OSDN Git Service

android-x86/external-alsa-lib.git
15 years agoMake snd_pcm_hw_params_names static to pcm_params.c .
Diego E. 'Flameeyes' Pettenò [Fri, 21 Nov 2008 19:28:56 +0000 (20:28 +0100)]
Make snd_pcm_hw_params_names static to pcm_params.c .

Signed-off-by: Diego E. 'Flameeyes' Pettenò <flameeyes@gmail.com>
15 years agoMake seq, rawmidi and control operation structures static const.
Diego E. 'Flameeyes' Pettenò [Fri, 21 Nov 2008 19:26:12 +0000 (20:26 +0100)]
Make seq, rawmidi and control operation structures static const.

Since they are never changed it does not make sense to have them in
the writeable .data section, just make sure to add const to the ops
member in the structure definitions so that there are no extra
warnings added.

Signed-off-by: Diego E. 'Flameeyes' Pettenò <flameeyes@gmail.com>
15 years agoMake some static tables and strings constants.
Diego E. 'Flameeyes' Pettenò [Fri, 21 Nov 2008 00:21:32 +0000 (01:21 +0100)]
Make some static tables and strings constants.

By doing this we move them from the .data section to .rodata setion,
or from .data.rel to .data.rel.ro.

The .rodata section is mapped directly from the on-disk file, which is
always a save, while .data.rel.ro is mapped directly when using
prelink, which is a save in a lot of cases.

Signed-off-by: Diego E. 'Flameeyes' Pettenò <flameeyes@gmail.com>
15 years agoMake sure that python libraries are passed through LIBADD.
Diego E. 'Flameeyes' Pettenò [Fri, 21 Nov 2008 00:19:17 +0000 (01:19 +0100)]
Make sure that python libraries are passed through LIBADD.

Also avoid an indirection by using $(PYTHON_LIBS) and
$(PYTHON_INCLUDES) directly.

Signed-off-by: Diego E. 'Flameeyes' Pettenò <flameeyes@gmail.com>
15 years agoCheck for --no-undefined linker flag and use it.
Diego E. 'Flameeyes' Pettenò [Fri, 21 Nov 2008 00:16:37 +0000 (01:16 +0100)]
Check for --no-undefined linker flag and use it.

This adds extra safety that the built libraries will have all the
correct dependencies linked in.

Signed-off-by: Diego E. 'Flameeyes' Pettenò <flameeyes@gmail.com>
15 years agoAdd the attributes.m4 macro file from xine/lscube.
Diego E. 'Flameeyes' Pettenò [Fri, 21 Nov 2008 00:08:42 +0000 (01:08 +0100)]
Add the attributes.m4 macro file from xine/lscube.

This is a shared macro file that is currently maintained in both xine
and lscube repositories and contains a series of utility macros to
check compiler and linker features.

Signed-off-by: Diego E. 'Flameeyes' Pettenò <flameeyes@gmail.com>
15 years agoDon't use AC_CANONICAL_SYSTEM, only use AC_CANONICAL_HOST.
Diego E. 'Flameeyes' Pettenò [Fri, 21 Nov 2008 00:02:55 +0000 (01:02 +0100)]
Don't use AC_CANONICAL_SYSTEM, only use AC_CANONICAL_HOST.

Since alsa-lib is not a tool generating architecture code, the target
definition does not matter, instead use $host and $build properly.

See http://blog.flameeyes.eu/2008/10/11 for a detailed explanation of
the problem and the fix.

Signed-off-by: Diego E. 'Flameeyes' Pettenò <flameeyes@gmail.com>
15 years agoFix segfault with invalid meter plugin option
Takashi Iwai [Wed, 19 Nov 2008 15:14:27 +0000 (16:14 +0100)]
Fix segfault with invalid meter plugin option

snd_pcm_meter_add_scope_conf() may cause a segfault when pcm_scope_type
isn't defined.

Initialize type_conf properly to avoid it.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoAdd linear plugin wrapping iec958 PCM for ice1724-based boards
Takashi Iwai [Wed, 19 Nov 2008 13:59:03 +0000 (14:59 +0100)]
Add linear plugin wrapping iec958 PCM for ice1724-based boards

The ice1724-based cards can handle only 32bit while the apps almost
expet 16bit format for SPDIF I/O.  This prevents the default config
working on many apps like mplayer, xine, etc.

This patch simply adds the least automatic conversion by linear plugin.
Note that "plug" isn't used here.  Otherwise we get a problem of the
routing (plug over plug is buggy).

Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoFix volume/switch updates for global/simple mixer elements
Takashi Iwai [Mon, 10 Nov 2008 16:39:11 +0000 (17:39 +0100)]
Fix volume/switch updates for global/simple mixer elements

Fixed a long-standing bug that the values of global or simple mixer
elements aren't updated when dir = SM_CAPT is given.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoRelease v1.0.18
Jaroslav Kysela [Wed, 29 Oct 2008 12:42:10 +0000 (13:42 +0100)]
Release v1.0.18

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoMerge branch 'master' of git@git.alsa-project.org:alsa-lib
Jaroslav Kysela [Thu, 23 Oct 2008 07:58:10 +0000 (09:58 +0200)]
Merge branch 'master' of git@git.alsa-project.org:alsa-lib

15 years agoAdd snd_pcm_avail() and snd_pcm_avail_delay() prototypes
Jaroslav Kysela [Thu, 23 Oct 2008 07:57:11 +0000 (09:57 +0200)]
Add snd_pcm_avail() and snd_pcm_avail_delay() prototypes

Prototypes for new functions from commit
fe5391c9da2212dc594d1ee7f366e55432af0454
were lost somehow.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoAdd iec958 PCM definition for PS3
Takashi Iwai [Mon, 20 Oct 2008 09:25:12 +0000 (11:25 +0200)]
Add iec958 PCM definition for PS3

Added the iec958 PCM definition for PS3.
Since it's a new feature, the definition is marked as optional.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoFix sort-out of non-existing devices in namehint
Takashi Iwai [Fri, 17 Oct 2008 15:51:45 +0000 (17:51 +0200)]
Fix sort-out of non-existing devices in namehint

get_dev_name() checks wrongly the device_output, and it doesn't check
properly the case when only device is set and device_input and output
are unset.  This resulted in listing of non-existing HDMI device, for
example.  The patch fixes both issues.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoIncrease the components array size
Takashi Iwai [Tue, 7 Oct 2008 10:08:29 +0000 (12:08 +0200)]
Increase the components array size

Increase the components array size from 80 to 128 chars.
The string there is supposed to be null-terminated, so it can cover
the smaller array size in the older version, too.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoFix return value of snd_ctl_hw_subscribe_events()
Takashi Iwai [Tue, 30 Sep 2008 12:44:12 +0000 (14:44 +0200)]
Fix return value of snd_ctl_hw_subscribe_events()

snd_ctl_subscribe_events() must return 0 when succeeded.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agoReplace some assert() with runtime checks
Takashi Iwai [Tue, 30 Sep 2008 12:43:14 +0000 (14:43 +0200)]
Replace some assert() with runtime checks

assert() for sanity checks that can happen in runtime isn't a good idea.
Replaced it with the real check.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
15 years agotest/pcm: Fix error messages in async callback and add snd_pcm_start() condition
Jaroslav Kysela [Mon, 15 Sep 2008 13:14:58 +0000 (15:14 +0200)]
test/pcm: Fix error messages in async callback and add snd_pcm_start() condition

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoFix comment typos for snd_pcm_read[in]()
Jaroslav Kysela [Tue, 9 Sep 2008 17:56:17 +0000 (19:56 +0200)]
Fix comment typos for snd_pcm_read[in]()

From: William Estrada <MrUmunhum@popdial.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoRelease v1.0.18rc3
Jaroslav Kysela [Tue, 9 Sep 2008 16:55:49 +0000 (18:55 +0200)]
Release v1.0.18rc3

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agopcm_softvol plugin: remove access type change for refine
Jaroslav Kysela [Tue, 9 Sep 2008 11:07:39 +0000 (13:07 +0200)]
pcm_softvol plugin: remove access type change for refine

The softvol plugin does in-place sample recalculation. The access type
cannot be changed like in linear plugins. Remove access type change
in refine functions.

This bug was detected with PulseAudio on Sony PS3 platform.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoAdd snd_pcm_avail() and snd_pcm_avail_delay() functions. Make snd_pcm_hwsync() deprec...
Jaroslav Kysela [Mon, 1 Sep 2008 09:38:53 +0000 (11:38 +0200)]
Add snd_pcm_avail() and snd_pcm_avail_delay() functions. Make snd_pcm_hwsync() deprecated.

