Takashi Iwai [Sun, 5 Jan 2020 14:47:16 +0000 (15:47 +0100)]
ALSA: Allow const arrays for legacy resource management helpers
Declare the arrays passed to the helper functions for legacy resources
(mostly for ISA drivers) as const, so that each caller can make its
static data as const for minor optimizations, too.
Takashi Iwai [Sun, 5 Jan 2020 08:11:19 +0000 (09:11 +0100)]
ALSA: hda/realtek - Apply mic mute LED quirk for Dell E7xx laptops, too
Dell E7xx laptops have also mic mute LED that is driven by the
dell-laptop platform driver. Bind it with the capture control as
already done for other models.
A caveat is that the fixup hook for the mic mute LED has to be applied
at last, otherwise it results in the invalid override of the callback.
Takashi Iwai [Sat, 4 Jan 2020 11:00:57 +0000 (12:00 +0100)]
ALSA: sh: Fix unused variable warnings
Remove unused variables that are left over after the conversion of new
PCM ops:
sound/sh/sh_dac_audio.c:166:26: warning: unused variable 'runtime'
sound/sh/sh_dac_audio.c:186:26: warning: unused variable 'runtime'
sound/sh/sh_dac_audio.c:205:26: warning: unused variable 'runtime'
Takashi Iwai [Sat, 4 Jan 2020 08:35:56 +0000 (09:35 +0100)]
ALSA: control: Add verification for kctl accesses
The current implementation of ALSA control API fully relies on the
callbacks of each driver, and there is no verification of the values
passed via API. This patch is an attempt to improve the situation
slightly by adding the validation code for the values stored via info
and get callbacks.
The patch adds a new kconfig, CONFIG_SND_CTL_VALIDATION. It depends
on CONFIG_SND_DEBUG and off as default since the validation would
require a slight overhead including the additional call of info
callback at each get callback invocation.
When this config is enabled, the values stored by each info callback
invocation are verified, namely:
- Whether the info type is valid
- Whether the number of enum items is non-zero
- Whether the given info count is within the allowed boundary
Similarly, the values stored at each get callback are verified as
well:
- Whether the values are within the given range
- Whether the values are aligned with the given step
- Whether any further changes are seen in the data array over the
given info count
The last point helps identifying a possibly invalid data type access,
typically a case where the info callback declares the type being
SNDRV_CTL_ELEM_TYPE_ENUMERATED while the get/put callbacks store
the values in value.integer.value[] array.
When a validation fails, the ALSA core logs an error message including
the device and the control ID, and the API call also returns an
error. So, with the new validation turned on, the driver behavior
difference may be visible on user-space, too -- it's intentional,
though, so that we can catch an error more clearly.
The patch also introduces a new ctl access type,
SNDRV_CTL_ELEM_ACCESS_SKIP_CHECK. A driver may pass this flag with
other access bits to indicate that the ctl element won't be verified.
It's useful when a driver code is specially written to access the data
greater than info->count size by some reason. For example, this flag
is actually set now in HD-audio HDMI codec driver which needs to clear
the data array in the case of the disconnected monitor.
Also, the PCM channel-map helper code is slightly modified to avoid
the false-positive hit by this validation code, too.
Takashi Iwai [Fri, 3 Jan 2020 08:16:49 +0000 (09:16 +0100)]
ALSA: drivers: Constify snd_kcontrol_new items
Most of snd_kcontrol_new definitions are read-only and passed as-is.
Let's declare them as const for further optimization.
Constify snd_kcontrol_new items
There should be no functional changes by this patch.
Takashi Iwai [Fri, 3 Jan 2020 08:16:45 +0000 (09:16 +0100)]
ALSA: seq: Constify struct snd_midi_op
Change the argument of snd_midi_process_event() to receive a const
snd_midi_op pointer and its callers respectively. This allows further
optimizations.
There should be no functional changes by this patch.
Takashi Iwai [Fri, 3 Jan 2020 08:16:38 +0000 (09:16 +0100)]
ALSA: ac97: Treat snd_ac97_bus_ops as const
This is a preliminary patch to allow const for snd_ac97_bus_ops
definitions in each driver's code. The ops reference is read-only,
hence it can be declared as const for further optimization.
There should be no functional changes by this patch.
