The silence size cannot exceed the silence threshold. Move the check
from the manual condition to the common code. This may happen for
small ring buffers (where the 1/10th second is too large).
Suggested-by: Oswald Buddenhagen <oswald.buddenhagen@gmx.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Add possibility to return -EINTR instead waiting for the event. The
applications may want to handle -EINTR condition themselves.
BugLink: https://github.com/alsa-project/alsa-lib/issues/228
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The snd_pcm_wait() function is called also internally from
the various plugins to wait for the drain with -1 and from i/o
routines in pcm.c.
Define two special negative timeout values to distinguish the
drain and i/o wait and calculate the maximal timeout according
the wait place.
Fixes: https://github.com/alsa-project/alsa-lib/issues/228
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The application may not require to touch the playback
sample stream for the drain operation at all. In this case,
the application is responsible to setup a silencing mechanism
for the playback or another graceful stop (like using the
rewind operation).
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Handle the driver informing us that it is not necessary to set up
silencing upon playback draining. This will be the case for drivers
which are guaranteed to not read any samples beyond the application
pointer.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Wrap last_commit_ptr using boundary. Was just wrapped to 0, which is
correct only if the buffer size, and hence the boundary, is an integer
multiple of the period size.
Fixes: 467d69c5bc ("Fix CPU hog with combination of rate plugin")
Fixes: 29041c5220 ("fix infinite draining of the rate plugin in SND_PCM_NONBLOCK mode")
Link: https://lore.kernel.org/alsa-devel/20221230164814.901457-1-consult.awy@gmail.com/
Signed-off-by: Alan Young <consult.awy@gmail.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The fast_ops callback invocation must always pass the fast_op_arg
as the pcm argument. Plugins expect that.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The previous fix in aa4f56c3 was not correct. The root of the cause is
implementation in snd_pcm_may_wait_for_avail_min() inline function
where the improper pcm argument is passed to the fast_ops function.
Fixes: aa4f56c3 ("pcm: rate - fix the crash in snd_pcm_rate_may_wait_for_avail_min()")
Fixes: d9dbb57b ("pcm: rate - rewrite the may_wait_for_avail_min callback for the rate plugin")
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The size argument in snd_pcm_rate_commit_area can be smaller
than cont in drain. Return the original code. It should
not make things worse. Perhaps, we can fill the rest of
the period with silence in this case.
Fixes: https://github.com/alsa-project/alsa-lib/issues/274
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
due to undefined reference to 'environ' error, ucm_exec.c uses workaround.
Fixes: https://github.com/alsa-project/alsa-lib/pull/250
Signed-off-by: SASANO Takayoshi <uaa@uaa.org.uk>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
It is expected to return the correct numbers for get_periods and
get_buffer_time requests. Copy the appropriate fields in the
hw_params call.
Fixes: https://github.com/alsa-project/alsa-lib/issues/258
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Those symbols should not be exported:
snd_pcm_direct_poll_descriptors
snd_pcm_direct_check_xrun
snd_pcm_direct_slave_recover
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Allow three configuration types:
rate 48000 # single rate
rate [ 48000 ] # single rate (2nd)
rate [ 44100 48000 ] # range
Fixes: https://github.com/alsa-project/alsa-lib/pull/191
Fixes: ac04cb63 ("hw: add "min_rate" and "max_rate" as alternatives to single "rate" parameter")
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Shuffle the code to avoid special conditions using the plugin type
in the generic plugin code. The rate plugin has the own
may_wait_for_avail_min callback implementation now.
Fixes: d21e0e01 ("pcm: plugin - fix avail_min calculation on rate plugin")
Fixes: https://github.com/alsa-project/alsa-lib/pull/218
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
commit 88e4ae27, ff1f669d introduced a dynamic recalculation of the slave's
avail_min value.
The calculated avail_min setting did not take into account, that the avail_min value
depends on the used sampling rate and must be adapted accordingly
if the slave is using a different sampling rate.
That leads to too large/too small calculated avail_min settings and inaccurate
period wake-up events if a rate converter plugin is used.
This patch is adapting the avail_min calculation to consider a different
sampling rate between actual pcm and it's slave.
