- Alsa-devel - mailman.alsa-project.org

[PATCH] ALSA: control - add the missing prev_lops2 initialization
by Jaroslav Kysela 01 Apr '21

01 Apr '21

As static analysis reported, the prev_lops2 should contain the previous lops2 pointer in snd_ctl_disconnect_layer(). Link: https://lore.kernel.org/alsa-devel/96e9bd5c-c8db-0db8-b393-fbf4a047dc80@can… Fixes: 3f0638a0333b ("ALSA: control - add layer registration routines") Signed-off-by: Jaroslav Kysela <perex(a)perex.cz> --- sound/core/control.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/sound/core/control.c b/sound/core/control.c index 6825ca75daf5..20d707d4ef40 100644 --- a/sound/core/control.c +++ b/sound/core/control.c @@ -2074,7 +2074,7 @@ void snd_ctl_disconnect_layer(struct snd_ctl_layer_ops *lops) struct snd_ctl_layer_ops *lops2, *prev_lops2; down_write(&snd_ctl_layer_rwsem); - for (lops2 = snd_ctl_layer, prev_lops2 = NULL; lops2; lops2 = lops2->next) + for (lops2 = snd_ctl_layer, prev_lops2 = NULL; lops2; lops2 = lops2->next) { if (lops2 == lops) { if (!prev_lops2) snd_ctl_layer = lops->next; @@ -2082,6 +2082,8 @@ void snd_ctl_disconnect_layer(struct snd_ctl_layer_ops *lops) prev_lops2->next = lops->next; break; } + prev_lops2 = lops2; + } up_write(&snd_ctl_layer_rwsem); } EXPORT_SYMBOL_GPL(snd_ctl_disconnect_layer); -- 2.30.2

2 1

[PATCH v4 0/5] soundwire: add static port map support
by Srinivas Kandagatla 01 Apr '21

01 Apr '21

In some cases, SoundWire device ports are statically mapped to Controller ports during design, however there is no way to expose this information to the controller. Controllers like Qualcomm ones use this info to setup static bandwidth parameters for those ports. A generic port allocation is not possible in this cases! This patch adds a new member m_port_map to SoundWire device so that it can populate the static master port map and share it with controller to be able to setup correct bandwidth parameters. As a user of this feature this patchset also adds new bindings for wsa881x smart speaker which has 4 ports which are statically mapped to the 3 output and 1 input port of the controller. Tested it on DB845c and SM8250 MTP. thanks, srini Changes since v3: - updated kernel doc for more clarity on m_port_map Srinivas Kandagatla (5): soundwire: add static port mapping support soundwire: qcom: update port map allocation bit mask soundwire: qcom: add static port map support ASoC: dt-bindings: wsa881x: add bindings for port mapping ASoC: codecs: wsa881x: add static port map support .../bindings/sound/qcom,wsa881x.yaml | 9 ++++++ drivers/soundwire/qcom.c | 31 +++++++++++++++---- include/linux/soundwire/sdw.h | 2 ++ sound/soc/codecs/wsa881x.c | 7 +++++ 4 files changed, 43 insertions(+), 6 deletions(-) -- 2.21.0

4 9

[PATCH] soundwire: stream: fix memory leak in stream config error path
by Bard Liao 01 Apr '21

01 Apr '21

From: Rander Wang <rander.wang(a)intel.com> When stream config is failed, master runtime will release all slave runtime in the slave_rt_list, but slave runtime is not added to the list at this time. This patch frees slave runtime in the config error path to fix the memory leak. Fixes: bbe7379d8040a ("soundwire: Add support for SoundWire stream management") Signed-off-by: Rander Wang <rander.wang(a)intel.com> Reviewed-by: Keyon Jie <yang.jie(a)intel.com> Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski(a)linux.intel.com> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Signed-off-by: Bard Liao <yung-chuan.liao(a)linux.intel.com> --- drivers/soundwire/stream.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/soundwire/stream.c b/drivers/soundwire/stream.c index 9c064b672745..1eaedaaba094 100644 --- a/drivers/soundwire/stream.c +++ b/drivers/soundwire/stream.c @@ -1375,8 +1375,16 @@ int sdw_stream_add_slave(struct sdw_slave *slave, } ret = sdw_config_stream(&slave->dev, stream, stream_config, true); - if (ret) + if (ret) { + /* + * sdw_release_master_stream will release s_rt in slave_rt_list in + * stream_error case, but s_rt is only added to slave_rt_list + * when sdw_config_stream is successful, so free s_rt explicitly + * when sdw_config_stream is failed. + */ + kfree(s_rt); goto stream_error; + } list_add_tail(&s_rt->m_rt_node, &m_rt->slave_rt_list); -- 2.17.1

