It was only implemented in OpenGL, though the option was visible in both
backends, leading to memory leaks if you enabled it in DirectX.
And it wasn't particularly useful as a debug feature as it only showed
where in the EFB the copies were taken from, not what format it was, or
what the copy was used for, or what content was in the EFB at that point
in time.
Also, it stretched the copy regions relative to the window, so the
on-screen regions don't even line up with the window unless the game used
the full EFB (some pal games) and you game image stretched to the full
window.
This 'absolutely not' a read of uninitialized memory is '100% not'
the cause of our non-deterministic behavior on the Intel NUC.
If there was a such an error, it would show up on all FifoCi
backends equally. This is 'probably' unrelated to the fact that
the Intel NUC is the only fifoci runner not running under virtualization.
If there were two commands in the buffer at once, it would only run the
first because of an error in UpdateInput.
If you sent the command "SET C" it would segfault because of a logic
issue in ParseCommand.
This addresses a bit of thread unsafety mentioned in a comment, and
fixes a 'ScheduleEvent_Threadsafe from main thread' message.
To make this work nicely, make PauseAndLock call DeclareAsCPUThread -
i.e. while you have the CPU thread locked, you can consider yourself the
CPU thread.
Ubuntu 14.04 has miniupnpc 1.6.3
REGEX was failing because the string was empty, causing cmake to
error and not generate a Makefile.
This allows systems with older versions of miniupnpc to fall back
to the statically linked version in externals.
091db36 added a new delay for disc commands in an attempt to fix issues.
It fixed Sonic Riders crashing after the third level... or so we thought.
091db36 also happened changed the order disc reads happen in so that the
data is copied to memory before the emulated delay is finished. This was
inaccurate and caused an audio problem in Resident Evil 3, leading to the
old order being restored in 8cc6e5c. Now that the order is correct,
Sonic Riders is broken again, despite 091db36's delay still existing.
We're more or less back to square one - nobody knows what's broken, and
nobody knows how to fix it. This commit restores SUDTR to Sonic Rider's
game INI so that it'll work out of the box in 5.0 just like in 4.0.
This check is not useful anymore since that website is not running their own
buildbots anymore. It's also trivially bypassable -- it just happened to work
when the builds were fully automated over there.
flacs