- 23 Sep, 2023 1 commit
-
-
Kees Cook authored
Prepare for the coming implementation by GCC and Clang of the __counted_by attribute. Flexible array members annotated with __counted_by can have their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS (for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family functions). As found with Coccinelle[1], add __counted_by for struct hda_conn_list. [1] https://github.com/kees/kernel-tools/blob/trunk/coccinelle/examples/counted_by.cocci Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.com> Cc: Cezary Rojewski <cezary.rojewski@intel.com> Cc: alsa-devel@alsa-project.org Signed-off-by:
Kees Cook <keescook@chromium.org> Link: https://lore.kernel.org/r/20230922175042.work.547-kees@kernel.orgSigned-off-by:
Takashi Iwai <tiwai@suse.de>
-
- 22 Sep, 2023 4 commits
-
-
Stefan Binding authored
When the CS35L41 amp is requested to mute using the ACPI notification mechanism, userspace is not notified that the amp is muted. To allow userspace to know about the mute, add an ALSA control which tracks the forced mute override. This control does not track the overall mute state of the amp, since the amp is only unmuted during playback anyway, instead it tracks the mute override request from the ACPI notification. Signed-off-by:
Stefan Binding <sbinding@opensource.cirrus.com> Link: https://lore.kernel.org/r/20230921162849.1988124-5-sbinding@opensource.cirrus.comSigned-off-by:
-
Stefan Binding authored
Some laptops require a hardware based mute system, where when a hotkey is pressed, it forces the amp to be muted. For CS35L41, when the hotkey is pressed, an acpi notification is sent to the CS35L41 Device Node. The driver needs to handle this notification and call a _DSM function to retrieve the mute state. Since the amp is only muted during playback, the driver will only mute or unmute if playback is occurring, otherwise it will save the mute state for when playback starts. This uses the ACPI Notification mechanism, where a handler has been registered in the component master, which notifies each amp through the component binding. Signed-off-by:
-
Stefan Binding authored
For systems which have support for ACPI notifications, add a mechanism to register a handler for ACPI notifications and then call the acpi_notify api on the bound components. Registering a handler in the Realtek HDA driver, allows a single handler to be registered, which then calls into all the components, rather than attempting to register the same handler multiple times, once for each component. Signed-off-by:
-
Stefan Binding authored
Some systems support a notification from ACPI, which can be used for different things. Only one handler can be registered for the acpi notification, but all amps need to receive that notification, we can register a single handler inside the component master, so that it can then notify through the component framework. This is required to support mute notifications from ACPI. Signed-off-by:
-
- 20 Sep, 2023 2 commits
-
-
Nícolas F. R. A. Prado authored
When parsing the configs, keep track of card configurations that match the current system but haven't matched any card, and report those as test failures as they represent that a card which was expected to be present on the system is missing. This allows the configuration files to not only be used to detect missing PCM devices (which is currently possible) but also that the soundcard hasn't been registered at all. Signed-off-by:
Nícolas F. R. A. Prado <nfraprado@collabora.com> Reviewed-by:
Jaroslav Kysela <perex@perex.cz> Reviewed-by:
Mark Brown <broonie@kernel.org> Link: https://lore.kernel.org/r/20230919152702.100617-1-nfraprado@collabora.comSigned-off-by:
-
Richard Fitzgerald authored
The KUnit test for cirrus_scodec uses GPIO library functions so select GPIOLIB in Kconfig. This fixes the ld failures on builds that didn't already select GPIOLIB. ld: vmlinux.o: in function `cirrus_scodec_test_gpio_get': sound/pci/hda/cirrus_scodec_test.c:40: undefined reference to `gpiochip_get_data' ld: vmlinux.o: in function `cirrus_scodec_test_gpio_probe': sound/pci/hda/cirrus_scodec_test.c:94: undefined reference to `gpiochip_generic_request' ld: sound/pci/hda/cirrus_scodec_test.c:94: undefined reference to `gpiochip_generic_free' ld: sound/pci/hda/cirrus_scodec_test.c:95: undefined reference to `devm_gpiochip_add_data_with_key' Signed-off-by:
Richard Fitzgerald <rf@opensource.cirrus.com> Reported-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202309201646.NnjfKPWk-lkp@intel.com/ Fixes: 2144833e ("ALSA: hda: cirrus_scodec: Add KUnit test") Link: https://lore.kernel.org/r/20230920090338.29345-1-rf@opensource.cirrus.comSigned-off-by:
-
- 18 Sep, 2023 6 commits
-
-
Richard Fitzgerald authored