Fixes: https://github.com/alsa-project/alsa-lib/pull/218
Signed-off-by: Andreas Pape <apape@de.adit-jv.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Generic mixing function for 32bits pcm has used 8bits right shift for
pre-scaling. But this is generating wrong result if pcm data is
negative value because return value type of bswap_32() is unsigned int.
This patch adds type cast bswap_32() result to signed int.
Fixes: https://github.com/alsa-project/alsa-lib/pull/222
Signed-off-by: Katsuhiro Suzuki <katsuhiro@katsuster.net>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Continue the work in commit ba86ac55 ("conf: Use LFS calls when reading
config files") and fix the PCM LADSPA code, too.
Fixes: https://github.com/alsa-project/alsa-lib/pull/223
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The HDMI drivers set an uniform PCM names. Use ELD (EDID) to obtain
the HDMI device name and send this string to applications for a better
user experience.
Example (aplay -l):
card 1: PCH [HDA Intel PCH], device 8: HDMI 2 [HDMI 2]
Subdevices: 1/1
vs improved:
card 1: PCH [HDA Intel PCH], device 8: HDMI 2 [Philips 272P4]
Subdevices: 1/1
Fixes: https://github.com/alsa-project/alsa-lib/issues/209
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
The xrun/suspend may happen at any time and we should check it right
after the slave hwptr update (but before the actual sync_ptr update in
direct pcm side). Otherwise the hwptr value may be screwed and get
unexpected large read/write.
Reported-by: S.J. Wang <shengjiu.wang@nxp.com>
Acked-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The check of slave PCM state is always done before the client's
recoveries count check, so let's merge them to the common helper.
Also rename the helper function to snd_pcm_direct_check_xrun() as it's
checking both slave and client states now.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The current resume handling in PCM direct plugins don't treat multiple
clients properly: once after the slave PCM gets resumed by one
client, the access from others at a later point is seen as already
running although the internal state isn't updated and becomes
inconsistent. This may end up a negative size, which eventually hangs
up.
This patch is an attempt to improve the handling for resume. Now the
suspended state is treated similarly like XRUN; namely, we keep the
slave PCM "recoveries" count that is modified at each time the slave
PCM XRUN happens, so that we can check the inconsistency against the
client's state. As a differentiation to XRUN, we set the highest bit
of recoveries count when the slave stream hits SUSPENDED state. This
bit is referred at comparing with clients, and the client's state is
updated to either XRUN or SUSPENDED depending on this bit.
Along with this change, the actual resume is done in
snd_pcm_direct_slave_recover(), and snd_pcm_direct_resume() rather
calls this internally.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Change the snd_pcm_direct_client_chk_xrun() function to return the
current XRUN state via an error code instead of the state change.
This allows the caller more straightforwardly returning its error, and
also covers the case where XRUN has been set but the function gets
called twice.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
In the case that:
* the buffer size is not an integer multiple of the period size, and
* drain must flush a partial period located before the end of the buffer
but without a full period available, where
* these conditions may pertain to the source or slave pcm buffer, and
* because rate conversion is always done on a full period,
it is necessary to check that both a full source period is available
before source pcm buffer wrap and a full slave period is available
before slave pcm buffer wrap in order to use the simple, single-commit
implementation in snd_pcm_rate_commit_area().
The alternative fix would be to change snd_pcm_rate_write_areas1() to
take size and slave_size parameters. This would be more disruptive to
the code base, tricky to get right, and is unnecessary given that
snd_pcm_mmap_commit() only commits the partial period of actually valid
converted samples.
Fixes: 3047f8fa5a ("Fix possible problems of playback drain with rate plugin")
Signed-off-by: Alan Young <consult.awy@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This reverts partially commit b9a4997e92.
It seems that we have have some users for this very specific function.
Mark it deprecated and keep the softvol implementation separate,
so we can remove this function easily in future.
Fixes: https://github.com/alsa-project/alsa-lib/issues/186
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
When doing rewind or forward operations, application pointer needs to
be properly adjusted, in the same way when doing commit/write
operations.
Fixes: https://github.com/alsa-project/alsa-lib/pull/161
Signed-off-by: Sava Jakovljev <sava.jakovljev@teufel.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Fix and cleanup snd_pcm_direct_check_interleave() function.
Add requested / hardware channel check and use goto when the interleaved
Fixes: https://github.com/alsa-project/alsa-lib/issues/117
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