2 1

[PATCH] soundwire: qcom: wait for fifo space to be available before read/write
by Srinivas Kandagatla 01 Apr '21

01 Apr '21

If we write registers very fast we can endup in a situation where some of the writes will be dropped without any notice. So wait for the fifo space to be available before reading/writing the soundwire registers. Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla(a)linaro.org> --- drivers/soundwire/qcom.c | 74 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 74 insertions(+) diff --git a/drivers/soundwire/qcom.c b/drivers/soundwire/qcom.c index 6affa3cd4039..9b45717577f2 100644 --- a/drivers/soundwire/qcom.c +++ b/drivers/soundwire/qcom.c @@ -24,6 +24,8 @@ #define SWRM_COMP_CFG_IRQ_LEVEL_OR_PULSE_MSK BIT(1) #define SWRM_COMP_CFG_ENABLE_MSK BIT(0) #define SWRM_COMP_PARAMS 0x100 +#define SWRM_COMP_PARAMS_WR_FIFO_DEPTH GENMASK(14, 10) +#define SWRM_COMP_PARAMS_RD_FIFO_DEPTH GENMASK(19, 15) #define SWRM_COMP_PARAMS_DOUT_PORTS_MASK GENMASK(4, 0) #define SWRM_COMP_PARAMS_DIN_PORTS_MASK GENMASK(9, 5) #define SWRM_INTERRUPT_STATUS 0x200 @@ -51,6 +53,8 @@ #define SWRM_CMD_FIFO_CMD 0x308 #define SWRM_CMD_FIFO_FLUSH 0x1 #define SWRM_CMD_FIFO_STATUS 0x30C +#define SWRM_RD_CMD_FIFO_CNT_MASK GENMASK(20, 16) +#define SWRM_WR_CMD_FIFO_CNT_MASK GENMASK(12, 8) #define SWRM_CMD_FIFO_CFG_ADDR 0x314 #define SWRM_CONTINUE_EXEC_ON_CMD_IGNORE BIT(31) #define SWRM_RD_WR_CMD_RETRIES 0x7 @@ -104,6 +108,7 @@ #define SWR_BROADCAST_CMD_ID 0x0F #define SWR_MAX_CMD_ID 14 #define MAX_FIFO_RD_RETRY 3 +#define SWR_OVERFLOW_RETRY_COUNT 30 struct qcom_swrm_port_config { u8 si; @@ -147,6 +152,8 @@ struct qcom_swrm_ctrl { int (*reg_read)(struct qcom_swrm_ctrl *ctrl, int reg, u32 *val); int (*reg_write)(struct qcom_swrm_ctrl *ctrl, int reg, int val); u32 slave_status; + u32 wr_fifo_depth; + u32 rd_fifo_depth; }; struct qcom_swrm_data { @@ -238,6 +245,63 @@ static u32 swrm_get_packed_reg_val(u8 *cmd_id, u8 cmd_data, return val; } +static int swrm_wait_for_rd_fifo_avail(struct qcom_swrm_ctrl *swrm) +{ + u32 fifo_outstanding_cmd, value; + u8 fifo_retry_count = SWR_OVERFLOW_RETRY_COUNT; + + /* Check for fifo underflow during read */ + swrm->reg_read(swrm, SWRM_CMD_FIFO_STATUS, &value); + fifo_outstanding_cmd = FIELD_GET(SWRM_RD_CMD_FIFO_CNT_MASK, value); + + /* Check number of outstanding commands in fifo before read */ + if (fifo_outstanding_cmd) + return 0; + + do { + usleep_range(500, 