August 2021 - Alsa-devel - mailman.alsa-project.org

[PATCH] ASoC: intel: skylake: Drop superfluous mmap callback
by Takashi Iwai 02 Aug '21

02 Aug '21

skl_platform_soc_mmap() just calls the standard mmap helper, hence it's superfluous. Let's drop it. Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/soc/intel/skylake/skl-pcm.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/sound/soc/intel/skylake/skl-pcm.c b/sound/soc/intel/skylake/skl-pcm.c index b1ca64d2f7ea..c604200de80e 100644 --- a/sound/soc/intel/skylake/skl-pcm.c +++ b/sound/soc/intel/skylake/skl-pcm.c @@ -1214,13 +1214,6 @@ static snd_pcm_uframes_t skl_platform_soc_pointer( return bytes_to_frames(substream->runtime, pos); } -static int skl_platform_soc_mmap(struct snd_soc_component *component, - struct snd_pcm_substream *substream, - struct vm_area_struct *area) -{ - return snd_pcm_lib_default_mmap(substream, area); -} - static u64 skl_adjust_codec_delay(struct snd_pcm_substream *substream, u64 nsec) { @@ -1460,7 +1453,6 @@ static const struct snd_soc_component_driver skl_component = { .trigger = skl_platform_soc_trigger, .pointer = skl_platform_soc_pointer, .get_time_info = skl_platform_soc_get_time_info, - .mmap = skl_platform_soc_mmap, .pcm_construct = skl_platform_soc_new, .module_get_upon_open = 1, /* increment refcount when a pcm is opened */ }; -- 2.26.2

3 5

[PATCH 0/5] ASoC: Fix reference to PCM buffer address
by Takashi Iwai 02 Aug '21

02 Aug '21

Hi, while updating the memory allocation API for WC pages, I casually stumbled on quite a few drivers dealing with the PCM buffer address wrongly via substream->dma_buffer.addr. This patch series corrects those usages. Takashi === Takashi Iwai (5): ASoC: amd: Fix reference to PCM buffer address ASoC: intel: atom: Fix reference to PCM buffer address ASoC: xilinx: Fix reference to PCM buffer address ASoC: uniphier: Fix reference to PCM buffer address ASoC: kirkwood: Fix reference to PCM buffer address sound/soc/amd/acp-pcm-dma.c | 2 +- sound/soc/amd/raven/acp3x-pcm-dma.c | 2 +- sound/soc/amd/renoir/acp3x-pdm-dma.c | 2 +- sound/soc/intel/atom/sst-mfld-platform-pcm.c | 3 +-- sound/soc/kirkwood/kirkwood-dma.c | 26 ++++++++++++++------ sound/soc/uniphier/aio-dma.c | 2 +- sound/soc/xilinx/xlnx_formatter_pcm.c | 4 +-- 7 files changed, 25 insertions(+), 16 deletions(-) -- 2.26.2

