summaryrefslogtreecommitdiffstats
path: root/sound/core
Commit message (Collapse)AuthorAgeFilesLines
* ALSA: pcm: Don't leave PREPARED state after drainingTakashi Iwai2015-02-231-0/+2
| | | | | | | | | | When a PCM draining is performed to an empty stream that has been already in PREPARED state, the current code just ignores and leaves as it is, although the drain is supposed to set all such streams to SETUP state. This patch covers that overlooked case. Cc: <stable@vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* ALSA: seq: potential out of bounds in do_control()Dan Carpenter2015-02-121-0/+3
| | | | | | | | | | | | | | | | | Smatch complains that "control" is user specifigy and needs to be capped. The call tree to understand this warning is quite long. snd_seq_write() <-- get the event from the user snd_seq_client_enqueue_event() snd_seq_deliver_event() deliver_to_subscribers() snd_seq_deliver_single_event() snd_opl3_oss_event_input() snd_midi_process_event() do_control() Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* ALSA: control: fix failure to return new numerical ID in 'replace' event dataTakashi Sakamoto2015-02-091-0/+1
| | | | | | | | | In 'replace' event data, numerical ID of control is always invalid. This commit fix this bug so as the event data has renewed numerical ID for control. Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* ALSA: pcm: allow for trigger_tstamp snapshot in .triggerPierre-Louis Bossart2015-02-091-1/+3
| | | | | | | | | Don't use generic snapshot of trigger_tstamp if low-level driver or hardware can get a more precise value for better audio/system time synchronization. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* ALSA: pcm: don't override timestamp unconditionallyPierre-Louis Bossart2015-02-091-1/+4
| | | | | | | | | | | | timestamp in RUNNING mode is already taken in update_hw_ptr routine, getting a new timestamp introduces offset between hw_ptr, audio_tstamp and system time Add else condition to read timestamp as fallback and only when enabled Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* ALSA: control: fix failure to return numerical ID in 'add' eventTakashi Sakamoto2015-02-081-0/+1
| | | | | | | | | Currently when adding a new control, the assigned numerical ID is not set for event data, thus userspace applications cannot realize it just by event data. Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* ALSA: Add a helper to add a new attribute group to cardTakashi Iwai2015-02-061-6/+25
| | | | | | | | | | | | | | | For assigning sysfs entries for a card device from the driver, introduce a new helper function, snd_card_add_dev_attr(). In this way, we can avoid the possible race between the device registration and the sysfs addition / removal. The driver can pass a new attribute group to add freely. This has to be called before snd_card_register(). Currently, up to two extra groups can be added. More than that, it'll return an error. Signed-off-by: Takashi Iwai <tiwai@suse.de>
* Merge tag 'asoc-v3.20-2' of ↵Takashi Iwai2015-02-051-0/+85
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-next ASoC: Updates for v3.20 More updates for v3.20: - Lots of refactoring from Lars-Peter Clausen, moving drivers to more data driven initialization and rationalizing a lot of DAPM usage. - Much improved handling of CDCLK clocks on Samsung I2S controllers. - Lots of driver specific cleanups and feature improvements. - CODEC support for TI PCM514x and TLV320AIC3104 devices. - Board support for Tegra systems with Realtek RT5677. Conflicts: sound/soc/intel/sst-mfld-platform-pcm.c
| * Merge remote-tracking branch 'asoc/topic/pcm512x' into asoc-nextMark Brown2015-02-041-0/+85
| |\
| | * ALSA: pcm: Add snd_interval_ranges() and snd_pcm_hw_constraint_ranges()Peter Rosin2015-01-281-0/+85
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add helper functions to allow drivers to specify several disjoint ranges for a variable. In particular, there is a codec (PCM512x) that has a hole in its supported range of rates, due to PLL and divider restrictions. This is like snd_pcm_hw_constraint_list(), but for ranges instead of points. Signed-off-by: Peter Rosin <peda@axentia.se> Reviewed-by: Lars-Peter Clausen <lars@metafoo.de> Reviewed-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Mark Brown <broonie@kernel.org>
* | | Merge branch 'topic/snd-device' into for-nextTakashi Iwai2015-02-039-255/+237
|\ \ \
| * | | ALSA: Drop snd_get_device() helperTakashi Iwai2015-02-021-43/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since the device is no longer hidden but embedded into each component, we no longer need snd_get_device(). Let's drop it and relevant codes. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: Simplify snd_device_register() variantsTakashi Iwai2015-02-028-86/+51
