summaryrefslogtreecommitdiff
path: root/sound/soc/samsung/i2s.c
AgeCommit message (Collapse)AuthorFilesLines
2019-02-12ASoC: samsung: i2s: Fix prescaler setting for the secondary DAISylwester Nawrocki1-1/+7
Make sure i2s->rclk_srcrate is properly initialized also during playback through the secondary DAI. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Acked-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2019-02-07ASoC: samsung: Prevent clk_get_rate() calls in atomic contextSylwester Nawrocki1-5/+5
This patch moves clk_get_rate() call from trigger() to hw_params() callback to avoid calling sleeping clk API from atomic context and prevent deadlock as indicated below. Before this change clk_get_rate() was being called with same spinlock held as the one passed to the clk API when registering clocks exposed by the I2S driver. [ 82.109780] BUG: sleeping function called from invalid context at kernel/locking/mutex.c:908 [ 82.117009] in_atomic(): 1, irqs_disabled(): 128, pid: 1554, name: speaker-test [ 82.124235] 3 locks held by speaker-test/1554: [ 82.128653] #0: cc8c5328 (snd_pcm_link_rwlock){...-}, at: snd_pcm_stream_lock_irq+0x20/0x38 [ 82.137058] #1: ec9eda17 (&(&substream->self_group.lock)->rlock){..-.}, at: snd_pcm_ioctl+0x900/0x1268 [ 82.146417] #2: 6ac279bf (&(&pri_dai->spinlock)->rlock){..-.}, at: i2s_trigger+0x64/0x6d4 [ 82.154650] irq event stamp: 8144 [ 82.157949] hardirqs last enabled at (8143): [<c0a0f574>] _raw_read_unlock_irq+0x24/0x5c [ 82.166089] hardirqs last disabled at (8144): [<c0a0f6a8>] _raw_read_lock_irq+0x18/0x58 [ 82.174063] softirqs last enabled at (8004): [<c01024e4>] __do_softirq+0x3a4/0x66c [ 82.181688] softirqs last disabled at (7997): [<c012d730>] irq_exit+0x140/0x168 [ 82.188964] Preemption disabled at: [ 82.188967] [<00000000>] (null) [ 82.195728] CPU: 6 PID: 1554 Comm: speaker-test Not tainted 5.0.0-rc5-00192-ga6e6caca8f03 #191 [ 82.204302] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) [ 82.210376] [<c0111a54>] (unwind_backtrace) from [<c010d8f4>] (show_stack+0x10/0x14) [ 82.218084] [<c010d8f4>] (show_stack) from [<c09ef004>] (dump_stack+0x90/0xc8) [ 82.225278] [<c09ef004>] (dump_stack) from [<c0152980>] (___might_sleep+0x22c/0x2c8) [ 82.232990] [<c0152980>] (___might_sleep) from [<c0a0a2e4>] (__mutex_lock+0x28/0xa3c) [ 82.240788] [<c0a0a2e4>] (__mutex_lock) from [<c0a0ad80>] (mutex_lock_nested+0x1c/0x24) [ 82.248763] [<c0a0ad80>] (mutex_lock_nested) from [<c04923dc>] (clk_prepare_lock+0x78/0xec) [ 82.257079] [<c04923dc>] (clk_prepare_lock) from [<c049538c>] (clk_core_get_rate+0xc/0x5c) [ 82.265309] [<c049538c>] (clk_core_get_rate) from [<c0766b18>] (i2s_trigger+0x490/0x6d4) [ 82.273369] [<c0766b18>] (i2s_trigger) from [<c074fec4>] (soc_pcm_trigger+0x100/0x140) [ 82.281254] [<c074fec4>] (soc_pcm_trigger) from [<c07378a0>] (snd_pcm_do_start+0x2c/0x30) [ 82.289400] [<c07378a0>] (snd_pcm_do_start) from [<c07376cc>] (snd_pcm_action_single+0x38/0x78) [ 82.298065] [<c07376cc>] (snd_pcm_action_single) from [<c073a450>] (snd_pcm_ioctl+0x910/0x1268) [ 82.306734] [<c073a450>] (snd_pcm_ioctl) from [<c0292344>] (do_vfs_ioctl+0x90/0x9ec) [ 82.314443] [<c0292344>] (do_vfs_ioctl) from [<c0292cd4>] (ksys_ioctl+0x34/0x60) [ 82.321808] [<c0292cd4>] (ksys_ioctl) from [<c0101000>] (ret_fast_syscall+0x0/0x28) [ 82.329431] Exception stack(0xeb875fa8 to 0xeb875ff0) [ 82.334459] 5fa0: 00033c18 b6e31000 00000004 00004142 00033d80 00033d80 [ 82.342605] 5fc0: 00033c18 b6e31000 00008000 00000036 00008000 00000000 beea38a8 00008000 [ 82.350748] 5fe0: b6e3142c beea384c b6da9a30 b6c9212c [ 82.355789] [ 82.357245] ====================================================== [ 82.363397] WARNING: possible circular locking dependency detected [ 82.369551] 5.0.0-rc5-00192-ga6e6caca8f03 #191 Tainted: G W [ 82.376395] ------------------------------------------------------ [ 82.382548] speaker-test/1554 is trying to acquire lock: [ 82.387834] 6d2007f4 (prepare_lock){+.+.}, at: clk_prepare_lock+0x78/0xec [ 82.394593] [ 82.394593] but task is already holding lock: [ 82.400398] 6ac279bf (&(&pri_dai->spinlock)->rlock){..-.