- 29 Dec, 2023 38 commits
-
-
Geoffrey D. Bennett authored
Gen 2/3 devices have a single notification value for "input other" changes. Gen 4 has separate notification values for level, pad, air, and phantom power changes. Therefore, split the input_other_updated field and the scarlett2_update_input_other() function into the four components so that they can be handled separately later. Signed-off-by:
Geoffrey D. Bennett <g@b4.vu> Link: https://lore.kernel.org/r/a1a1d190659d56689792aa20ceeb53a6175171ad.1703444932.git.g@b4.vuSigned-off-by:
Takashi Iwai <tiwai@suse.de>
-
Geoffrey D. Bennett authored
db_scale_scarlett2_gain is the TLV for the output volume controls. Gen 4 has software-controllable input gain controls, so rename this to db_scale_scarlett2_volume so we can use that name for the inputs. Signed-off-by:
-
Geoffrey D. Bennett authored
Add a #define for SCARLETT2_MIX_MAX (max of mixer inputs * outputs) as that will be used again soon. Signed-off-by:
-
Geoffrey D. Bennett authored
Refactor scarlett2_usb_get_mix(), moving the scarlett2_mixer_values[] lookup into scarlett2_mixer_value_to_db(). Signed-off-by:
-
Geoffrey D. Bennett authored
Currently-supported interfaces with a mixer have per-channel volume controls, but this changes in Gen 4. Add a check so that the Playback Volume and associated controls don't get created unless the SCARLETT2_CONFIG_LINE_OUT_VOLUME config item is present. Signed-off-by:
-
Geoffrey D. Bennett authored
By splitting config set gen2 into gen2a/b (for 6i6/18i8 vs 18i20), and gen3b into gen3b/c (for 4i4/8i6 vs 18i8/18i20), we can use scarlett2_has_config_item() instead of the per-device line_out_hw_vol. As Gen 4 has a master volume control but no SW/HW switches, check for both SCARLETT2_CONFIG_MASTER_VOLUME and SCARLETT2_CONFIG_SW_HW_SWITCH as needed, even though for Gen 2 and Gen 3 the former implies the latter. Signed-off-by:
-
Geoffrey D. Bennett authored
Scarlett Gen 2 and Gen 3 devices combine volume and dim/mute notifications. The Scarlett 4i4 Gen 4 has volume change notification but no dim/mute control so split dim_mute_update out from vol_update. Signed-off-by:
-
Geoffrey D. Bennett authored
The struct scarlett2_usb_volume_status matched the config space layout of a few volume controls that could be read together and were in fixed locations between Gen 2 and Gen 3 devices. Gen 4 devices have removed, moved, and new related controls, so this needs to be cleaned up. By adding SCARLETT2_CONFIG_MASTER_VOLUME (the only config item that didn't already have its own entry, because it is read-only), we can remove: - struct scarlett2_usb_volume_state, - #define SCARLETT2_USB_VOLUME_STATUS_OFFSET, and - scarlett2_usb_get_volume_status() and replace with calls to scarlett2_usb_get_config(). Signed-off-by:
-
Geoffrey D. Bennett authored
Rather than looking up the analogue and mixer I/O counts repeatedly in info->port_count[SCARLETT2_PORT_TYPE_*][SCARLETT2_PORT_*], save those numbers in private variables. Signed-off-by:
-
Geoffrey D. Bennett authored
num_mux_srcs and num_mux_dsts will fit into a u8, so change the type. More similar counts are coming soon. Signed-off-by:
-
Geoffrey D. Bennett authored
The notification values were previously #define'd, and checked with a series of if() statements calling functions. Replace with an array of masks/callback function pointers, and a pointer to that array in the scarlett2_config_set definitions. Signed-off-by:
-
Geoffrey D. Bennett authored
Add missing blank line before comment. For consistency with other functions that have few parameters, move the parameters onto the same line as the function name. Signed-off-by:
-
Geoffrey D. Bennett authored
