[Bug 41833] New: DirectShow applications: Playback doesn' t end when playing some MP3 files if winegstreamer is disabled

https://bugs.winehq.org/show_bug.cgi?id=41833

Bug ID: 41833
           Summary: DirectShow applications: Playback doesn't end when
                    playing some MP3 files if winegstreamer is disabled
           Product: Wine
           Version: 1.9.24
          Hardware: x86-64
                OS: Linux
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: quartz
          Assignee: [e-mail]
          Reporter: [e-mail]
      Distribution: ---

When the test program at https://bugs.winehq.org/show_bug.cgi?id=41290 plays
"16. Night Sky.mp3" at http://redmine.audacious-media-player.org/issues/628,
WaitForCompletion() doesn't return. Built-in quartz.dll displays many
"fixme:quartz:parse_header Not a valid header: XX:XX:XX:XX" messages and after
this it displays "fixme:quartz:FileAsyncReader_WaitForNext Returned: 258
(00000000)" infinitely.

The MP3 BGM files in "Mystic Star"
(http://www.vector.co.jp/soft/dl/win95/game/se509140.html) have the same
problem (MP3 BGMs don't loop).

Workaround: Use native quartz.dll

[Bug 41833] DirectShow applications: Playback doesn' t finish when playing MP3 files that end in bad data if winegstreamer is disabled

https://bugs.winehq.org/show_bug.cgi?id=41833

Bruno Jesus  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |download
             Status|UNCONFIRMED                 |NEW
     Ever confirmed|0                           |1
            Summary|DirectShow applications:    |DirectShow applications:
                   |Playback doesn't end when   |Playback doesn't finish
                   |playing some MP3 files if   |when playing MP3 files that
                   |winegstreamer is disabled   |end in bad data if
                   |                            |winegstreamer is disabled
                 CC|                            |[e-mail]
                URL|                            |http://ftp.vector.co.jp/66/
                   |                            |17/3135/MysticStar.zip

--- Comment #1 from Bruno Jesus  ---
I can confirm this issue. Wild guesses follow.

It looks like quartz keeps track of the position the file is now and where it
should end. The problem with broken MP3 is that they end before what quartz
expect, in this case quartz keeps trying to pump data but it is already past
the end of the buffer [2]. Line 204 hits false (there is no else) and it
proceeds as if nothing wrong happened and the function will keep returning hr =
0 (good) forever.

A side note: It looks very suspicious to me that the return from
WaitForMultipleObjects is compared with the number of samples [2], because in
the situation of the bug the function returns WAIT_TIMEOUT.

[1] https://source.winehq.org/source/dlls/quartz/mpegsplit.c#0195
[2] https://source.winehq.org/source/dlls/quartz/filesource.c#1191

[Bug 41833] DirectShow applications: Playback doesn' t finish when playing MP3 files that end in bad data if winegstreamer is disabled

https://bugs.winehq.org/show_bug.cgi?id=41833

--- Comment #2 from Bruno Jesus  ---
Created attachment 56252
  --> https://bugs.winehq.org/attachment.cgi?id=56252
compiled test sample

Easiest way to reproduce is to unpack the game and run playfile-dshow.exe
Music\NR_theme9.mp3 which is the shortest music in the game (48 seconds).

[Bug 41833] DirectShow applications: Playback doesn' t finish when playing MP3 files that end in bad data if winegstreamer is disabled

https://bugs.winehq.org/show_bug.cgi?id=41833

--- Comment #3 from Andrew Eikum  ---
(In reply to Bruno Jesus from comment #1)
> I can confirm this issue. Wild guesses follow.
>
> It looks like quartz keeps track of the position the file is now and where
> it should end. The problem with broken MP3 is that they end before what
> quartz expect, in this case quartz keeps trying to pump data but it is
> already past the end of the buffer [2]. Line 204 hits false (there is no
> else) and it proceeds as if nothing wrong happened and the function will
> keep returning hr = 0 (good) forever.

Good analysis. That code could definitely use better error handling.
> A side note: It looks very suspicious to me that the return from
> WaitForMultipleObjects is compared with the number of samples [2], because
> in the situation of the bug the function returns WAIT_TIMEOUT.

Yeah, that's definitely broken. More bad error handling, and that code looks
like a rats nest to try to de-tangle.

[Bug 41833] DirectShow applications: Playback doesn't finish when playing MP3 files that end in bad data if winegstreamer is disabled

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=41833

[e-mail] changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
           See Also|                            |https://bugs.winehq.org/sho
                   |                            |w_bug.cgi?id=48292

[Bug 41833] DirectShow applications: Playback doesn't finish when playing MP3 files that end in bad data if winegstreamer is disabled

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=41833

[e-mail] changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |WONTFIX

--- Comment #4 from [e-mail] ---
winegstreamer is needed for MP3 playback since
https://source.winehq.org/git/wine.git/commitdiff/40a4c782b7f77534535dfa7bb2...,
closing.

[Bug 41833] DirectShow applications: Playback doesn't finish when playing MP3 files that end in bad data if winegstreamer is disabled

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=41833

Austin English  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #5 from Austin English  ---
Closing.