As proposed in http://mailman.alsa-project.org/pipermail/alsa-devel/2008-June/008558.html
the snd_pcm_avail() and snd_pcm_avail_delay() functions are now available
to get accurate stream position in a straight way. The snd_pcm_avail_delay()
function was added to ensure full sync between avail and delay values. It's
actually implemented using delay() + avail_update() calls but it might be
changed in future.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoalsa-lib: Don't call the simple mixer interface "legacy"
Lennart Poettering [Sun, 31 Aug 2008 01:42:56 +0000 (03:42 +0200)]
alsa-lib: Don't call the simple mixer interface "legacy"

The simpler mixer interface is not legacy, so the header file shouldn't claim
otherwise.

Fixes bug#3569

From: Lennart Poettering <mznyfn@0pointer.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoalsa-lib: Reword doxygen doc for snd_pcm_delay()
Lennart Poettering [Sun, 31 Aug 2008 01:42:20 +0000 (03:42 +0200)]
alsa-lib: Reword doxygen doc for snd_pcm_delay()

snd_pcm_delay() is for synchronization purposes, it returns the overall latency
of the stream, not just the latency induced by the hardware playback buffer.
The documentation is a bit misleading about this fact, and some people already
misunderstood it. So let's reword this to make the explanation clearer and
explicit.

This is a result of the discussions of the thread this mail belongs to:

http://mailman.alsa-project.org/pipermail/alsa-devel/2008-June/008456.html

From: Lennart Poettering <mznyfn@0pointer.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoalsa-lib: Make sure SND_PCM_NO_xxx flags don't get lost when nonblocking mode is...
Lennart Poettering [Sun, 31 Aug 2008 01:41:48 +0000 (03:41 +0200)]
alsa-lib: Make sure SND_PCM_NO_xxx flags don't get lost when nonblocking mode is enabled

The plug PCM copies the 'mode' field from the slave PCM. If blocking mode is
enabled for the plug PCM the mode is subsequently overwritten with the original
requested 'mode'. If non-blocking mode is requested this does not happen.

Because the hw PCM synthesizes the 'mode' from the actual file descriptor flags
no SND_PCM_NO_xxx will ever be set for it. This has the effect that the 'mode'
of the plug PCM will also not include those flags anymore -- unless they are
overwritten as mentioned above. This basically means SND_PCM_NO_xxx is ignored
for plug:hw:4711 style device strings opened in non-blocking mode.

You can easily test this with "aplay --channels 7 --disable-channels -f S16_LE
-r 44100 -D plug:hw:0" on a device that cannot do 7 channels. Normally this
call should fail, however if you add "-N" to the command line this call will
succeed.

This patch simply copies the SND_PCM_NO_xxx flags back into the 'mode' field in
case we don't overwrite it with the original anyway.

Probably closes bug 3571 for good.

From: Lennart Poettering <mznyfn@0pointer.de>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
15 years agoRelease v1.0.17a
Jaroslav Kysela [Fri, 15 Aug 2008 11:28:55 +0000 (13:28 +0200)]
Release v1.0.17a

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
16 years agoIEC958 definition for consumer status channel update
Pawel MOLL [Fri, 1 Aug 2008 10:23:50 +0000 (11:23 +0100)]
IEC958 definition for consumer status channel update

Updated IEC958 consumer status channel definitions according
to the third edition of IEC60958-3 spec.

Signed-off-by: Pawel Moll <pawel.moll@st.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
16 years agoFix non-tread mode with PCM direct plugins
Takashi Iwai [Thu, 31 Jul 2008 10:18:08 +0000 (12:18 +0200)]
Fix non-tread mode with PCM direct plugins

tread=0 wasn't set properly for very old ALSA drivers with no timer
tread support.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoFix interleave check in pcm_direct.c
Takashi Iwai [Tue, 29 Jul 2008 12:41:51 +0000 (14:41 +0200)]
Fix interleave check in pcm_direct.c

Fix the check of interleaved format.  It checked a totally bogus value
as PCM format.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agofix return value of snd_pcm_rewind()/snd_pcm_forward() to return how much actually...
Lennart Poettering [Fri, 18 Jul 2008 19:24:38 +0000 (21:24 +0200)]
fix return value of snd_pcm_rewind()/snd_pcm_forward() to return how much actually has been rewound, instead of what actually could have rewound

Make snd_pcm_plugin_rewind()/_forward() actually return how much has
been rewound/forwarded instead of how much could have been
rewounded/forwarded. This makes the code actually do what the
documentation of snd_pcm_rewind() suggests.

Signed-off-by: Lennart Poettering <lennart@poettering.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agofix type of internally used sframes variable, to avoid unnecessary casts
Lennart Poettering [Fri, 18 Jul 2008 19:22:50 +0000 (21:22 +0200)]
fix type of internally used sframes variable, to avoid unnecessary casts

This minor patch fixes the type of the sframes variable in
snd_pcm_plugin_forward(). With this fix we need to cast less and the
code is less confusing.