Takashi Iwai [Fri, 3 Jan 2020 08:16:19 +0000 (09:16 +0100)]
ALSA: core: Treat snd_device_ops as const
This is a preliminary patch to allow const for snd_device_ops
definitions in each driver's code. The ops reference is read-only,
hence it can be declared as const for further optimization.
There should be no functional changes by this patch.
Takashi Iwai [Mon, 30 Dec 2019 21:27:42 +0000 (22:27 +0100)]
ALSA: uapi: Add linux/types.h include back (but carefully)
A few uapi/sound/*.h headers have been corrected for recovering from
the compile errors with the existing user-space code (alsa-lib) by the
recent commits. OTOH, these introduced another regression, as now
linux/types.h inclusion became mandatory for the uapi header checks.
As a compromise, this patch re-adds linux/types.h inclusions again,
but conditionally not to break other non-standard user-space stuff
again.
Fixes: 2e4688676392 ("ALSA: emu10k1: Make uapi/emu10k1.h compilable again") Fixes: d63e63d42107 ("ALSA: hdsp: Make uapi/hdsp.h compilable again") Fixes: 4fa406caf950 ("ALSA: hdspm: Drop linux/types.h inclusion in uapi header") Reported-by: kbuild test robot <lkp@intel.com> Link: https://lore.kernel.org/r/20191230212742.28925-1-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai@suse.de>
Chris Chiu [Mon, 30 Dec 2019 03:11:18 +0000 (11:11 +0800)]
ALSA: hda/realtek - Enable the bass speaker of ASUS UX431FLC
ASUS reported that there's an bass speaker in addition to internal
speaker and it uses DAC 0x02. It was not enabled in the commit 436e25505f34 ("ALSA: hda/realtek - Enable internal speaker of ASUS
UX431FLC") which only enables the amplifier and the front speaker.
This commit enables the bass speaker on top of the aforementioned
work to improve the acoustic experience.
Fixes: 436e25505f34 ("ALSA: hda/realtek - Enable internal speaker of ASUS UX431FLC") Signed-off-by: Chris Chiu <chiu@endlessm.com> Signed-off-by: Jian-Hong Pan <jian-hong@endlessm.com> Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/20191230031118.95076-1-chiu@endlessm.com Signed-off-by: Takashi Iwai <tiwai@suse.de>
Takashi Iwai [Wed, 18 Dec 2019 19:26:06 +0000 (20:26 +0100)]
ALSA: ice1724: Fix sleep-in-atomic in Infrasonic Quartet support code
Jia-Ju Bai reported a possible sleep-in-atomic scenario in the ice1724
driver with Infrasonic Quartet support code: namely, ice->set_rate
callback gets called inside ice->reg_lock spinlock, while the callback
in quartet.c holds ice->gpio_mutex.
This patch fixes the invalid call: it simply moves the calls of
ice->set_rate and ice->set_mclk callbacks outside the spinlock.
The Corsair Virtuoso RGB Wireless is a USB headset with a mic and a
sidetone feature. Label its mixer appropriately instead of all
"Headset", so that applications such as Pulseaudio don't just move
the sidetone control when they intend the main Headset control.
Takashi Iwai [Thu, 26 Dec 2019 10:03:53 +0000 (11:03 +0100)]
ALSA: control: Fix incompatible protocol error
The recent change to bump the ALSA control API protocol version from
2.0.7 to 2.1.0 caused a regression on user-space; while the user-space
expects both the major and the minor versions to be identical with the
supported numbers, we changed the minor number from 0 to 1.
For recovering from the incompatibility, this patch changes the
protocol version again to 2.0.8, which is compatible, but yet higher
than the original number 2.0.7, indicating that the protocol change.
Fixes: bd3eb4e87eb3 ("ALSA: ctl: bump protocol version up to v2.1.0") Reported-by: Paul Menzel <pmenzel@molgen.mpg.de> Tested-by: Paul Menzel <pmenzel@molgen.mpg.de> Link: https://lore.kernel.org/r/s5h1rsr769i.wl-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai@suse.de>
Takashi Sakamoto [Mon, 23 Dec 2019 09:33:47 +0000 (18:33 +0900)]
ALSA: ctl: allow TLV read operation for callback type of element in locked case
A design of ALSA control core allows applications to execute three
operations for TLV feature; read, write and command. Furthermore, it
allows driver developers to process the operations by two ways; allocated
array or callback function. In the former, read operation is just allowed,
thus developers uses the latter when device driver supports variety of
models or the target model is expected to dynamically change information
stored in TLV container.