3 9

[PATCH 00/10] soundwire/ASoC: add mockup codec support
by Bard Liao 02 Aug '21

02 Aug '21

Adding mockup SoundWire codec is useful to debug driver/topology changes without having any actual device connected. Bard Liao (2): soundwire: stream: don't abort bank switch on Command_Ignored/-ENODATA soundwire: stream: don't program mockup device ports Pierre-Louis Bossart (8): ASoC: codecs: add SoundWire mockup device support ASoC: soc-acpi: cnl: add table for SoundWire mockup devices ASoC: soc-acpi: tgl: add table for SoundWire mockup devices ASoC: Intel: boards: sof_sdw: add SoundWire mockup codecs for tests soundwire: add flag to ignore all command/control for mockup devices soundwire: bus: squelch error returned by mockup devices soundwire: cadence: add debugfs interface for PDI loopbacks soundwire: cadence: override PDI configurations to create loopback drivers/soundwire/bus.c | 10 +- drivers/soundwire/cadence_master.c | 174 ++++++++-- drivers/soundwire/cadence_master.h | 3 + drivers/soundwire/stream.c | 5 +- include/linux/soundwire/sdw.h | 3 + sound/soc/codecs/Kconfig | 18 + sound/soc/codecs/Makefile | 2 + sound/soc/codecs/sdw-mockup.c | 312 ++++++++++++++++++ sound/soc/intel/boards/Kconfig | 1 + sound/soc/intel/boards/sof_sdw.c | 41 +++ sound/soc/intel/common/Makefile | 3 +- .../intel/common/soc-acpi-intel-cnl-match.c | 15 + .../common/soc-acpi-intel-sdw-mockup-match.c | 166 ++++++++++ .../common/soc-acpi-intel-sdw-mockup-match.h | 17 + .../intel/common/soc-acpi-intel-tgl-match.c | 23 ++ 15 files changed, 754 insertions(+), 39 deletions(-) create mode 100644 sound/soc/codecs/sdw-mockup.c create mode 100644 sound/soc/intel/common/soc-acpi-intel-sdw-mockup-match.c create mode 100644 sound/soc/intel/common/soc-acpi-intel-sdw-mockup-match.h -- 2.17.1

4 15

[PATCH] soundwire: cadence: add paranoid check on self-clearing bits
by Bard Liao 02 Aug '21

