ASoC: Intel: avs: Allow for NHLT configuration override
Subsystems (3)
Cezary Rojewski
maintainercover ASoC: Intel: avs: Allow for NHLT configuration override
Small set of changes providing new feature which the driver is already utilizing on the market - for its Long-Term-Support (LTS) devices.
The goal is to cover systems which shipped with invalid Non HDAudio Link Table (NHLT) - not just the descriptors (headers), but cases where the hardware configuration is invalid too. The table is part of the ACPI tree and forcing BIOS updates is not a feasible solution. With the override, the topology file can carry the hardware configuration instead.
From AudioDSP perspective, only gateway-related modules e.g.: Copier care about the procedure. To ensure correct order of operations when initializing such modules, the overrides take precedence over what's currently there in the NHLT.
Cezary Rojewski (2):
ASoC: Intel: avs: Allow the topology to carry NHLT data
ASoC: Intel: avs: Honor NHLT override when setting up a path
include/uapi/sound/intel/avs/tokens.h | 6 ++ sound/soc/intel/avs/path.c | 13 ++- sound/soc/intel/avs/topology.c | 113 +++++++++++++++++++++++++- sound/soc/intel/avs/topology.h | 8 ++ 4 files changed, 133 insertions(+), 7 deletions(-)
Signature · 1 line
2.25.1
Cezary Rojewski
maintainerPATCH 1/2 ASoC: Intel: avs: Allow the topology to carry NHLT data
Typically the hardware configuration for I2S and DMIC devices resides in the Non-HDAudio Link Table (NHLT) that is part of the ACPI tree. As the NHLTs existing in the field are not always perfect, workaround mechanisms are provided to patch them.
Currently the avs-driver is utilizing the ->blob_fmt override (see topology.h and struct avs_tplg_modcfg_ext) when there is a valid entry within a NHLT to configure the hardware for specific format but its descriptor (header) is invalid.
A separate case is when there is no correct hardware configuration at all within the NHLT available in the system. Patching the header won't help and forcing ad-hoc BIOS updates for dated system is not feasible. Allowing the topology to carry the data is the solution of choice as replacing a userspace file that is part of /lib/firmware/intel/ is less invasive than BIOS update and solves the problem.
--- include/uapi/sound/intel/avs/tokens.h | 5 ++ sound/soc/intel/avs/topology.c | 106 +++++++++++++++++++++++++- sound/soc/intel/avs/topology.h | 7 ++ 3 files changed, 115 insertions(+), 3 deletions(-)
include/uapi/sound/intel/avs/tokens.h +5 -0
@@ -21,6 +21,7 @@ enum avs_tplg_token {AVS_TKN_MANIFEST_NUM_BINDINGS_U32 =8 ,AVS_TKN_MANIFEST_NUM_CONDPATH_TMPLS_U32 =9 ,AVS_TKN_MANIFEST_NUM_INIT_CONFIGS_U32 =10 , +AVS_TKN_MANIFEST_NUM_NHLT_CONFIGS_U32 =11 ,/* struct avs_tplg_library */ AVS_TKN_LIBRARY_ID_U32 = 101,
@@ -160,6 +161,10 @@ enum avs_tplg_token {AVS_TKN_INIT_CONFIG_ID_U32 =2401 ,AVS_TKN_INIT_CONFIG_PARAM_U8 =2402 ,AVS_TKN_INIT_CONFIG_LENGTH_U32 =2403 , + +/* struct avs_tplg_nhlt_config */ +AVS_TKN_NHLT_CONFIG_ID_U32 =2501 , + AVS_TKN_NHLT_CONFIG_SIZE_U32 = 2502, };#endif
sound/soc/intel/avs/topology.c +103 -3
156 lines not on this page · raw message
sound/soc/intel/avs/topology.h +7 -0
@@ -37,6 +37,8 @@ struct avs_tplg {u32 num_condpath_tmpls;struct avs_tplg_init_config *init_configs;u32 num_init_configs; +struct avs_tplg_nhlt_config *nhlt_configs; +u32 num_nhlt_configs;struct list_head path_tmpl_list; };
@@ -175,6 +177,11 @@ struct avs_tplg_init_config {void *data; }; +struct avs_tplg_nhlt_config { +u32 id; +struct acpi_nhlt_config *blob; +}; +struct avs_tplg_path {u32 id;
Signature · 1 line
2.25.1
Cezary Rojewski
maintainerPATCH 2/2 ASoC: Intel: avs: Honor NHLT override when setting up a path
In reply to Cezary RojewskiIn case topology provides NHLT configuration, use it instead of relying on the table in ACPI tree. Only gateway-related modules e.g.: Copier care about the process. For those the order of fetching for hardware configuration becomes:
1) check if NHLT override is set,
2) check if NHLT descriptor override is set,
3) use NHLT from ACPI directly
Such approach ensures no conflicts exist between 1) and 2) and that 1) always takes precedence.
