beatloop rolls occasionally reset the playhead to the start

[Owen Williams]

This is a really hard-to-reproduce party stopper: Every so often a beatloop roll will cause the track to seek to the beginning and stop completely. I'm not sure, but I think it might happen when the loop is disabled, not when the loop is active.

[Owen Williams (ywwg)]

it's probably some sort of race with slipenabled and slipposition getting all messed up.

[Owen Williams (ywwg)]

This might do it https://github.com/mixxxdj/mixxx/pull/326

[Owen Williams (ywwg)]

My best theory for this bug:

in slotControlSlip, the enabled value was set before the toggled value. If the timing was just wrong, processSlip could be executed just between those two calls. Perhaps this could cause m_dSlipPosition to not be set to the current playposition, so when slip was released it would rewind to the beginning. But I'm not sure that's right, because playback would also stop, and there's nothing in this codepath that would cause playback to stop. Also I'd have thought that Toggled would be on for the next loop, so it would still reset the slip position.

So I'm going to leave this bug open for a little while and really hammer the rolling loops in my testing :)

[Owen Williams (ywwg)]

erm, I mean when I see this bug, playback stops, but with this theory I don't know why the track would stop.

[Owen Williams (ywwg)]

I haven't seen this in a while, so I'm fairly sure it's fixed.

[RJ Skerry-Ryan (rryan)]

let's re-open if it returns

[Swiftb0y (swiftb0y)]

Mixxx now uses GitHub for bug tracking. This bug has been migrated to:
https://github.com/mixxxdj/mixxx/issues/7565