platform/kernel/linux-starfive.git
5 years agoALSA: firewire: use managed-resource of fw unit device for private data
Takashi Sakamoto [Tue, 2 Oct 2018 23:21:50 +0000 (08:21 +0900)]
ALSA: firewire: use managed-resource of fw unit device for private data

At present, private data of each driver in ALSA firewire stack is
allocated/freed by kernel slab allocator for corresponding unit on
IEEE 1394 bus. In this case, resource-managed slab allocator is
available to release memory object automatically just before releasing
device structure for the unit. This idea can prevent runtime from
memory leak due to programming mistakes.

This commit uses the allocator for the private data. These drivers
already use reference counter to maintain lifetime of device structure
for the unit by a pair of fw_unit_get()/fw_unit_put(). The private data
is safely released in a callback of 'struct snd_card.private_free().

Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoMerge branch 'for-linus' into for-next
Takashi Iwai [Thu, 4 Oct 2018 05:52:58 +0000 (07:52 +0200)]
Merge branch 'for-linus' into for-next

Back-merge 4.19-devel branch into 4.20 for applying FireWire patches
cleanly.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/realtek - Cannot adjust speaker's volume on Dell XPS 27 7760
Kai-Heng Feng [Thu, 4 Oct 2018 03:39:42 +0000 (11:39 +0800)]
ALSA: hda/realtek - Cannot adjust speaker's volume on Dell XPS 27 7760

The issue is the same as commit dd9aa335c880 ("ALSA: hda/realtek - Can't
adjust speaker's volume on a Dell AIO"), the output requires to connect
to a node with Amp-out capability.

Applying the same fixup ALC298_FIXUP_SPK_VOLUME can fix the issue.

BugLink: https://bugs.launchpad.net/bugs/1775068
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: intel8x0: Fix fall-through annotations
Gustavo A. R. Silva [Wed, 3 Oct 2018 10:38:36 +0000 (12:38 +0200)]
ALSA: intel8x0: Fix fall-through annotations

Replace "fallthru" with a proper "fall through" annotation.

This fix is part of the ongoing efforts to enabling
-Wimplicit-fallthrough

Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: sb8: Fix fall-through annotations
Gustavo A. R. Silva [Tue, 2 Oct 2018 09:56:36 +0000 (11:56 +0200)]
ALSA: sb8: Fix fall-through annotations

Replace "fallthru" with a proper "fall through" annotation.

This fix is part of the ongoing efforts to enabling
-Wimplicit-fallthrough

Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: firewire-lib: add PCM rules to obsolete PCM constraints based on LCM of SYT_INT...
Takashi Sakamoto [Sun, 30 Sep 2018 19:11:49 +0000 (04:11 +0900)]
ALSA: firewire-lib: add PCM rules to obsolete PCM constraints based on LCM of SYT_INTERVAL

In blocking mode of IEC 61883-1/6, when one isochronous packet includes
data for events, the data is for the same number of events as
SYT_INTERVAL decided according to sampling transmission frequency (SFC).

IEC 61883-1/6 engine of ALSA firewire stack applies constraints of
period and buffer size of PCM intermediate buffer of PCM substream.
At present, this constraint is designed to round the size up/down to
32 frames. This value comes from the least common multiple (LCM) of
SYT_INTERVAL. Although this looks to work well, in lower sampling
rate, applications are not allowed to set size of period quite near
period time constraint (at present 5 msec per period).

This commit adds PCM rules for period/buffer size and rate to obsoletes
the constraints based on LCM.

Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add ZxR exit commands
Connor McAdams [Sun, 30 Sep 2018 03:03:26 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Add ZxR exit commands

This patch adds exit operations for the Sound Blaster ZxR.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add ZxR 600 ohm gain control
Connor McAdams [Sun, 30 Sep 2018 03:03:25 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Add ZxR 600 ohm gain control

This patch adds a control for 600 ohm gain on the Sound Blaster ZxR.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Remove input select enum for ZxR
Connor McAdams [Sun, 30 Sep 2018 03:03:24 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Remove input select enum for ZxR

This patch removes the input select control for the ZxR, as it only has
one input option, rear microphone.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add ZxR input/output select commands
Connor McAdams [Sun, 30 Sep 2018 03:03:23 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Add ZxR input/output select commands

This patch adds commands for selecting input and output on the Sound
Blaster ZxR. The ZxR has no front panel header, and has line-in on the
separate daughter board, so it only does rear-mic.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add ZxR DSP post-download commands
Connor McAdams [Sun, 30 Sep 2018 03:03:22 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Add ZxR DSP post-download commands

This patch adds commands for setting up the ZxR after the DSP is
downloaded. The ZxR already shares most of the post-download commands
from the regular Sound Blaster Z.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add ZxR init commands
Connor McAdams [Sun, 30 Sep 2018 03:03:21 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Add ZxR init commands

This patch adds init commands for the main Sound Blaster ZxR card.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add DBpro hda_codec_ops
Connor McAdams [Sun, 30 Sep 2018 03:03:20 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Add DBpro hda_codec_ops

This patch adds separate hda_codec_ops for the DBPro daughter board, as
it behaves more like a generic HDA codec than the other ca0132 cards,
despite having a ca0132 on board.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add ZxR pincfg
Connor McAdams [Sun, 30 Sep 2018 03:03:19 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Add ZxR pincfg

This patch adds a pincfg for the ZxR, and defines which pins are used
for both.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add ZxR quirks + new quirk check function
Connor McAdams [Sun, 30 Sep 2018 03:03:18 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Add ZxR quirks + new quirk check function

This patch adds quirk ID's for the ZxR and it's daughter board, the
DBPro. It also adds a function for determining the quirk for each board
through HDA subsytem ID's instead of PCI subsystem ID's.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Fix surround sound with output effects
Connor McAdams [Sun, 30 Sep 2018 03:03:17 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Fix surround sound with output effects

This patch fixes an issue where if surround sound was the selected
output and output effects were enabled, the sound wasn't sent to all
channels correctly.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Fix AE-5 control type
Connor McAdams [Sun, 30 Sep 2018 03:03:16 +0000 (23:03 -0400)]
ALSA: hda/ca0132 - Fix AE-5 control type

This patch corrects the control type of the additional AE-5 controls
added in a previous patch from HDA_INPUT to HDA_OUTPUT.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: timer: fix wrong comment to refer to 'SNDRV_TIMER_PSFLG_*'
Takashi Sakamoto [Sat, 29 Sep 2018 14:20:51 +0000 (23:20 +0900)]
ALSA: timer: fix wrong comment to refer to 'SNDRV_TIMER_PSFLG_*'

ALSA timer core has a comment referring to 'SNDRV_MIXER_PSFLG_*' in
a definition of 'struct snd_timer_params' of UAPI header. I can see
this in initial state of ALSA timer core, at least in
'alsa-driver-0.4.0.tar.gz'.

This commit fixes the comment.

Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda: Fix mismatch for register mask and value in ext controller.
Keyon Jie [Fri, 28 Sep 2018 09:38:59 +0000 (17:38 +0800)]
ALSA: hda: Fix mismatch for register mask and value in ext controller.

E.g. for snd_hdac_ext_bus_link_power_up(), we should set mask to be
AZX_MLCTL_SPA(it was 0), and AZX_MLCTL_SPA as value to power up it,
here correct it and several similar mismatches.

Signed-off-by: Keyon Jie <yang.jie@linux.intel.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add AE-5 exit function
Connor McAdams [Tue, 18 Sep 2018 18:33:43 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add AE-5 exit function

This patch adds exit commands for the AE-5.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add AE-5 specific controls
Connor McAdams [Tue, 18 Sep 2018 18:33:42 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add AE-5 specific controls

This patch adds controls for the AE-5's headphone gain setting, and the
DAC's interpolation filter setting.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add input selection commands for AE-5
Connor McAdams [Tue, 18 Sep 2018 18:33:41 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add input selection commands for AE-5

This patch adds the input selection commands for the Sound BlasterX
AE-5.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add output set commands for AE-5
Connor McAdams [Tue, 18 Sep 2018 18:33:40 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add output set commands for AE-5

This patch adds output selection commands for the AE-5.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Clean up ca0132_alt_out_select
Connor McAdams [Tue, 18 Sep 2018 18:33:39 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Clean up ca0132_alt_out_select

This patch cleans up ca0132_alt_out_select by moving the card specific
output commands into a separate function. As more cards are added, the
function ca0132_alt_out_select is going to get more bloated with these,
so moving into a separate function tries to keep that from happening.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add DSP setup functions for AE-5
Connor McAdams [Tue, 18 Sep 2018 18:33:38 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add DSP setup functions for AE-5

This patch adds DSP setup functions for the AE-5.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Merge post-dsp functions + cleanup
Connor McAdams [Tue, 18 Sep 2018 18:33:37 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Merge post-dsp functions + cleanup

This patch cleans up some of the formatting of the post-dsp load setup
functions, and also merges some of the sub functions into individual
ones.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Change firmware name and usage
Connor McAdams [Tue, 18 Sep 2018 18:33:36 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Change firmware name and usage

The Recon3D, AE-5, Z and ZxR all share the same firmware file. Rename
this from the specific "ctefx-sbz.bin" to "ctefx-desktop.bin" and set
the AE-5 and Recon3D to use it as well.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add AE-5 regular init setup
Connor McAdams [Tue, 18 Sep 2018 18:33:35 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add AE-5 regular init setup

This patch adds AE-5 specific stuff to the ca0132_alt_init function.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add AE-5 pre-init and ca0113 functions
Connor McAdams [Tue, 18 Sep 2018 18:33:34 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add AE-5 pre-init and ca0113 functions

This patch adds AE-5 pre-init functions that happen before the main
ca0132_alt_init, and gives functions related to the ca0113 a ca0113
prefix instead of ca0132. It also adds functions to write to the 8051's
SFRs, and to write the special ca0113 commands for the AE-5.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Change ca0132_mmio_init for AE-5
Connor McAdams [Tue, 18 Sep 2018 18:33:33 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Change ca0132_mmio_init for AE-5

This patch adds the unique writes for the AE-5 on startup to
ca0132_mmio_init. The other cards share some addresses written to, but
use different values.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Set AE-5 bools and select mixer
Connor McAdams [Tue, 18 Sep 2018 18:33:32 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Set AE-5 bools and select mixer

This patch sets the bool values for the AE-5, as well as selects the
mixer it will use.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add pincfg for AE-5
Connor McAdams [Tue, 18 Sep 2018 18:33:31 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add pincfg for AE-5

This patch adds the pincfg for the Sound BlasterX AE-5, and cleans up
the function it's assigned in.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Add quirk for Sound BlasterX AE-5
Connor McAdams [Tue, 18 Sep 2018 18:33:30 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Add quirk for Sound BlasterX AE-5

This patch adds the PCI subsys ID quirk for the Sound BlasterX AE-5.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda/ca0132 - Define new verbs and control params
Connor McAdams [Tue, 18 Sep 2018 18:33:29 +0000 (14:33 -0400)]
ALSA: hda/ca0132 - Define new verbs and control params

This patch defines some new verbs found from reverse engineering of the
onboard 8051 CPU, and a control param found there as well. This clears
up usage of these verbs in other parts of the driver, and removes their
usage where they're now known to be unnecessary.

Signed-off-by: Connor McAdams <conmanx360@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: xen-front: Refine indentations and constify snd_pcm_ops
Nick Simonov [Wed, 19 Sep 2018 21:47:10 +0000 (00:47 +0300)]
ALSA: xen-front: Refine indentations and constify snd_pcm_ops

snd_pcm_ops are not supposed to change. So mark the
non-const structs as const. Also, refine indentation
to ncrease readability.

Signed-off-by: Nick Simonov <nicksimonovv@gmail.com>
Reviewed-by: Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda: Fix the audio-component completion timeout
Takashi Iwai [Tue, 18 Sep 2018 16:21:11 +0000 (18:21 +0200)]
ALSA: hda: Fix the audio-component completion timeout

The timeout of audio component binding was incorrectly specified in
msec, not in jiffies, which results in way too shorter timeout than
expected.

Along with fixing it, add the information print about the binding
failure to show the unexpected situation more clearly.

Fixes: a57942bfdd61 ("ALSA: hda: Make audio component support more generic")
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoMerge tag 'asoc-v4.19-rc4' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie...
Takashi Iwai [Mon, 17 Sep 2018 16:59:21 +0000 (18:59 +0200)]
Merge tag 'asoc-v4.19-rc4' of https://git./linux/kernel/git/broonie/sound into for-linus

ASoC: Fixes for v4.19

This is the usual set of small fixes scatterd around various drivers,
plus one fix for DAPM and a UAPI build fix.  There's not a huge amount
that stands out here relative to anything else.

5 years agoALSA: fireworks: fix memory leak of response buffer at error path
Takashi Sakamoto [Mon, 17 Sep 2018 08:26:41 +0000 (17:26 +0900)]
ALSA: fireworks: fix memory leak of response buffer at error path

After allocating memory object for response buffer, ALSA fireworks
driver has leak of the memory object at error path.

This commit releases the object at the error path.

Fixes: 7d3c1d5901aa('ALSA: fireworks: delayed registration of sound card')
Cc: <stable@vger.kernel.org> # v4.7+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: oxfw: fix memory leak of discovered stream formats at error path
Takashi Sakamoto [Mon, 17 Sep 2018 08:26:20 +0000 (17:26 +0900)]
ALSA: oxfw: fix memory leak of discovered stream formats at error path

After finishing discover of stream formats, ALSA OXFW driver has memory
leak of allocated memory object at error path.

This commit releases the memory object at the error path.