Signed-off-by: Lennart Poettering <lennart@poettering.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoclamp snd_pcm_rewind()/snd_pcm_forward() into the right direction
Lennart Poettering [Fri, 18 Jul 2008 19:21:23 +0000 (21:21 +0200)]
clamp snd_pcm_rewind()/snd_pcm_forward() into the right direction

The clamping of the input parameter in snd_pcm_rewind()/_forward() is
in the wrong direction.

Signed-off-by: Lennart Poettering <lennart@poettering.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoAdd boolean (mute) functionality to softvol plugin
Takashi Iwai [Wed, 16 Jul 2008 10:37:51 +0000 (12:37 +0200)]
Add boolean (mute) functionality to softvol plugin

When the resolution is set to 2, a boolean control is created as a
mute switch instead of a volume control.
Also, fixed the possible zero-division error.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoRelease v1.0.17
Jaroslav Kysela [Mon, 14 Jul 2008 08:57:56 +0000 (10:57 +0200)]
Release v1.0.17

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
16 years agotest/pcm.c: Fix SND_PCM_FORMAT_S24 support
Takashi Iwai [Thu, 10 Jul 2008 16:45:23 +0000 (18:45 +0200)]
test/pcm.c: Fix SND_PCM_FORMAT_S24 support

The program uses snd_pcm_format_width() wrongly to calculate the sample
size.  It must be snd_pcm_format_physical_width() instead.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoFix type-punning in test/pcm.c
Takashi Iwai [Thu, 10 Jul 2008 16:44:15 +0000 (18:44 +0200)]
Fix type-punning in test/pcm.c

The cast won't work well with strict aliasing.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agopcm: fix comment for snd_pcm_avail_update()
Jaroslav Kysela [Mon, 16 Jun 2008 11:23:22 +0000 (13:23 +0200)]
pcm: fix comment for snd_pcm_avail_update()

In some cases, value might be used for r/w ops, too.

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
16 years agoRelease v1.0.17rc2
Jaroslav Kysela [Mon, 16 Jun 2008 07:40:42 +0000 (09:40 +0200)]
Release v1.0.17rc2

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
16 years agoFix segfault with dmix of 3-bytes formats
Josh Green [Sat, 7 Jun 2008 08:10:49 +0000 (10:10 +0200)]
Fix segfault with dmix of 3-bytes formats

The i386 and x86-64 dmix may cause segfaults when 3-bytes formats are used
due to btsl asm code, which may overcome the buffer end-boundary.
The patch changes btsl to btsw so that it doesn't happen.

ALSA bug#3341:
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3341

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoPCM: allow mmap-access conversion in plug
Takashi Iwai [Fri, 6 Jun 2008 15:10:41 +0000 (17:10 +0200)]
PCM: allow mmap-access conversion in plug

The plug plugin has a long-standing problem that it can handle only
slaves that support mmap because of format/rate/access conversions
(these corresponding plugins work only with mmap).

This patch adds the support of automatic mmap->rw conversion via
mmap_emul plugin.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoExport __snd_pcm_mmap_emul_open()
Takashi Iwai [Fri, 6 Jun 2008 15:09:07 +0000 (17:09 +0200)]
Export __snd_pcm_mmap_emul_open()

Export __snd_pcm_mmap_emul_open() for plug layer.  This isn't exported
globally, though.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agopcm_mmap_emul: clean up
Takashi Iwai [Fri, 6 Jun 2008 15:07:45 +0000 (17:07 +0200)]
pcm_mmap_emul: clean up

A little bit of code clean up.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agopcm_mmap_emul: Fix invalid check
Takashi Iwai [Fri, 6 Jun 2008 15:05:56 +0000 (17:05 +0200)]
pcm_mmap_emul: Fix invalid check

The check in snd_pcm_mmap_emul_refine() is bogus and buggy.