510); + swrm->reg_read(swrm, SWRM_CMD_FIFO_STATUS, &value); + fifo_outstanding_cmd = FIELD_GET(SWRM_RD_CMD_FIFO_CNT_MASK, value); + if (fifo_outstanding_cmd > 0) + break; + } while (fifo_retry_count--); + + if (fifo_outstanding_cmd == 0) { + dev_err_ratelimited(swrm->dev, "%s err read underflow\n", __func__); + return -ENOMEM; + } + + return 0; +} + +static int swrm_wait_for_wr_fifo_avail(struct qcom_swrm_ctrl *swrm) +{ + u32 fifo_outstanding_cmd, value; + u8 fifo_retry_count = SWR_OVERFLOW_RETRY_COUNT; + + /* Check for fifo overflow during write */ + swrm->reg_read(swrm, SWRM_CMD_FIFO_STATUS, &value); + fifo_outstanding_cmd = FIELD_GET(SWRM_WR_CMD_FIFO_CNT_MASK, value); + + /* Check number of outstanding commands in fifo before write */ + if (fifo_outstanding_cmd != swrm->wr_fifo_depth) + return 0; + + do { + usleep_range(500, 510); + swrm->reg_read(swrm, SWRM_CMD_FIFO_STATUS, &value); + fifo_outstanding_cmd = FIELD_GET(SWRM_WR_CMD_FIFO_CNT_MASK, value); + if (fifo_outstanding_cmd < swrm->wr_fifo_depth) + break; + } while (fifo_retry_count--); + + if (fifo_outstanding_cmd == swrm->wr_fifo_depth) { + dev_err_ratelimited(swrm->dev, "%s err write overflow\n", __func__); + return -ENOMEM; + } + + return 0; +} static int qcom_swrm_cmd_fifo_wr_cmd(struct qcom_swrm_ctrl *swrm, u8 cmd_data, u8 dev_addr, u16 reg_addr) @@ -256,6 +320,9 @@ static int qcom_swrm_cmd_fifo_wr_cmd(struct qcom_swrm_ctrl *swrm, u8 cmd_data, dev_addr, reg_addr); } + if (swrm_wait_for_wr_fifo_avail(swrm)) + return SDW_CMD_FAIL_OTHER; + /* Its assumed that write is okay as we do not get any status back */ swrm->reg_write(swrm, SWRM_CMD_FIFO_WR_CMD, val); @@ -295,6 +362,9 @@ static int qcom_swrm_cmd_fifo_rd_cmd(struct qcom_swrm_ctrl *swrm, /* wait for FIFO RD CMD complete to avoid overflow */ usleep_range(250, 255); + if (swrm_wait_for_rd_fifo_avail(swrm)) + return SDW_CMD_FAIL_OTHER; + do { swrm->reg_read(swrm, SWRM_CMD_FIFO_RD_FIFO_ADDR, &cmd_data); rval[0] = cmd_data & 0xFF; @@ -586,6 +656,10 @@ static int qcom_swrm_init(struct qcom_swrm_ctrl *ctrl) SWRM_INTERRUPT_STATUS_RMSK); } ctrl->slave_status = 0; + ctrl->reg_read(ctrl, SWRM_COMP_PARAMS, &val); + ctrl->rd_fifo_depth = FIELD_GET(SWRM_COMP_PARAMS_RD_FIFO_DEPTH, val); + ctrl->wr_fifo_depth = FIELD_GET(SWRM_COMP_PARAMS_WR_FIFO_DEPTH, val); + return 0; } -- 2.21.0

3 2

[PATCH 0/2] kbl_da7219_max9357a machine changes for wov and MST
by vamshi.krishna.gopal＠intel.com 31 Mar '21