Fixes: 6c29230e2a5f ('ALSA: oxfw: delayed registration of sound card')
Cc: <stable@vger.kernel.org> # v4.7+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: oxfw: fix memory leak for model-dependent data at error path
Takashi Sakamoto [Mon, 17 Sep 2018 08:26:08 +0000 (17:26 +0900)]
ALSA: oxfw: fix memory leak for model-dependent data at error path

After allocating model-dependent data, ALSA OXFW driver has memory leak
of the data at error path.

This commit releases the data at the error path.

Fixes: 6c29230e2a5f ('ALSA: oxfw: delayed registration of sound card')
Cc: <stable@vger.kernel.org> # v4.7+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: bebob: fix memory leak for M-Audio FW1814 and ProjectMix I/O at error path
Takashi Sakamoto [Mon, 17 Sep 2018 08:25:24 +0000 (17:25 +0900)]
ALSA: bebob: fix memory leak for M-Audio FW1814 and ProjectMix I/O at error path

After allocating model-dependent data for M-Audio FW1814 and ProjectMix
I/O, ALSA bebob driver has memory leak at error path.

This commit releases the allocated data at the error path.

Fixes: 04a2c73c97eb('ALSA: bebob: delayed registration of sound card')
Cc: <stable@vger.kernel.org> # v4.7+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: atiixp: fix fall-through annotations
Gustavo A. R. Silva [Thu, 13 Sep 2018 18:43:47 +0000 (13:43 -0500)]
ALSA: atiixp: fix fall-through annotations

Replace "fallthru" with a proper "fall through" annotation.

This fix is part of the ongoing efforts to enabling
-Wimplicit-fallthrough

Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda - Enable runtime PM only for discrete GPU
Takashi Iwai [Mon, 10 Sep 2018 14:20:25 +0000 (16:20 +0200)]
ALSA: hda - Enable runtime PM only for discrete GPU

The recent change of vga_switcheroo allowed the runtime PM for
HD-audio on AMD GPUs, but this also resulted in a regression.  When
the HD-audio controller driver gets runtime-suspended, HD-audio link
is turned off, and the hotplug notification is ignored.  This leads to
the inconsistent audio state (the connection isn't notified and ELD is
ignored).

The best fix would be to implement the proper ELD notification via the
audio component, but it's still not ready.  As a quick workaround,
this patch adds the check of runtime_idle and allows the runtime
suspend only when the vga_switcheroo is bound with discrete GPU.
That is, a system with a single GPU and APU would be again without
runtime PM to keep the HD-audio link for the hotplug notification and
ELD read out.

Also, the codec->auto_runtime_pm flag is set only for the discrete GPU
at the time GPU gets bound via vga_switcheroo (i.e. only dGPU is
forcibly runtime-PM enabled), so that APU can still get the ELD
notification.

For identifying which GPU is bound, a new vga_switcheroo client
callback, gpu_bound, is implemented.  The vga_switcheroo simply calls
this when GPU is bound, and tells whether it's dGPU or APU.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=200945
Fixes: 07f4f97d7b4b ("vga_switcheroo: Use device link for HDA controller")
Reported-by: Jian-Hong Pan <jian-hong@endlessm.com>
Tested-by: Jian-Hong Pan <jian-hong@endlessm.com>
Acked-by: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: oxfw: fix memory leak of private data
Takashi Sakamoto [Thu, 13 Sep 2018 12:31:18 +0000 (21:31 +0900)]
ALSA: oxfw: fix memory leak of private data

Although private data of sound card instance is usually allocated in the
tail of the instance, drivers in ALSA firewire stack allocate the private
data before allocating the instance. In this case, the private data
should be released explicitly at .private_free callback of the instance.

This commit fixes memory leak following to the above design.

Fixes: 6c29230e2a5f ('ALSA: oxfw: delayed registration of sound card')
Cc: <stable@vger.kernel.org> # v4.7+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: firewire-tascam: fix memory leak of private data
Takashi Sakamoto [Thu, 13 Sep 2018 12:31:05 +0000 (21:31 +0900)]
ALSA: firewire-tascam: fix memory leak of private data

Although private data of sound card instance is usually allocated in the
tail of the instance, drivers in ALSA firewire stack allocate the private
data before allocating the instance. In this case, the private data
should be released explicitly at .private_free callback of the instance.

This commit fixes memory leak following to the above design.

Fixes: b610386c8afb ('ALSA: firewire-tascam: deleyed registration of sound card')
Cc: <stable@vger.kernel.org> # v4.7+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: firewire-digi00x: fix memory leak of private data
Takashi Sakamoto [Thu, 13 Sep 2018 12:30:34 +0000 (21:30 +0900)]
ALSA: firewire-digi00x: fix memory leak of private data

Although private data of sound card instance is usually allocated in the
tail of the instance, drivers in ALSA firewire stack allocate the private
data before allocating the instance. In this case, the private data
should be released explicitly at .private_free callback of the instance.

This commit fixes memory leak following to the above design.

Fixes: 86c8dd7f4da3 ('ALSA: firewire-digi00x: delayed registration of sound card')
Cc: <stable@vger.kernel.org> # v4.7+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: intel8x0m: Register irq handler after register initializations
Takashi Iwai [Tue, 28 Aug 2018 14:39:10 +0000 (16:39 +0200)]
ALSA: intel8x0m: Register irq handler after register initializations

The interrupt handler has to be acquired after the other resource
initialization when allocated with IRQF_SHARED.  Otherwise it's
triggered before the resource gets ready, and may lead to unpleasant
behavior.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agosound: don't call skl_init_chip() to reset intel skl soc
Yu Zhao [Tue, 11 Sep 2018 21:15:16 +0000 (15:15 -0600)]
sound: don't call skl_init_chip() to reset intel skl soc

Internally, skl_init_chip() calls snd_hdac_bus_init_chip() which
1) sets bus->chip_init to prevent multiple entrances before device
is stopped; 2) enables interrupt.

We shouldn't use it for the purpose of resetting device only because
1) when we really want to initialize device, we won't be able to do
so; 2) we are ready to handle interrupt yet, and kernel crashes when
interrupt comes in.

Rename azx_reset() to snd_hdac_bus_reset_link(), and use it to reset
device properly.

Fixes: 60767abcea3d ("ASoC: Intel: Skylake: Reset the controller in probe")
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agosound: enable interrupt after dma buffer initialization
Yu Zhao [Tue, 11 Sep 2018 21:14:04 +0000 (15:14 -0600)]
sound: enable interrupt after dma buffer initialization

In snd_hdac_bus_init_chip(), we enable interrupt before
snd_hdac_bus_init_cmd_io() initializing dma buffers. If irq has
been acquired and irq handler uses the dma buffer, kernel may crash
when interrupt comes in.

Fix the problem by postponing enabling irq after dma buffer
initialization. And warn once on null dma buffer pointer during the
initialization.

