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

[PATCH] [6.4.y] ASoC: SOF: intel: hda: Clean up link DMA for IPC3 during stop
by Jaroslav Kysela 21 Aug '23

21 Aug '23

From: Ranjani Sridharan <ranjani.sridharan(a)linux.intel.com> commit 90219f1bd273055f1dc1d7bdc0965755b992c045 upstream. With IPC3, we reset hw_params during the stop trigger, so we should also clean up the link DMA during the stop trigger. Cc: <stable(a)vger.kernel.org> # 6.4.x Fixes: 1bf83fa6654c ("ASoC: SOF: Intel: hda-dai: Do not perform DMA cleanup during stop") Closes: https://github.com/thesofproject/linux/issues/4455 Closes: https://github.com/thesofproject/linux/issues/4482 Closes: https://bugzilla.kernel.org/show_bug.cgi?id=217673 Signed-off-by: Ranjani Sridharan <ranjani.sridharan(a)linux.intel.com> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Reviewed-by: Rander Wang <rander.wang(a)intel.com> Reviewed-by: Péter Ujfalusi <peter.ujfalusi(a)linux.intel.com> Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski(a)linux.intel.com> Reviewed-by: Bard Liao <yung-chuan.liao(a)linux.intel.com> Signed-off-by: Peter Ujfalusi <peter.ujfalusi(a)linux.intel.com> Link: https://lore.kernel.org/r/20230808110627.32375-1-peter.ujfalusi@linux.intel… Signed-off-by: Mark Brown <broonie(a)kernel.org> Note that many recent Intel based laptops are affected. Added missing code for 6.4 kernels to keep the fix simple not depending on the other changes. This commit is present in 6.5 tree already. Signed-off-by: Jaroslav Kysela <perex(a)perex.cz> --- sound/soc/sof/intel/hda-dai-ops.c | 13 ++++++++++++- sound/soc/sof/intel/hda-dai.c | 8 ++++---- sound/soc/sof/intel/hda.h | 2 ++ 3 files changed, 18 insertions(+), 5 deletions(-) diff --git a/sound/soc/sof/intel/hda-dai-ops.c b/sound/soc/sof/intel/hda-dai-ops.c index 4b39cecacd68..5938046f46b2 100644 --- a/sound/soc/sof/intel/hda-dai-ops.c +++ b/sound/soc/sof/intel/hda-dai-ops.c @@ -289,16 +289,27 @@ static const struct hda_dai_widget_dma_ops hda_ipc4_chain_dma_ops = { static int hda_ipc3_post_trigger(struct snd_sof_dev *sdev, struct snd_soc_dai *cpu_dai, struct snd_pcm_substream *substream, int cmd) { + struct hdac_ext_stream *hext_stream = hda_get_hext_stream(sdev, cpu_dai, substream); struct snd_soc_dapm_widget *w = snd_soc_dai_get_widget(cpu_dai, substream->stream); + struct snd_soc_pcm_runtime *rtd = asoc_substream_to_rtd(substream); + struct snd_soc_dai *codec_dai = asoc_rtd_to_codec(rtd, 0); switch (cmd) { case SNDRV_PCM_TRIGGER_SUSPEND: case SNDRV_PCM_TRIGGER_STOP: { struct snd_sof_dai_config_data data = { 0 }; + int ret; data.dai_data = DMA_CHAN_INVALID; - return hda_dai_config(w, SOF_DAI_CONFIG_FLAGS_HW_FREE, &data); + ret = hda_dai_config(w, SOF_DAI_CONFIG_FLAGS_HW_FREE, &data); + if (ret < 0) + return ret; + + if (cmd == SNDRV_PCM_TRIGGER_STOP) + return hda_link_dma_cleanup(substream, hext_stream, cpu_dai, codec_dai); + + break; } case SNDRV_PCM_TRIGGER_PAUSE_PUSH: return hda_dai_config(w, SOF_DAI_CONFIG_FLAGS_PAUSE, NULL); diff --git a/sound/soc/sof/intel/hda-dai.c b/sound/soc/sof/intel/hda-dai.c index 44a5d94c5050..8a76320c3b99 100644 --- a/sound/soc/sof/intel/hda-dai.c +++ b/sound/soc/sof/intel/hda-dai.c @@ -91,10 +91,10 @@ hda_dai_get_ops(struct snd_pcm_substream *substream, struct snd_soc_dai *cpu_dai return sdai->platform_private; } -static int hda_link_dma_cleanup(struct snd_pcm_substream *substream, - struct hdac_ext_stream *hext_stream, - struct snd_soc_dai *cpu_dai, - struct snd_soc_dai *codec_dai) +int hda_link_dma_cleanup(struct snd_pcm_substream *substream, + struct hdac_ext_stream *hext_stream, + struct snd_soc_dai *cpu_dai, + struct snd_soc_dai *codec_dai) { struct snd_sof_dev *sdev = snd_soc_component_get_drvdata(cpu_dai->component); const struct hda_dai_widget_dma_ops *ops = hda_dai_get_ops(substream, cpu_dai); diff --git a/sound/soc/sof/intel/hda.h b/sound/soc/sof/intel/hda.h index c4befacde23e..94c738eae751 100644 --- a/sound/soc/sof/intel/hda.h +++ b/sound/soc/sof/intel/hda.h @@ -942,5 +942,7 @@ const struct hda_dai_widget_dma_ops * hda_select_dai_widget_ops(struct snd_sof_dev *sdev, struct snd_sof_widget *swidget); int hda_dai_config(struct snd_soc_dapm_widget *w, unsigned int flags, struct snd_sof_dai_config_data *data); +int hda_link_dma_cleanup(struct snd_pcm_substream *substream, struct hdac_ext_stream *hext_stream, + struct snd_soc_dai *cpu_dai, struct snd_soc_dai *codec_dai); #endif -- 2.41.0

2 1

[PATCH v3 1/2] ALSA: hda/tas2781: Add tas2781 HDA driver
by Shenghao Ding 21 Aug '23