}, at: i2s_trigger+0x64/0x6d4 [ 82.408197] [ 82.408197] which lock already depends on the new lock. [ 82.416343] [ 82.416343] the existing dependency chain (in reverse order) is: [ 82.423795] [ 82.423795] -> #1 (&(&pri_dai->spinlock)->rlock){..-.}: [ 82.430472] clk_mux_set_parent+0x34/0xb8 [ 82.434975] clk_core_set_parent_nolock+0x1c4/0x52c [ 82.440347] clk_set_parent+0x38/0x6c [ 82.444509] of_clk_set_defaults+0xc8/0x308 [ 82.449186] of_clk_add_provider+0x84/0xd0 [ 82.453779] samsung_i2s_probe+0x408/0x5f8 [ 82.458376] platform_drv_probe+0x48/0x98 [ 82.462879] really_probe+0x224/0x3f4 [ 82.467037] driver_probe_device+0x70/0x1c4 [ 82.471716] bus_for_each_drv+0x44/0x8c [ 82.476049] __device_attach+0xa0/0x138 [ 82.480382] bus_probe_device+0x88/0x90 [ 82.484715] deferred_probe_work_func+0x6c/0xbc [ 82.489741] process_one_work+0x200/0x740 [ 82.494246] worker_thread+0x2c/0x4c8 [ 82.498408] kthread+0x128/0x164 [ 82.502131] ret_from_fork+0x14/0x20 [ 82.506204] (null) [ 82.508976] [ 82.508976] -> #0 (prepare_lock){+.+.}: [ 82.514264] __mutex_lock+0x60/0xa3c [ 82.518336] mutex_lock_nested+0x1c/0x24 [ 82.522756] clk_prepare_lock+0x78/0xec [ 82.527088] clk_core_get_rate+0xc/0x5c [ 82.531421] i2s_trigger+0x490/0x6d4 [ 82.535494] soc_pcm_trigger+0x100/0x140 [ 82.539913] snd_pcm_do_start+0x2c/0x30 [ 82.544246] snd_pcm_action_single+0x38/0x78 [ 82.549012] snd_pcm_ioctl+0x910/0x1268 [ 82.553345] do_vfs_ioctl+0x90/0x9ec [ 82.557417] ksys_ioctl+0x34/0x60 [ 82.561229] ret_fast_syscall+0x0/0x28 [ 82.565477] 0xbeea384c [ 82.568421] [ 82.568421] other info that might help us debug this: [ 82.568421] [ 82.576394] Possible unsafe locking scenario: [ 82.576394] [ 82.582285] CPU0 CPU1 [ 82.586792] ---- ---- [ 82.591297] lock(&(&pri_dai->spinlock)->rlock); [ 82.595977] lock(prepare_lock); [ 82.601782] lock(&(&pri_dai->spinlock)->rlock); [ 82.608975] lock(prepare_lock); [ 82.612268] [ 82.612268] *** DEADLOCK *** Fixes: 647d04f8e07a ("ASoC: samsung: i2s: Ensure the RCLK rate is properly determined") Reported-by: Krzysztof Kozłowski <krzk@kernel.org> Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2018-08-03ASoC: samsung: i2s: Mark expected switch fall-throughGustavo A. R. Silva1-0/+1
In preparation to enabling -Wimplicit-fallthrough, mark switch cases where we are expecting to fall through. Addresses-Coverity-ID: 1381093 ("Missing break in switch") Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2018-03-12ASoC: samsung: i2s: Fix rclk_srcrate handlingSylwester Nawrocki1-1/+1
As the RCLK clock may be updated through the common clk API before each snd_soc_dai_ops::trigger call, it is not enough to update i2s->rclk_srcrate only once after it has been initially set to 0. To avoid wrong PSR values we always get RCLK frequency from the CLK_I2S_RCLK_SRC clock, when that clock is available. Fixes: e1417fdf3011 "ASoC: samsung: i2s: Ensure the RCLK rate is properly determined" Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2018-02-14ASoC: samsung: i2s: Define the parameters list for SAMSUNG_I2S_OPCLKSylwester Nawrocki1-1/+1
The SAMSUNG_I2S_OPCLK is not currently used by any card driver thus we can safely change semantics of 'dir' argument of the I2S set_sysclk() callback. Now an enumeration is exported instead of directly using register bit field values. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Acked-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2018-02-14ASoC: samsung: i2s: Prevent external abort on exynos5433 I2S1 accessSylwester Nawrocki1-2/+7
It seems both PCLK_I2S1 and SCLK_I2S1 clocks need to be enabled before I2S1 control registers can be accessed on exynos5433. If SCLK clock is disabled an exception is triggered. To fix this parent clock of the RCLK_SRC clock is assigned to pri_dai->op_clk so required gate clock is handled by the runtime PM ops. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Acked-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2018-02-14ASoC: samsung: i2s: Ensure names of supplied clocks are uniqueSylwester Nawrocki1-9/+19
