January 2023 - Alsa-devel - mailman.alsa-project.org

[PATCH v4 0/4] Minor SoundWire clean ups
by Charles Keepax 09 Jan '23

09 Jan '23

Just some minor tidy ups and preparation for starting to upstream some Cirrus SoundWire devices. The first three patches are pretty trivial, the last patch which moves the remaining core over to using the no_pm functions could probably use some careful review. Changes since v3: - Return an error and add a WARN_ONCE if the build stubs are used. Changes since v2: - Fix up a memory leak of a buffer on the error path - Added some reviewed by's Thanks, Charles *** BLURB HERE *** Charles Keepax (3): soundwire: Provide build stubs for common functions soundwire: debugfs: Switch to sdw_read_no_pm soundwire: stream: Move remaining register accesses over to no_pm Simon Trimmer (1): soundwire: bus: export sdw_nwrite_no_pm and sdw_nread_no_pm functions drivers/soundwire/bus.c | 10 ++-- drivers/soundwire/debugfs.c | 13 ++++- drivers/soundwire/stream.c | 30 +++++----- include/linux/soundwire/sdw.h | 107 ++++++++++++++++++++++++++++++---- 4 files changed, 129 insertions(+), 31 deletions(-) -- 2.30.2

2 5

[PATCH] soundwire: cadence: remove unused sdw_cdns_master_ops declaration
by Gaosheng Cui 09 Jan '23

09 Jan '23

sdw_cdns_master_ops has been removed since commit c91605f48938 ("soundwire: Remove cdns_master_ops"), so remove it. Signed-off-by: Gaosheng Cui <cuigaosheng1(a)huawei.com> --- drivers/soundwire/cadence_master.h | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/soundwire/cadence_master.h b/drivers/soundwire/cadence_master.h index ca9e805bab88..273eca3cb0b7 100644 --- a/drivers/soundwire/cadence_master.h +++ b/drivers/soundwire/cadence_master.h @@ -142,7 +142,6 @@ struct sdw_cdns { /* Exported symbols */ int sdw_cdns_probe(struct sdw_cdns *cdns); -extern struct sdw_master_ops sdw_cdns_master_ops; irqreturn_t sdw_cdns_irq(int irq, void *dev_id); irqreturn_t sdw_cdns_thread(int irq, void *dev_id); -- 2.25.1

2 1

[PATCH v2 0/2] ASoC/soundwire: revisit support for clock registers
by Bard Liao 09 Jan '23

09 Jan '23

SoundWire clock base and scale registers are only supported for SDCA devices. That's fine, but that leaves SoundWire 1.2 with optional registers not supported. This is a corner case that needs to be supported. The change is mainly on soundwire. But rt1318-sdw.c was applied to ASoC tree recently and not in SoundWire tree yet. @Vinod, Are you good if we go throutgh ASoC tree? Or @Mark can provide a tag and we can go through SoundWire tree? changes: v2: - remove is_sdca flag from rt1318-sdw codec driver, too. Pierre-Louis Bossart (2): ASoC/soundwire: remove is_sdca boolean property soundwire: enable optional clock registers for SoundWire 1.2 devices drivers/soundwire/bus.c | 11 ++++++----- include/linux/soundwire/sdw.h | 6 ++++-- sound/soc/codecs/rt1316-sdw.c | 1 - sound/soc/codecs/rt1318-sdw.c | 1 - sound/soc/codecs/rt711-sdca-sdw.c | 1 - 5 files changed, 10 insertions(+), 10 deletions(-) -- 2.25.1

4 9

[PATCH v2 0/3] soundwire: cadence: Fix oversized FIFO size define
by Richard Fitzgerald 09 Jan '23