The core also allows applications to lock any element so that the other
applications can't perform write operation to the element for element
value and TLV information. When the element is locked, write and command
operation for TLV information are prohibited as well as element value.
Any read operation should be allowed in the case.
At present, when an element has callback function for TLV information,
TLV read operation returns EPERM if the element is locked. On the
other hand, the read operation is success when an element has allocated
array for TLV information. In both cases, read operation is success for
element value expectedly.
This commit fixes the bug. This change can be backported to v4.14
kernel or later.
Takashi Iwai [Mon, 23 Dec 2019 22:18:16 +0000 (23:18 +0100)]
ALSA: hda - Apply sync-write workaround to old Intel platforms, too
Klaus Ethgen reported occasional high CPU usages in his system that
seem caused by HD-audio driver. The perf output revealed that it's
in the unsolicited event handling in the workqueue, and the problem
seems triggered by some communication stall between the controller and
the codec at the runtime or system resume.
Actually a similar phenomenon was seen in the past for other Intel
platforms, and we already applied the workaround to enforce sync-write
for CORB/RIRB verbs for Skylake and newer chipsets (commit 2756d9143aa5 "ALSA: hda - Fix intermittent CORB/RIRB stall on Intel
chips"). Fortunately, the same workaround is applicable to the old
chipset, and the experiment showed the positive effect.
Based on the experiment result, this patch enables the sync-write
workaround for all Intel chipsets. The only reason I hesitated to
apply this workaround was about the possibly slightly higher CPU usage.
But if the lack of sync causes a much severer problem even for quite
old chip, we should think this would be necessary for all Intel chips.
Takashi Sakamoto [Mon, 23 Dec 2019 02:39:21 +0000 (11:39 +0900)]
ALSA: ctl: bump protocol version up to v2.1.0
In a development period for v5.6 kernel, some changes are introduced to
structures in ALSA control interface:
- 'tstamp' member is removed from 'struct snd_ctl_elem_value
- 'TSTAMP' flag is removed from a set of access flags for 'struct
snd_ctl_elem_info'
- 'dimen' member is removed from 'struct snd_ctl_elem_info
Although these changes were introduced with enough consideration for
backward compatibility, they include slightly lose of it. This commit
bumps protocol version of ALSA control interface up to v2.1.0.
Takashi Sakamoto [Mon, 23 Dec 2019 02:39:20 +0000 (11:39 +0900)]
ALSA: ctl: remove dimen member from elem_info structure
The 'dimen' member of 'struct snd_ctl_elem_info' is designed to deliver
information to use an array of value as multi-dimensional values. This
feature is used just by echoaudio PCI driver, and fortunately it's not
used by the other applications than 'echomixer' in alsa-tools.
In a previous commit, usage of 'dimen' member is removed from echoaudio
PCI driver. Nowadays no driver/application use the feature.
Takashi Sakamoto [Mon, 23 Dec 2019 02:39:19 +0000 (11:39 +0900)]
ALSA: pci: echoaudio: remove usage of dimen menber of elem_value structure
In a couple of years ago, 'echomixer' userspace application was revised
not to use 'dimen' member of 'struct snd_ctl_elem_info'.
This commit removes usage of 'dimen' member from echoaudio PCI driver so
that no implementation uses the member.
Reference: 275353bb684e ("ALSA: echoaudio: purge contradictions between dimension matrix members and total number of members")
Reference: 51db452df07b ("Revert "ALSA: echoaudio: purge contradictions between dimension matrix members and total number of members") Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp> Link: https://lore.kernel.org/r/20191223023921.8151-3-o-takashi@sakamocchi.jp Signed-off-by: Takashi Iwai <tiwai@suse.de>
Takashi Sakamoto [Mon, 23 Dec 2019 02:39:18 +0000 (11:39 +0900)]
ALSA: ctl: remove unused macro for timestamping of elem_value
In a former commit, 'tstamp' member was removed from 'struct
snd_ctl_elem_value' in a middle way toward solution of Y2038 issue. In a
protocol of ALSA control interface, this member is designed to deliver
timestamp information in the value structure when the target element
supports SNDRV_CTL_ELEM_ACCESS_TIMESTAMP flag.