02 Aug '21

From: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> The Cadence IP exposes a small number of self-clearing bits in the MCP_CONTROL and MCP_CONFIG_UPDATE registers. We currently do not check that those bits are indeed cleared, e.g. during resume operations. That could lead to resuming peripheral devices too early. In addition, if we happen to read these registers, update one of the fields and write the register back, we may be writing stale data that might have been cleared in hardware. These sort of race conditions could lead to e.g. doing a hw_reset twice or stopping a clock that just restarted. There is no clear way of avoiding these potential race conditions other than making sure that these registers fields are cleared before any read-modify-write sequence. If we detect this sort of errors, we only log them since there is no clear recovery possible. The only way out is likely to restart the IP with a suspend/resume cycle. Note that the checks are performed before updating the registers, as well as after the Intel 'sync go' sequence in multi-link mode. That should cover both the start and end of suspend/resume hardware configurations. The Multi-Master mode gates the configuration updates until the 'sync go' signal is asserted, so we only check on init and after the end of the 'sync go' sequence. The duration of the usleep_range() was defined by the GSYNC frequency used in multi-master mode. With a 4kHz frequency, any configuration change might be deferred by up to 250us. Extending the range to 1000-1500us should guarantee that the configuration change is completed without any significant impact on the overall resume time. Suggested-by: Bard Liao <yung-chuan.liao(a)linux.intel.com> Signed-off-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/cadence_master.c | 47 ++++++++++++++++++++++++++++++ drivers/soundwire/cadence_master.h | 4 +++ drivers/soundwire/intel.c | 14 +++++++++ 3 files changed, 65 insertions(+) diff --git a/drivers/soundwire/cadence_master.c b/drivers/soundwire/cadence_master.c index c7372519452b..0b7f037e6cd0 100644 --- a/drivers/soundwire/cadence_master.c +++ b/drivers/soundwire/cadence_master.c @@ -979,6 +979,49 @@ static void cdns_update_slave_status_work(struct work_struct *work) } +/* paranoia check to make sure self-cleared bits are indeed cleared */ +void sdw_cdns_check_self_clearing_bits(struct sdw_cdns *cdns, const char *string, + bool initial_delay, int reset_iterations) +{ + u32 mcp_control; + u32 mcp_config_update; + int i; + + if (initial_delay) + usleep_range(1000, 1500); + + mcp_control = cdns_readl(cdns, CDNS_MCP_CONTROL); + + /* the following bits should be cleared immediately */ + if (mcp_control & CDNS_MCP_CONTROL_CMD_RST) + dev_err(cdns->dev, "%s failed: MCP_CONTROL_CMD_RST is not cleared\n", string); + if (mcp_control & CDNS_MCP_CONTROL_SOFT_RST) + dev_err(cdns->dev, "%s failed: MCP_CONTROL_SOFT_RST is not cleared\n", string); + if (mcp_control & CDNS_MCP_CONTROL_SW_RST) + dev_err(cdns->dev, "%s failed: MCP_CONTROL_SW_RST is not cleared\n", string); + if (mcp_control & CDNS_MCP_CONTROL_CLK_STOP_CLR) + dev_err(cdns->dev, "%s failed: MCP_CONTROL_CLK_STOP_CLR is not cleared\n", string); + mcp_config_update = cdns_readl(cdns, CDNS_MCP_CONFIG_UPDATE); + if (mcp_config_update & CDNS_MCP_CONFIG_UPDATE_BIT) + dev_err(cdns->dev, "%s failed: MCP_CONFIG_UPDATE_BIT is not cleared\n", string); + + i = 0; + while (mcp_control & CDNS_MCP_CONTROL_HW_RST) { + if (i == reset_iterations) { + dev_err(cdns->dev, "%s failed: MCP_CONTROL_HW_RST is not cleared\n", string); + break; + } + + dev_dbg(cdns->dev, "%s: MCP_CONTROL_HW_RST is not cleared at iteration %d\n", string, i); + i++; + + usleep_range(1000, 1500); + mcp_control = cdns_readl(cdns, CDNS_MCP_CONTROL); + } + +} +EXPORT_SYMBOL(sdw_cdns_check_self_clearing_bits); + /* * init routines */ @@ -1256,6 +1299,8 @@ int sdw_cdns_init(struct sdw_cdns *cdns) cdns_init_clock_ctrl(cdns); + sdw_cdns_check_self_clearing_bits(cdns, __func__, false, 0); + /* reset msg_count to default value of FIFOLEVEL */ cdns->msg_count = cdns_readl(cdns, CDNS_MCP_FIFOLEVEL); @@ -1500,6 +1545,8 @@ int sdw_cdns_clock_stop(struct sdw_cdns *cdns, bool block_wake) struct sdw_slave *slave; int ret; + sdw_cdns_check_self_clearing_bits(cdns, __func__, false, 0); + /* Check suspend status */ if (sdw_cdns_is_clock_stop(cdns)) { dev_dbg(cdns->dev, "Clock is already stopped\n"); diff --git a/drivers/soundwire/cadence_master.h b/drivers/soundwire/cadence_master.h index 6c039d456ba8..e587aede63bf 100644 --- a/drivers/soundwire/cadence_master.h +++ b/drivers/soundwire/cadence_master.h @@ -187,4 +187,8 @@ int cdns_bus_conf(struct sdw_bus *bus, struct sdw_bus_params *params); int cdns_set_sdw_stream(struct snd_soc_dai *dai, void *stream, bool pcm, int direction); + +void sdw_cdns_check_self_clearing_bits(struct sdw_cdns *cdns, const char *string, + bool initial_delay, int reset_iterations); + #endif /* __SDW_CADENCE_H */ diff --git a/drivers/soundwire/intel.c b/drivers/soundwire/intel.c index 15668d6fecd6..fb9c23e13206 100644 --- a/drivers/soundwire/intel.c +++ b/drivers/soundwire/intel.c @@ -23,6 +23,7 @@ #include "intel.h" #define INTEL_MASTER_SUSPEND_DELAY_MS 3000 +#define INTEL_MASTER_RESET_ITERATIONS 10 /* * debug/config flags for the Intel SoundWire Master. @@ -1393,6 +1394,8 @@ int intel_link_startup(struct auxiliary_device *auxdev) goto err_interrupt; } } + sdw_cdns_check_self_clearing_bits(cdns, __func__, + true, INTEL_MASTER_RESET_ITERATIONS); /* Register DAIs */ ret = intel_register_dai(sdw); @@ -1709,6 +1712,8 @@ static int __maybe_unused intel_resume(struct device *dev) return ret; } } + sdw_cdns_check_self_clearing_bits(cdns, __func__, + true, INTEL_MASTER_RESET_ITERATIONS); /* * after system resume, the pm_runtime suspend() may kick in @@ -1793,6 +1798,9 @@ static int __maybe_unused intel_resume_runtime(struct device *dev) return ret; } } + sdw_cdns_check_self_clearing_bits(cdns, "intel_resume_runtime TEARDOWN", + true, INTEL_MASTER_RESET_ITERATIONS); + } else if (clock_stop_quirks & SDW_INTEL_CLK_STOP_BUS_RESET) { ret = intel_init(sdw); if (ret) { @@ -1866,6 +1874,9 @@ static int __maybe_unused intel_resume_runtime(struct device *dev) } } } + sdw_cdns_check_self_clearing_bits(cdns, "intel_resume_runtime BUS_RESET", + true, INTEL_MASTER_RESET_ITERATIONS); + } else if (!clock_stop_quirks) { clock_stop0 = sdw_cdns_is_clock_stop(&sdw->cdns); @@ -1889,6 +1900,9 @@ static int __maybe_unused intel_resume_runtime(struct device *dev) dev_err(dev, "unable to resume master during resume\n"); return ret; } + + sdw_cdns_check_self_clearing_bits(cdns, "intel_resume_runtime no_quirks", + true, INTEL_MASTER_RESET_ITERATIONS); } else { dev_err(dev, "%s clock_stop_quirks %x unsupported\n", __func__, clock_stop_quirks); -- 2.17.1