31 Mar '21

From: Vamshi Krishna Gopal <vamshi.krishna.gopal(a)intel.com> Hello, This patch series about creating dailink for Wake on voice functionality and also adding MST route changes Mac Chiang (1): ASoc: Intel: board: add BE DAI link for WoV Vamshi Krishna Gopal (1): ASoC: Intel: kbl: Add MST route change to kbl machine drivers sound/soc/intel/boards/kbl_da7219_max98357a.c | 42 +++++++++++++++++-- 1 file changed, 38 insertions(+), 4 deletions(-) -- 2.17.1

4 8

[PATCH][next] soundwire: qcom: Fix a u8 comparison with less than zero
by Colin King 31 Mar '21

31 Mar '21

From: Colin Ian King <colin.king(a)canonical.com> Variable devnum is being checked for a less than zero error return however the comparison will always be false because devnum is an 8 bit unsigned integer. Fix this by making devnum an int. Also there is no need to iniitialize devnum with zero as this value is no read, so remove the redundant assignment. Addresses-Coverity: ("Unsigned compared against 0") Fixes: c7d49c76d1d5 ("soundwire: qcom: add support to new interrupts") Signed-off-by: Colin Ian King <colin.king(a)canonical.com> --- drivers/soundwire/qcom.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soundwire/qcom.c b/drivers/soundwire/qcom.c index b08ecb9b418c..ec86c4e53fdb 100644 --- a/drivers/soundwire/qcom.c +++ b/drivers/soundwire/qcom.c @@ -428,7 +428,7 @@ static irqreturn_t qcom_swrm_irq_handler(int irq, void *dev_id) struct qcom_swrm_ctrl *swrm = dev_id; u32 value, intr_sts, intr_sts_masked, slave_status; u32 i; - u8 devnum = 0; + int devnum; int ret = IRQ_HANDLED; swrm->reg_read(swrm, SWRM_INTERRUPT_STATUS, &intr_sts); -- 2.30.2

2 1

[PATCH v2 1/2] ASoC: tlv320aic32x4: Increase maximum register in regmap
by Annaliese McDermond 31 Mar '21

31 Mar '21

AIC32X4_REFPOWERUP was added as a register, but the maximum register value in the regmap and regmap range was not correspondingly increased. This caused an error when this register was attempted to be written. Fixes: ec96690de82c ("ASoC: tlv320aic32x4: Enable fast charge") Cc: stable(a)vger.kernel.org Signed-off-by: Annaliese McDermond <nh6z(a)nh6z.net> --- sound/soc/codecs/tlv320aic32x4.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/tlv320aic32x4.c b/sound/soc/codecs/tlv320aic32x4.c index f04f88c8d425..1ac3b3b12dc6 100644 --- a/sound/soc/codecs/tlv320aic32x4.c +++ b/sound/soc/codecs/tlv320aic32x4.c @@ -577,12 +577,12 @@ static const struct regmap_range_cfg aic32x4_regmap_pages[] = { .window_start = 0, .window_len = 128, .range_min = 0, - .range_max = AIC32X4_RMICPGAVOL, + .range_max = AIC32X4_REFPOWERUP, }, }; const struct regmap_config aic32x4_regmap_config = { - .max_register = AIC32X4_RMICPGAVOL, + .max_register = AIC32X4_REFPOWERUP, .ranges = aic32x4_regmap_pages, .num_ranges = ARRAY_SIZE(aic32x4_regmap_pages), }; -- 2.27.0

1 0

[PATCH v2 2/2] ASoC: tlv320aic32x4: Register clocks before registering component
by Annaliese McDermond 31 Mar '21