09 Jan '23

As determined by experimentation and asking a hardware person, the FIFO in the Cadence IP is actually only 8 entries long, not 32. This is fixed in patch #1. As a bonus, patches #2 and #3 fix two other things I noticed while debugging this. Changes since v1: - Rewrite commit message of patch #1 - Only reduce response_buf to 34 (32 + 2) - Trim RX_FIFO_AVAIL to length of response_buf instead of expected FIFO size Richard Fitzgerald (3): soundwire: cadence: Don't overflow the command FIFOs soundwire: cadence: Remove wasted space in response_buf soundwire: cadence: Drain the RX FIFO after an IO timeout drivers/soundwire/cadence_master.c | 46 +++++++++++++++++++----------- drivers/soundwire/cadence_master.h | 13 ++++++++- 2 files changed, 41 insertions(+), 18 deletions(-) -- 2.30.2

3 5

[PATCH 0/3] ALSA: firewire-lib: keep history to process isochronous packet
by Takashi Sakamoto 09 Jan '23

09 Jan '23

Hi, This patchset is preparation for computation of extra delay in runtime of PCM substream. Current implementation uses list of packet descriptor to process isochronous packets. The packet descriptors are overwritten every time to process packets, while the history of packet descriptors is useful when computing gap between current isochronous cycle and the latest isochronous cycle in which isochronous packet is processed or scheduled. Conveniently, circular linked list was added to Linux kernel v5.19 at a commit 2fbdf45d7d26 ("list: Add list_next_entry_circular() and list_prev_entry_circular()"). This patchset changes data structure from list to the circular linked list for the packet descriptors. Takashi Sakamoto (3): ALSA: firewire-lib: use circular linked list to enumerate packet descriptors ALSA: firewire-lib: use circular linked list for context payload processing layer ALSA: firewire-lib: store history to process isochronous packet sound/firewire/amdtp-am824.c | 18 +++--- sound/firewire/amdtp-stream.c | 61 ++++++++++++------- sound/firewire/amdtp-stream.h | 17 +++++- sound/firewire/digi00x/amdtp-dot.c | 18 +++--- sound/firewire/fireface/amdtp-ff.c | 18 +++--- sound/firewire/motu/amdtp-motu.c | 48 ++++++++------- .../motu/motu-command-dsp-message-parser.c | 11 ++-- .../motu/motu-register-dsp-message-parser.c | 11 ++-- sound/firewire/motu/motu.h | 8 +-- sound/firewire/tascam/amdtp-tascam.c | 18 +++--- 10 files changed, 139 insertions(+), 89 deletions(-) -- 2.37.2

2 4

[PATCH v3] ALSA: control-led: use strscpy in set_led_id()
by Jaroslav Kysela 09 Jan '23

09 Jan '23

The use of strncpy() in the set_led_id() was incorrect. The len variable should use 'min(sizeof(buf2) - 1, count)' expression. Use strscpy() function to simplify things and handle the error gracefully. Fixes: a135dfb5de15 ("ALSA: led control - add sysfs kcontrol LED marking layer") Reported-by: yang.yang29(a)zte.com.cn Link: https://lore.kernel.org/alsa-devel/202301091945513559977@zte.com.cn/ Cc: <stable(a)vger.kernel.org> Signed-off-by: Jaroslav Kysela <perex(a)perex.cz> v1..v2: - remove min() expression which strips the last char v2..v3: - BugLink -> Link - add Fixes tag --- sound/core/control_led.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/sound/core/control_led.c b/sound/core/control_led.c index f975cc85772b..3cadd40100f3 100644 --- a/sound/core/control_led.c +++ b/sound/core/control_led.c @@ -530,12 +530,11 @@ static ssize_t set_led_id(struct snd_ctl_led_card *led_card, const char *buf, si bool attach) { char buf2[256], *s, *os; - size_t len = max(sizeof(s) - 1, count); struct snd_ctl_elem_id id; int err; - strncpy(buf2, buf, len); - buf2[len] = '\0'; + if (strscpy(buf2, buf, sizeof(buf2)) < 0) + return -E2BIG; memset(&id, 0, sizeof(id)); id.iface = SNDRV_CTL_ELEM_IFACE_MIXER; s = buf2; -- 2.39.0