Since the changed access type is took back at snd_pcm_mmap_emul_refine
again, it has to check the availability of mmap at each time.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoFix compile warnings in pcm_hw.c
Takashi Iwai [Fri, 6 Jun 2008 14:59:47 +0000 (16:59 +0200)]
Fix compile warnings in pcm_hw.c

Two trivial compile warning fixes:

- Add a missing return to snd_pcm_hw_clear_timer_queue()
- params->info is no long but int

The second one might have hit already on 64bit machine, but alas,
no one didn't notice it.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoFix cast warning
Takashi Iwai [Fri, 6 Jun 2008 14:58:02 +0000 (16:58 +0200)]
Fix cast warning

int64_t and long long isn't strictly identical, and thus gcc gives us
a heartful warning.  Suppress the warning by a pointer cast.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
16 years agoRelease v1.0.17rc1
Jaroslav Kysela [Tue, 3 Jun 2008 08:18:14 +0000 (10:18 +0200)]
Release v1.0.17rc1

Signed-off-by: Jaroslav Kysela <perex@perex.cz>
16 years agohgcompile -> gitcompile
Jaroslav Kysela [Tue, 3 Jun 2008 08:17:48 +0000 (10:17 +0200)]
hgcompile -> gitcompile

16 years agoremoved .hg files and renamed hgcompile to gitcompile
Jaroslav Kysela [Tue, 20 May 2008 11:29:44 +0000 (13:29 +0200)]
removed .hg files and renamed hgcompile to gitcompile

16 years agoImplemented snd_pcm_sw_params_(set|get)_period_event for interrupt wakeup like behaviour
Jaroslav Kysela [Fri, 9 May 2008 14:02:02 +0000 (16:02 +0200)]
Implemented snd_pcm_sw_params_(set|get)_period_event for interrupt wakeup like behaviour

Actually, PCM timer is used as source for poll(). It might be optimized
in the kernel code later.

16 years agoimplemented snd_pcm_rewindable() and snd_pcm_forwardable(), removed can_rewind and...
Jaroslav Kysela [Mon, 21 Apr 2008 10:46:50 +0000 (12:46 +0200)]
implemented snd_pcm_rewindable() and snd_pcm_forwardable(), removed can_rewind and can_forward

16 years agoimplemented integer volume <-> dB volume conversion functions for simple mixer
Jaroslav Kysela [Mon, 21 Apr 2008 09:44:20 +0000 (11:44 +0200)]
implemented integer volume <-> dB volume conversion functions for simple mixer

16 years agoadded snd_pcm_hw_params_is_monotonic/can_forward/can_rewind functions
Jaroslav Kysela [Fri, 11 Apr 2008 12:11:04 +0000 (14:11 +0200)]
added snd_pcm_hw_params_is_monotonic/can_forward/can_rewind functions

16 years agoPCM API - explain more trigger timestamp
Jaroslav Kysela [Thu, 10 Apr 2008 07:46:10 +0000 (09:46 +0200)]
PCM API - explain more trigger timestamp

16 years agofix compilation in pcm/pcm_hw.c - monotonic clock
Jaroslav Kysela [Thu, 10 Apr 2008 07:08:08 +0000 (09:08 +0200)]
fix compilation in pcm/pcm_hw.c - monotonic clock

16 years agopcsp: remove downsampling
Stas Sergeev [Sat, 22 Mar 2008 16:40:57 +0000 (17:40 +0100)]
pcsp: remove downsampling

apply softvol before plug as softvol doesn't support U8 as of now.
This also improves the sound quality.

Signed-off-by: Stas Sergeev <stsp@aknet.ru>
16 years agoUse defaults.pcm.file_format for the default file format of file plugin
Takashi Iwai [Mon, 17 Mar 2008 15:11:57 +0000 (16:11 +0100)]
Use defaults.pcm.file_format for the default file format of file plugin

Use "defaults.pcm.file_format" for the default file format of
file plugin.  It's set to "raw" as default for compatibility.

16 years agoAdd the support of WAV format in PCM file plugin
Takashi Iwai [Mon, 17 Mar 2008 15:02:56 +0000 (16:02 +0100)]
Add the support of WAV format in PCM file plugin

Added the support of WAV format in PCM file plugin.
The infile is still only in raw format.

16 years agoUse slave PCM as a timing-source for file ifile
Takashi Iwai [Fri, 14 Mar 2008 13:03:54 +0000 (14:03 +0100)]
Use slave PCM as a timing-source for file ifile

When ifile option is used for the file plugin, it ignores the slave PCM
and just feeds the input data.
This patch changes the behavior a bit - it uses the slave PCM as the
timing source (just read and throw data away) so that the input data
can be read in the right sample rate.

16 years agoAdd truncate option to PCM file plugin
Takashi Iwai [Wed, 12 Mar 2008 15:30:26 +0000 (16:30 +0100)]
Add truncate option to PCM file plugin

Addeed a new option "truncate" to indicate the behavior of creating
the output file.  When it's true (the default), the file is overwritten
and truncated at creation.  When false, the plugin tries to open a
unique file with a number suffix.

