we
> need to provide our own version of portmidi in our ppa and depend Mixxx
> on that.
>
> @rryan can you adopt this bug?
>
> --
> You received this bug notification because you are a member of Mixxx
> Development Team, which is subscribed to Mixxx.
> https://bugs.launchpad.net/bugs/1527888
>
> Title:
> Ubuntu: portmidi calls exit(-1) in error conditions
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/mixxx/+bug/1527888/+subscriptions
>

RJ Ryan (rryan) wrote :

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=765375

On Sat, Dec 19, 2015, 1:01 PM RJ Ryan <email address hidden> wrote:

> We don't do this on our windows/Mac builds. I filed a bug with Ubuntu
> years ago to get this fixed and it was! Annoying that this has reemerged.
>
> On Sat, Dec 19, 2015, 9:40 AM Daniel Schürmann <email address hidden>
> wrote:
>
>> We have to check if his is the case on our buildserver as well and we
>> need to provide our own version of portmidi in our ppa and depend Mixxx
>> on that.
>>
>> @rryan can you adopt this bug?
>>
>> --
>> You received this bug notification because you are a member of Mixxx
>> Development Team, which is subscribed to Mixxx.
>> https://bugs.launchpad.net/bugs/1527888
>>
>> Title:
>> Ubuntu: portmidi calls exit(-1) in error conditions
>>
>> To manage notifications about this bug go to:
>> https://bugs.launchpad.net/mixxx/+bug/1527888/+subscriptions
>>
>

Daniel Schürmann (daschuer) wrote :

This is the Ubuntu ubstream bug:
https://bugs.launchpad.net/mixxx/+bug/1097286
I have added a patch to it.
I think we need to provide a our own builds until it is fixed.
All Ubuntu versions are effected.

Changed in mixxx:
status: New → Confirmed
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.