Reviewed-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoRevert "ASoC: Intel: Skylake: Acquire irq after RIRB allocation"
Yu Zhao [Tue, 11 Sep 2018 21:12:46 +0000 (15:12 -0600)]
Revert "ASoC: Intel: Skylake: Acquire irq after RIRB allocation"

This reverts commit 12eeeb4f4733bbc4481d01df35933fc15beb8b19.

The patch doesn't fix accessing memory with null pointer in
skl_interrupt().

There are two problems: 1) skl_init_chip() is called twice, before
and after dma buffer is allocate. The first call sets bus->chip_init
which prevents the second from initializing bus->corb.buf and
rirb.buf from bus->rb.area. 2) snd_hdac_bus_init_chip() enables
interrupt before snd_hdac_bus_init_cmd_io() initializing dma buffers.
There is a small window which skl_interrupt() can be called if irq
has been acquired. If so, it crashes when using null dma buffer
pointers.

Will fix the problems in the following patches. Also attaching the
crash for future reference.

[   16.949148] general protection fault: 0000 [#1] PREEMPT SMP KASAN PTI
<snipped>
[   16.950903] Call Trace:
[   16.950906]  <IRQ>
[   16.950918]  skl_interrupt+0x19e/0x2d6 [snd_soc_skl]
[   16.950926]  ? dma_supported+0xb5/0xb5 [snd_soc_skl]
[   16.950933]  __handle_irq_event_percpu+0x27a/0x6c8
[   16.950937]  ? __irq_wake_thread+0x1d1/0x1d1
[   16.950942]  ? __do_softirq+0x57a/0x69e
[   16.950944]  handle_irq_event_percpu+0x95/0x1ba
[   16.950948]  ? _raw_spin_unlock+0x65/0xdc
[   16.950951]  ? __handle_irq_event_percpu+0x6c8/0x6c8
[   16.950953]  ? _raw_spin_unlock+0x65/0xdc
[   16.950957]  ? time_cpufreq_notifier+0x483/0x483
[   16.950959]  handle_irq_event+0x89/0x123
[   16.950962]  handle_fasteoi_irq+0x16f/0x425
[   16.950965]  handle_irq+0x1fe/0x28e
[   16.950969]  do_IRQ+0x6e/0x12e
[   16.950972]  common_interrupt+0x7a/0x7a
[   16.950974]  </IRQ>
<snipped>
[   16.951031] RIP: snd_hdac_bus_update_rirb+0x19b/0x4cf [snd_hda_core] RSP: ffff88015c807c08
[   16.951036] ---[ end trace 58bf9ece1775bc92 ]---

Fixes: 2eeeb4f4733b ("ASoC: Intel: Skylake: Acquire irq after RIRB allocation")
Signed-off-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoALSA: hda: fix unused variable warning
Anders Roxell [Tue, 11 Sep 2018 14:18:36 +0000 (16:18 +0200)]
ALSA: hda: fix unused variable warning

When CONFIG_X86=n function azx_snoop doesn't use the variable chip it
only returns true.

sound/pci/hda/hda_intel.c: In function â€˜dma_alloc_pages’:
sound/pci/hda/hda_intel.c:2002:14: warning: unused variable â€˜chip’ [-Wunused-variable]
  struct azx *chip = bus_to_azx(bus);
              ^~~~

Create a inline function of azx_snoop.

Fixes: a41d122449be ("ALSA: hda - Embed bus into controller object")
Signed-off-by: Anders Roxell <anders.roxell@linaro.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: emu10k1: fix possible info leak to userspace on SNDRV_EMU10K1_IOCTL_INFO
Willy Tarreau [Sat, 8 Sep 2018 06:12:21 +0000 (08:12 +0200)]
ALSA: emu10k1: fix possible info leak to userspace on SNDRV_EMU10K1_IOCTL_INFO

snd_emu10k1_fx8010_ioctl(SNDRV_EMU10K1_IOCTL_INFO) allocates
memory using kmalloc() and partially fills it by calling
snd_emu10k1_fx8010_info() before returning the resulting
structure to userspace, leaving uninitialized holes. Let's
just use kzalloc() here.

BugLink: http://blog.infosectcbr.com.au/2018/09/linux-kernel-infoleaks.html
Signed-off-by: Willy Tarreau <w@1wt.eu>
Cc: Jann Horn <jannh@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoASoC: cs4265: fix MMTLR Data switch control
Sébastien Szymanski [Thu, 6 Sep 2018 09:16:00 +0000 (11:16 +0200)]
ASoC: cs4265: fix MMTLR Data switch control

The MMTLR bit is in the CS4265_SPDIF_CTL2 register at address 0x12 bit 0
and not at address 0x0 bit 1. Fix this.

Signed-off-by: Sébastien Szymanski <sebastien.szymanski@armadeus.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org
5 years agoASoC: AMD: Ensure reset bit is cleared before configuring
Akshu Agrawal [Mon, 10 Sep 2018 08:06:30 +0000 (13:36 +0530)]
ASoC: AMD: Ensure reset bit is cleared before configuring

HW register descriptions says:
"DMA Channel Reset...Software must confirm that this bit is
cleared before reprogramming any of the channel configuration registers."
There could be cases where dma stop errored out leaving dma channel
in reset state. We need to ensure that before the start of another dma,
channel is out of the reset state.

Signed-off-by: Akshu Agrawal <akshu.agrawal@amd.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoALSA: fireface: fix memory leak in ff400_switch_fetching_mode()
Takashi Sakamoto [Sun, 9 Sep 2018 13:25:52 +0000 (22:25 +0900)]
ALSA: fireface: fix memory leak in ff400_switch_fetching_mode()

An allocated memory forgets to be released.

Fixes: 76fdb3a9e13 ('ALSA: fireface: add support for Fireface 400')
Cc: <stable@vger.kernel.org> # 4.12+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: bebob: use address returned by kmalloc() instead of kernel stack for streaming...
Takashi Sakamoto [Sun, 9 Sep 2018 13:25:12 +0000 (22:25 +0900)]
ALSA: bebob: use address returned by kmalloc() instead of kernel stack for streaming DMA mapping

When executing 'fw_run_transaction()' with 'TCODE_WRITE_BLOCK_REQUEST',
an address of 'payload' argument is used for streaming DMA mapping by
'firewire_ohci' module if 'size' argument is larger than 8 byte.
Although in this case the address should not be on kernel stack, current
implementation of ALSA bebob driver uses data in kernel stack for a cue
to boot M-Audio devices. This often brings unexpected result, especially
for a case of CONFIG_VMAP_STACK=y.

This commit fixes the bug.

Reference: https://bugzilla.kernel.org/show_bug.cgi?id=201021
Reference: https://forum.manjaro.org/t/firewire-m-audio-410-driver-wont-load-firmware/51165
Fixes: a2b2a7798fb6('ALSA: bebob: Send a cue to load firmware for M-Audio Firewire series')
Cc: <stable@vger.kernel.org> # v3.16+
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: pcm: Update hardware pointer before start capture
Ricardo Biehl Pasquali [Fri, 7 Sep 2018 19:58:54 +0000 (16:58 -0300)]
ALSA: pcm: Update hardware pointer before start capture

This ensures the transfer loop won't waste a run to read
the few frames (if any) between start and hw_ptr update.
It will wait for the next interrupt with wait_for_avail().