The global behavior of "file" and "tee" PCMs is defined via
defaults.pcm.file_truncate option.  You can overwrite it in ~/.asoundrc.

16 years agodmix skipping first set of samples
Takashi Iwai [Mon, 10 Mar 2008 11:02:26 +0000 (12:02 +0100)]
dmix skipping first set of samples

There was a change in alsa-lib 1.0.16 which looks like it was designed to
make dmix skip samples in the case of underruns, but it causes the first
sample to be skipped since dmix->slave_hw_ptr == dmix->slave_appl_ptr.
The following patch fixes this and fixes a small typo in the comment.

From: Mike Gorse <mgorse@mgorse.dhs.org>

16 years agoFix for alsa-lib cross-compilation problems with ALSA_CONFIG_DIR and ALSA_PLUGIN_DIR
Pawel MOLL [Fri, 7 Mar 2008 16:18:16 +0000 (17:18 +0100)]
Fix for alsa-lib cross-compilation problems with ALSA_CONFIG_DIR and ALSA_PLUGIN_DIR

"./configure" options for selecting ALSA configuration (default
/usr/share/alsa) and plugin (/usr/lib/alsa-lib) directories introduced
by alsa-hg/alsa-lib changeset 2284 cause problems with cross-compilation
and packaging - there is no way to redefine them in runtime, during
installation phase.

This patch adds a level of indirection between constants and their
usage - alsaconfigdir for ALSA_CONFIG_DIR and alsaplugindir for
ALSA_PLUGIN_DIR - which can be redefined during "make install" stage.

Signed-off-by: Pawel MOLL <pawel.moll@st.com>
16 years agoAdd PCM "hdmi"
Takashi Iwai [Fri, 7 Mar 2008 16:15:10 +0000 (17:15 +0100)]
Add PCM "hdmi"

Added the new PCM "hdmi" for HDA-Intel.
It's still experimental.

16 years agoDon't show non-existing devices in snd_device_name_hint()
Takashi Iwai [Fri, 7 Mar 2008 16:01:56 +0000 (17:01 +0100)]
Don't show non-existing devices in snd_device_name_hint()

Suppress the non-existing devices in snd_device_name_hint().

16 years agoFix device number assignment in hints
Takashi Iwai [Fri, 7 Mar 2008 16:00:45 +0000 (17:00 +0100)]
Fix device number assignment in hints

Handle the device number properly if given in hints.
The current code resets the device number to -1 wrongly.

16 years agoFix the build with old glibc
Takashi Iwai [Fri, 7 Mar 2008 13:35:25 +0000 (14:35 +0100)]
Fix the build with old glibc

The old systems don't support CLOCK_MONOTONIC although clock_gettime() API
itself exists.  This causes compile errors.

16 years agoIEC958 definitions for consumer status channel, byte 4
Pawel MOLL [Fri, 29 Feb 2008 11:42:57 +0000 (12:42 +0100)]
IEC958 definitions for consumer status channel, byte 4

Added definition for byte 4 of SPDIF channel status, according to
second edition of IEC 60958-3 (consumer) spec.

Signed-off-by: Pawel MOLL <pawel.moll@st.com>
16 years agoFix the state in snd_pcm_ioplug_pause()
Takashi Iwai [Mon, 25 Feb 2008 14:11:46 +0000 (15:11 +0100)]
Fix the state in snd_pcm_ioplug_pause()

The states[] in snd_pcm_ioplug_pause() has wrong values.  They should be
swapped.  ALSA bug#3796:
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3796

16 years agoAdd test binaries to hgignore
Takashi Iwai [Mon, 25 Feb 2008 13:58:45 +0000 (14:58 +0100)]
Add test binaries to hgignore

16 years agoAdd doxygen.cfg to hgignore
Takashi Iwai [Mon, 25 Feb 2008 13:57:31 +0000 (14:57 +0100)]
Add doxygen.cfg to hgignore

16 years agoCreate doxgen.cfg dynamically
Takashi Iwai [Mon, 25 Feb 2008 13:56:51 +0000 (14:56 +0100)]
Create doxgen.cfg dynamically

The patch from bug#3799
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3799

The "doc" target in doc/Makefile.am assumes the build directory and
the source directory are the same and fails if they differ ... this is
because the doxygen.cfg contains relative paths (such as ../) that
only work when building in the source tree

The attached patch against hg replaces all relative paths with
@top_srcdir@ and changes doxygen.cfg to a generated file so that
configure replaces @top_srcdir@ with the appropriate configure-time
path.

