On Friday 10 October 2008 12:21:03 Stéphane Letz wrote:
>
> Le 9 oct. 08 à 23:18, Yair K. a écrit :
>
> Thanks a lot! Should be fixed in jack2 SVN now.
Looking at the diff, I think you missed some things:
First, the behaviour where the ioctl returns a different result than desired is
not limited to SNDCTL_DSP_SPEED. Most ioctls that communicates directly with
the device can do this, so it's also possible to have SNDCTL_DSP_CHANNELS and
SNDCTL_DSP_SETFMT return a different result (e.g. non-vmix device which is
stereo-only, or only supporting 16 bit formats). Also, You are checking the
fragment size via SNDCTL_DSP_GETBLKSIZE, but perhaps a warning would be
preferable to a fatal error, in the case '-b' ("Ignore hardware period size")
argument is used?
Second, you've dropped the "case 64:" part (it matched AFMT_FLOAT before) in
SetSampleFormat(), but you have kept that case at some of the other functions.
Yours,
Yair K.
_______________________________________________
oss-devel mailing list
oss-devel@mailman.opensound.com
http://mailman.opensound.com/mailman/listinfo/oss-devel