2 1

[PATCH] soundwire: intel: remove DAI startup/shutdown
by Bard Liao 09 Jan '23

09 Jan '23

From: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> The only thing these DAI startup/shutdown callbacks do is play with pm_runtime reference counts. This is not wrong, but it's not necessary at all. At the ASoC core level, only the component matters for pm_runtime. The ASoC core already calls pm_runtime_get_sync() in snd_soc_pcm_component_pm_runtime_get(), before the DAI startup callback is invoked. None of the SoundWire codec drivers rely on pm_runtime helpers in their DAI startup/shutdown either. This adds to the evidence that only the component, or more precisely the device specified when registering a component, should deal with pm_runtime transitions. Beyond the code cleanup, this move prepares for the addition of link power management in the auxiliary device startup/resume/suspend callbacks. The DAI callbacks can by-design assume that the device is already pm_runtime active. Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Reviewed-by: Ranjani Sridharan <ranjani.sridharan(a)linux.intel.com> Reviewed-by: Rander Wang <rander.wang(a)intel.com> Reviewed-by: Péter Ujfalusi <peter.ujfalusi(a)linux.intel.com> Signed-off-by: Bard Liao <yung-chuan.liao(a)linux.intel.com> --- drivers/soundwire/intel.c | 27 --------------------------- 1 file changed, 27 deletions(-) diff --git a/drivers/soundwire/intel.c b/drivers/soundwire/intel.c index bc9c50bacc49..2651767272c7 100644 --- a/drivers/soundwire/intel.c +++ b/drivers/soundwire/intel.c @@ -778,22 +778,6 @@ static int intel_post_bank_switch(struct sdw_intel *sdw) * DAI routines */ -static int intel_startup(struct snd_pcm_substream *substream, - struct snd_soc_dai *dai) -{ - struct sdw_cdns *cdns = snd_soc_dai_get_drvdata(dai); - int ret; - - ret = pm_runtime_resume_and_get(cdns->dev); - if (ret < 0 && ret != -EACCES) { - dev_err_ratelimited(cdns->dev, - "pm_runtime_resume_and_get failed in %s, ret %d\n", - __func__, ret); - return ret; - } - return 0; -} - static int intel_hw_params(struct snd_pcm_substream *substream, struct snd_pcm_hw_params *params, struct snd_soc_dai *dai) @@ -954,15 +938,6 @@ intel_hw_free(struct snd_pcm_substream *substream, struct snd_soc_dai *dai) return 0; } -static void intel_shutdown(struct snd_pcm_substream *substream, - struct snd_soc_dai *dai) -{ - struct sdw_cdns *cdns = snd_soc_dai_get_drvdata(dai); - - pm_runtime_mark_last_busy(cdns->dev); - pm_runtime_put_autosuspend(cdns->dev); -} - static int intel_pcm_set_sdw_stream(struct snd_soc_dai *dai, void *stream, int direction) { @@ -1088,12 +1063,10 @@ static int intel_component_dais_suspend(struct snd_soc_component *component) } static const struct snd_soc_dai_ops intel_pcm_dai_ops = { - .startup = intel_startup, .hw_params = intel_hw_params, .prepare = intel_prepare, .hw_free = intel_hw_free, .trigger = intel_trigger, - .shutdown = intel_shutdown, .set_stream = intel_pcm_set_sdw_stream, .get_stream = intel_get_sdw_stream, }; -- 2.25.1

2 1

[PATCH v2] ALSA: control-led: use strscpy in set_led_id()
by Jaroslav Kysela 09 Jan '23

09 Jan '23

The use of strncpy() in the set_led_id() was incorrect. The len variable should use 'min(sizeof(buf2) - 1, count)' expression. Use strscpy() function to simplify things and handle the error gracefully. Reported-by: yang.yang29(a)zte.com.cn BugLink: https://lore.kernel.org/alsa-devel/202301091945513559977@zte.com.cn/ Cc: <stable(a)vger.kernel.org> Signed-off-by: Jaroslav Kysela <perex(a)perex.cz> v1..v2: - remove min() expression which strips the last char --- sound/core/control_led.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/sound/core/control_led.c b/sound/core/control_led.c index f975cc85772b..3cadd40100f3 100644 --- a/sound/core/control_led.c +++ b/sound/core/control_led.c @@ -530,12 +530,11 @@ static ssize_t set_led_id(struct snd_ctl_led_card *led_card, const char *buf, si bool attach) { char buf2[256], *s, *os; - size_t len = max(sizeof(s) - 1, count); struct snd_ctl_elem_id id; int err; - strncpy(buf2, buf, len); - buf2[len] = '\0'; + if (strscpy(buf2, buf, sizeof(buf2)) < 0) + return -E2BIG; memset(&id, 0, sizeof(id)); id.iface = SNDRV_CTL_ELEM_IFACE_MIXER; s = buf2; -- 2.39.0