21 Aug '23

Create tas2781 side codec HDA driver for Lenovo Laptops. The quantity of the speakers has been define in ACPI. All of the tas2781s in the laptop will be aggregated as one audio speaker. The code supports realtek codec as the primary codec. Code offers several controls for digtial/analog gain setting during playback, and other for eq params setting in case of different audio profiles, such as music, voice, movie, etc. Signed-off-by: Shenghao Ding <shenghao-ding(a)ti.com> --- Changes in v3: - oprimize the indentation - fix the ambiguous error message - The check of the given value to kcontrol's put func is in the sound/tas2781-comlib.c - In tasdevice_info_programs, corrected max value - Seperate snd_kcontrol_new tas2781_dsp_controls into two standalone kcontrol. - All the controls set as const static - Add static for page_array & rgno_array - Remove uncessary blank lines - Add descriptions for tas2781_save_calibration - remove global addr handling in the code - checking subid in switch statement in function tas2781_hda_bind - add force firmware load Kcontrol - rename the kcontrol name to be more undertandable - remove Superfluous cast in tasdevice_fw_ready - correct weird line break in function tas2781_acpi_get_i2c_resource - correct Referencing adev after acpi_dev_put() in tas2781_hda_read_acpi - As to checking the given value in tasdevice_set_profile_id, it seems done by the tasdevice_info_profile - replace strcpy with strscpy in tas2781_hda_read_acpi - rewrite the subid judgement - Add tiwai(a)suse.de into Cc list - remove the cast in tas2781_acpi_get_i2c_resource - remove else in tas2781_acpi_get_i2c_resource - fix the return value in tasdevice_set_profile_id - remove unneeded NL in tasdevice_config_get - Unifiy the comment style - remove ret = 0 in tasdevice_fw_ready - remove ret in tas2781_save_calibration - remove unused ret in tas2781_hda_playback - add force firmware load Kcontrol --- sound/pci/hda/Kconfig | 15 + sound/pci/hda/Makefile | 2 + sound/pci/hda/tas2781_hda_i2c.c | 858 ++++++++++++++++++++++++++++++++ 3 files changed, 875 insertions(+) create mode 100644 sound/pci/hda/tas2781_hda_i2c.c diff --git a/sound/pci/hda/Kconfig b/sound/pci/hda/Kconfig index 886255a03e8b..e66257277492 100644 --- a/sound/pci/hda/Kconfig +++ b/sound/pci/hda/Kconfig @@ -130,6 +130,21 @@ config SND_HDA_SCODEC_CS35L41_SPI comment "Set to Y if you want auto-loading the side codec driver" depends on SND_HDA=y && SND_HDA_SCODEC_CS35L41_SPI=m +config SND_HDA_SCODEC_TAS2781_I2C + tristate "Build TAS2781 HD-audio side codec support for I2C Bus" + depends on I2C + depends on ACPI + depends on SND_SOC + select SND_SOC_TAS2781_COMLIB + select SND_SOC_TAS2781_FMWLIB + select CRC32_SARWATE + help + Say Y or M here to include TAS2781 I2C HD-audio side codec support + in snd-hda-intel driver, such as ALC287. + +comment "Set to Y if you want auto-loading the side codec driver" + depends on SND_HDA=y && SND_HDA_SCODEC_TAS2781_I2C=m + config SND_HDA_CODEC_REALTEK tristate "Build Realtek HD-audio codec support" select SND_HDA_GENERIC diff --git a/sound/pci/hda/Makefile b/sound/pci/hda/Makefile index 00d306104484..1c76609690d6 100644 --- a/sound/pci/hda/Makefile +++ b/sound/pci/hda/Makefile @@ -32,6 +32,7 @@ snd-hda-scodec-cs35l41-objs := cs35l41_hda.o snd-hda-scodec-cs35l41-i2c-objs := cs35l41_hda_i2c.o snd-hda-scodec-cs35l41-spi-objs := cs35l41_hda_spi.o snd-hda-cs-dsp-ctls-objs := hda_cs_dsp_ctl.o +snd-hda-scodec-tas2781-i2c-objs := tas2781_hda_i2c.o # common driver obj-$(CONFIG_SND_HDA) := snd-hda-codec.o @@ -56,6 +57,7 @@ obj-$(CONFIG_SND_HDA_SCODEC_CS35L41) += snd-hda-scodec-cs35l41.o obj-$(CONFIG_SND_HDA_SCODEC_CS35L41_I2C) += snd-hda-scodec-cs35l41-i2c.o obj-$(CONFIG_SND_HDA_SCODEC_CS35L41_SPI) += snd-hda-scodec-cs35l41-spi.o obj-$(CONFIG_SND_HDA_CS_DSP_CONTROLS) += snd-hda-cs-dsp-ctls.o +obj-$(CONFIG_SND_HDA_SCODEC_TAS2781_I2C) += snd-hda-scodec-tas2781-i2c.o # this must be the last entry after codec drivers; # otherwise the codec patches won't be hooked before the PCI probe diff --git a/sound/pci/hda/tas2781_hda_i2c.c b/sound/pci/hda/tas2781_hda_i2c.c new file mode 100644 index 000000000000..4ccbe12ad451 --- /dev/null +++ b/sound/pci/hda/tas2781_hda_i2c.c @@ -0,0 +1,858 @@ +// SPDX-License-Identifier: GPL-2.0 +// +// TAS2781 HDA I2C driver +// +// Copyright 2023 Texas Instruments, Inc. +// +// Author: Shenghao Ding <shenghao-ding(a)ti.com> + +#include <linux/acpi.h> +#include <linux/crc8.h> +#include <linux/crc32.h> +#include <linux/efi.h> +#include <linux/firmware.h> +#include <linux/i2c.h> +#include <linux/mod_devicetable.h> +#include <linux/module.h> +#include <linux/pm_runtime.h> +#include <linux/regmap.h> +#include <sound/hda_codec.h> +#include <sound/soc.h> +#include <sound/tas2781.h> +#include <sound/tlv.h> +#include <sound/tas2781-tlv.h> + +#include "hda_local.h" +#include "hda_auto_parser.h" +#include "hda_component.h" +#include "hda_jack.h" +#include "hda_generic.h" + +#define TASDEVICE_SPEAKER_CALIBRATION_SIZE 20 + +/* No standard control callbacks for SNDRV_CTL_ELEM_IFACE_CARD + * Define two controls, one is Volume control callbacks, the other is + * flag setting control callbacks. + */ + +/* Volume control callbacks for tas2781 */ +#define ACARD_SINGLE_RANGE_EXT_TLV(xname, xreg, xshift, xmin, xmax, xinvert, \ + xhandler_get, xhandler_put, tlv_array) \ +{ .iface = SNDRV_CTL_ELEM_IFACE_CARD, .name = (xname),\ + .access = SNDRV_CTL_ELEM_ACCESS_TLV_READ |\ + SNDRV_CTL_ELEM_ACCESS_READWRITE,\ + .tlv.p = (tlv_array), \ + .info = snd_soc_info_volsw_range, \ + .get = xhandler_get, .put = xhandler_put, \ + .private_value = (unsigned long)&(struct soc_mixer_control) \ + {.reg = xreg, .rreg = xreg, .shift = xshift, \ + .rshift = xshift, .min = xmin, .max = xmax, \ + .invert = xinvert} } + +/* Flag control callbacks for tas2781 */ +#define ACARD_SINGLE_BOOL_EXT(xname, xdata, xhandler_get, xhandler_put) \ +{ .iface = SNDRV_CTL_ELEM_IFACE_CARD, .name = xname, \ + .info = snd_ctl_boolean_mono_info, \ + .get = xhandler_get, .put = xhandler_put, \ + .private_value = xdata } + +enum calib_data { + R0_VAL = 0, + INV_R0, + R0LOW, + POWER, + TLIM, + CALIB_MAX +}; + +static int tas2781_get_i2c_res(struct acpi_resource *ares, void *data) +{ + struct tasdevice_priv *tas_priv = data; + struct acpi_resource_i2c_serialbus *sb; + + if (i2c_acpi_get_i2c_resource(ares, &sb)) { + if (tas_priv->ndev < TASDEVICE_MAX_CHANNELS && + sb->slave_address != TAS2781_GLOBAL_ADDR) { + tas_priv->tasdevice[tas_priv->ndev].dev_addr = + (unsigned int)sb->slave_address; + tas_priv->ndev++; + } + } + return 