In order to support multiple instances of the I2S IP block the platform device name is prepended to each clock registered by the driver. The clock-output-names property is now not used, this should not cause any issues as, for example, CDCLK clock is referenced through DT 'clocks' property, not by name. This change allows to have both I2S0 and I2S1 enabled simultaneously on exynos5433 and working properly when #clock-cells property is specified in respective DT nodes. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Acked-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2018-02-12ASoC: samsung: i2s: Ensure the RCLK rate is properly determinedSylwester Nawrocki1-2/+11
If the RCLK mux clock configuration is specified in DT and no set_sysclk() callback is used in the sound card driver the sclk_srcrate field will remain set to 0, leading to an incorrect PSR divider setting. To fix this the frequency value is retrieved from the CLK_I2S_RCLK_SRC clock, so the actual RCLK mux selection is taken into account. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Acked-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2017-11-11Merge remote-tracking branches 'asoc/topic/samsung', 'asoc/topic/sgtl5000', ↵Mark Brown1-0/+4
'asoc/topic/stm32', 'asoc/topic/sunxi' and 'asoc/topic/tas571x' into asoc-next
2017-10-18ASoC: samsung: i2s: disable secondary DAI until it gets fixedMarek Szyprowski1-0/+1
Secondary DAI in Exynos I2S driver is not used by any of the currently supported boards and it causes problems due to some limitations in the ASoC code. Disable it until it gets proper support both by board-specific and ASoC core code. Also disable IDMA support, which relies on secondary DAI presence. This patch fixes following kernel warning: samsung-i2s 3830000.i2s: ASoC: Failed to create component debugfs directory samsung-i2s 3830000.i2s: ASoC: Failed to create component debugfs directory ------------[ cut here ]------------ WARNING: CPU: 3 PID: 82 at fs/proc/generic.c:330 proc_register+0xec/0x10c proc_dir_entry 'sub0/prealloc' already registered Modules linked in: CPU: 3 PID: 82 Comm: kworker/3:1 Not tainted 4.14.0-rc5-next-20171017 #3089 Hardware name: SAMSUNG EXYNOS (Flattened Device Tree) Workqueue: events deferred_probe_work_func [<c0110114>] (unwind_backtrace) from [<c010c900>] (show_stack+0x10/0x14) [<c010c900>] (show_stack) from [<c083e664>] (dump_stack+0x90/0xc8) [<c083e664>] (dump_stack) from [<c011d2b8>] (__warn+0xd4/0x100) [<c011d2b8>] (__warn) from [<c011d384>] (warn_slowpath_fmt+0x38/0x48) [<c011d384>] (warn_slowpath_fmt) from [<c0271268>] (proc_register+0xec/0x10c) [<c0271268>] (proc_register) from [<c027130c>] (proc_create_data+0x84/0xc8) [<c027130c>] (proc_create_data) from [<c061afbc>] (snd_info_register+0x64/0xcc) [<c061afbc>] (snd_info_register) from [<c062a6e0>] (snd_pcm_lib_preallocate_pages1+0x78/0x1a0) [<c062a6e0>] (snd_pcm_lib_preallocate_pages1) from [<c063eef4>] (dmaengine_pcm_new+0xa0/0x1ec) [<c063eef4>] (dmaengine_pcm_new) from [<c062b9f8>] (snd_soc_platform_drv_pcm_new+0x1c/0x28) [<c062b9f8>] (snd_soc_platform_drv_pcm_new) from [<c063d54c>] (soc_new_pcm+0x2f4/0x4f4) [<c063d54c>] (soc_new_pcm) from [<c063107c>] (snd_soc_register_card+0xc4c/0xdc4) [<c063107c>] (snd_soc_register_card) from [<c063db30>] (devm_snd_soc_register_card+0x34/0x70) [<c063db30>] (devm_snd_soc_register_card) from [<c064af60>] (asoc_simple_card_probe+0x230/0x47c) [<c064af60>] (asoc_simple_card_probe) from [<c047f8fc>] (platform_drv_probe+0x50/0xb0) [<c047f8fc>] (platform_drv_probe) from [<c047dee0>] (driver_probe_device+0x2a0/0x46c) [<c047dee0>] (driver_probe_device) from [<c047c0bc>] (bus_for_each_drv+0x44/0x8c) [<c047c0bc>] (bus_for_each_drv) from [<c047db50>] (__device_attach+0xa0/0x134) [<c047db50>] (__device_attach) from [<c047cf7c>] (bus_probe_device+0x88/0x90) [<c047cf7c>] (bus_probe_device) from [<c047d484>] (deferred_probe_work_func+0x3c/0x168) [<c047d484>] (deferred_probe_work_func) from [<c01371f8>] (process_one_work+0x188/0x41c) [<c01371f8>] (process_one_work) from [<c01374b4>] (process_scheduled_works+0x28/0x38) [<c01374b4>] (process_scheduled_works) from [<c01376d4>] (worker_thread+0x210/0x4dc) [<c01376d4>] (worker_thread) from [<c013d9cc>] (kthread+0x128/0x164) [<c013d9cc>] (kthread) from [<c0108848>] (ret_from_fork+0x14/0x2c) ---[ end trace bad8db6ee771d094 ]-- Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2017-09-19ASoC: samsung: i2s: Use specific name for i2s daisJaechul Lee1-0/+4
