tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: recording /dev/sound vs pause



On 2020-03-24, Tetsuya Isaki wrote:
> At Fri, 20 Mar 2020 16:39:34 +0900,
> Tetsuya Isaki wrote:
> > I found a strange behavior on /dev/sound.
> > 
> > On NetBSD7, opening /dev/audio always initializes as pause=false,
> > and it forces recording start.
> > 
> > Opening /dev/sound inherits sticky pause.  If the sticky
> > pause = true, /dev/sound initializes as pause=true and then
> > recording doesn't start.  It will be expected behavior.
> > 
> > On the other hand, if the sticky pause = false, /dev/sound doesn't
> > initialize pause (not "initialize with false") and therefore
> > recording doesn't start.
> > 
> > In result, on NetBSD7 (or prior):
> > opening /dev/audio always starts recording and
> > opening /dev/sound always doesn't start recording (regardless of
> > sticky pause).
> > 
> > On NetBSD9, both of opening /dev/{audio,sound} doesn't start
> > recording now.  And it should be fixed.
> > But is this NetBSD7 behavior expected?  (And should I keep this
> > (strange) behavior in new audio?)
> 
> I'll treat this /dev/sound's behavior as a bug and fix it as follows:
> - opening /dev/audio (pause is always false in this case) always starts
>   recording.
>   (NetBSD7 did so, but -9/HEAD doesn't currently, this is my bug)
> - opening /dev/sound (if the sticky pause = false) starts recording.
>   (NetBSD7 and -9/HEAD didn't so, this is an old bug)
> - opening /dev/sound (if the sticky pause = true) doesn't start recording.
>   (no changes)

That is the behaviour I would expect, thanks!

-- 
Kind regards,

Yorick Hardy


Home | Main Index | Thread Index | Old Index