Assuming people like being able to do `make -f Makefile.am doc` on
an unconfigured and having it work, i added a small `test&&sed`
that'll generate an appropriate default doxygen.cfg for them.

16 years agomark snd_seq_client_info_{get,set}_event_filter deprecated
Aldrin Martoq [Fri, 22 Feb 2008 16:51:05 +0000 (17:51 +0100)]
mark snd_seq_client_info_{get,set}_event_filter deprecated

Mark snd_seq_client_info_{get,set}_event_filter deprecated

Signed-off-by: Aldrin Martoq <amartoq@dcc.uchile.cl>
16 years agouse snd_seq_client_info_event_filter_*() functions
Aldrin Martoq [Fri, 22 Feb 2008 16:50:01 +0000 (17:50 +0100)]
use snd_seq_client_info_event_filter_*() functions

Change snd_seq_set_client_event_filter to use the new
snd_seq_client_info_event_filter_* API

Signed-off-by: Aldrin Martoq <amartoq@dcc.uchile.cl>
16 years agoadd a test code for snd_seq_client_info_event_filter_*()
Aldrin Martoq [Fri, 22 Feb 2008 16:49:01 +0000 (17:49 +0100)]
add a test code for snd_seq_client_info_event_filter_*()

Added test code for
snd_seq_client_info_event_filter_{clear,add,del,check}

Signed-off-by: Aldrin Martoq <amartoq@dcc.uchile.cl>
16 years agoadd snd_seq_client_info_event_filter_*() functions
Aldrin Martoq [Fri, 22 Feb 2008 16:46:50 +0000 (17:46 +0100)]
add snd_seq_client_info_event_filter_*() functions

Added snd_seq_client_info_event_filter_{clear,add,del,check} to alsa
sequencer API

Signed-off-by: Aldrin Martoq <amartoq@dcc.uchile.cl>
16 years agoadd snd_seq_unset_bit()
Aldrin Martoq [Fri, 22 Feb 2008 16:42:56 +0000 (17:42 +0100)]
add snd_seq_unset_bit()

Added snd_seq_unset_bit() to alsa sequencer API

Signed-off-by: Aldrin Martoq <amartoq@dcc.uchile.cl>
16 years agoFix snd_seq_change_bit()
Takashi Iwai [Thu, 21 Feb 2008 11:31:18 +0000 (12:31 +0100)]
Fix snd_seq_change_bit()

snd_seq_change_bit() doesn't change but only set.

From: Aldrin Martoq <amartoq@dcc.uchile.cl>

16 years agoAdd atomic operation for super-H(sh3,4) architectures
Takashi Iwai [Mon, 18 Feb 2008 12:15:55 +0000 (13:15 +0100)]
Add atomic operation for super-H(sh3,4) architectures

This patch adds atomic operation for super-H(sh3,sh4) architecture.
See ALSA bug#3789
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3789

16 years agoAdd surround71 definition to NFORCE.conf
Takashi Iwai [Mon, 18 Feb 2008 11:25:02 +0000 (12:25 +0100)]
Add surround71 definition to NFORCE.conf

Now the board with ALC850 can work with 8-channel outputs.

16 years agoFix conflict of obsoleted snd_pcm_hw_* definitions
Takashi Iwai [Mon, 11 Feb 2008 13:19:32 +0000 (14:19 +0100)]
Fix conflict of obsoleted snd_pcm_hw_* definitions

When only SND_PCM_OLD_HW_PARAMS_API is defined but no
SND_PCM_OLD_SW_PARAMS_API, the declerations of some obsoleted functions
conflict.  Although the apps should define both at the same time, it's
not good to break.  Fixed the ifdef now.

16 years agoAdded tag v1.0.16 for changeset b1dfac5a4d55
Jaroslav Kysela [Tue, 5 Feb 2008 09:08:10 +0000 (10:08 +0100)]
Added tag v1.0.16 for changeset b1dfac5a4d55

16 years agorelease 1.0.16
Jaroslav Kysela [Tue, 5 Feb 2008 09:08:09 +0000 (10:08 +0100)]
release 1.0.16
Patch-level: Merged

16 years agoFix build with --disable-hwdep and co
Takashi Iwai [Mon, 4 Feb 2008 15:00:16 +0000 (16:00 +0100)]
Fix build with --disable-hwdep and co

control.h has function declarations with hwdep or rawmidi types
that aren't included when built without the corresponding supports.

Add ifdef appropriately to fix this.