Use the new scarlett2_usb_activate_config() helper function rather than preparing the request manually and calling scarlett2_usb(). Signed-off-by:
-
Geoffrey D. Bennett authored
Pull out common code from scarlett2_usb_set_config() and create scarlett2_usb_set_data() and scarlett2_usb_activate_config(). Signed-off-by:
-
Geoffrey D. Bennett authored
Update scarlett2_usb_get_config() and scarlett2_usb_set_config() to make sure that the config_item_num is valid for the device. Signed-off-by:
-
Geoffrey D. Bennett authored
Replace array index into config sets with a pointer to a config set. Copy the config_set pointer to the scarlett2_data struct. This simplifies both the definition and use of the config sets. Signed-off-by:
-
Geoffrey D. Bennett authored
Add struct scarlett2_config_set so that data which is common to all devices in a config set can be stored there rather than in the model-specific data. Accordingly, rename scarlett2_config_items[] to scarlett2_config_sets[]. Signed-off-by:
-
Geoffrey D. Bennett authored
Currently the presence of a mixer is determined by checking if the device uses the GEN_3A config set. Add scarlett2_has_mixer() function which checks for the presence of mux_assignment entries instead. Signed-off-by:
-
Geoffrey D. Bennett authored
Allow for the phantom persistence config item to not exist. This is needed for the Scarlett Gen 4 series. Signed-off-by:
-
Geoffrey D. Bennett authored
Rather than assuming the standalone switch is present for all devices with a mixer, instead check for the presence of the SCARLETT2_CONFIG_STANDALONE_SWITCH config item. Signed-off-by:
-
Geoffrey D. Bennett authored
Rather than storing has_msd_mode in the per-device structure, infer this from the presence of the SCARLETT2_CONFIG_MSD_SWITCH entry in the device's configuration set. Signed-off-by:
-
Geoffrey D. Bennett authored
This commit removes the explicit integer assignments from the enums. The actual values matter little, and not assigning explicit values makes it easier to modify the longer lists in the future. Signed-off-by:
-
Geoffrey D. Bennett authored
Add ops.write to the hwdep interface. Once the upgrade firmware flash segment has been erased, writes to the hwdep fd are permitted, and translated to SCARLETT2_USB_WRITE_SEGMENT commands to the device. Signed-off-by:
-
Geoffrey D. Bennett authored
Add ioctls: - SCARLETT2_IOCTL_SELECT_FLASH_SEGMENT - SCARLETT2_IOCTL_ERASE_FLASH_SEGMENT - SCARLETT2_IOCTL_GET_ERASE_PROGRESS The settings or the firmware flash segment can be selected and then erased (asynchronous operation), and the erase progress can be monitored. If the erase progress is not monitored, then subsequent hwdep operations will block until the erase is complete. Once the erase is started, ALSA controls that communicate with the device will all return -EBUSY, and the device must be rebooted. Signed-off-by:
-
Geoffrey D. Bennett authored
Add skeleton hwdep/ioctl interface, beginning with SCARLETT2_IOCTL_PVERSION and SCARLETT2_IOCTL_REBOOT. Signed-off-by:
-
Geoffrey D. Bennett authored
Call SCARLETT2_USB_INFO_FLASH and SCARLETT2_USB_INFO_SEGMENT to find the App_Settings and App_Upgrade flash segment numbers, and store them in the scarlett2_data struct. These will be used later to implement reset to factory defaults and firmware upgrade functions. Signed-off-by:
-
Geoffrey D. Bennett authored
Add #defines for SCARLETT2_USB_* needed for firmware upgrade: reboot, info-flash, info-segment, erase-segment, get-erase, and write-segment. Signed-off-by:
-
Geoffrey D. Bennett authored
As scarlett2_meter_ctl_get() uses meter_level_map[], the data_mutex should be locked while accessing it. Signed-off-by:
