The initialization procedure consist from default initialization and
state restoration. When restore fails for a reason or the config file
contains inaccurate information, then init is called.
Also, if something fails (init or restore), a file specified using -r
option can be created with description what failed. It can be useful
when user space decides to notify user to set the sound driver properly.
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
- change SYSFS_DEVICE to CONFIG{sysfs_device}
- INCLUDE key now handles also directories
- RESULT key can assign value now
- EXIT="return" operation returns from included file immediately
Default 00main and hda configuration files changes:
- handle preinit and postinit directories
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Added iecset support for some of consumer status channel bits,
new in the third edition od IEC60958-3 spec.
Signed-off-by: Pawel Moll <pawel.moll@st.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Here is a small patch for alsaconf, which removes an useless dependency
on `which' by using a bash built-in instead : `type -p'. I encountered
the problem of the missing `which' while using alsaconf on a clfs-built
linux system. It is useless to install `which' when we already have
`type -p', and alsaconf already needs bash, so this does not replace
a dependency by another one, but really removes a dependency for alsaconf.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Use snd_ctl_elem_type_name() to detect the value of the type comment
instead of using hardcoded strings.
The types list now also includes the BYTES type (Debian bug 481515).
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
This patch makes speaker-test fill the buffers with properly coded data on
both big- and little-endian processors.
Signed-off-by: Giuliano Pochini <pochini@shiny.it>
speaker-test doesn't work well when you choose a little-endian format on a
big-endian processor, or the opposite. Yes, I know about plughw:, but for
debugging purposes it may not be an option. The following patch add proper
support for S32 and S16 support for but LE and BE processors:
- The "if (bits-per-sample)" construct was replaced by case (format).
- Support for S16_BE, S32_BE formats was added.
- S16_LE and S32_LE were made compatible with big-endian processors.
- NB: The pink noise generator wasn't changed (I'll do if this patch is OK).
From: Giuliano Pochini <pochini@shiny.it>
The recent ncurses package provides ncurses*-config program to give the
proper cflags and libs. Let's use them if available.
Right now, the ncurses version (5) is hard-coded. It should be better
to be variable as well, but it'd be messy. Hope the ncurses version won't
change rapidly.
Added the support of stereo VU-meter.
Enabled via -Vs option.
The new option, -V, can be used to enable the VU-meter. Now
VU-meter can be enabled even without -vv.
WAV files with more than 2 channels or with more than 16 bits per samples
can be saved with "extensible format" chunk
(see http://msdn2.microsoft.com/en-us/library/ms713496(VS.85).aspx).
For instance, sox, when converting data to 24- or 32-bits per sample format
uses this format, and aplay was unable to play such file. Now the problem
is solved :-)
Signed-off-by: Pawel MOLL <pawel.moll@st.com>
This patch modifies aplay/arecord to support playing/capturing IEEE float
32-bit WAV files. Tested on HDA hardware in both stereo and multi-channel
modes. Added the WAV file constant for Dolby AC-3 S/PDIF passthrough to
formats.h for future use when AC-3 passthrough is better supported.
Signed-off-by: Andrew Paprocki <andrew@ishiboo.com>
When using alsactl to save or restore the card settings, it currently
skips over controls that don't have the appropriate read/write
permissions. It should also skip over inactive controls, otherwise it
will get an error when it tries to access that control, and will fail to
save the card state (or fully restore it.)
From: Dave Dillow <dave@thedillows.org>
snd_pcm_drain() doesn't block when running with O_NONBLOCK.
Reset the non-blocking mode before calling snd_pcm_drain() properly
(and restore again for any further operations).
Set -F option as default for restore. There are still too many systems
that are too lazy to set -F option...
Added the new -P option to back to the old behavior.