--- include/uapi/sound/intel/avs/tokens.h | 1 + sound/soc/intel/avs/path.c | 13 +++++++++---- sound/soc/intel/avs/topology.c | 7 +++++++ sound/soc/intel/avs/topology.h | 1 + 4 files changed, 18 insertions(+), 4 deletions(-)
include/uapi/sound/intel/avs/tokens.h +1 -0
@@ -125,6 +125,7 @@ enum avs_tplg_token {AVS_TKN_MOD_KCONTROL_ID_U32 =1707 ,AVS_TKN_MOD_INIT_CONFIG_NUM_IDS_U32 =1708 ,AVS_TKN_MOD_INIT_CONFIG_ID_U32 =1709 , +AVS_TKN_MOD_NHLT_CONFIG_ID_U32 =1710 ,/* struct avs_tplg_path_template */ AVS_TKN_PATH_TMPL_ID_U32 = 1801,
sound/soc/intel/avs/path.c +9 -4
@@ -210,9 +210,11 @@ int avs_path_set_constraint(struct avs_dev *adev, struct avs_tplg_path_templatecontinue ; } - blob =avs_nhlt_config_or_default (adev, module_template); -if (IS_ERR (blob)) -continue ; +if (!module_template->nhlt_config) { + blob =avs_nhlt_config_or_default (adev, module_template); +if (IS_ERR (blob)) +continue ; + } rlist[i] = path_template->fe_fmt->sampling_freq; clist[i] = path_template->fe_fmt->num_channels;
@@ -382,7 +384,10 @@ static int avs_fill_gtw_config(struct avs_dev *adev, struct avs_copier_gtw_cfg *struct acpi_nhlt_config *blob;size_t gtw_size; - blob =avs_nhlt_config_or_default (adev, t); +if (t->nhlt_config) + blob = t->nhlt_config->blob; +else + blob =avs_nhlt_config_or_default (adev, t);if (IS_ERR (blob))return PTR_ERR (blob);
sound/soc/intel/avs/topology.c +7 -0
@@ -350,6 +350,7 @@ AVS_DEFINE_PTR_PARSER(modcfg_base, struct avs_tplg_modcfg_base, modcfgs_base); AVS_DEFINE_PTR_PARSER(modcfg_ext, struct avs_tplg_modcfg_ext, modcfgs_ext); AVS_DEFINE_PTR_PARSER(pplcfg, struct avs_tplg_pplcfg, pplcfgs); AVS_DEFINE_PTR_PARSER(binding, struct avs_tplg_binding, bindings); +AVS_DEFINE_PTR_PARSER(nhlt_config, struct avs_tplg_nhlt_config, nhlt_configs);static int parse_audio_format_bitfield (struct snd_soc_component *comp,void *elem,void *object,u32 offset)
@@ -1200,6 +1201,12 @@ static const struct avs_tplg_token_parser module_parsers[] = { .offset = offsetof(struct avs_tplg_module, num_config_ids), .parse = avs_parse_byte_token, }, + { + .token = AVS_TKN_MOD_NHLT_CONFIG_ID_U32, + .type = SND_SOC_TPLG_TUPLE_TYPE_WORD, + .offset = offsetof(struct avs_tplg_module, nhlt_config), + .parse = avs_parse_nhlt_config_ptr, + }, };static const struct avs_tplg_token_parser init_config_parsers[] = {
sound/soc/intel/avs/topology.h +1 -0
@@ -223,6 +223,7 @@ struct avs_tplg_module {u32 ctl_id;u32 num_config_ids;u32 *config_ids; +struct avs_tplg_nhlt_config *nhlt_config;struct avs_tplg_pipeline *owner;/* Pipeline modules management. */
Signature · 1 line
2.25.1
Mark Brown
maintainerQuoted 11 lines · Cezary Rojewski
Small set of changes providing new feature which the driver is already utilizing on the market - for its Long-Term-Support (LTS) devices.
The goal is to cover systems which shipped with invalid Non HDAudio Link Table (NHLT) - not just the descriptors (headers), but cases where the hardware configuration is invalid too. The table is part of the ACPI tree and forcing BIOS updates is not a feasible solution. With the override, the topology file can carry the hardware configuration instead.
[...]
Applied to
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-next
Thanks!
[1/2] ASoC: Intel: avs: Allow the topology to carry NHLT data
commit: dd9896d41fdf1050934d6a46a1c5ca2164284e72
[2/2] ASoC: Intel: avs: Honor NHLT override when setting up a path
commit: d5c8b7902a41625ea328b52c78ebe750fbf6fef7All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted.
You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed.
If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced.
Please add any relevant lists and maintainers to the CCs when replying to this mail.
Thanks,
Mark