1; +} + +static int tas2781_read_acpi(struct tasdevice_priv *p, const char *hid) +{ + struct acpi_device *adev; + struct device *physdev; + LIST_HEAD(resources); + const char *sub; + int ret; + + adev = acpi_dev_get_first_match_dev(hid, NULL, -1); + if (!adev) { + dev_err(p->dev, + "Failed to find an ACPI device for %s\n", hid); + return -ENODEV; + } + + ret = acpi_dev_get_resources(adev, &resources, tas2781_get_i2c_res, p); + if (ret < 0) + goto err; + + acpi_dev_free_resource_list(&resources); + strscpy(p->dev_name, hid, sizeof(p->dev_name)); + physdev = get_device(acpi_get_first_physical_node(adev)); + acpi_dev_put(adev); + + /* No side-effect to the playback even if subsystem_id is NULL*/ + sub = acpi_get_subsystem_id(ACPI_HANDLE(physdev)); + if (IS_ERR(sub)) + sub = NULL; + + p->acpi_subsystem_id = sub; + + put_device(physdev); + + return 0; + +err: + dev_err(p->dev, "read acpi error, ret: %d\n", ret); + put_device(physdev); + + return ret; +} + +static void tas2781_hda_playback_hook(struct device *dev, int action) +{ + struct tasdevice_priv *tas_priv = dev_get_drvdata(dev); + + dev_dbg(tas_priv->dev, "%s: action = %d\n", __func__, action); + switch (action) { + case HDA_GEN_PCM_ACT_OPEN: + pm_runtime_get_sync(dev); + mutex_lock(&tas_priv->codec_lock); + tasdevice_tuning_switch(tas_priv, 0); + mutex_unlock(&tas_priv->codec_lock); + break; + case HDA_GEN_PCM_ACT_CLOSE: + mutex_lock(&tas_priv->codec_lock); + tasdevice_tuning_switch(tas_priv, 1); + mutex_unlock(&tas_priv->codec_lock); + + pm_runtime_mark_last_busy(dev); + pm_runtime_put_autosuspend(dev); + break; + default: + dev_dbg(tas_priv->dev, "Playback action not supported: %d\n", + action); + break; + } +} + +static int tasdevice_info_profile(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + + uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER; + uinfo->count = 1; + uinfo->value.integer.min = 0; + uinfo->value.integer.max = tas_priv->rcabin.ncfgs - 1; + + return 0; +} + +static int tasdevice_get_profile_id(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + + ucontrol->value.integer.value[0] = tas_priv->rcabin.profile_cfg_id; + + return 0; +} + +static int tasdevice_hda_clamp(int val, int max) +{ + if (val > max) + val = max; + + if (val < 0) + val = 0; + return val; +} + +static int tasdevice_set_profile_id(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + int nr_profile = ucontrol->value.integer.value[0]; + int max = tas_priv->rcabin.ncfgs - 1; + int val, ret = 0; + + val = tasdevice_hda_clamp(nr_profile, max); + + if (tas_priv->rcabin.profile_cfg_id != nr_profile) { + tas_priv->rcabin.profile_cfg_id = nr_profile; + ret = 1; + } + + return ret; +} + +static int tasdevice_info_programs(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + struct tasdevice_fw *tas_fw = tas_priv->fmw; + + uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER; + uinfo->count = 1; + uinfo->value.integer.min = 0; + uinfo->value.integer.max = tas_fw->nr_programs - 1; + + return 0; +} + +static int tasdevice_info_config(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + struct tasdevice_fw *tas_fw = tas_priv->fmw; + + uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER; + uinfo->count = 1; + uinfo->value.integer.min = 0; + uinfo->value.integer.max = tas_fw->nr_configurations - 1; + + return 0; +} + +static int tasdevice_program_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + + ucontrol->value.integer.value[0] = tas_priv->cur_prog; + + return 0; +} + +static int tasdevice_program_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + struct tasdevice_fw *tas_fw = tas_priv->fmw; + int nr_program = ucontrol->value.integer.value[0]; + int max = tas_fw->nr_programs - 1; + int val, ret = 0; + + val = tasdevice_hda_clamp(nr_program, max); + + if (tas_priv->cur_prog != nr_program) { + tas_priv->cur_prog = nr_program; + ret = 1; + } + + return ret; +} + +static int tasdevice_config_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + + ucontrol->value.integer.value[0] = tas_priv->cur_conf; + + return 0; +} + +static int tasdevice_config_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + struct tasdevice_fw *tas_fw = tas_priv->fmw; + int nr_config = ucontrol->value.integer.value[0]; + int max = tas_fw->nr_configurations - 1; + int val, ret = 0; + + val = tasdevice_hda_clamp(nr_config, max); + + if (tas_priv->cur_conf != nr_config) { + tas_priv->cur_conf = nr_config; + ret = 1; + } + + return ret; +} + +/* + * tas2781_digital_getvol - get the volum control + * @kcontrol: control pointer + * @ucontrol: User data + * Customer Kcontrol for tas2781 is primarily for regmap booking, paging + * depends on internal regmap mechanism. + * tas2781 contains book and page two-level register map, especially + * book switching will set the register BXXP00R7F, after switching to the + * correct book, then leverage the mechanism for paging to access the + * register. + */ +static int tas2781_digital_getvol(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + struct soc_mixer_control *mc = + (struct soc_mixer_control *)kcontrol->private_value; + + return tasdevice_digital_getvol(tas_priv, ucontrol, mc); +} + +static int tas2781_amp_getvol(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + struct soc_mixer_control *mc = + (struct soc_mixer_control *)kcontrol->private_value; + + return tasdevice_amp_getvol(tas_priv, ucontrol, mc); +} + +static int tas2781_digital_putvol(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + struct soc_mixer_control *mc = + (struct soc_mixer_control *)kcontrol->private_value; + + /* The check of the given value is in tasdevice_digital_putvol. */ + return tasdevice_digital_putvol(tas_priv, ucontrol, mc); +} + +static int tas2781_amp_putvol(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + struct soc_mixer_control *mc = + (struct soc_mixer_control *)kcontrol->private_value; + + /* The check of the given value is in tasdevice_amp_putvol. */ + return tasdevice_amp_putvol(tas_priv, ucontrol, mc); +} + +static int tas2781_force_fwload_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + + ucontrol->value.integer.value[0] = (int)tas_priv->force_fwload_status; + dev_dbg(tas_priv->dev, "%s : Force FWload %s\n", __func__, + tas_priv->force_fwload_status ? "ON" : "OFF"); + + return 0; +} + +static int tas2781_force_fwload_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct tasdevice_priv *tas_priv = snd_kcontrol_chip(kcontrol); + bool change, val = (bool)ucontrol->value.integer.value[0]; + + if (tas_priv->force_fwload_status == val) + change = false; + else { + change = true; + tas_priv->force_fwload_status = val; + } + dev_dbg(tas_priv->dev, "%s : Force FWload %s\n", __func__, + tas_priv->force_fwload_status ? "ON" : "OFF"); + + return change; +} + +static const struct snd_kcontrol_new tas2781_snd_controls[] = { + ACARD_SINGLE_RANGE_EXT_TLV("Speaker Analog Gain", TAS2781_AMP_LEVEL, + 1, 0, 20, 0, tas2781_amp_getvol, + tas2781_amp_putvol, amp_vol_tlv), + ACARD_SINGLE_RANGE_EXT_TLV("Speaker Digital Gain", TAS2781_DVC_LVL, + 0, 0, 200, 1, tas2781_digital_getvol, + tas2781_digital_putvol, dvc_tlv), + ACARD_SINGLE_BOOL_EXT("Speaker Force Firmware Load", 0, + tas2781_force_fwload_get, tas2781_force_fwload_put), +}; + +static const struct snd_kcontrol_new tas2781_prof_ctrl = { + .name = "Speaker Profile Id", + .iface = SNDRV_CTL_ELEM_IFACE_CARD, + .info = tasdevice_info_profile, + .get = tasdevice_get_profile_id, + .put = tasdevice_set_profile_id, +}; + +static const struct snd_kcontrol_new tas2781_dsp_prog_ctrl = { + .name = "Speaker Program Id", + .iface = SNDRV_CTL_ELEM_IFACE_CARD, + .info = tasdevice_info_programs, + .get = tasdevice_program_get, + .put = tasdevice_program_put, +}; + +static const struct snd_kcontrol_new tas2781_dsp_conf_ctrl = { + .name = "Speaker Config Id", + .iface = SNDRV_CTL_ELEM_IFACE_CARD, + .info = tasdevice_info_config, + .get = tasdevice_config_get, + .put = tasdevice_config_put, +}; + +static void tas2781_apply_calib(struct tasdevice_priv *tas_priv) +{ + static const unsigned char page_array[CALIB_MAX] = { + 0x17, 0x18, 0x18, 0x0d, 0x18 + }; + static const unsigned char rgno_array[CALIB_MAX] = { + 0x74, 0x0c, 0x14, 0x3c, 0x7c + }; + unsigned char *data; + int i, j, rc; + + for (i = 0; i < tas_priv->ndev; i++) { + data = tas_priv->cali_data.data + + i * TASDEVICE_SPEAKER_CALIBRATION_SIZE; + for (j = 0; j < CALIB_MAX; j++) { + rc = tasdevice_dev_bulk_write(tas_priv, i, + TASDEVICE_REG(0, page_array[j], rgno_array[j]), + &(data[4 * j]), 4); + if (rc < 0) + dev_err(tas_priv->dev, + "chn %d calib %d bulk_wr err = %d\n", + i, j, rc); + } + } +} + +/* Update the calibrate data, including speaker impedance, f0, etc, into algo. + * Calibrate data is done by manufacturer in the factory. These data are used + * by Algo for calucating the speaker temperature, speaker membrance excursion + * and f0 in real time during playback. + */ +static int tas2781_save_calibration(struct tasdevice_priv *tas_priv) +{ + efi_guid_t efi_guid = EFI_GUID(0x02f9af02, 0x7734, 0x4233, 0xb4, 0x3d, + 0x93, 0xfe, 0x5a, 0xa3, 0x5d, 0xb3); + static efi_char16_t efi_name[] = L"CALI_DATA"; + struct tm *tm = &tas_priv->tm; + unsigned int attr, crc; + unsigned int *tmp_val; + efi_status_t status; + + /* Lenovo devices */ + if (tas_priv->catlog_id == LENOVO) + efi_guid = EFI_GUID(0x1f52d2a1, 0xbb3a, 0x457d, 0xbc, 0x09, + 0x43, 0xa3, 0xf4, 0x31, 0x0a, 0x92); + + tas_priv->cali_data.total_sz = 0; + /* Get real size of UEFI variable */ + status = efi.get_variable(efi_name, &efi_guid, &attr, + &tas_priv->cali_data.total_sz, tas_priv->cali_data.data); + if (status == EFI_BUFFER_TOO_SMALL) { + /* Allocate data buffer of data_size bytes */ + tas_priv->cali_data.data = devm_kzalloc(tas_priv->dev, + tas_priv->cali_data.total_sz, GFP_KERNEL); + if (!tas_priv->cali_data.data) + return -ENOMEM; + /* Get variable contents into buffer */ + status = efi.get_variable(efi_name, &efi_guid, &attr, + &tas_priv->cali_data.total_sz, + tas_priv->cali_data.data); + if (status != EFI_SUCCESS) + return -EINVAL; + } + + tmp_val = (unsigned int *)tas_priv->cali_data.data; + + crc = crc32(~0, tas_priv->cali_data.data, 84) ^ ~0; + dev_dbg(tas_priv->dev, "cali crc 0x%08x PK tmp_val 0x%08x\n", + crc, tmp_val[21]); + + if (crc == tmp_val[21]) { + time64_to_tm(tmp_val[20], 0, tm); + dev_dbg(tas_priv->dev, "%4ld-%2d-%2d, %2d:%2d:%2d\n", + tm->tm_year, tm->tm_mon, tm->tm_mday, + tm->tm_hour, tm->tm_min, tm->tm_sec); + tas2781_apply_calib(tas_priv); + } else + tas_priv->cali_data.total_sz = 0; + + return 0; +} + +static void tasdev_fw_ready(const struct firmware *fmw, void *context) +{ + struct tasdevice_priv *tas_priv = context; + struct hda_codec *codec = tas_priv->codec; + int i, ret; + + pm_runtime_get_sync(tas_priv->dev); + mutex_lock(&tas_priv->codec_lock); + + ret = tasdevice_rca_parser(tas_priv, fmw); + if (ret) + goto out; + + ret = snd_ctl_add(codec->card, + snd_ctl_new1(&tas2781_prof_ctrl, tas_priv)); + if (ret) { + dev_err(tas_priv->dev, + "Failed to add KControl %s = %d\n", + tas2781_prof_ctrl.name, ret); + goto out; + } + + for (i = 0; i < ARRAY_SIZE(tas2781_snd_controls); i++) { + ret = snd_ctl_add(codec->card, + snd_ctl_new1(&tas2781_snd_controls[i], tas_priv)); + if (ret) { + dev_err(tas_priv->dev, + "Failed to add KControl %s = %d\n", + tas2781_snd_controls[i].name, ret); + goto out; + } + } + + tasdevice_dsp_remove(tas_priv); + + tas_priv->fw_state = TASDEVICE_DSP_FW_PENDING; + scnprintf(tas_priv->coef_binaryname, 64, "TAS2XXX%04X.bin", + codec->core.subsystem_id & 0xffff); + ret = tasdevice_dsp_parser(tas_priv); + if (ret) { + dev_err(tas_priv->dev, "dspfw load %s error\n", + tas_priv->coef_binaryname); + tas_priv->fw_state = TASDEVICE_DSP_FW_FAIL; + goto out; + } + + ret = snd_ctl_add(codec->card, + snd_ctl_new1(&tas2781_dsp_prog_ctrl, tas_priv)); + if (ret) { + dev_err(tas_priv->dev, + "Failed to add KControl %s = %d\n", + tas2781_dsp_prog_ctrl.name, ret); + goto out; + } + + ret = snd_ctl_add(codec->card, + snd_ctl_new1(&tas2781_dsp_conf_ctrl, tas_priv)); + if (ret) { + dev_err(tas_priv->dev, + "Failed to add KControl %s = %d\n", + tas2781_dsp_conf_ctrl.name, ret); + goto out; + } + + tas_priv->fw_state = TASDEVICE_DSP_FW_ALL_OK; + tasdevice_prmg_load(tas_priv, 0); + + /* If calibrated data occurs error, dsp will still works with default + * calibrated data inside algo. + */ + tas2781_save_calibration(tas_priv); + +out: + if (tas_priv->fw_state == TASDEVICE_DSP_FW_FAIL) { + /*If DSP FW fail, kcontrol won't be created */ + tasdevice_config_info_remove(tas_priv); + tasdevice_dsp_remove(tas_priv); + } + mutex_unlock(&tas_priv->codec_lock); + if (fmw) + release_firmware(fmw); + pm_runtime_mark_last_busy(tas_priv->dev); + pm_runtime_put_autosuspend(tas_priv->dev); +} + +static int tas2781_hda_bind(struct device *dev, struct device *master, + void *master_data) +{ + struct tasdevice_priv *tas_priv = dev_get_drvdata(dev); + struct hda_component *comps = master_data; + struct hda_codec *codec; + unsigned int subid; + int ret; + + if (!comps || tas_priv->index < 0 || + tas_priv->index >= HDA_MAX_COMPONENTS) + return -EINVAL; + + comps = &comps[tas_priv->index]; + if (comps->dev) + return -EBUSY; + + codec = comps->codec; + subid = codec->core.subsystem_id >> 16; + + switch (subid) { + case 0x17aa: + tas_priv->catlog_id = LENOVO; + break; + default: + tas_priv->catlog_id = OTHERS; + break; + } + + pm_runtime_get_sync(dev); + + comps->dev = dev; + + strscpy(comps->name, dev_name(dev), sizeof(comps->name)); + + ret = tascodec_init(tas_priv, codec, tasdev_fw_ready); + if (ret) + return ret; + + comps->playback_hook = tas2781_hda_playback_hook; + + pm_runtime_mark_last_busy(dev); + pm_runtime_put_autosuspend(dev); + + return 0; +} + +static void tas2781_hda_unbind(struct device *dev, + struct device *master, void *master_data) +{ + struct tasdevice_priv *tas_priv = dev_get_drvdata(dev); + struct hda_component *comps = master_data; + + if (comps[tas_priv->index].dev == dev) + memset(&comps[tas_priv->index], 0, sizeof(*comps)); + + tasdevice_config_info_remove(tas_priv); + tasdevice_dsp_remove(tas_priv); + + tas_priv->fw_state = TASDEVICE_DSP_FW_PENDING; +} + +static const struct component_ops tas2781_hda_comp_ops = { + .bind = tas2781_hda_bind, + .unbind = tas2781_hda_unbind, +}; + +static void tas2781_hda_remove(struct device *dev) +{ + struct tasdevice_priv *tas_priv = dev_get_drvdata(dev); + + pm_runtime_get_sync(tas_priv->dev); + pm_runtime_disable(tas_priv->dev); + + component_del(tas_priv->dev, &tas2781_hda_comp_ops); + + pm_runtime_put_noidle(tas_priv->dev); + + tasdevice_remove(tas_priv); +} + +static int tas2781_hda_i2c_probe(struct i2c_client *clt) +{ + struct tasdevice_priv *tas_priv; + const char *device_name; + int ret; + + if (strstr(dev_name(&clt->dev), "TIAS2781")) + device_name = "TIAS2781"; + else + return -ENODEV; + + tas_priv = tasdevice_kzalloc(clt); + if (!tas_priv) + return -ENOMEM; + + tas_priv->irq_info.irq = clt->irq; + ret = tas2781_read_acpi(tas_priv, device_name); + if (ret) + return dev_err_probe(tas_priv->dev, ret, + "Platform not supported\n"); + + ret = tasdevice_init(tas_priv); + if (ret) + goto err; + + pm_runtime_set_autosuspend_delay(tas_priv->dev, 3000); + pm_runtime_use_autosuspend(tas_priv->dev); + pm_runtime_mark_last_busy(tas_priv->dev); + pm_runtime_set_active(tas_priv->dev); + pm_runtime_get_noresume(tas_priv->dev); + pm_runtime_enable(tas_priv->dev); + + pm_runtime_put_autosuspend(tas_priv->dev); + + ret = component_add(tas_priv->dev, &tas2781_hda_comp_ops); + if (ret) { + dev_err(tas_priv->dev, "Register component failed: %d\n", ret); + pm_runtime_disable(tas_priv->dev); + goto err; + } + + tas2781_reset(tas_priv); +err: + if (ret) + tas2781_hda_remove(&clt->dev); + return ret; +} + +static void tas2781_hda_i2c_remove(struct i2c_client *clt) +{ + tas2781_hda_remove(&clt->dev); +} + +static int tas2781_runtime_suspend(struct device *dev) +{ + struct tasdevice_priv *tas_priv = dev_get_drvdata(dev); + int i; + + dev_dbg(tas_priv->dev, "Runtime Suspend\n"); + + mutex_lock(&tas_priv->codec_lock); + + if (tas_priv->playback_started) { + tasdevice_tuning_switch(tas_priv, 1); + tas_priv->playback_started = false; + } + + for (i = 0; i < tas_priv->ndev; i++) { + tas_priv->tasdevice[i].cur_book = -1; + tas_priv->tasdevice[i].cur_prog = -1; + tas_priv->tasdevice[i].cur_conf = -1; + } + + regcache_cache_only(tas_priv->regmap, true); + regcache_mark_dirty(tas_priv->regmap); + + mutex_unlock(&tas_priv->codec_lock); + + return 0; +} + +static int tas2781_runtime_resume(struct device *dev) +{ + struct tasdevice_priv *tas_priv = dev_get_drvdata(dev); + unsigned long calib_data_sz = + tas_priv->ndev * TASDEVICE_SPEAKER_CALIBRATION_SIZE; + int ret; + + dev_dbg(tas_priv->dev, "Runtime Resume\n"); + + mutex_lock(&tas_priv->codec_lock); + + regcache_cache_only(tas_priv->regmap, false); + ret = regcache_sync(tas_priv->regmap); + if (ret) { + dev_err(tas_priv->dev, + "Failed to restore register cache: %d\n", ret); + goto out; + } + + tasdevice_prmg_load(tas_priv, tas_priv->cur_prog); + + /* If calibrated data occurs error, dsp will still works with default + * calibrated data inside algo. + */ + if (tas_priv->cali_data.total_sz > calib_data_sz) + tas2781_apply_calib(tas_priv); + +out: + mutex_unlock(&tas_priv->codec_lock); + + return ret; +} + +static int tas2781_system_suspend(struct device *dev) +{ + struct tasdevice_priv *tas_priv = dev_get_drvdata(dev); + int ret; + + dev_dbg(tas_priv->dev, "System Suspend\n"); + + ret = pm_runtime_force_suspend(dev); + if (ret) + return ret; + + /* Shutdown chip before system suspend */ + regcache_cache_only(tas_priv->regmap, false); + tasdevice_tuning_switch(tas_priv, 1); + regcache_cache_only(tas_priv->regmap, true); + regcache_mark_dirty(tas_priv->regmap); + + /* + * Reset GPIO may be shared, so cannot reset here. + * However beyond this point, amps may be powered down. + */ + return 0; +} + +static int tas2781_system_resume(struct device *dev) +{ + struct tasdevice_priv *tas_priv = dev_get_drvdata(dev); + unsigned long calib_data_sz = + tas_priv->ndev * TASDEVICE_SPEAKER_CALIBRATION_SIZE; + int i, ret; + + dev_dbg(tas_priv->dev, "System Resume\n"); + + ret = pm_runtime_force_resume(dev); + if (ret) + return ret; + + mutex_lock(&tas_priv->codec_lock); + + for (i = 0; i < tas_priv->ndev; i++) { + tas_priv->tasdevice[i].cur_book = -1; + tas_priv->tasdevice[i].cur_prog = -1; + tas_priv->tasdevice[i].cur_conf = -1; + } + tas2781_reset(tas_priv); + tasdevice_prmg_load(tas_priv, tas_priv->cur_prog); + + /* If calibrated data occurs error, dsp will still work with default + * calibrated data inside algo. + */ + if (tas_priv->cali_data.total_sz > calib_data_sz) + tas2781_apply_calib(tas_priv); + mutex_unlock(&tas_priv->codec_lock); + + return 0; +} + +static const struct dev_pm_ops tas2781_hda_pm_ops = { + RUNTIME_PM_OPS(tas2781_runtime_suspend, tas2781_runtime_resume, NULL) + SYSTEM_SLEEP_PM_OPS(tas2781_system_suspend, tas2781_system_resume) +}; + +static const struct i2c_device_id tas2781_hda_i2c_id[] = { + { "tas2781-hda", 0 }, + {} +}; + +static const struct acpi_device_id tas2781_acpi_hda_match[] = { + {"TIAS2781", 0 }, + {} +}; +MODULE_DEVICE_TABLE(acpi, tas2781_acpi_hda_match); + +static struct i2c_driver tas2781_hda_i2c_driver = { + .driver = { + .name = "tas2781-hda", + .acpi_match_table = tas2781_acpi_hda_match, + .pm = &tas2781_hda_pm_ops, + }, + .id_table = tas2781_hda_i2c_id, + .probe_new = tas2781_hda_i2c_probe, + .remove = tas2781_hda_i2c_remove, +}; +module_i2c_driver(tas2781_hda_i2c_driver); + +MODULE_DESCRIPTION("TAS2781 HDA Driver"); +MODULE_AUTHOR("Shenghao Ding, TI, <shenghao-ding(a)ti.com>"); +MODULE_LICENSE("GPL"); +MODULE_IMPORT_NS(SND_SOC_TAS2781_FMWLIB); -- 2.34.1