Signed-off-by: Ricardo Biehl Pasquali <pasqualirb@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoASoC: rsnd: don't fallback to PIO mode when -EPROBE_DEFER
Kuninori Morimoto [Thu, 6 Sep 2018 03:21:47 +0000 (03:21 +0000)]
ASoC: rsnd: don't fallback to PIO mode when -EPROBE_DEFER

Current rsnd driver will fallback to PIO mode if it can't get DMA
handler. But, DMA might return -EPROBE_DEFER when probe timing.
This driver always fallback to PIO mode especially from
commit ac6bbf0cdf4206c ("iommu: Remove IOMMU_OF_DECLARE") because
of this reason.

The DMA driver will be probed later, but sound driver might be
probed as PIO mode in such case. This patch fixup this issue.
Then, -EPROBE_DEFER is not error. Thus, let's don't indicate error
message in such case.
And it needs to call rsnd_adg_remove() individually if probe failed,
because it registers clk which should be unregister.

Maybe PIO fallback feature itself is not needed,
but let's keep it so far.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoASoC: rsnd: adg: care clock-frequency size
Kuninori Morimoto [Thu, 6 Sep 2018 03:21:33 +0000 (03:21 +0000)]
ASoC: rsnd: adg: care clock-frequency size

ADG has buffer over flow bug if DT has more than 3 clock-frequency.
This patch fixup this issue, and uses first 2 values.

clock-frequency = <x y>; /* this is OK */
clock-frequency = <x y z>; /* this is NG */

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoASoC: uniphier: change status to orphan
Katsuhiro Suzuki [Wed, 5 Sep 2018 08:31:37 +0000 (17:31 +0900)]
ASoC: uniphier: change status to orphan

Since I'm leaving from Socionext, I'll unable to access specification
documents of this hardware (these are not public). So change the
state to orphan until someone will maintain this driver.

Signed-off-by: Katsuhiro Suzuki <suzuki.katsuhiro@socionext.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoMerge branch 'topic/pcm-indirect-fixes' into for-next
Takashi Iwai [Tue, 4 Sep 2018 18:23:29 +0000 (20:23 +0200)]
Merge branch 'topic/pcm-indirect-fixes' into for-next

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: mips: Cleanup indirect PCM helper usages
Takashi Iwai [Sun, 2 Sep 2018 08:26:08 +0000 (10:26 +0200)]
ALSA: mips: Cleanup indirect PCM helper usages

We shouldn't set up the indirect PCM parameters at trigger but they
should be set at prepare.  Also, remove a useless debug message, too.

Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: cs46xx: Use SNDRV_PCM_INFO_SYNC_APPLPTR info flag
Takashi Iwai [Sun, 2 Sep 2018 08:23:22 +0000 (10:23 +0200)]
ALSA: cs46xx: Use SNDRV_PCM_INFO_SYNC_APPLPTR info flag

The recently introduced PCM info flag assures the call of ack ops at
each applptr change, and this is mandatory for the indirect PCM
helpers.

Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: emu10k1: Use SNDRV_PCM_INFO_SYNC_APPLPTR info flag
Takashi Iwai [Sun, 2 Sep 2018 08:23:07 +0000 (10:23 +0200)]
ALSA: emu10k1: Use SNDRV_PCM_INFO_SYNC_APPLPTR info flag

The recently introduced PCM info flag assures the call of ack ops at
each applptr change, and this is mandatory for the indirect PCM
helpers.

Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: mips: Use SNDRV_PCM_INFO_SYNC_APPLPTR info flag
Takashi Iwai [Sun, 2 Sep 2018 08:22:37 +0000 (10:22 +0200)]
ALSA: mips: Use SNDRV_PCM_INFO_SYNC_APPLPTR info flag

The recently introduced PCM info flag assures the call of ack ops at
each applptr change, and this is mandatory for the indirect PCM
helpers.

Also, with the proper ack callback, we need no longer prefill at
trigger start.  The relevant code can be killed.

Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: rme32: Use SNDRV_PCM_INFO_SYNC_APPLPTR info flag
Takashi Iwai [Sun, 2 Sep 2018 08:22:13 +0000 (10:22 +0200)]
ALSA: rme32: Use SNDRV_PCM_INFO_SYNC_APPLPTR info flag

The recently introduced PCM info flag assures the call of ack ops at
each applptr change, and this is mandatory for the indirect PCM
helpers.

Also, with the proper ack callback, we need no longer prefill at
trigger start.  The relevant code can be killed.

Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda: Fix several mismatch for register mask and value
Keyon Jie [Mon, 3 Sep 2018 02:47:09 +0000 (10:47 +0800)]
ALSA: hda: Fix several mismatch for register mask and value

E.g. for snd_hdac_ext_link_clear_stream_id(), we should set (1 << stream)
as mask, and 0 as value, here correct it and several similar mismatches.

And, here also remove unreadable register_mask usage for those mask value
updating.

Signed-off-by: Keyon Jie <yang.jie@linux.intel.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: aoa: Convert to using %pOFn instead of device_node.name
Rob Herring [Tue, 28 Aug 2018 15:44:31 +0000 (10:44 -0500)]
ALSA: aoa: Convert to using %pOFn instead of device_node.name

In preparation to remove the node name pointer from struct device_node,
convert printf users to use the %pOFn format specifier.

Signed-off-by: Rob Herring <robh@kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoASoC: rsnd: fixup not to call clk_get/set under non-atomic
Jiada Wang [Mon, 3 Sep 2018 07:08:58 +0000 (07:08 +0000)]
ASoC: rsnd: fixup not to call clk_get/set under non-atomic

Clocking operations clk_get/set_rate, are non-atomic,
they shouldn't be called in soc_pcm_trigger() which is atomic.

Following issue was found due to execution of clk_get_rate() causes
sleep in soc_pcm_trigger(), which shouldn't be blocked.

We can reproduce this issue by following
> enable CONFIG_DEBUG_ATOMIC_SLEEP=y
> compile, and boot
> mount -t debugfs none /sys/kernel/debug
> while true; do cat /sys/kernel/debug/clk/clk_summary > /dev/null; done &
> while true; do aplay xxx; done

This patch adds support to .prepare callback, and moves non-atomic
clocking operations to it. As .prepare is non-atomic, it is always
called before trigger_start/trigger_stop.