Add specific dais name when components are registered. Component and dai name will follow their parent dev name, if the name isn't described. In case of this driver, each dais will have same name like '11440000.i2s0' by fmt_single_name function. The problem having same name is that TM2 machine driver can't detect capture devices correctly. Machine driver doesn't know which one is proper to use for cpu dai. The driver just selects to use 'samsung-i2c-sec' that doesn't have capture functionality because the component of samsung-i2s-sec is located in the first of the component_list. I add dai name like 'samsung-i2s', 'samsung-i2s-sec' for each dais. The reason why adding dai id to 1 is that it doesn't allow to use particular dai name in case of when I use 0 for dai id. Signed-off-by: Jaechul Lee <jcsing.lee@samsung.com> Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2017-09-04ASoC: samsung: i2s: Fix error handling path in i2s_set_sysclk()Christophe Jaillet1-1/+4
If 'clk_prepare_enable()' fails, we must 'put' the corresponding clock. Othewise, there is a resource leak. Fixes: f5c97c7b0438 ("ASoC: samsung: i2s: Handle return value of clk_prepare_enable.") Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Signed-off-by: Mark Brown <broonie@kernel.org>
2017-08-15ASoC: samsung: i2s: Null pointer dereference on samsung_i2s_removeAnton Vasilyev1-5/+1
If (quirks & QUIRK_SEC_DAI == 0) then samsung_i2s_probe() doesn't allocate sec_dai and pri_dai->sec_dai remains Null, but samsung_i2s_remove() performs pri_dai->sec_dai dereference in any case. The patch removes useless reinitialization of sec_dai at samsung_i2s_remove(), because resources are under devm control. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by: Anton Vasilyev <vasilyev@ispras.ru> Acked-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2017-07-26ASoC: samsung: i2s: Handle return value of clk_prepare_enable.Arvind Yadav1-4/+15
clk_prepare_enable() can fail here and we must check its return value. Signed-off-by: Arvind Yadav <arvind.yadav.cs@gmail.com> Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2017-07-17ASoC: samsung: i2s: Support more resolution ratesJaechul Lee1-7/+13
This driver can support more frequencies over 96KHz. There are no reasons to limit the frequency range below 96KHz. If codecs/amps or something else can't support higher resolution rates, the constraints would be set rates properly because each drivers have its own limits. I added the 'pcm_rates' field to the dai_data to be set rates by the compatibilities. As a result, rates will be set each devices respectively. For example of exynos5433, rates will be set from 8KHz to 192KHz. Signed-off-by: Jaechul Lee <jcsing.lee@samsung.com> Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2017-03-07ASoC: samsung: Remove extra blank linesCalin Cruceru1-1/+0
This was reported by checkpatch.pl Signed-off-by: Calin Cruceru <calin@rosedu.org> Signed-off-by: Mark Brown <broonie@kernel.org>
2017-02-19Merge remote-tracking branches 'asoc/topic/rt5665', 'asoc/topic/rt5677', ↵Mark Brown1-109/+94
'asoc/topic/samsung', 'asoc/topic/simple' and 'asoc/topic/sunxi' into asoc-next
2017-01-23ASoC: Revert "samsung: Remove unneeded initialization of chan_name"Sylwester Nawrocki1-0/+3
This reverts commit cdaf9af1eaeb539e32bfd6da6310b41ad6c3ba23 which breaks I2S support on the non-DT Samsung SoC platforms, since the default "tx", "rx" DMA channel names for playback and capture streams or custom channel names in struct snd_dmaengine_pcm_config are supported in the ASoC dmaengine module only for devicetree booting case. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-12-31ASoC: samsung: i2s: Provide I2S device for registered clocksMarek Szyprowski1-3/+3