4 12

[PATCH 1/3] ASoC: dt-bindings: fsl_easrc: Add support for imx8mp-easrc
by Adam Ford 21 Aug '23

21 Aug '23

The i.MX8MP appears to have the same easrc support as the Nano, so add imx8mp as an option with a fallback to imx8mn. Signed-off-by: Adam Ford <aford173(a)gmail.com> diff --git a/Documentation/devicetree/bindings/sound/fsl,easrc.yaml b/Documentation/devicetree/bindings/sound/fsl,easrc.yaml index bdde68a1059c..2d53b3b10f2c 100644 --- a/Documentation/devicetree/bindings/sound/fsl,easrc.yaml +++ b/Documentation/devicetree/bindings/sound/fsl,easrc.yaml @@ -14,7 +14,11 @@ properties: pattern: "^easrc@.*" compatible: - const: fsl,imx8mn-easrc + oneOf: + - items: + - enum: + - fsl,imx8mp-easrc + - const: fsl,imx8mn-easrc reg: maxItems: 1 -- 2.39.2

3 9

[PATCH] iommu: Remove the device_lock_assert() from __iommu_probe_device()
by Jason Gunthorpe 21 Aug '23

21 Aug '23

It turns out several drivers are calling of_dma_configure() outside the expected bus_type.dma_configure op. This ends up being mis-locked and triggers a lockdep assertion, or instance: iommu_probe_device_locked+0xd4/0xe4 of_iommu_configure+0x10c/0x200 of_dma_configure_id+0x104/0x3b8 a6xx_gmu_init+0x4c/0xccc [msm] a6xx_gpu_init+0x3ac/0x770 [msm] adreno_bind+0x174/0x2ac [msm] component_bind_all+0x118/0x24c msm_drm_bind+0x1e8/0x6c4 [msm] try_to_bring_up_aggregate_device+0x168/0x1d4 __component_add+0xa8/0x170 component_add+0x14/0x20 dsi_dev_attach+0x20/0x2c [msm] dsi_host_attach+0x9c/0x144 [msm] devm_mipi_dsi_attach+0x34/0xb4 lt9611uxc_attach_dsi.isra.0+0x84/0xfc [lontium_lt9611uxc] lt9611uxc_probe+0x5c8/0x68c [lontium_lt9611uxc] i2c_device_probe+0x14c/0x290 really_probe+0x148/0x2b4 __driver_probe_device+0x78/0x12c driver_probe_device+0x3c/0x160 __device_attach_driver+0xb8/0x138 bus_for_each_drv+0x84/0xe0 __device_attach+0xa8/0x1b0 device_initial_probe+0x14/0x20 bus_probe_device+0xb0/0xb4 deferred_probe_work_func+0x8c/0xc8 process_one_work+0x1ec/0x53c worker_thread+0x298/0x408 kthread+0x124/0x128 ret_from_fork+0x10/0x20 It is subtle and was never documented or enforced, but there has always been an assumption that of_dma_configure_id() is not concurrent. It makes several calls into the iommu layer that require this, including dev_iommu_get(). The majority of cases have been preventing concurrency using the device_lock(). Thus the new lock debugging added exposes an existing problem in drivers. On inspection this looks like a theoretical locking problem as generally the cases are already assuming they are the exclusive (single threaded) user of the target device. Sadly, there are deeper technical problems with all of the places doing this. There are several problemetic patterns: 1) Probe a driver on device A and then steal device B and use it as part of the driver operation. Since no driver was probed to device B it means we never called bus_type.dma_configure and thus the drivers hackily try to open code this. Unfortunately nothing prevents another driver from binding to device B and creating total chaos. eg vfio bind triggered by userspace 2) Probe a driver on device A and then create a new platform driver B for a fwnode that doesn't have one, then do #1 This has the same essential problem as #1, the new device is never probed so the hack call to of_dma_configure() is needed to setup DMA, and we are at risk of something else trying to use the device. 3) Probe a driver on device A but the of_node was incorrect for DMA so fix it by figuring out the right node and calling of_dma_configure() This will blow up in the iommu code if the driver is unprobed because the bus_type now assumes that dma_configure and dma_cleanup are strictly paired. Since dma_configure will have done the wrong thing due to the missing of_node, dma_cleanup will be unpaired and iommu_device_unuse_default_domain() will blow up. Further the driver operating on device A will not be protected against changes to the iommu domain since it never called iommu_device_use_default_domain() At least this case will not throw a lockdep warning as probe is done under the device_lock already. While it is tempting to just add a device_lock() around the few missing places, this doesn't solve the above underlying problems with those callers. For this reason remove the locking assertion in the iommu code so that theses driver problems continue to remain hidden. Add a comment to all the problematic call sites to encourage fixing and discourage future cut & paste expanding the problem. Add documentation to of_dma_configure_id() explaining the locking. Fixes: b19ca5922dff ("iommu: Do not attempt to re-lock the iommu device when probing") Reported-by: Marek Szyprowski <m.szyprowski(a)samsung.com> Closes: https://lore.kernel.org/all/d090f196-a5c2-b188-31bf-e42553d8d251@samsung.co… Signed-off-by: Jason Gunthorpe <jgg(a)nvidia.com> --- drivers/bcma/main.c | 4 ++++ drivers/dma/qcom/hidma_mgmt.c | 4 ++++ drivers/gpu/drm/etnaviv/etnaviv_drv.c | 4 ++++ drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 8 ++++++++ drivers/gpu/drm/sun4i/sun4i_backend.c | 4 ++++ drivers/gpu/drm/sun4i/sun8i_mixer.c | 4 ++++ drivers/gpu/drm/vc4/vc4_drv.c | 4 ++++ drivers/gpu/host1x/bus.c | 4 ++++ drivers/iommu/iommu.c | 12 ++++++++---- drivers/media/platform/qcom/venus/firmware.c | 4 ++++ drivers/media/platform/samsung/s5p-mfc/s5p_mfc.c | 4 ++++ drivers/net/wireless/ath/ath10k/snoc.c | 4 ++++ drivers/net/wireless/ath/ath11k/ahb.c | 4 ++++ drivers/of/device.c | 3 +++ sound/soc/bcm/bcm63xx-pcm-whistler.c | 4 ++++ 15 files changed, 67 insertions(+), 4 deletions(-) This is my suggestion to the issue, I don't like removing the locking assertion but if we can't have debug kernels reporting existing bugs then this would be the next best option. I've CC'd all the driver owners, if people have better explanations for what their drivers are doing please share :) diff --git a/drivers/bcma/main.c b/drivers/bcma/main.c index 7061d3ee836a15..4df97000d7e0a7 100644 --- a/drivers/bcma/main.c +++ b/drivers/bcma/main.c @@ -208,6 +208,10 @@ static void bcma_of_fill_device(struct device *parent, core->irq = bcma_of_get_irq(parent, core, 0); + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ of_dma_configure(&core->dev, node, false); } diff --git a/drivers/dma/qcom/hidma_mgmt.c b/drivers/dma/qcom/hidma_mgmt.c index 05e96b31d87169..2efd97747d37b3 100644 --- a/drivers/dma/qcom/hidma_mgmt.c +++ b/drivers/dma/qcom/hidma_mgmt.c @@ -390,6 +390,10 @@ static int __init hidma_mgmt_of_populate_channels(struct device_node *np) goto out; } new_pdev->dev.of_node = child; + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ of_dma_configure(&new_pdev->dev, child, true); /* * It is assumed that calling of_msi_configure is safe on diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c index 31a7f59ccb49ee..22729c66e39a65 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c @@ -635,6 +635,10 @@ static int etnaviv_pdev_probe(struct platform_device *pdev) * device as the GPU we found. This assumes that all Vivante * GPUs in the system share the same DMA constraints. */ + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ if (first_node) of_dma_configure(&pdev->dev, first_node, true); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index 