2 2

[PATCH] ALSA: usb-audio: Always initialize fixed_rate in snd_usb_find_implicit_fb_sync_format()
by Jaroslav Kysela 09 Jan '23

09 Jan '23

Handle the fallback code path, too. Fixes: fd28941cff1c ("ALSA: usb-audio: Add new quirk FIXED_RATE for JBL Quantum810 Wireless") BugLink: https://lore.kernel.org/alsa-devel/Y7frf3N%2FxzvESEsN@kili/ Reported-by: Dan Carpenter <error27(a)gmail.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Jaroslav Kysela <perex(a)perex.cz> --- sound/usb/implicit.c | 3 ++- sound/usb/pcm.c | 2 ++ 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/sound/usb/implicit.c b/sound/usb/implicit.c index 41ac7185b42b..4727043fd745 100644 --- a/sound/usb/implicit.c +++ b/sound/usb/implicit.c @@ -471,7 +471,7 @@ snd_usb_find_implicit_fb_sync_format(struct snd_usb_audio *chip, subs = find_matching_substream(chip, stream, target->sync_ep, target->fmt_type); if (!subs) - return sync_fmt; + goto end; high_score = 0; list_for_each_entry(fp, &subs->fmt_list, list) { @@ -485,6 +485,7 @@ snd_usb_find_implicit_fb_sync_format(struct snd_usb_audio *chip, } } + end: if (fixed_rate) *fixed_rate = snd_usb_pcm_has_fixed_rate(subs); return sync_fmt; diff --git a/sound/usb/pcm.c b/sound/usb/pcm.c index 99a66d0ef5b2..d2c652aa1385 100644 --- a/sound/usb/pcm.c +++ b/sound/usb/pcm.c @@ -163,6 +163,8 @@ bool snd_usb_pcm_has_fixed_rate(struct snd_usb_substream *subs) struct snd_usb_audio *chip = subs->stream->chip; int rate = -1; + if (!subs) + return false; if (!(chip->quirk_flags & QUIRK_FLAG_FIXED_RATE)) return false; list_for_each_entry(fp, &subs->fmt_list, list) { -- 2.39.0

2 1

[PATCH] ALSA: control-led: use strscpy in set_led_id()
by Jaroslav Kysela 09 Jan '23

09 Jan '23

The use of strncpy() in the set_led_id() was incorrect. The len variable should use 'min(sizeof(buf2) - 1, count)' expression. Use strscpy() function to simplify things and handle the error gracefully. Reported-by: yang.yang29(a)zte.com.cn BugLink: https://lore.kernel.org/alsa-devel/202301091945513559977@zte.com.cn/ Cc: <stable(a)vger.kernel.org> Signed-off-by: Jaroslav Kysela <perex(a)perex.cz> --- sound/core/control_led.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/sound/core/control_led.c b/sound/core/control_led.c index f975cc85772b..45c8eb5700c1 100644 --- a/sound/core/control_led.c +++ b/sound/core/control_led.c @@ -530,12 +530,11 @@ static ssize_t set_led_id(struct snd_ctl_led_card *led_card, const char *buf, si bool attach) { char buf2[256], *s, *os; - size_t len = max(sizeof(s) - 1, count); struct snd_ctl_elem_id id; int err; - strncpy(buf2, buf, len); - buf2[len] = '\0'; + if (strscpy(buf2, buf, min(sizeof(buf2), count)) < 0) + return -E2BIG; memset(&id, 0, sizeof(id)); id.iface = SNDRV_CTL_ELEM_IFACE_MIXER; s = buf2; -- 2.39.0

1 1