This patch adds pointer to I2S device to clk_register_* functions. This in the future allow clock framework to ensure proper runtime state of the I2S device during all operations on the clocks provided by I2S module. Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-12-31ASoC: samsung: i2s: Let runtime PM operations to control op_clk tooMarek Szyprowski1-0/+5
This patch adds handling of parent operational clock to runtime PM callbacks. This way it is ensured that when I2S module is in runtime suspended state, all its parent clocks are disabled and unprepared. Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-12-31ASoC: samsung: i2s: Move saving and restoring regs to runtime pm operationsMarek Szyprowski1-14/+12
This patch moves saving and restoring I2S registers to runtime PM operations, what prepares the driver to operate with audio power domain. When support for audio power domain is enabled and the domain is being turned off, the I2S module will loose its context (registers), so runtime callbacks have to handle it. System sleep suspend/resume operation are implemented on top of runtime PM operations with generic pm_runtime_force_suspend/resume helpers. Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-12-31ASoC: samsung: i2s: Ensure proper runtime PM state of I2S deviceMarek Szyprowski1-8/+46
This patch adds calls to pm_runtime_get/put to ensure that any access to I2S registers is done with proper (active) runtime PM state of I2S device. Till now the driver enabled runtime PM, but didn't manage the state during driver operation. The driver worked fine only because the runtime PM callbacks managed device clock, which was enabled all the time because of the additional enable call in the driver's probe function. Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-12-31ASoC: samsung: i2s: Remove virtual device for secondary DAIMarek Szyprowski1-76/+26
For some unknown (maybe historical?) reasons support for secondary I2S DAI was implemented by adding additional virtual platform device, which was then probed again with the main I2S driver. This pattern is really hard to follow and provides no benefits, so lets remove this hack and register both DAIs during linear probe of Exynos I2S controller driver. Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-12-15ASoC: samsung: Remove tests of member addressKrzysztof Kozlowski1-8/+2
The driver was checking for non-NULL address of struct's members: - s3c_audio_pdata->type (union), - s3c_audio_pdata->type.i2s (embedded struct). This is pointless as these will be always non-NULL. The 's3c_audio_pdata' is always initialized in static memory so it will be zeroed. Additionally the 'type' member was an union with only one member. It is safe to reorganize the structures to get rid of useless union and checks for addresses to fix the coccinelle warning: >> sound/soc/samsung/i2s.c:1270:2-4: ERROR: test of a variable/field address Reported-by: kbuild test robot <fengguang.wu@intel.com> Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Reviewed-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-10-28Merge branch 'fix/samsung' of ↵Mark Brown1-9/+10
git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into asoc-samsung
2016-10-28ASoC: samsung: get access to DMA engine early to defer probe properlyMarek Szyprowski1-9/+10
ASoC Samsung sub-drivers tried to get access to their DMA engine controllers as a last step in driver probe. If a DMA engine was not available yet, samsung_asoc_dma_platform_register() function ended in -EPROBE_DEFER, but the driver already registered its component to ASoC core. This patch moves samsung_asoc_dma_platform_register() call before registering any components, to the common place, where driver was gathering all needed resources. In case of Samsung Exynos i2s driver the issue was even worse. The driver managed already to register its secondary DAI platform device before even getting the DMA engine access. That together with -EPROBE_DEFER error code from samsung_i2s_probe() immediately triggered another round of deferred probe retry and in turn endless loop of driver probing. This patch fixes broken boot on Odroid XU3 and other Exynos5422-based boards. Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Reviewed-by: Javier Martinez Canillas <javier@osg.samsung.com> Reviewed-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-10-25Merge branch 'topic/dmaengine' of ↵Mark Brown1-3/+0