31 Mar '21

Clock registration must be performed before the component is registered. aic32x4_component_probe attempts to get all the clocks right off the bat. If the component is registered before the clocks there is a race condition where the clocks may not be registered by the time aic32x4_componet_probe actually runs. Fixes: d1c859d314d8 ("ASoC: codec: tlv3204: Increased maximum supported channels") Cc: stable(a)vger.kernel.org Signed-off-by: Annaliese McDermond <nh6z(a)nh6z.net> --- sound/soc/codecs/tlv320aic32x4.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sound/soc/codecs/tlv320aic32x4.c b/sound/soc/codecs/tlv320aic32x4.c index 1ac3b3b12dc6..b689f26fc4be 100644 --- a/sound/soc/codecs/tlv320aic32x4.c +++ b/sound/soc/codecs/tlv320aic32x4.c @@ -1243,6 +1243,10 @@ int aic32x4_probe(struct device *dev, struct regmap *regmap) if (ret) goto err_disable_regulators; + ret = aic32x4_register_clocks(dev, aic32x4->mclk_name); + if (ret) + goto err_disable_regulators; + ret = devm_snd_soc_register_component(dev, &soc_component_dev_aic32x4, &aic32x4_dai, 1); if (ret) { @@ -1250,10 +1254,6 @@ int aic32x4_probe(struct device *dev, struct regmap *regmap) goto err_disable_regulators; } - ret = aic32x4_register_clocks(dev, aic32x4->mclk_name); - if (ret) - goto err_disable_regulators; - return 0; err_disable_regulators: -- 2.27.0

1 0

re: ALSA: control - add layer registration routines
by Colin Ian King 31 Mar '21

31 Mar '21

Hi, Static analysis on linux-next with Coverity has detected a potential issue in the following commit: commit 3f0638a0333bfdd0549985aa620f2ab69737af47 Author: Jaroslav Kysela <perex(a)perex.cz> Date: Wed Mar 17 18:29:41 2021 +0100 ALSA: control - add layer registration routines The static analysis is as follows: 2072 void snd_ctl_disconnect_layer(struct snd_ctl_layer_ops *lops) 2073 { 2074 struct snd_ctl_layer_ops *lops2, *prev_lops2; 2075 2076 down_write(&snd_ctl_layer_rwsem); assignment: Assigning: prev_lops2 = NULL. 2077 for (lops2 = snd_ctl_layer, prev_lops2 = NULL; lops2; lops2 = lops2->next) 2078 if (lops2 == lops) { null: At condition prev_lops2, the value of prev_lops2 must be NULL. dead_error_condition: The condition !prev_lops2 must be true. 2079 if (!prev_lops2) 2080 snd_ctl_layer = lops->next; 2081 else 'Constant' variable guards dead code (DEADCODE) dead_error_line: Execution cannot reach this statement: prev_lops2->next = lops->next;. Local variable prev_lops2 is assigned only once, to a constant value, making it effectively constant throughout its scope. If this is not the intent, examine the logic to see if there is a missing assignment that would make prev_lops2 not remain constant. 2082 prev_lops2->next = lops->next; 2083 break; 2084 } 2085 up_write(&snd_ctl_layer_rwsem); 2086 } I couldn't quite figure out the original intent of the prev_lops use, so I'd thought I'd report this issue as the code does look incorrect. Colin

2 1

[PATCH V2] soundwire: qcom: use signed variable for error return
by Vinod Koul 31 Mar '21

31 Mar '21

We get warning of using a unsigned variable being compared to less than zero. The comparison is correct as it checks for errors from previous call to qcom_swrm_get_alert_slave_dev_num(), so we should use a signed variable here. While at it, drop the superfluous initialization as well drivers/soundwire/qcom.c: qcom_swrm_irq_handler() warn: impossible condition '(devnum < 0) => (0-255 < 0)' Reported-by: kernel test robot <lkp(a)intel.com> Signed-off-by: Vinod Koul <vkoul(a)kernel.org> --- drivers/soundwire/qcom.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soundwire/qcom.c b/drivers/soundwire/qcom.c index b08ecb9b418c..ec86c4e53fdb 100644 --- a/drivers/soundwire/qcom.c +++ b/drivers/soundwire/qcom.c @@ -428,7 +428,7 @@ static irqreturn_t qcom_swrm_irq_handler(int irq, void *dev_id) struct qcom_swrm_ctrl *swrm = dev_id; u32 value, intr_sts, intr_sts_masked, slave_status; u32 i; - u8 devnum = 0; + int devnum; int ret = IRQ_HANDLED; swrm->reg_read(swrm, SWRM_INTERRUPT_STATUS, &intr_sts); -- 2.26.3

3 2