Actually, the feature is neither used by kernel space nor user space,
especiall alsa-lib has no API for the feature. Therefore it's reasonable
to remove both of them. Practically, the timestamp information
corresponds to no information about type of clock ID. It can bring
confusions to applications.
Alex Deucher [Sat, 21 Dec 2019 00:17:02 +0000 (19:17 -0500)]
ALSA: hda/hdmi - fix atpx_present when CLASS is not VGA
You can't use PCI_BASE_CLASS with pci_get_class(). This
happens to work by luck on devices with PCI_CLASS_DISPLAY_VGA, but
misses PCI_CLASS_DISPLAY_OTHER. Add a check for those as well.
Takashi Iwai [Fri, 20 Dec 2019 16:15:55 +0000 (17:15 +0100)]
ALSA: uapi: Drop unneeded typedefs
We kept some typedefs in uapi/sound/*.h so that the programs in
alsa-tools can be built. Now that alsa-lib takes these and applies
the workarounds in its own, we don't need these typedefs any longer in
the kernel uapi side. Let's drop them.
Takashi Iwai [Fri, 20 Dec 2019 15:34:15 +0000 (16:34 +0100)]
ALSA: uapi: Drop asound.h inclusion from asoc.h
The asound.h isn't always available while asoc.h itself is distributed
in alsa-lib package. So we need to avoid the unnecessary inclusion of
asound.h from there.
Takashi Iwai [Fri, 20 Dec 2019 15:34:14 +0000 (16:34 +0100)]
ALSA: uapi: Fix typos and header inclusion in asound.h
The recent changes in uapi/asoundlib.h caused some build errors in
alsa-lib side because of a typo and the new included files.
Basically asound.h is supposed to be usable also on non-Linux systems,
so we've tried to avoid the Linux-specific include files.
This patch is an attempt to recover from those changes.
Fixes: 3ddee7f88aaf ("ALSA: Avoid using timespec for struct snd_pcm_status") Fixes: 80fe7430c708 ("ALSA: add new 32-bit layout for snd_pcm_mmap_status/control") Link: https://lore.kernel.org/r/20191220153415.2740-5-tiwai@suse.de Signed-off-by: Takashi Iwai <tiwai@suse.de>
Takashi Iwai [Fri, 20 Dec 2019 15:34:13 +0000 (16:34 +0100)]
ALSA: hdspm: Drop linux/types.h inclusion in uapi header
The hdspm.h uapi header has been used also from non-Linux or platforms
that don't have linux/*.h. It was OK in the past because alsa-lib
contained the modified version of this header file, but now it tries
to the verbatim copy, so it broke the build. This fixes it again.
Takashi Iwai [Fri, 20 Dec 2019 15:34:12 +0000 (16:34 +0100)]
ALSA: hdsp: Make uapi/hdsp.h compilable again
Recently alsa-lib updated its content of sound/hdsp.h just by copying
the latest Linus kernel uapi/*.h, and this broke the build of
alsa-tools programs. We used to modify the headers so that they can
be built without asoundlib.h and linux kernel headers, and the
verbatim copy doesn't work as is.
This patch removes again the linux/types.h inclusion and drop __user
prefix that broke the build and adjusts the corresponding code.
Takashi Iwai [Fri, 20 Dec 2019 15:34:11 +0000 (16:34 +0100)]
ALSA: emu10k1: Make uapi/emu10k1.h compilable again
Recently we updated the content in alsa-lib uapi header files by just
copying from the latest Linus kernel uapi/*.h, and noticed that it
broke the build of some alsa-tools programs. The reason is that we
used to have a modified version in the past, so that the program can
be built without referring to the unexported stuff like
snd_ctl_elem_id or __user prefix.
This patch attempts to restore that, i.e. dropping the stuff that
can't be referred in the user-space. For adapting the changes in
uapi/emu10k1.h, the emu10k1 driver code is also slightly modified.
Most of changes are pointer cast.
Make sure to check the return value of usb_altnum_to_altsetting() to
avoid dereferencing a NULL pointer when the requested alternate settings
is missing.
The format altsetting number may come from a quirk table and there does
not seem to be any other validation of it (the corresponding index is
checked however).