16 years agoRemove obsolete instr check in configure
Takashi Iwai [Mon, 4 Feb 2008 14:59:07 +0000 (15:59 +0100)]
Remove obsolete instr check in configure

16 years agoFix wrong exported functions
Takashi Iwai [Mon, 4 Feb 2008 13:15:44 +0000 (14:15 +0100)]
Fix wrong exported functions

The internal functions (as alias of 0.9.0rc4 variants) must be
exported but they were wrongly listed in Versions file.
Fixed the function names now.

16 years agoAdd the missing card alias for Prodigy71Hifi
Takashi Iwai [Fri, 1 Feb 2008 11:00:39 +0000 (12:00 +0100)]
Add the missing card alias for Prodigy71Hifi

See ALSA bug#3735
https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3735

16 years agoAdded tag v1.0.16rc2 for changeset 06fcc64b20b3
Jaroslav Kysela [Tue, 29 Jan 2008 08:30:53 +0000 (09:30 +0100)]
Added tag v1.0.16rc2 for changeset 06fcc64b20b3

16 years agorelease 1.0.16rc2
Jaroslav Kysela [Tue, 29 Jan 2008 08:30:53 +0000 (09:30 +0100)]
release 1.0.16rc2
Patch-level: Merged

16 years agoAllow auto-config for dsnoop and dshare plugins
Takashi Iwai [Wed, 23 Jan 2008 11:26:22 +0000 (12:26 +0100)]
Allow auto-config for dsnoop and dshare plugins

16 years agoRemove obsolete instr directory
Takashi Iwai [Wed, 23 Jan 2008 07:06:11 +0000 (08:06 +0100)]
Remove obsolete instr directory

16 years agoAdded tag v1.0.16rc1 for changeset 81eabb0e6c10
Jaroslav Kysela [Mon, 21 Jan 2008 09:42:47 +0000 (10:42 +0100)]
Added tag v1.0.16rc1 for changeset 81eabb0e6c10

16 years agorelease 1.0.16rc1
Jaroslav Kysela [Mon, 21 Jan 2008 09:42:47 +0000 (10:42 +0100)]
release 1.0.16rc1
Patch-level: Merged

16 years agoFix a memory leak in PCM hook plugin
Takashi Iwai [Fri, 18 Jan 2008 13:43:56 +0000 (14:43 +0100)]
Fix a memory leak in PCM hook plugin

16 years agodmix - Enable auto format detection as default
Takashi Iwai [Fri, 18 Jan 2008 12:42:51 +0000 (13:42 +0100)]
dmix - Enable auto format detection as default

The direct plugins have the automatic format-detection feature but it
wasn't enabled properly in the interface.  Now you can pass the format
"unchanged" to make the plugin detect a proper format.

This will change the default format of some drivers, such as, HD-audio.

16 years agoImplement missing htimestamp callbacks
Takashi Iwai [Tue, 15 Jan 2008 13:29:34 +0000 (14:29 +0100)]
Implement missing htimestamp callbacks

Implemented the missing htimestamp callbacks for ioplug, rate and null
plugins.

16 years agoAllow pcm slave string references for direct plugins (bug#2893).
Jaroslav Kysela [Tue, 15 Jan 2008 13:07:19 +0000 (14:07 +0100)]
Allow pcm slave string references for direct plugins (bug#2893).

16 years agoAdd missing remix_areas_* for x86-64
Takashi Iwai [Mon, 14 Jan 2008 16:24:14 +0000 (17:24 +0100)]
Add missing remix_areas_* for x86-64

The remix_areas_* were missing the dmix x86-64 code.  Added now.

16 years agoAdd SND_PCM_TSTAMP_MMAP back
Takashi Iwai [Mon, 14 Jan 2008 16:15:13 +0000 (17:15 +0100)]
Add SND_PCM_TSTAMP_MMAP back

SND_PCM_TSTAMP_MMAP is used (blidnly) by portaudio, unfortunately.
Re-added it not to break API.

16 years agofix memory leak in snd_config_update_r error path
Clemens Ladisch [Mon, 14 Jan 2008 07:53:06 +0000 (08:53 +0100)]
fix memory leak in snd_config_update_r error path

Do not forget to free the memory for the file name when a file to be
read by snd_config_update_r() cannot be accessed.

16 years agopcm dmix plugin: fix generic direct remixing
Clemens Ladisch [Mon, 14 Jan 2008 07:52:16 +0000 (08:52 +0100)]
pcm dmix plugin: fix generic direct remixing

In the case of the sum buffer being uninitialized, the source sample
must be negated not only when writing to the sum but also when writing
to the destination.