2 1

[sound:topic/asoc-pcm-addr-fixes 20/24] sound/soc/amd/renoir/acp3x-pdm-dma.c:245:18: error: 'runtime' undeclared
by kernel test robot 01 Aug '21

01 Aug '21

tree: https://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git topic/asoc-pcm-addr-fixes head: 3ed85d6b39dbdab30d06d6c686739e12c0793cde commit: c4bb3c6064991e5668115533a840609b5dedfb8e [20/24] ASoC: amd: Fix reference to PCM buffer address config: x86_64-randconfig-a006-20210728 (attached as .config) compiler: gcc-10 (Ubuntu 10.3.0-1ubuntu1~20.04) 10.3.0 reproduce (this is a W=1 build): # https://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git/commit/?id=… git remote add sound https://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git git fetch --no-tags sound topic/asoc-pcm-addr-fixes git checkout c4bb3c6064991e5668115533a840609b5dedfb8e # save the attached .config to linux build tree mkdir build_dir make W=1 O=build_dir ARCH=x86_64 SHELL=/bin/bash sound/soc/amd/renoir/ If you fix the issue, kindly add following tag as appropriate Reported-by: kernel test robot <lkp(a)intel.com> All errors (new ones prefixed by >>): sound/soc/amd/renoir/acp3x-pdm-dma.c: In function 'acp_pdm_dma_hw_params': >> sound/soc/amd/renoir/acp3x-pdm-dma.c:245:18: error: 'runtime' undeclared (first use in this function) 245 | rtd->dma_addr = runtime->dma_addr; | ^~~~~~~ sound/soc/amd/renoir/acp3x-pdm-dma.c:245:18: note: each undeclared identifier is reported only once for each function it appears in vim +/runtime +245 sound/soc/amd/renoir/acp3x-pdm-dma.c 232 233 static int acp_pdm_dma_hw_params(struct snd_soc_component *component, 234 struct snd_pcm_substream *substream, 235 struct snd_pcm_hw_params *params) 236 { 237 struct pdm_stream_instance *rtd; 238 size_t size, period_bytes; 239 240 rtd = substream->runtime->private_data; 241 if (!rtd) 242 return -EINVAL; 243 size = params_buffer_bytes(params); 244 period_bytes = params_period_bytes(params); > 245 rtd->dma_addr = runtime->dma_addr; 246 rtd->num_pages = (PAGE_ALIGN(size) >> PAGE_SHIFT); 247 config_acp_dma(rtd, substream->stream); 248 init_pdm_ring_buffer(MEM_WINDOW_START, size, period_bytes, 249 rtd->acp_base); 250 return 0; 251 } 252 --- 0-DAY CI Kernel Test Service, Intel Corporation https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

