[Bug 47989] FL Studio 20 hangs on splash screen

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=47989

ajduck at outlook.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ajduck at outlook.com

--- Comment #3 from ajduck at outlook.com ---
I can confirm this does happen with FL Studio, but it's strange, it only
happens (or doesn't) in certain conditions.

If I run FL directly from the command line with simply "wine FL64.exe", it
happens for me, but if I run it from my distro's application menu which (in the
.desktop file) has a more complex set of commands for Exec (basic outline:
/usr/bin/env sh -c "winepath [...] | xargs -0 'wine FL64.exe'"), it doesn't
hang.

It used to hang even with that, and I had to add "&> output.log" somewhere at
the end of the Exec command (due to its nesting) in order for it to not hang at
start up. Actually that makes it work now when running from the terminal, if I
run 'bash -c "wine FL64.exe" &> freezetest.log"'. Try that to see if it runs.

I've also noticed that it can sometimes hang or not hang possibly depending on
the distro and what audio driver is being used, e.g. pulseaudio, FL Studio ASIO
or wineasio with JACK. I once noticed that when I changed the buffer size for
JACK to very high (e.g. 8192 = higher latency) or low (e.g. 16 = lower latency)
numbers the hanging either stopped or started. I can't remember exactly what it
was doing beforehand however, and this behaviour doesn't happen any more for
me.

It almost seems like some sort of race condition to me because of all this
weird behaviour.

The RtlpWaitForCriticalSection hang seems to be a wider issue with FL. It's the
same thing that's outputted when FL hangs when you try to show the GUI
(embedded in an internal FL window) of a plugin "bridged" for compatibility
with another bit range (32bit -> 64bit or 64bit -> 32bit). VST hosts can't load
up VST plugins with another bit range without this bridging process. That
particular issue is detailed in bug #44827.

-- 
Do not reply to this email, post in Bugzilla using the
above URL to reply.
You are receiving this mail because:
You are watching all bug changes.