Add a KUnit test for cirrus_scodec_get_speaker_id(). It is impractical to have enough hardware with every possible permutation of speaker id. So use a test harness to test all theoretically supported options. The test harness consists of: - a mock GPIO controller. - a mock struct device to represent the scodec driver - software nodes to provide the fwnode info that would normally come from ACPI. Signed-off-by:
-
Richard Fitzgerald authored
Add handling of the "spk-id-gpios" _DSD property. If present, the value indicated by the GPIOs is appended to the subsystem-id part of the firmware name to load the appropriate tunings for that speaker. Some manufacturers use multiple sources of speakers, which need different tunings for best performance. On these models the type of speaker fitted is indicated by the values of one or more GPIOs. The number formed by the GPIOs identifies the tuning required. The speaker ID is only used in combination with a _SUB identifier because the value is only meaningful if the exact model is known. The code to get the speaker ID value has been implemented as a new library so that the cs35l41_hda driver can be switched in future to share common code. This library can be extended for other common functionality shared by Cirrus Logic amp drivers. Signed-off-by:
-
Geoffrey D. Bennett authored
This driver was originally developed for the Focusrite Scarlett Gen 2 series, but now also supports the Scarlett Gen 3 series, the Clarett 8Pre USB, and the Clarett+ 8Pre. The messages output by the driver on initialisation and error include the identifying text "Scarlett Gen 2/3", but this is no longer accurate, and writing "Scarlett Gen 2/3/Clarett USB/Clarett+" would be unwieldy. Add series_name field to the scarlett2_device_entry struct so that concise and accurate messages can be output. Signed-off-by:
Geoffrey D. Bennett <g@b4.vu> Link: https://lore.kernel.org/r/3774b9d35bf1fbdd6fdad9f3f4f97e9b82ac76bf.1694705811.git.g@b4.vuSigned-off-by:
-
Geoffrey D. Bennett authored
The Clarett 8Pre USB works the same as the Clarett+ 8Pre, only the USB ID is different. Tested-by:
Philippe Perrot <philippe@perrot-net.fr> Signed-off-by:
-
Geoffrey D. Bennett authored
By moving the USB IDs from the device_info struct into scarlett2_devices[], that will allow for devices with different USB IDs to share the same device_info. Tested-by:
-
Geoffrey D. Bennett authored
Early versions of this mixer driver did not work on all hardware, so out of caution the driver was disabled by default and had to be explicitly enabled with device_setup=1. Since commit 764fa6e6 ("ALSA: usb-audio: scarlett2: Fix device hang with ehci-pci") no more problems of this nature have been reported. Therefore, enable the driver by default but provide a new device_setup option to disable the driver in case that is needed. - device_setup value of 0 now means "enable" rather than "disable". - device_setup value of 1 is now ignored. - device_setup value of 4 now means "disable". Signed-off-by:
-
- 16 Sep, 2023 1 commit
-
-
Takashi Iwai authored
At the implementation of the optional proc fs in rawmidi, I forgot that rmidi->ops itself is optional and can be NULL. Add the proper NULL check for avoiding the Oops. Fixes: fa030f66 ("ALSA: ump: Additional proc output") Reported-and-tested-by:
Mark Hills <mark@xwax.org> Closes: https://lore.kernel.org/r/ef9118c3-a2eb-d0ff-1efa-cc5fb6416bde@xwax.org Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/20230916060725.11726-1-tiwai@suse.deSigned-off-by:
-
- 15 Sep, 2023 18 commits
-
-
Takashi Iwai authored
The filling of card->longname can be gracefully truncated, as it's only informative. Use scnprintf() and suppress the superfluous compile warning with -Wformat-truncation. Link: https://lore.kernel.org/r/20230915091313.5988-4-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The filling of card->longname can be gracefully truncated, as it's only informative. Use scnprintf() and suppress the superfluous compile warning with -Wformat-truncation. Link: https://lore.kernel.org/r/20230915091313.5988-3-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The filling of card->longname can be gracefully truncated, as it's only informative. Use scnprintf() and suppress the superfluous compile warning with -Wformat-truncation. Link: https://lore.kernel.org/r/20230915091313.5988-2-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
add_control_with_pfx() constructs a mixer name element with the fixed size, and it got compile warnings with -Wformat-truncation. Although the size overflow is very unlikely, let's have a sanity check of the string size and returns the error if it really doesn't fit instead of silent truncation. Link: https://lore.kernel.org/r/20230915082802.28684-14-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
CMIPCI driver got compile warnings with -Wformat-truncation at a couple of plain sprintf() usages. Use scnprintf() for filling the longname string for avoiding the warnings. Link: https://lore.kernel.org/r/20230915082802.28684-13-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The compile warnings at filling MIDI stream name strings are all false-positive; the number of streams can't go so high. For suppressing the warning, replace snprintf() with scnprintf(). As stated in the above, truncation doesn't matter. Link: https://lore.kernel.org/r/20230915082802.28684-12-tiwai@suse.deReviewed-by:
Takashi Sakamoto <o-takashi@sakamocchi.jp> Tested-by:
-
Takashi Iwai authored
The filling of card->longname can be gracefully truncated, as it's only informative. Use scnprintf() and suppress the superfluous compile warning with -Wformat-truncation. Link: https://lore.kernel.org/r/20230915082802.28684-11-tiwai@suse.deReviewed-by:
-
Takashi Iwai authored
The compile warning with -Wformat-truncation at xen_snd_front_cfg_card() is false-positive; the loop can be only for SNDRV_PCM_DEVICES which is at most 32. For suppressing the warning, replace snprintf() with scnprintf(). As stated in the above, truncation doesn't matter. Link: https://lore.kernel.org/r/20230915082802.28684-10-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The filling of card->longname can be gracefully truncated, as it's only informative. Use scnprintf() and suppress the superfluous compile warning with -Wformat-truncation. Link: https://lore.kernel.org/r/20230915082802.28684-9-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The filling of card->longname can be gracefully truncated, as it's only informative. Use scnprintf() and suppress the superfluous compile warning with -Wformat-truncation. Link: https://lore.kernel.org/r/20230915082802.28684-8-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The filling of card->longname can be gracefully truncated, as it's only informative. Use scnprintf() and suppress the superfluous compile warning with -Wformat-truncation. Link: https://lore.kernel.org/r/20230915082802.28684-7-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The warning with -Wformat-truncation at sscape_upload_microcode() is false-positive; the version number can be only a single digit, hence fitting with the given string size. For suppressing the warning, replace snprintf() with scnprintf(). As stated in the above, truncation doesn't matter. Link: https://lore.kernel.org/r/20230915082802.28684-6-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The filling of card->longname can be gracefully truncated, as it's only informative. Use scnprintf() and suppress the superfluous compile warning with -Wformat-truncation. Link: https://lore.kernel.org/r/20230915082802.28684-5-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The recent enablement of -Wformat-truncation leads to a false-positive warning for mixer_scarlett_gen2.c. For suppressing the warning, replace snprintf() with scnprintf(). As stated in the above, truncation doesn't matter. Link: https://lore.kernel.org/r/20230915082802.28684-4-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The compile warnings with -Wformat-truncation appearing at snd_seq_midisynth_probe() in seq_midi.c are false-positive; those must fit within the given string size. For suppressing the warning, replace snprintf() with scnprintf(). As stated in the above, truncation doesn't matter. Link: https://lore.kernel.org/r/20230915082802.28684-3-tiwai@suse.deSigned-off-by:
-
Takashi Iwai authored
The filling of a port name string got a warning with W=1 due to the potentially too long group name. Add the string precision to limit the size. Fixes: 81fd444a ("ALSA: seq: Bind UMP device") Link: https://lore.kernel.org/r/20230915082802.28684-2-tiwai@suse.deSigned-off-by:
-
Richard Fitzgerald authored
In cs35l56_hda_read_acpi() do not return if ACPI _SUB is missing. A missing _SUB means that the driver cannot load a system-specific firmware, because the firmware is identified by the _SUB. But it can fallback to a generic firmware. Unfortunately this was being handled by immediately returning 0, which would skip the remaining ACPI configuration in cs35l56_hda_read_acpi() and so it would not get the RESET GPIO. Signed-off-by:
-
Kailang Yang authored
Lenovo ThinkCentre M70q had boot up pop noise. Disable power save will solve pop issue. Signed-off-by:
Kailang Yang <kailang@realtek.com> Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/315900e2efef42fd9855eacfeb443abd@realtek.comSigned-off-by:
-
- 14 Sep, 2023 1 commit
-
-
Richard Fitzgerald authored
The final return in cs35l56_hda_posture_get() was returning the value of 'ret', but ret is always zero at this point. So this can be a simple 'return 0'. Signed-off-by:
-
- 13 Sep, 2023 4 commits
-
-
Peter Ujfalusi authored