git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into asoc-samsung
2016-10-25ASoC: samsung: Remove unneeded initialization of chan_nameSylwester Nawrocki1-3/+0
This patch updates the I2S drivers to always use chan_names[] field of struct snd_dmaengine_pcm_config for specifying DMA channel names, rather than using struct snd_dmaengine_dai_dma_data. This allows us to subsequently drop the SND_DMAENGINE_PCM_FLAG_CUSTOM_CHANNEL_NAME flag, now when the last use of that flag is removed. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-10-24ASoC: samsung: i2s: Fixup last IRQ unsafe spin lock callCharles Keepax1-2/+3
Unfortunately, I seem to have missed a case where an IRQ safe spinlock was required, in samsung_i2s_dai_remove, when I fixed up the other calls in this patch: 316fa9e09ad7 ("ASoC: samsung: Use IRQ safe spin lock calls") This causes a lockdep warning when unbinding and rebinding the audio card: [ 104.357664] CPU0 CPU1 [ 104.362174] ---- ---- [ 104.366692] lock(&(&pri_dai->spinlock)->rlock); [ 104.371372] local_irq_disable(); [ 104.377283] lock(&(&substream->self_group.lock)->rlock); [ 104.385259] lock(&(&pri_dai->spinlock)->rlock); [ 104.392469] <Interrupt> [ 104.395072] lock(&(&substream->self_group.lock)->rlock); [ 104.400710] [ 104.400710] *** DEADLOCK *** Fixes: ce8bcdbb61d9 ("ASoC: samsung: i2s: Protect more registers with a spinlock") Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com> Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org> Reviewed-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-08-24ASoC: samsung: i2s: Add missing clk_disable_unprepare() on error in ↵Wei Yongjun1-2/+6
samsung_i2s_probe() Add the missing clk_disable_unprepare() before return from samsung_i2s_probe() in the error handling case. Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com> Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-08-08ASoC: samsung: Drop usage of struct s3c_dma_params from i2s.cSylwester Nawrocki1-24/+24
struct s3c_dma_params already includes struct snd_dmaengine_dai_dma_data, there is no need for such an indirection so switch to using struct snd_dmaengine_dai_dma_data instead of struct s3c_dma_params. This also allows us to use snd_soc_dai_init_dma_data() function instead of the platform specific samsung_asoc_init_dma_data helper. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-07-21ASoC: samsung: Specify DMA channels through struct snd_dmaengine_pcm_configSylwester Nawrocki1-2/+3
The DMA channel names are specified through struct snd_dmaengine_pcm_config rather than using SND_DMAENGINE_PCM_FLAG_CUSTOM_CHANNEL_NAME flag when booting with devicetree in order to properly support deferred probing. Without this change the sound machine driver initialization can complete successfully with unavailable DMA resources. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-07-21ASoC: samsung: Fix error paths in the I2S driver's probe()Sylwester Nawrocki1-5/+20
Ensure they secondary DAI device is freed properly when asoc_dma_platform registration fails. This change is needed for proper deferred probe support and will help preventing situations when the CPU DAI's initialization completes without required DMA resources. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-05-30ASoC: samsung: Use of_device_get_match_data() helperSylwester Nawrocki1-17/+7
Simplify the code a little by using a standard function for getting the match data. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-05-30ASoC: samsung: Remove definition of an unused data structureSylwester Nawrocki1-4/+0
samsung_dai_type_pri is not referenced anywhere so remove it. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-05-30ASoC: samsung: Remove unused "samsung-i2sv4" platform_device_id entrySylwester Nawrocki1-3/+0
"samsung-i2sv4" identifier was previously used for the I2S device of the S5PV210 SoCs, it can be removed now when s5pv210 is a dt-only platform. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2016-02-19ASoC: samsung: Use IRQ safe spin lock callsCharles Keepax1-9/+12
