[openal] OpenAL Soft release soon-ish?
chris.kcat at gmail.com
Thu Aug 13 00:07:01 EDT 2015
I'm contemplating having an OpenAL Soft 1.17 release fairly soon, and
would like feedback on any relatively important issues for it.
Currently, I know of:
The HRTF extension:
I've not heard too much feedback from this, so I assume it's good
enough. I'd really prefer to not release 1.17 without an HRTF extension,
so if there's any changes you'd like to see, please speak up.
althrd_start race condition:
I admit I can't see where the race condition is. I can only guess there
might be some obscure WinAPI issue I'm not accounting for, but I don't
know what the problem is.
OpenSL backend callback race condition:
It seems setting the stream player to SL_PLAYSTATE_STOPPED and calling
the buffer queue's Clear method does not guarantee the callback is
finished and won't be called again (until playback resumes). I don't
know if this is a bug in OpenSL, or if there's some other function that
would ensure the callback is completely done with.
UT2k4 wants "DirectSound3D" on Windows:
I really would like to avoid hacks like this, but from the looks of it,
OpenAL Soft would have to accept "DirectSound3D" and treat it as the
default device (on Windows). Is there really no way to change the device
name the game wants?
Compile problems with defining _XOPEN_SOURCE and _POSIX_C_SOURCE:
Apparently FreeBSD and GNU disagree on how these macros should be used.
GNU defaults to enabling the subset of features offered by the language
being compiled for (C99) and needs the macros defined for anything more
(POSIX, XSI). FreeBSD defaults to having everything enabled with the
language, and uses those macros to restrict what's allowed, including
the types used by BSD's OSS headers. So I guess I'll need to check if
sys/soundcard.h can be included with those macros.
If anyone has any suggestions for these issues, or has anything more
I've forgotten that would be important for release, please let me know.
More information about the openal