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now that all callers have been replaced with snd_device_register_for_dev(), let's drop the obsolete device registration code and concentrate only on the code handling struct device directly. That said, - remove the old snd_device_register(), - rename snd_device_register_for_dev() with snd_device_register(), - drop superfluous arguments from snd_device_register(), - change snd_unregister_device() to pass the device pointer directly Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: compress: Embed struct deviceTakashi Iwai2015-02-021-5/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Like previous patches, this one embeds the struct device into struct snd_compr. As the dev field wasn't used beforehand, it's reused as the new device struct. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: seq: Handle the device directlyTakashi Iwai2015-02-021-2/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Like the previous change for the timer device, this patch changes the device management for the ALSA sequencer device using the struct device directly. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: timer: Handle the device directlyTakashi Iwai2015-02-021-2/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a relatively straightforward change, using the struct device directly for managing the ALSA timer device. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: timer: Propagate the error at initializationTakashi Iwai2015-02-021-10/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | ... instead of just printing errors. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: rawmidi: Use rawmidi device file for kernel messagesTakashi Iwai2015-02-021-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ... instead of card's device. This will be helpful to distinguish errors from multiple rawmidi devices on a single card. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: rawmidi: Embed struct deviceTakashi Iwai2015-02-021-7/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Like previous patches, this changes the device management for rawmidi, embedding the struct device into struct snd_rawmidi. The required change is more or less same as hwdep device. The currently unused dev field is reused as the new embedded struct field now. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: pcm: Embed struct deviceTakashi Iwai2015-02-021-23/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Like previous patches, at this time we embed the struct device into PCM object. However, this needs a bit more caution: struct snd_pcm doesn't own one device but two, for both playback and capture! Thus not struct snd_pcm but struct snd_pcm_str object contains the device. Along with this change, pcm->dev field is dropped for avoiding confusion. It was meant to point to a non-standard parent. But, since now we can touch each struct device directly, we can manipulate the parent field easily there, too. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: hwdep: Embed struct deviceTakashi Iwai2015-02-021-50/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Like the previous patch, this one embeds the device object into hwdep object. For a proper object lifecycle, it's freed in the release callback. This also allows us to create sysfs entries via passing to the groups field of the device without explicit function calls. Since each driver can see the device and touch its groups field directly, we don't need to delegate in hwdep core any longer. So, remove the groups field from snd_hwdep, and let the user (in this case only hda_hwdep.c) modify the device groups. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: control: Embed struct deviceTakashi Iwai2015-02-021-24/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch embeds a struct device for the control device into the card object and avoid the device creation at registration time. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: Add a helper to initialize deviceTakashi Iwai2015-02-021-0/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Introduce a new helper function snd_device_initialize() to initialize the device object for sound devices. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: control: Provide a helper to look for the preferred subdeviceTakashi Iwai2015-02-023-27/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of open-coding the search over the control file loop, provide a helper function for the preferred subdevice assigned to the current process. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
| * | | ALSA: Allow to pass the device object to snd_register_device*()Takashi Iwai2015-02-023-22/+37
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a preliminary patch for the further work on embedding struct device into each sound device instance. It changes snd_register_device*() helpers to receive the device object directly for skipping creating a device there. Reviewed-by: Jaroslav Kysela <perex@perex.cz> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | | ALSA: Include linux/uaccess.h and linux/bitopts.h instead of asm/*Takashi Iwai2015-01-281-1/+1
| | | | | | | | | | | | Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | | ALSA: Include linux/io.h instead of asm/io.hTakashi Iwai2015-01-283-3/+3
| | | | | | | | | | | | | | | | | | Nowadays it's recommended. Replace all in a shot. Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | | ALSA: seq: remove unused callback_all fieldClemens Ladisch2015-01-263-7/+2
| | | | | | | | | | | | | | | Signed-off-by: Clemens Ladisch <clemens@ladisch.de> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | | ALSA: seq: fix off-by-one error in port limit checkClemens Ladisch2015-01-261-1/+1
| | | | | | | | | | | | | | | Signed-off-by: Clemens Ladisch <clemens@ladisch.de> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | | ALSA: seq: correctly report maximum number of portsClemens Ladisch2015-01-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Due to SNDRV_SEQ_ADDRESS_BROADCAST, not all 256 port number values can be used. Signed-off-by: Clemens Ladisch <clemens@ladisch.de> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | | Merge branch 'for-linus' into for-nextTakashi Iwai2015-01-261-31/+0
|\ \ \ | |/ / | | | | | | | | | Sync with the latest 3.19-rc state for applying other ALSA sequencer core fixes.