BUG: sleeping function called from invalid context at kernel/locking/mutex.c:620
in_atomic(): 1, irqs_disabled(): 128, pid: 2242, name: aplay
INFO: lockdep is turned off.
irq event stamp: 5964
hardirqs last enabled at (5963): [<ffff200008e59e40>] mutex_lock_nested+0x6e8/0x6f0
hardirqs last disabled at (5964): [<ffff200008e623f0>] _raw_spin_lock_irqsave+0x24/0x68
softirqs last enabled at (5502): [<ffff200008081838>] __do_softirq+0x560/0x10c0
softirqs last disabled at (5495): [<ffff2000080c2e78>] irq_exit+0x160/0x25c
Preemption disabled at:[ 62.904063] [<ffff200008be4d48>] snd_pcm_stream_lock+0xb4/0xc0
CPU: 2 PID: 2242 Comm: aplay Tainted: G B C 4.9.54+ #186
Hardware name: Renesas Salvator-X board based on r8a7795 (DT)
Call trace:
[<ffff20000808fe48>] dump_backtrace+0x0/0x37c
[<ffff2000080901d8>] show_stack+0x14/0x1c
[<ffff2000086f4458>] dump_stack+0xfc/0x154
[<ffff2000081134a0>] ___might_sleep+0x57c/0x58c
[<ffff2000081136b8>] __might_sleep+0x208/0x21c
[<ffff200008e5980c>] mutex_lock_nested+0xb4/0x6f0
[<ffff2000087cac74>] clk_prepare_lock+0xb0/0x184
[<ffff2000087cb094>] clk_core_get_rate+0x14/0x54
[<ffff2000087cb0f4>] clk_get_rate+0x20/0x34
[<ffff20000113aa00>] rsnd_adg_ssi_clk_try_start+0x158/0x4f8 [snd_soc_rcar]
[<ffff20000113da00>] rsnd_ssi_init+0x668/0x7a0 [snd_soc_rcar]
[<ffff200001133ff4>] rsnd_soc_dai_trigger+0x4bc/0xcf8 [snd_soc_rcar]
[<ffff200008c1af24>] soc_pcm_trigger+0x2a4/0x2d4

Fixes: e7d850dd10f4 ("ASoC: rsnd: use mod base common method on SSI-parent")
Signed-off-by: Jiada Wang <jiada_wang@mentor.com>
Signed-off-by: Timo Wischer <twischer@de.adit-jv.com>
[Kuninori: tidyup for upstream]
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Hiroyuki Yokoyama <hiroyuki.yokoyama.vx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org
5 years agoALSA: rawmidi: Initialize allocated buffers
Takashi Iwai [Mon, 3 Sep 2018 13:16:43 +0000 (15:16 +0200)]
ALSA: rawmidi: Initialize allocated buffers

syzbot reported the uninitialized value exposure in certain situations
using virmidi loop.  It's likely a very small race at writing and
reading, and the influence is almost negligible.  But it's safer to
paper over this just by replacing the existing kvmalloc() with
kvzalloc().

Reported-by: syzbot+194dffdb8b22fc5d207a@syzkaller.appspotmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoASoC: q6routing: initialize data correctly
Srinivas Kandagatla [Mon, 3 Sep 2018 11:07:47 +0000 (12:07 +0100)]
ASoC: q6routing: initialize data correctly

Some of the router data fields are left as default zeros which are
valid dai ids, so initialize these to invalid value of -1.

Without intializing these correctly get_session_from_id() can return
incorrect session resulting in not closing the opened copp and messing
up with the copp ref count.

Fixes: e3a33673e845 ("ASoC: qdsp6: q6routing: Add q6routing driver")
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoASoC: tas6424: Save last fault register even when clear
Andrew F. Davis [Fri, 31 Aug 2018 15:14:05 +0000 (10:14 -0500)]
ASoC: tas6424: Save last fault register even when clear

When there is no fault bit set in a fault register we skip the fault
reporting section for that register. This also skips over saving that
registers value. We save the value so we will not double report an
error, but if an error clears then returns we will also not report it
as we did not save the all cleared register value. Fix this by saving
the fault register value in the all clear path.

Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org
5 years agoMerge tag 'hda-codec-h-move' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie...
Takashi Iwai [Thu, 30 Aug 2018 14:55:38 +0000 (16:55 +0200)]
Merge tag 'hda-codec-h-move' of git://git./linux/kernel/git/broonie/sound into for-next

ALSA: Move hda_codec.h to include/sound

For easier sharing with ASoC.

5 years agoALSA: hda: move hda_codec.h to include/sound
Pierre-Louis Bossart [Wed, 22 Aug 2018 20:24:57 +0000 (15:24 -0500)]
ALSA: hda: move hda_codec.h to include/sound

As suggested by Takashi, move this header file to make it easier
to include from e.g. the Intel Skylake driver in follow-up patches

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoALSA: hda - Fix cancel_work_sync() stall from jackpoll work
Takashi Iwai [Thu, 30 Aug 2018 13:13:16 +0000 (15:13 +0200)]
ALSA: hda - Fix cancel_work_sync() stall from jackpoll work

On AMD/ATI controllers, the HD-audio controller driver allows a bus
reset upon the error recovery, and its procedure includes the
cancellation of pending jack polling work as found in
snd_hda_bus_codec_reset().  This works usually fine, but it becomes a
problem when the reset happens from the jack poll work itself; then
calling cancel_work_sync() from the work being processed tries to wait
the finish endlessly.

As a workaround, this patch adds the check of current_work() and
applies the cancel_work_sync() only when it's not from the
jackpoll_work.

This doesn't fix the root cause of the reported error below, but at
least, it eases the unexpected stall of the whole system.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=200937
Cc: <stable@vger.kernel.org>
Cc: Lukas Wunner <lukas@wunner.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda - Clean up jackpoll_ms option handling
Takashi Iwai [Thu, 30 Aug 2018 05:58:50 +0000 (07:58 +0200)]
ALSA: hda - Clean up jackpoll_ms option handling

Currently the jackpoll_ms option value is passed indirectly by
referring to an array in chip->jackpoll_ms although each card needs to
see only the assigned value.  Also, the sanity check is done at each
time in get_jackpoll_interval() although basically jackpoll_ms option
is a read-only, hence we need to evaluate only once at probe time.

This patch is the code simplification about the above points: the jack
polling interval is directly set to chip->jackpoll_interval so that it
can be simply copied to each codec.

No functional change but only code reduction.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoASoC: sigmadsp: safeload should not have lower byte limit
Danny Smith [Thu, 23 Aug 2018 08:26:20 +0000 (10:26 +0200)]
ASoC: sigmadsp: safeload should not have lower byte limit

Fixed range in safeload conditional to allow safeload to up to 20 bytes,
without a lower limit.

Signed-off-by: Danny Smith <dannys@axis.com>
Acked-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoASoC: wm8804: Add ACPI support
Pierre-Louis Bossart [Thu, 23 Aug 2018 03:49:36 +0000 (22:49 -0500)]
ASoC: wm8804: Add ACPI support

HID made of either Wolfson/CirrusLogic PCI ID + 8804 identifier.

This helps enumerate the HifiBerry Digi+ HAT boards on the Up2 platform.