5deb79924897af..509bd8b7ec12ae 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -1492,6 +1492,10 @@ int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) gmu->dev = &pdev->dev; + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ of_dma_configure(gmu->dev, node, true); pm_runtime_enable(gmu->dev); @@ -1556,6 +1560,10 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) gmu->dev = &pdev->dev; + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ of_dma_configure(gmu->dev, node, true); /* Fow now, don't do anything fancy until we get our feet under us */ diff --git a/drivers/gpu/drm/sun4i/sun4i_backend.c b/drivers/gpu/drm/sun4i/sun4i_backend.c index b11dbd50d73e98..30cbefb6799e51 100644 --- a/drivers/gpu/drm/sun4i/sun4i_backend.c +++ b/drivers/gpu/drm/sun4i/sun4i_backend.c @@ -800,6 +800,10 @@ static int sun4i_backend_bind(struct device *dev, struct device *master, * for us, and DRM doesn't do per-device allocation either, so * we would need to fix DRM first... */ + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ ret = of_dma_configure(drm->dev, dev->of_node, true); if (ret) return ret; diff --git a/drivers/gpu/drm/sun4i/sun8i_mixer.c b/drivers/gpu/drm/sun4i/sun8i_mixer.c index 11d5244a5aa5fc..e26bea207599d3 100644 --- a/drivers/gpu/drm/sun4i/sun8i_mixer.c +++ b/drivers/gpu/drm/sun4i/sun8i_mixer.c @@ -399,6 +399,10 @@ static int sun8i_mixer_bind(struct device *dev, struct device *master, * DRM doesn't do per-device allocation either, so we * would need to fix DRM first... */ + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ ret = of_dma_configure(drm->dev, dev->of_node, true); if (ret) return ret; diff --git a/drivers/gpu/drm/vc4/vc4_drv.c b/drivers/gpu/drm/vc4/vc4_drv.c index 823395c23cc308..82496d8fb50f84 100644 --- a/drivers/gpu/drm/vc4/vc4_drv.c +++ b/drivers/gpu/drm/vc4/vc4_drv.c @@ -305,6 +305,10 @@ static int vc4_drm_bind(struct device *dev) node = of_find_matching_node_and_match(NULL, vc4_dma_range_matches, NULL); if (node) { + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ ret = of_dma_configure(dev, node, true); of_node_put(node); diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c index 4d16a3396c4abf..5c7ef738a7d7d7 100644 --- a/drivers/gpu/host1x/bus.c +++ b/drivers/gpu/host1x/bus.c @@ -475,6 +475,10 @@ static int host1x_device_add(struct host1x *host1x, device->dev.bus = &host1x_bus_type; device->dev.parent = host1x->dev; + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ of_dma_configure(&device->dev, host1x->dev->of_node, true); device->dev.dma_parms = &device->dma_parms; diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 18162049bd2294..eb623a611f571d 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -458,8 +458,15 @@ static int __iommu_probe_device(struct device *dev, struct list_head *group_list * Allow __iommu_probe_device() to be safely called in parallel, * both dev->iommu_group and the initial setup of dev->iommu are * protected this way. + * + * We cannot have the assertion yet because of a number of drivers + * improperly calling of_dma_configure() and expecting to attach + * the iommu outside the bus_type.dma_configure callback. This messes + * up our ownership tracking, does not exclude other drivers from + * binding an operating the device, and is technically mislocked. + * + * device_lock_assert(dev); */ - device_lock_assert(dev); if (!ops) return -ENODEV; @@ -527,8 +534,6 @@ int iommu_probe_device_locked(struct device *dev) const struct iommu_ops *ops; int ret; - device_lock_assert(dev); - ret = __iommu_probe_device(dev, NULL); if (ret) return ret; @@ -1202,7 +1207,6 @@ void iommu_group_remove_device(struct device *dev) __iommu_group_remove_device(dev); } device_unlock(dev); - } EXPORT_SYMBOL_GPL(iommu_group_remove_device); diff --git a/drivers/media/platform/qcom/venus/firmware.c b/drivers/media/platform/qcom/venus/firmware.c index cfb11c551167ce..f70360c5534d64 100644 --- a/drivers/media/platform/qcom/venus/firmware.c +++ b/drivers/media/platform/qcom/venus/firmware.c @@ -296,6 +296,10 @@ int venus_firmware_init(struct venus_core *core) pdev->dev.of_node = np; + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ ret = of_dma_configure(&pdev->dev, np, true); if (ret) { dev_err(core->dev, "dma configure fail\n"); diff --git a/drivers/media/platform/samsung/s5p-mfc/s5p_mfc.c b/drivers/media/platform/samsung/s5p-mfc/s5p_mfc.c index e30e54935d79a9..40fa7b62545d7a 100644 --- a/drivers/media/platform/samsung/s5p-mfc/s5p_mfc.c +++ b/drivers/media/platform/samsung/s5p-mfc/s5p_mfc.c @@ -1114,6 +1114,10 @@ static struct device *s5p_mfc_alloc_memdev(struct device *dev, * to be treated as valid DMA masters we need a bit of a hack to force * them to inherit the MFC node's DMA configuration. */ + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ of_dma_configure(child, dev->of_node, true); if (device_add(child) == 0) { diff --git a/drivers/net/wireless/ath/ath10k/snoc.c b/drivers/net/wireless/ath/ath10k/snoc.c index 26214c00cd0d7b..977e5c6186b131 100644 --- a/drivers/net/wireless/ath/ath10k/snoc.c +++ b/drivers/net/wireless/ath/ath10k/snoc.c @@ -1614,6 +1614,10 @@ static int ath10k_fw_init(struct ath10k *ar) pdev->dev.of_node = node; + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ ret = of_dma_configure(&pdev->dev, node, true); if (ret) { ath10k_err(ar, "dma configure fail: %d\n", ret); diff --git a/drivers/net/wireless/ath/ath11k/ahb.c b/drivers/net/wireless/ath/ath11k/ahb.c index 139da578831a44..a4d2a89ab96de0 100644 --- a/drivers/net/wireless/ath/ath11k/ahb.c +++ b/drivers/net/wireless/ath/ath11k/ahb.c @@ -1012,6 +1012,10 @@ static int ath11k_ahb_fw_resources_init(struct ath11k_base *ab) return PTR_ERR(pdev); } + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ ret = of_dma_configure(&pdev->dev, node, true); if (ret) { ath11k_err(ab, "dma configure fail: %d\n", ret); diff --git a/drivers/of/device.c b/drivers/of/device.c index 0f00f1b8070829..bc6f62ebcbd040 100644 --- a/drivers/of/device.c +++ b/drivers/of/device.c @@ -108,6 +108,9 @@ of_dma_set_restricted_buffer(struct device *dev, struct device_node *np) * If platform code needs to use its own special DMA configuration, it * can use a platform bus notifier and handle BUS_NOTIFY_ADD_DEVICE events * to fix up DMA configuration. + * + * This function should only be called from the bus_type.dma_configure callback + * and the driver_lock(dev) must be held. */ int of_dma_configure_id(struct device *dev, struct device_node *np, bool force_dma, const u32 *id) diff --git a/sound/soc/bcm/bcm63xx-pcm-whistler.c b/sound/soc/bcm/bcm63xx-pcm-whistler.c index 2c600b017524f3..d8e0520ca61155 100644 --- a/sound/soc/bcm/bcm63xx-pcm-whistler.c +++ b/sound/soc/bcm/bcm63xx-pcm-whistler.c @@ -354,6 +354,10 @@ static int bcm63xx_soc_pcm_new(struct snd_soc_component *component, i2s_priv = dev_get_drvdata(asoc_rtd_to_cpu(rtd, 0)->dev); + /* + * FIXME: of_dma_configure() should not be called outside + * bus_type.dma_configure + */ of_dma_configure(pcm->card->dev, pcm->card->dev->of_node, 1); ret = dma_coerce_mask_and_coherent(pcm->card->dev, DMA_BIT_MASK(32)); base-commit: 1bba4f1c0b4f2429e16da231a246ab4b99cc633c -- 2.41.0