| * | ALSA: seq-dummy: remove deadlock-causing events on closeClemens Ladisch2015-01-261-31/+0
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the last subscriber to a "Through" port has been removed, the subscribed destination ports might still be active, so it would be wrong to send "all sounds off" and "reset controller" events to them. The proper place for such a shutdown would be the closing of the actual MIDI port (and close_substream() in rawmidi.c already can do this). This also fixes a deadlock when dummy_unuse() tries to send events to its own port that is already locked because it is being freed. Reported-by: Peter Billam <peter@www.pjb.com.au> Cc: <stable@vger.kernel.org> Signed-off-by: Clemens Ladisch <clemens@ladisch.de> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: timer: Use setup_timer() and mod_timer()Takashi Iwai2015-01-191-3/+1
| | | | | | | | | | | | No functional change, refactoring with the standard helpers. Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: seq: Deletion of unnecessary checks before the function call ↵Markus Elfring2015-01-042-6/+3
| | | | | | | | | | | | | | | | | | | | | | | | "snd_midi_event_free" The snd_midi_event_free() function tests whether its argument is NULL and then returns immediately. Thus the test around the call is not needed. This issue was detected by using the Coccinelle software. Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: pcm: add SNDRV_PCM_TRIGGER_DRAIN triggerLibin Yang2014-12-311-1/+9
| | | | | | | | | | | | | | | | | | | | Add SNDRV_PCM_TRIGGER_DRAIN trigger for pcm drain. Some audio devices require notification of drain events in order to properly drain and shutdown an audio stream. Signed-off-by: Libin Yang <libin.yang@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | Merge branch 'topic/pcm-params' into for-nextTakashi Iwai2014-12-301-3/+3
|\ \
| * | ALSA: snd_pcm_oss_period_size: Use round{up,down}_pow_of_two()Lars-Peter Clausen2014-12-301-3/+3
| |/ | | | | | | | | | | | | | | | | | | Instead of opencoding them use the standard roundup_pow_of_two() and rounddown_pow_of_two() helper functions. This gets rids one of the few users of the custom ld2() function and also makes it a bit more obvious what the code does. Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: Add support for wildcard msbits constraintsLars-Peter Clausen2014-12-301-1/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently the msbits constraints requires to specify a specific sample format width for which the constraint should be applied. But often the number of most significant bits is not sample format specific, but rather a absolute limit. E.g. the PCM interface might accept 32-bit and 24-bit samples, but the DAC has a 16-bit resolution and throws away the LSBs. In this case for both 32-bit and 24-bit format msbits should be set to 16. This patch extends snd_pcm_hw_constraint_msbits() so that a wildcard constraint can be setup that is applied for all formats with a sample width larger than the specified msbits. Choosing the wildcard constraint is done by setting the sample width parameter of the function to 0. Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: Fix handling of multiple msbits constraints on the same runtimeLars-Peter Clausen2014-12-301-1/+1
|/ | | | | | | | | | | | | If the sound card is made up of discrete components, each with their own driver (e.g. like in the ASoC case), we might end up with multiple msbits constraint rules installed. Currently this will result in msbits being set to whatever the last rule set it to. This patch updates the behavior of the rule to choose the minimum (other than zero) of all the installed rules. Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* Merge branch 'for-linus' into for-nextTakashi Iwai2014-11-282-0/+10
|\ | | | | | | | | | | The commit [7a2e9ddc: ALSA: usb-audio: Add native DSD support for Denon/Marantz DACs] requires the new format definition that has landed only in for-next branch.