1 0

[PATCH] ALSA: memalloc: Fix regression with SNDRV_DMA_TYPE_CONTINUOUS
by Takashi Iwai 01 Aug '21

01 Aug '21

The recent code refactoring made the mmap of continuous pages to be done via the own helper snd_dma_continuous_mmap() with remap_pfn_range(). There I overlooked that dmab->addr isn't set for the allocation with SNDRV_DMA_TYPE_CONTINUOUS. This resulted always in an error at mmap with this buffer type on the system such as Intel SST Baytrail driver. This patch fixes the regression by passing the correct address. Fixes: 30b7ba6972d5 ("ALSA: core: Add continuous and vmalloc mmap ops") Reported-by: Hans de Goede <hdegoede(a)redhat.com> Link: https://lore.kernel.org/r/8d6674da-7d7b-803e-acc9-7de6cb1223fa@redhat.com Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/core/memalloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/core/memalloc.c b/sound/core/memalloc.c index 83b79edfa52d..439a358ecfe9 100644 --- a/sound/core/memalloc.c +++ b/sound/core/memalloc.c @@ -215,7 +215,7 @@ static int snd_dma_continuous_mmap(struct snd_dma_buffer *dmab, struct vm_area_struct *area) { return remap_pfn_range(area, area->vm_start, - dmab->addr >> PAGE_SHIFT, + page_to_pfn(virt_to_page(dmab->area)), area->vm_end - area->vm_start, area->vm_page_prot); } -- 2.26.2

1 0

[5.14 regression] "ALSA: core: Add continuous and vmalloc mmap ops" breaks Intel SST audio
by Hans de Goede 01 Aug '21

01 Aug '21

Hi All;, As discussed in the "5.14 regression, Intel SST (BYT) audio no longer works" thread, audio on X86 devices using the Intel SST driver no longer works with 5.14. After poking at this for a while I've found the culprit: 30b7ba6972d5 ("ALSA: core: Add continuous and vmalloc mmap ops") If I revert that single commit then everything works fine again with 5.14. This is with a Fedora 34 userspace using pipewire as audiodaemon Regards, Hans

2 3

[PATCH] ALSA: usb-audio: make array static const, makes object smaller
by Colin King 01 Aug '21

01 Aug '21

From: Colin Ian King <colin.king(a)canonical.com> Don't populate array names_to_check on the stack but instead it static. Makes the object code smaller by 56 bytes. Before: text data bss dec hex filename 103512 34380 0 137892 21aa4 ./sound/usb/mixer.o After: text data bss dec hex filename 103264 34572 0 137836 21a6c ./sound/usb/mixer.o gcc version 10.2.0) Signed-off-by: Colin Ian King <colin.king(a)canonical.com> --- sound/usb/mixer.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/sound/usb/mixer.c b/sound/usb/mixer.c index f4cdaf1ba44a..aec2499284a5 100644 --- a/sound/usb/mixer.c +++ b/sound/usb/mixer.c @@ -1572,8 +1572,9 @@ static size_t append_ctl_name(struct snd_kcontrol *kctl, const char *str) static void check_no_speaker_on_headset(struct snd_kcontrol *kctl, struct snd_card *card) { - const char *names_to_check[] = { - "Headset", "headset", "Headphone", "headphone", NULL}; + static const char *names_to_check[] = { + "Headset", "headset", "Headphone", "headphone", NULL + }; const char **s; bool found = false; -- 2.31.1

2 1