4 7

[PATCH] ASoC: amd: yc: Add VivoBook Pro 15 to quirks list for acp6x
by BrenoRCBrito 21 Aug '23

21 Aug '23

VivoBook Pro 15 Ryzen Edition uses Ryzen 6800H processor, and adding to quirks list for acp6x will enable internal mic. Signed-off-by: BrenoRCBrito <brenorcbrito(a)gmail.com> --- sound/soc/amd/yc/acp6x-mach.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/amd/yc/acp6x-mach.c b/sound/soc/amd/yc/acp6x-mach.c index a2fe3bd4f9a1..a2661dd637a5 100644 --- a/sound/soc/amd/yc/acp6x-mach.c +++ b/sound/soc/amd/yc/acp6x-mach.c @@ -248,6 +248,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "M3402RA"), } }, + { + .driver_data = &acp6x_card, + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK COMPUTER INC."), + DMI_MATCH(DMI_PRODUCT_NAME, "M6500RC"), + } + }, { .driver_data = &acp6x_card, .matches = { -- 2.40.1

2 1

[PATCH 0/4] ASoC: amd: acp: Add sound support for a line of HUAWEI laptops
by Marian Postevca 21 Aug '23

21 Aug '23

This series adds support for a line of HUAWEI laptops with AMD CPUs that connect using the ACP3x module to a ES8336 codec. The codec driver must be extended to support the S32 LE format and the MCLK div by 2 option. MCLK div by 2 is needed for one specific SKU which uses a 48Mhz MCLK which seems to be too high of a frequency for the codec and must be divided by 2. The acp legacy driver must also be extended by using callbacks so that the more complicated handling for this specific codec can be moved outside the more generic ACP code. The last patch tries to avoid anoying pop sounds when the speaker/headphones are enabled/disabled by delaying the handling of the GPIOs and using a mutex to avoid race conditions between the speaker power event callback and the trigger callback. Marian Postevca (4): ASoC: es8316: Enable support for S32 LE format and MCLK div by 2 ASoC: amd: acp: Add support for splitting the codec specific code from the ACP driver ASoC: amd: acp: Add machine driver that enables sound for systems with a ES8336 codec ASoC: amd: acp: Improve support for speaker power events sound/soc/amd/acp-config.c | 70 ++ sound/soc/amd/acp/Makefile | 2 +- sound/soc/amd/acp/acp-legacy-mach.c | 105 ++- sound/soc/amd/acp/acp-mach-common.c | 8 + sound/soc/amd/acp/acp-mach.h | 67 ++ sound/soc/amd/acp/acp-renoir.c | 4 + sound/soc/amd/acp/acp3x-es83xx/acp3x-es83xx.c | 615 ++++++++++++++++++ sound/soc/amd/acp/acp3x-es83xx/acp3x-es83xx.h | 12 + sound/soc/codecs/es8316.c | 21 +- sound/soc/codecs/es8316.h | 3 + 10 files changed, 886 insertions(+), 21 deletions(-) create mode 100644 sound/soc/amd/acp/acp3x-es83xx/acp3x-es83xx.c create mode 100644 sound/soc/amd/acp/acp3x-es83xx/acp3x-es83xx.h -- 2.39.1

4 25

[PATCH] ALSA: usb-audio: Fix init call orders for UAC1
by Takashi Iwai 21 Aug '23

21 Aug '23

There have been reports of USB-audio driver spewing errors at the probe time on a few devices like Jabra and Logitech. The suggested fix there couldn't be applied as is, unfortunately, because it'll likely break other devices. But, the patch suggested an interesting point: looking at the current init code in stream.c, one may notice that it does initialize differently from the device setup in endpoint.c. Namely, for UAC1, we should call snd_usb_init_pitch() and snd_usb_init_sample_rate() after setting the interface, while the init sequence at parsing calls them before setting the interface blindly. This patch changes the init sequence at parsing for UAC1 (and other devices that need a similar behavior) to be aligned with the rest of the code, setting the interface at first. And, this fixes the long-standing problems on a few UAC1 devices like Jabra / Logitech, as reported, too. Reported-and-tested-by: Joakim Tjernlund <joakim.tjernlund(a)infinera.com> Closes: https://lore.kernel.org/r/202bbbc0f51522e8545783c4c5577d12a8e2d56d.camel@in… Cc: <stable(a)vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/usb/stream.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/sound/usb/stream.c b/sound/usb/stream.c index f10f4e6d3fb8..3d4add94e367 100644 --- a/sound/usb/stream.c +++ b/sound/usb/stream.c @@ -1093,6 +1093,7 @@ static int __snd_usb_parse_audio_interface(struct snd_usb_audio *chip, int i, altno, err, stream; struct audioformat *fp = NULL; struct snd_usb_power_domain *pd = NULL; + bool set_iface_first; int num, protocol; dev = chip->dev; @@ -1223,11 +1224,19 @@ static int __snd_usb_parse_audio_interface(struct snd_usb_audio *chip, return err; } + set_iface_first = false; + if (protocol == UAC_VERSION_1 || + (chip->quirk_flags & QUIRK_FLAG_SET_IFACE_FIRST)) + set_iface_first = true; + /* try to set the interface... */ usb_set_interface(chip->dev, iface_no, 0); + if (set_iface_first) + usb_set_interface(chip->dev, iface_no, altno); snd_usb_init_pitch(chip, fp); snd_usb_init_sample_rate(chip, fp, fp->rate_max); - usb_set_interface(chip->dev, iface_no, altno); + if (!set_iface_first) + usb_set_interface(chip->dev, iface_no, altno); } return 0; } -- 2.35.3

1 0

[PATCH][next] ALSA: hda/realtek: Fix spelling mistake "powe" -> "power"
by Colin Ian King 21 Aug '23

21 Aug '23

There is a spelling mistake in a quirk entry. Fix it. Signed-off-by: Colin Ian King <colin.i.king(a)gmail.com> --- sound/pci/hda/patch_realtek.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 6707db55f9c5..54e17791c6a8 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -9967,7 +9967,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x17aa, 0x3869, "Lenovo Yoga7 14IAL7", ALC287_FIXUP_YOGA9_14IAP7_BASS_SPK_PIN), SND_PCI_QUIRK(0x17aa, 0x387d, "Yoga S780-16 pro Quad AAC", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x17aa, 0x387e, "Yoga S780-16 pro Quad YC", ALC287_FIXUP_TAS2781_I2C), - SND_PCI_QUIRK(0x17aa, 0x3881, "YB9 dual powe mode2 YC", ALC287_FIXUP_TAS2781_I2C), + SND_PCI_QUIRK(0x17aa, 0x3881, "YB9 dual power mode2 YC", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x17aa, 0x3884, "Y780 YG DUAL", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x17aa, 0x3886, "Y780 VECO DUAL", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x17aa, 0x38a7, "Y780P AMD YG dual", ALC287_FIXUP_TAS2781_I2C), -- 2.39.2

2 1

How to determine which files were read/parsed
by GitHub issues - opened 21 Aug '23

21 Aug '23

alsa-project/alsa-lib issue #345 was opened from bmariuszb: I would like to know which files were read by alsa doing `alsactl restore` or `alsactl init`. So far I know that there is a folder `/usr/share/alsa` and it has `init/00main`, and this is probably read first, because when I deleted whole alsa dir I got error about this file, then I don't know what is read. Also there is a file `/var/lib/alsa/asound.state` which from my understanding contains values for different controls that are visible in alsamixer. There is also "user" config file in `/etc/asound.conf` and optionally any user can create `~/.asound.conf`. Am I missing any other files? getting the list of files in order they are read for each card or all cards would be helpfull in debugging. For example by default I found out that my card HDA Intel PCH has two devices ALC1150 Analog and ALC1150 Digital. Digital is probably one of the six jacks in the back of my PC that are unplugged so I don't care about them right now. However Analog in my PC is connected to speakers and also from motherboard comes another cable to front jack panel that I plugged my headphones. I have a few problems with the default configuration: 1. Headphone slider in alsamixer, known as control doesn't work. 2. There are too many controls in Playback that are useless in my opinion. 3. I haven't found yet how headphone and speakers jacks where merged into one Analog device. I wonder if it's a software merge or a hardware thing. I would like to combine card 2: NVidia [HDA NVidia], device 3: HDMI 0 [X2483_2481] with the merged card 1 device 0 (headphones and speakers) and I was wondering if it would be better to merge all 3 devices at once. Correct me if I'm wrong, but from my understanding you have to connect different devices to be able to dynamically switch between them, because normally every process, e.g browser connects to a pcm device at startup and you can't change it later. So the actual audio switching is unmuting one thing and muting rest of them. All "things" are connected to one pcm, right? Please correct me if I'm wrong or use wrong terminology. ![image](https://github.com/alsa-project/alsa-lib/assets/57236506/2d73bb21-e9ee-4172-8e8e-04785cf0a887) Issue URL : https://github.com/alsa-project/alsa-lib/issues/345 Repository URL: https://github.com/alsa-project/alsa-lib

1 0

[PATCH 1/2] ALSA: hda/tas2781: Fix acpi device refcount leak at tas2781_read_acpi()
by Takashi Iwai 20 Aug '23

20 Aug '23

The error path at tas2781_read_acpi() doesn't release the acpi_device adev but releases another device physdev instead. This results in a refcount leak. Fix it by replacing with the right object. Fixes: 5be27f1e3ec9 ("ALSA: hda/tas2781: Add tas2781 HDA driver") Reported-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Closes: https://lore.kernel.org/r/9f910785-e856-1539-e3e4-c9817af5fe67@linux.intel.… Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/pci/hda/tas2781_hda_i2c.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/hda/tas2781_hda_i2c.c b/sound/pci/hda/tas2781_hda_i2c.c index 35dafc4aec4f..0968ae915fd0 100644 --- a/sound/pci/hda/tas2781_hda_i2c.c +++ b/sound/pci/hda/tas2781_hda_i2c.c @@ -118,7 +118,7 @@ static int tas2781_read_acpi(struct tasdevice_priv *p, const char *hid) err: dev_err(p->dev, "read acpi error, ret: %d\n", ret); - put_device(physdev); + acpi_dev_put(adev); return ret; } -- 2.35.3

1 1