The kctl->id.name can be directly passed to snd_usb_copy_string_desc() and if the string has been fetched the suffix can be appended with the append_ctl_name() call. The temporary name string becomes redundant and can be removed. This change will also fixes the following compiler warning/error (W=1): sound/usb/mixer.c: In function ‘parse_audio_unit’: sound/usb/mixer.c:1972:29: error: ‘ Validity’ directive output may be truncated writing 9 bytes into a region of size between 1 and 44 [-Werror=format-truncation=] 1972 | "%s Validity", name); | ^~~~~~~~~ In function ‘parse_clock_source_unit’, inlined from ‘parse_audio_unit’ at sound/usb/mixer.c:2892:10: sound/usb/mixer.c:1971:17: note: ‘snprintf’ output between 10 and 53 bytes into a destination of size 44 1971 | snprintf(kctl->id.name, sizeof(kctl->id.name), | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1972 | "%s Validity", name); | ~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors The warnings got brought to light by a recent patch upstream: commit 6d4ab2e9 ("extrawarn: enable format and stringop overflow warnings in W=1") Signed-off-by:Peter Ujfalusi <peter.ujfalusi@linux.intel.com> Link: https://lore.kernel.org/r/20230913093933.24564-1-peter.ujfalusi@linux.intel.comSigned-off-by:
-
Knyazev Arseniy authored
UX3402VA and UX3402ZA models require different hex values, so comibining them into one model is incorrect. Fixes: 491a4ccd ("ALSA: hda/realtek: Add quirk for ASUS Zenbook using CS35L41") Signed-off-by:
Knyazev Arseniy <poseaydone@ya.ru> Link: https://lore.kernel.org/r/20230913053343.119798-1-poseaydone@ya.ruSigned-off-by:
-
Peter Ujfalusi authored
Use consistently u8 for sdw link index. The id is limited to 4, u8 is adequate in size to store it. This change will also fixes the following compiler warning/error (W=1): sound/hda/intel-sdw-acpi.c: In function ‘sdw_intel_acpi_scan’: sound/hda/intel-sdw-acpi.c:34:35: error: ‘-subproperties’ directive output may be truncated writing 14 bytes into a region of size between 7 and 17 [-Werror=format-truncation=] 34 | "mipi-sdw-link-%d-subproperties", i); | ^~~~~~~~~~~~~~ In function ‘is_link_enabled’, inlined from ‘sdw_intel_scan_controller’ at sound/hda/intel-sdw-acpi.c:106:8, inlined from ‘sdw_intel_acpi_scan’ at sound/hda/intel-sdw-acpi.c:180:9: sound/hda/intel-sdw-acpi.c:33:9: note: ‘snprintf’ output between 30 and 40 bytes into a destination of size 32 33 | snprintf(name, sizeof(name), | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~ 34 | "mipi-sdw-link-%d-subproperties", i); | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors The warnings got brought to light by a recent patch upstream: commit 6d4ab2e9 ("extrawarn: enable format and stringop overflow warnings in W=1") Signed-off-by:Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Link: https://lore.kernel.org/r/20230912162617.29178-1-peter.ujfalusi@linux.intel.comSigned-off-by:
-
Richard Fitzgerald authored
Do not allow the CS35L56 to be put into its lowest power "hibernation" mode. This only affects I2C because "hibernation" is already disabled on SPI. Recent firmwares need a different wake-up sequence. Until that sequence has been specified, the chip "hibernation" mode must be disabled otherwise it can intermittently fail to wake. Signed-off-by:
-
- 12 Sep, 2023 3 commits
-
-
Peter Ujfalusi authored
There is no need to use temporary string for the debugfs directory name as we can use the device name of the card. This change will also fixes the following compiler warning/error (W=1): sound/core/init.c: In function ‘snd_card_init’: sound/core/init.c:367:28: error: ‘%d’ directive writing between 1 and 10 bytes into a region of size 4 [-Werror=format-overflow=] 367 | sprintf(name, "card%d", idx); | ^~ sound/core/init.c:367:23: note: directive argument in the range [0, 2147483646] 367 | sprintf(name, "card%d", idx); | ^~~~~~~~ sound/core/init.c:367:9: note: ‘sprintf’ output between 6 and 15 bytes into a destination of size 8 367 | sprintf(name, "card%d", idx); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors The idx is guarantied to be less than SNDRV_CARDS (max 256 or 8) by the code in snd_card_init(), however the compiler does not see that. The warnings got brought to light by a recent patch upstream: commit 6d4ab2e9 ("extrawarn: enable format and stringop overflow warnings in W=1") Suggested-by:Arnd Bergmann <arnd@arndb.de> Suggested-by:
-
Kailang Yang authored
If system has two speakers and one connect to 0x14 pin, use this function will disable it. Fixes: e43252db ("ALSA: hda/realtek - ALC287 I2S speaker platform support") Signed-off-by:
-
Takashi Iwai authored
ALSA sequencer core still delivers events to the disabled UMP Group, leaving this handling to the device. But it's rather risky and it's easy to imagine that such an unexpected event may screw up the device firmware. This patch avoids the superfluous event deliveries by setting the group_filter of the UMP client as default, and evaluate the group_filter properly at delivery from non-UMP clients. The grouop_filter is updated upon the dynamic UMP Function Block updates, so that it follows the change of the disabled UMP Groups, too. Fixes: d2b70607 ("ALSA: seq: Add UMP group filter") Link: https://lore.kernel.org/r/20230912085144.32534-1-tiwai@suse.deSigned-off-by:
-