Lockdep warns of a potential lock inversion, i2s->lock is held numerous times whilst we are under the substream lock (snd_pcm_stream_lock). If we use the IRQ unsafe spin lock calls, you can also end up locking snd_pcm_stream_lock whilst under i2s->lock (if an IRQ happens whilst we are holding i2s->lock). This could result in deadlock. [ 18.147001] CPU0 CPU1 [ 18.151509] ---- ---- [ 18.156022] lock(&(&pri_dai->spinlock)->rlock); [ 18.160701] local_irq_disable(); [ 18.166622] lock(&(&substream->self_group.lock)->rlock); [ 18.174595] lock(&(&pri_dai->spinlock)->rlock); [ 18.181806] <Interrupt> [ 18.184408] lock(&(&substream->self_group.lock)->rlock); [ 18.190045] [ 18.190045] *** DEADLOCK *** This patch changes to using the irq safe spinlock calls, to avoid this issue. Fixes: ce8bcdbb61d9 ("ASoC: samsung: i2s: Protect more registers with a spinlock") Signed-off-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com> Tested-by: Anand Moon <linux.amoon@gmail.com> Signed-off-by: Mark Brown <broonie@kernel.org> Cc: stable@vger.kernel.org
2015-11-21ASoC: samsung: pass filter function as pointerArnd Bergmann1-3/+8
As we are now passing the filter data as pointers to the drivers, we can take the final step and also pass the filter function the same way. I'm keeping this change separate, as there it's less obvious that this is a net win. Upsides of this are: - The ASoC drivers are completely independent from the DMA engine implementation, which simplifies the Kconfig logic and in theory allows the same sound drivers to be built in a kernel that supports different kinds of dmaengine drivers. - Consistency with other subsystems and drivers On the other hand, we have a few downsides: - The s3c24xx-dma driver now needs to be built-in for the ac97 platform device to be instantiated on s3c2440. - samsung_dmaengine_pcm_config cannot be marked 'const' any more because the filter function pointer needs to be set at runtime. This is safe as long we don't have multiple different DMA engines in thet same system at runtime, but is nonetheless ugly. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-11-18ASoC: samsung: pass DMA channels as pointersArnd Bergmann1-21/+5
ARM64 allmodconfig produces a bunch of warnings when building the samsung ASoC code: sound/soc/samsung/dmaengine.c: In function 'samsung_asoc_init_dma_data': sound/soc/samsung/dmaengine.c:53:32: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] playback_data->filter_data = (void *)playback->channel; sound/soc/samsung/dmaengine.c:60:31: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] capture_data->filter_data = (void *)capture->channel; We could easily shut up the warning by adding an intermediate cast, but there is a bigger underlying problem: The use of IORESOURCE_DMA to pass data from platform code to device drivers is dubious to start with, as what we really want is a pointer that can be passed into a filter function. Note that on s3c64xx, the pl08x DMA data is already a pointer, but gets cast to resource_size_t so we can pass it as a resource, and it then gets converted back to a pointer. In contrast, the data we pass for s3c24xx is an index into a device specific table, and we artificially convert that into a pointer for the filter function. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-05-01ASoC: samsung: Constify platform_device_idKrzysztof Kozlowski1-1/+1
The platform_device_id is not modified by the driver and core uses it as const. Signed-off-by: Krzysztof Kozlowski <k.kozlowski.k@gmail.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Add clock provider for the I2S internal clocksSylwester Nawrocki1-20/+93