-
Geoffrey D. Bennett authored
Ensure the value passed to scarlett2_mixer_ctl_put() is between 0 and SCARLETT2_MIXER_MAX_VALUE so we don't attempt to access outside scarlett2_mixer_values[]. Signed-off-by:
-
Geoffrey D. Bennett authored
The *_ctl_get() functions which call scarlett2_update_*() were not checking the return value. Fix to check the return value and pass to the caller. Signed-off-by:
-
Geoffrey D. Bennett authored
scarlett2_usb_set_config() calls scarlett2_usb_get() but was not checking the result. Return the error if it fails rather than continuing with an invalid value. Signed-off-by:
-
Geoffrey D. Bennett authored
scarlett2_config_save() was ignoring the return value from scarlett2_usb(). As this function is not called from user-space we can't return the error, so call usb_audio_err() instead. Signed-off-by:
-
Geoffrey D. Bennett authored
Update MAINTAINERS and "enabled" message with GitHub repository links. Signed-off-by:
-
Geoffrey D. Bennett authored
Add missing conversion from little-endian data to CPU-endian in scarlett2_usb_get_meter_levels(). Fixes: 3473185f ("ALSA: scarlett2: Remap Level Meter values") Signed-off-by:
-
Gergo Koteles authored
Remove sound controls in hda_unbind to make module loadable after module unload. Add a driver specific struct (tas2781_hda) to store the controls. This patch depends on patch: ALSA: hda/tas2781: do not use regcache Fixes: 5be27f1e ("ALSA: hda/tas2781: Add tas2781 HDA driver") CC: stable@vger.kernel.org Signed-off-by:
Gergo Koteles <soyer@irl.hu> Link: https://lore.kernel.org/r/362aa3e2f81b9259a3e5222f576bec5debfc5e88.1703204848.git.soyer@irl.huSigned-off-by:
-
Gergo Koteles authored
allow driver specific driver data in tas2781-hda-i2c and tas2781-i2c Fixes: ef3bcde7 ("ASoC: tas2781: Add tas2781 driver") CC: stable@vger.kernel.org Signed-off-by:
-
Gergo Koteles authored
Correct typos. Signed-off-by:
-
Gergo Koteles authored
There are two problems with using regcache in this module. The amplifier has 3 addressing levels (BOOK, PAGE, REG). The firmware contains blocks that must be written to BOOK 0x8C. The regcache doesn't know anything about BOOK, so regcache_sync writes invalid values to the actual BOOK. The module handles 2 or more separate amplifiers. The amplifiers have different register values, and the module uses only one regmap/regcache for all the amplifiers. The regcache_sync only writes the last amplifier used. The module successfully restores all the written register values (RC profile, program, configuration, calibration) without regcache. Remove regcache functions and set regmap cache_type to REGCACHE_NONE. Link: https://lore.kernel.org/r/21a183b5a08cb23b193af78d4b1114cc59419272.1701906455.git.soyer@irl.hu/ Fixes: 5be27f1e ("ALSA: hda/tas2781: Add tas2781 HDA driver") Acked-by:
Mark Brown <broonie@kernel.org> CC: stable@vger.kernel.org Signed-off-by:
-
- 21 Dec, 2023 2 commits
-
-
Stefan Binding authored
If CONFIG_SPI is not set in the kernel, there is no point in trying to set the chip selects. We can selectively compile it. Fixes: 8c4c216d ("ALSA: hda: cs35l41: Add config table to support many laptops without _DSD") Reported-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202312192256.lJelQEoZ-lkp@intel.com/Signed-off-by:
Stefan Binding <sbinding@opensource.cirrus.com> Link: https://lore.kernel.org/r/20231219162232.790358-3-sbinding@opensource.cirrus.comSigned-off-by:
-
Stefan Binding authored
Initialise the variables to NULL so that they cannot be uninitialised when devm_kfree is called. Found by static analysis. Fixes: 8c4c216d ("ALSA: hda: cs35l41: Add config table to support many laptops without _DSD") Signed-off-by:
-