| * ALSA: pcm: Add big-endian DSD sample formats and fix XMOS DSD sample formatJussi Laako2014-11-212-0/+10
| | | | | | | | | | | | | | | | | | This patch fixes XMOS DSD sample format to DSD_U32_BE and also adds DSD_U16_BE and DSD_U32_BE sample formats. Signed-off-by: Jussi Laako <jussi@sonarnerd.net> Acked-by: Jurgen Kramer <gtmkramer@xs4all.nl> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: core: Deletion of unnecessary checks before two function callsMarkus Elfring2014-11-212-8/+4
| | | | | | | | | | | | | | | | | | | | The functions snd_seq_oss_timer_delete() and vunmap() perform also input parameter validation. Thus the test around the call is not needed. This issue was detected by using the Coccinelle software. Signed-off-by: Markus Elfring <elfring@users.sourceforge.net> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: pcm: Fix document for snd_pcm_stop_xrun()Takashi Iwai2014-11-101-1/+0
| | | | | | | | | | | | | | | | | | | | Fix a copy & paste error: Warning(sound/core/pcm_native.c:1112): Excess function parameter 'state' description in 'snd_pcm_stop_xrun' The state argument was dropped from snd_pcm_stop_xrun(). Reported-by: kbuild test robot <fengguang.wu@intel.com> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: pcm: Add snd_pcm_stop_xrun() helperTakashi Iwai2014-11-091-0/+23
| | | | | | | | | | | | | | | | | | | | | | Add a new helper function snd_pcm_stop_xrun() to the standard sequnce lock/snd_pcm_stop(XRUN)/unlock by a single call, and replace the existing open codes with this helper. The function checks the PCM running state to prevent setting the wrong state, too, for more safety. Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: snd_ctl_activate_id(): Fix index look-upLars-Peter Clausen2014-11-071-1/+1
| | | | | | | | | | | | | | | | We want to know the offset for the id that was passed to the function, not the offset of the first id of the control (which is always 0). Signed-off-by: Lars-Peter Clausen <lars@metafoo.de> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: pcm: Update the state properly before notificationTakashi Iwai2014-11-061-4/+4
| | | | | | | | | | | | | | | | | | | | Some state changes (e.g. snd_pcm_stop()) sets the runtime state after calling snd_timer_notify(). This is basically racy, since the notification may wakes up the user even before the state change. Although the possibility is low, we should set the state before the notifications. Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | Merge branch 'for-linus' into for-nextTakashi Iwai2014-11-051-0/+2
|\ \ | |/ | | | | | | | | | | | | This merges the USB-audio disconnect fix and resolves the conflicts so that we can continue working on development of usb-audio stuff. Conflicts: sound/usb/card.c
| * ALSA: pcm: Zero-clear reserved fields of PCM status ioctl in compat modeTakashi Iwai2014-10-281-0/+2
| | | | | | | | | | | | | | | | | | | | | | In compat mode, we copy each field of snd_pcm_status struct but don't touch the reserved fields, and this leaves uninitialized values there. Meanwhile the native ioctl does zero-clear the whole structure, so we should follow the same rule in compat mode, too. Reported-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Cc: <stable@vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: pcm: Add xrun_injection proc entryTakashi Iwai2014-11-041-0/+33
| | | | | | | | | | | | | | | | | | | | | | | | This patch adds a new proc entry for PCM substreams to inject an XRUN. When a PCM substream is running and any value is written to its xrun_injection proc file, the driver triggers XRUN. This is a useful feature for debugging XRUN and error handling code paths. Note that this entry is enabled only when CONFIG_SND_PCM_XRUN_DEBUG is set. Signed-off-by: Takashi Iwai <tiwai@suse.de>
* | ALSA: pcm: Replace PCM hwptr tracking with tracepointsTakashi Iwai2014-11-043-122/+137
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ALSA PCM core has a mechanism tracking the PCM hwptr updates for analyzing XRUNs. But its log is limited (up to 10) and its log output is a kernel message, which is hard to handle. In this patch, the hwptr logging is moved to the tracing infrastructure instead of its own. Not only the hwptr updates but also XRUN and hwptr errors are recorded on the trace log, so that user can see such events at the exact timing. The new "snd_pcm" entry will appear in the tracing events: # ls -F /sys/kernel/debug/tracing/events/snd_pcm enable filter hw_ptr_error/ hwptr/ xrun/ The hwptr is for the regular hwptr update events. An event trace looks like: aplay-26187 [004] d..3 4012.834761: hwptr: pcmC0D0p/sub0: POS: pos=488, old=0, base=0, period=1024, buf=16384 "POS" shows the hwptr update by the explicit position update call and "IRQ" means the hwptr update by the interrupt, i.e. snd_pcm_period_elapsed() call. The "pos" is the passed ring-buffer offset by the caller, "old" is the previous hwptr, "base" is the hwptr base position, "period" and "buf" are period- and buffer-size of the target PCM substream. (Note that the hwptr position displayed here isn't the ring-buffer offset. It increments up to the PCM position boundary.) The XRUN event appears similarly, but without "pos" field. The hwptr error events appear with the PCM identifier and its reason string, such as "Lost interrupt?". The XRUN and hwptr error reports on kernel message are still left, can be turned on/off via xrun_debug proc like before. But the bit 3, 4, 5 and 6 bits of xrun_debug proc are dropped by this patch. Also, along with the change, the message strings have been reformatted to be a bit more consistent. Last but not least, the hwptr reporting is enabled only when CONFIG_SND_PCM_XRUN_DEBUG is set. Signed-off-by: Takashi Iwai <tiwai@suse.de>
OpenPOWER on IntegriCloud