The scripts at https://github.com/thesofproject/acpi-scripts can be
used to add the ACPI initrd overlays.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoASoC: rt5682: Change DAC/ADC volume scale
Shuming Fan [Fri, 24 Aug 2018 02:52:19 +0000 (10:52 +0800)]
ASoC: rt5682: Change DAC/ADC volume scale

The step of DAC/ADC volume scale changes from 0.375dB to 0.75dB

Signed-off-by: Shuming Fan <shumingf@realtek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org
5 years agoASoC: max98373: Added 10ms sleep after amp software reset
Ryan Lee [Fri, 24 Aug 2018 01:37:08 +0000 (18:37 -0700)]
ASoC: max98373: Added 10ms sleep after amp software reset

Signed-off-by: Ryan Lee <ryans.lee@maximintegrated.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
5 years agoALSA: intel_hdmi: Use the new non-cached allocation
Takashi Iwai [Wed, 8 Aug 2018 20:38:36 +0000 (22:38 +0200)]
ALSA: intel_hdmi: Use the new non-cached allocation

The HDMI LPE audio driver requires the non-cached page allocations for
its buffers.  With the recent support of SNDRV_DMA_TYPE_DEV_UC type,
we can reduce lots of codes in the driver side and let the memalloc
core doing it properly.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: intel8x0: Use the new non-cached allocation for 440MX workaround
Takashi Iwai [Wed, 8 Aug 2018 20:31:52 +0000 (22:31 +0200)]
ALSA: intel8x0: Use the new non-cached allocation for 440MX workaround

intel8x0 driver requires the non-cached pages for 440MX workaround,
and this can be implemented more easily with the new memalloc type,
SNDRV_DMA_TYPE_DEV_UC.  This allows us to reduce lots of code.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda: Remove substream allocation/free ops
Takashi Iwai [Wed, 8 Aug 2018 15:12:58 +0000 (17:12 +0200)]
ALSA: hda: Remove substream allocation/free ops

Since we dropped the memory page fiddling in the own allocators in
hda_intel.c, the substream allocation and free ops in both hda_intel.c
and hda_tegra.c became nothing but the simple calls of the standard
snd_pcm_lib helpers.  As both are identical, there is no longer need
for indirect calls via ops; it's a good opportunity for removing ops
and simplifying the codes.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda: Use new non-cached allocation for non-snoop mode
Takashi Iwai [Wed, 8 Aug 2018 20:23:30 +0000 (22:23 +0200)]
ALSA: hda: Use new non-cached allocation for non-snoop mode

Now the ALSA memory allocator helper supports the new non-cached
pages, let's use the new type, SNDRV_DMA_TYPE_DEV_UC_SG, for HD-audio
driver.  This allows us to reduce lots of codes.

As another positive side-effect by this patch, the long-standing issue
with non-snoop mode playing in the non-mmap mode is fixed.  The core
memalloc helper does the proper pgprot setup for non-cached pages for
vmap(), which was missing in the past.

Reported-and-tested-by: Hans Hu <HansHu@zhaoxin.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hda: Check the non-cached stream buffers more explicitly
Takashi Iwai [Sat, 11 Aug 2018 21:33:34 +0000 (23:33 +0200)]
ALSA: hda: Check the non-cached stream buffers more explicitly

Introduce a new flag, uc_buffer, to indicate that the controller
requires the non-cached pages for stream buffers, either as a
chip-specific requirement or specified via snoop=0 option.
This improves the code-readability.

Also, this patch fixes the incorrect behavior for C-Media chip where
the stream buffers were never handled as non-cached due to the check
of driver_type even if you pass snoop=0 option.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: memalloc: Add non-cached buffer type
Takashi Iwai [Wed, 8 Aug 2018 15:01:00 +0000 (17:01 +0200)]
ALSA: memalloc: Add non-cached buffer type

In some cases (mainly for x86), we need the DMA coherent buffer with
non-cached pages.  Although this has been done in each driver side
like HD-audio and intel8x0, it can be done cleaner in the core memory
allocator.

This patch adds the new types, SNDRV_DMA_TYPE_DEV_UC and
SNDRV_DMA_TYPE_DEV_UC_SG, for allocating such non-cached buffer
pages.  On non-x86 architectures, they work as same as the standard
SNDRV_DMA_TYPE_DEV and *_SG.

One additional change by this move is that we can assure to pass the
non-cached pgprot to the vmapped buffer, too.  It eventually fixes the
case like non-snoop mode without mmap access on HD-audio.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: memalloc: Simplify snd_malloc_dev_pages() calls
Takashi Iwai [Fri, 10 Aug 2018 12:43:37 +0000 (14:43 +0200)]
ALSA: memalloc: Simplify snd_malloc_dev_pages() calls

snd_malloc_dev_pages() and snd_free_dev_pages() are local functions
and the parameters passed there are all contained in snd_dma_buffer
object.  As a code-simplification, pass snd_dma_buffer object and
assign the address there like other allocators do (except for
snd_malloc_pages() which is called from outside, hence we can't change
easily).

Only code refactoring, no functional changes.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: memalloc: Don't align the size to power-of-two
Takashi Iwai [Wed, 8 Aug 2018 14:56:46 +0000 (16:56 +0200)]
ALSA: memalloc: Don't align the size to power-of-two

The size passed to dma_alloc_coherent() doesn't have to be aligned
with power-of-two, rather it should be the raw size.  As a minor
optimization, remove the size adjustment in the current code.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: seq: Do error checks at creating system ports
Takashi Iwai [Tue, 28 Aug 2018 10:49:43 +0000 (12:49 +0200)]
ALSA: seq: Do error checks at creating system ports

snd_seq_system_client_init() doesn't check the errors returned from
its port creations.  Let's do it properly and handle the error paths.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: seq: add error check in snd_seq_system_client_init()
Dan Carpenter [Tue, 28 Aug 2018 08:46:34 +0000 (11:46 +0300)]
ALSA: seq: add error check in snd_seq_system_client_init()

Static checkers complain that snd_seq_create_kernel_client() can return
-EBUSY here so we need to have some error handling.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: pcm: signedness bug in snd_pcm_plug_alloc()
Dan Carpenter [Mon, 27 Aug 2018 09:21:45 +0000 (12:21 +0300)]
ALSA: pcm: signedness bug in snd_pcm_plug_alloc()

The "frames" variable is unsigned so the error handling doesn't work
properly.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: hdspm: fix spelling mistake "Initializeing" -> "Initializing"
Colin Ian King [Sun, 26 Aug 2018 22:09:52 +0000 (23:09 +0100)]
ALSA: hdspm: fix spelling mistake "Initializeing" -> "Initializing"

Trivial fix to spelling mistake in dev_dbg message and also remove
extraneous white space and repeated question marks.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoALSA: pcm: Return 0 when size < start_threshold in capture
Ricardo Biehl Pasquali [Sat, 25 Aug 2018 19:53:23 +0000 (16:53 -0300)]
ALSA: pcm: Return 0 when size < start_threshold in capture

In __snd_pcm_lib_xfer(), when capture, if state is PREPARED
and size is less than start_threshold nothing can be done.
As there is no error, 0 is returned.