This patch adds clock provider (currently only for DT platforms) for the CODECLKO (CDCLK) gate, RCLKSRC mux and RCLK pre-scaler divider divider clock. Those all tree clock are only available in the IIS Multi Audio Interface (I2S0), the regular IIS Bus Interface has only CDCLK gate clock. The motivation behind this patch is to expose the I2S internal clocks which are currently controlled through set_sysclk() through the clk API, so dedicated sound machine driver per each board can be avoided. The intention is also to fix the CDCLK gating issue reported by Daniel Drake: http://mailman.alsa-project.org/pipermail/alsa-devel/2014-September/081753.html This patch also reverts commit b97c60abf9a561f86ae71bd741add02673cc1 ("ASoC: samsung-i2s: Maintain CDCLK settings across i2s_{shutdown/ startup}") The problem that commit attempted to solve only affects the Odroid X2/U3, which doesn't configure the CDCLK clock in struct snd_soc_dai_ops hw_params callback and the issue should be now resolved by using clk API, i.e. having the codec enabling/ disabling the CDCLK clock as required. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Protect more registers with a spinlockSylwester Nawrocki1-30/+51
Ensure the I2SMOD, I2SPSR registers, which are also exposed through clk API are only accessed with the i2s->spinlock spinlock held. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Add spinlock in place of local_irq_* callsSylwester Nawrocki1-5/+13
It seems this driver hasn't been updated for SMP, as local_irq_save/ local_irq_restore don't provide proper protection of read/modify/write of the device's registers on such systems. Introduce a spinlock serializing access to the register region, it will be helpful later when I2SMOD, I2SPSR registers are made also accessible through the clk API. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Remove an unneeded goto usageSylwester Nawrocki1-9/+8
The usage of this goto seems unjustified, use if/else statement instead. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Add get_other_dai helper functionSylwester Nawrocki1-9/+15
The code to get pointer to the other DAI is repeated multiple times. Add a helper function and use it instead. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Move clk enable to the platform driver probe()Sylwester Nawrocki1-14/+11
The clk_prepare_enable() call on the "iis" clock is moved to happen earlier in the DAI platform device driver's probe() callback, so the I2S registers can be safely accessed through the clk API, after the clk supplier is registered in the platform device probe(). After this patch the "iis" clock is kept enabled since the (primary) I2S platform device probe() and until the platform device driver remove() call. This is similar to gating the clock in the snd_soc_dai probe() and remove() callbacks. Normally, in addition to that we should mark the device as PM runtime active, so if runtime PM is enabled it can idle the device by turning off the clock. Correcting this issue is left for a separate patch series, as we need to ensure the BUSCLK clock is always enabled when required. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Move clk_get() to platform driver probe()Sylwester Nawrocki1-12/+7
Acquire the I2S interface clock in driver probe() callback as it's a per-device not a per-DAI clock. While at it switch to the resource managed clk_get(). Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Request memory region in driver probe()Sylwester Nawrocki1-38/+7
The memory mapped registers region is common for both DAIs so request it in the I2S platform device driver's probe for the platform device corresponding to the primary DAI, rather than in the ASoC DAI's probe callback. While at it switch to devm_ioremap_resource(). This also drops the hard coded (0x100) register region size in the driver. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: Add return value checks in probe()Sylwester Nawrocki1-4/+8
These functions may fail so let's properly report any errors. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
2015-01-14ASoC: samsung: i2s: samsung_i2s_get_driver_data() cleanupSylwester Nawrocki1-5/+4
Tidy up the samsung_i2s_get_driver_data() function by using IS_ENABLE() instead of #ifdef and add missing braces for the 'else' part. Also ensure we are not dereferencing NULL 'match' pointer. Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com> Signed-off-by: Mark Brown <broonie@kernel.org>
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-27 21:55:16 +0300