Signed-off-by: Ricardo Biehl Pasquali <pasqualirb@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
5 years agoLinux 4.19-rc1
Linus Torvalds [Sun, 26 Aug 2018 21:11:59 +0000 (14:11 -0700)]
Linux 4.19-rc1

5 years agoMerge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 20:39:05 +0000 (13:39 -0700)]
Merge branch 'timers-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull timer update from Thomas Gleixner:
 "New defines for the compat time* types so they can be shared between
  32bit and 64bit builds. Not used yet, but merging them now allows the
  actual conversions to be merged through different maintainer trees
  without dependencies

  We still have compat interfaces for 32bit on 64bit even with the new
  2038 safe timespec/val variants because pointer size is different. And
  for the old style timespec/val interfaces we need yet another 'compat'
  interface for both 32bit native and 32bit on 64bit"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  y2038: Provide aliases for compat helpers

5 years agoMerge branch 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax
Linus Torvalds [Sun, 26 Aug 2018 18:48:42 +0000 (11:48 -0700)]
Merge branch 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax

Pull IDA updates from Matthew Wilcox:
 "A better IDA API:

      id = ida_alloc(ida, GFP_xxx);
      ida_free(ida, id);

  rather than the cumbersome ida_simple_get(), ida_simple_remove().

  The new IDA API is similar to ida_simple_get() but better named.  The
  internal restructuring of the IDA code removes the bitmap
  preallocation nonsense.

  I hope the net -200 lines of code is convincing"

* 'ida-4.19' of git://git.infradead.org/users/willy/linux-dax: (29 commits)
  ida: Change ida_get_new_above to return the id
  ida: Remove old API
  test_ida: check_ida_destroy and check_ida_alloc
  test_ida: Convert check_ida_conv to new API
  test_ida: Move ida_check_max
  test_ida: Move ida_check_leaf
  idr-test: Convert ida_check_nomem to new API
  ida: Start new test_ida module
  target/iscsi: Allocate session IDs from an IDA
  iscsi target: fix session creation failure handling
  drm/vmwgfx: Convert to new IDA API
  dmaengine: Convert to new IDA API
  ppc: Convert vas ID allocation to new IDA API
  media: Convert entity ID allocation to new IDA API
  ppc: Convert mmu context allocation to new IDA API
  Convert net_namespace to new IDA API
  cb710: Convert to new IDA API
  rsxx: Convert to new IDA API
  osd: Convert to new IDA API
  sd: Convert to new IDA API
  ...

5 years agoMerge tag 'gcc-plugins-v4.19-rc1-fix' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 18:41:08 +0000 (11:41 -0700)]
Merge tag 'gcc-plugins-v4.19-rc1-fix' of git://git./linux/kernel/git/kees/linux

Pull gcc plugin fix from Kees Cook:
 "Lift gcc test into Kconfig. This is for better behavior when the
  kernel is built with Clang, reported by Stefan Agner"

* tag 'gcc-plugins-v4.19-rc1-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  gcc-plugins: Disable when building under Clang

5 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 18:25:21 +0000 (11:25 -0700)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf updates from Thomas Gleixner:
 "Kernel:
   - Improve kallsyms coverage
   - Add x86 entry trampolines to kcore
   - Fix ARM SPE handling
   - Correct PPC event post processing

  Tools:
   - Make the build system more robust
   - Small fixes and enhancements all over the place
   - Update kernel ABI header copies
   - Preparatory work for converting libtraceevnt to a shared library
   - License cleanups"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (100 commits)
  tools arch: Update arch/x86/lib/memcpy_64.S copy used in 'perf bench mem memcpy'
  tools arch x86: Update tools's copy of cpufeatures.h
  perf python: Fix pyrf_evlist__read_on_cpu() interface
  perf mmap: Store real cpu number in 'struct perf_mmap'
  perf tools: Remove ext from struct kmod_path
  perf tools: Add gzip_is_compressed function
  perf tools: Add lzma_is_compressed function
  perf tools: Add is_compressed callback to compressions array
  perf tools: Move the temp file processing into decompress_kmodule
  perf tools: Use compression id in decompress_kmodule()
  perf tools: Store compression id into struct dso
  perf tools: Add compression id into 'struct kmod_path'
  perf tools: Make is_supported_compression() static
  perf tools: Make decompress_to_file() function static
  perf tools: Get rid of dso__needs_decompress() call in __open_dso()
  perf tools: Get rid of dso__needs_decompress() call in symbol__disassemble()
  perf tools: Get rid of dso__needs_decompress() call in read_object_code()
  tools lib traceevent: Change to SPDX License format
  perf llvm: Allow passing options to llc in addition to clang
  perf parser: Improve error message for PMU address filters
  ...

5 years agoMerge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 17:13:21 +0000 (10:13 -0700)]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Thomas Gleixner:

 - Correct the L1TF fallout on 32bit and the off by one in the 'too much
   RAM for protection' calculation.

 - Add a helpful kernel message for the 'too much RAM' case

 - Unbreak the VDSO in case that the compiler desides to use indirect
   jumps/calls and emits retpolines which cannot be resolved because the
   kernel uses its own thunks, which does not work for the VDSO. Make it
   use the builtin thunks.

 - Re-export start_thread() which was unexported when the 32/64bit
   implementation was unified. start_thread() is required by modular
   binfmt handlers.

 - Trivial cleanups

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/speculation/l1tf: Suggest what to do on systems with too much RAM
  x86/speculation/l1tf: Fix off-by-one error when warning that system has too much RAM
  x86/kvm/vmx: Remove duplicate l1d flush definitions
  x86/speculation/l1tf: Fix overflow in l1tf_pfn_limit() on 32bit
  x86/process: Re-export start_thread()
  x86/mce: Add notifier_block forward declaration
  x86/vdso: Fix vDSO build if a retpoline is emitted

5 years agoMerge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 26 Aug 2018 16:55:28 +0000 (09:55 -0700)]
Merge branch 'irq-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull irq update from Thomas Gleixner:
 "A small set of updats/fixes for the irq subsystem:

   - Allow GICv3 interrupts to be configured as wake-up sources to
     enable wakeup from suspend

   - Make the error handling of the STM32 irqchip init function work

   - A set of small cleanups and improvements"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/gic-v3: Allow interrupt to be configured as wake-up sources
  irqchip/tango: Set irq handler and data in one go
  dt-bindings: irqchip: renesas-irqc: Document r8a774a1 support
  irqchip/s3c24xx: Remove unneeded comparison of unsigned long to 0
  irqchip/stm32: Fix init error handling
  irqchip/bcm7038-l1: Hide cpu offline callback when building for !SMP