The clock must be running for the zero-crossing mute functionality.
However, it must be gated for VDD-SYS to be turned off during system
suspend. Disable it in the suspend callback, after everything has
already been muted, to avoid pops when muting/unmuting outputs.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Acked-by: Chen-Yu Tsai <wens@csie.org>
Link: https://lore.kernel.org/r/20200726025334.59931-3-samuel@sholland.org
Signed-off-by: Mark Brown <broonie@kernel.org>
The same enable bits are currently used for both the "Left/Right ADC"
and the "Left/Right ADC Mixer" widgets. This happens to work in practice
because the widgets are always enabled/disabled at the same time, but
each register bit should only be associated with a single widget.
To keep symmetry with the DAC widgets, keep the bits on the ADC widgets,
and remove them from the ADC Mixer widgets.
Fixes: 42371f327d ("ASoC: sunxi: Add new driver for Allwinner A64 codec's analog path controls")
Reported-by: Ondrej Jirman <megous@megous.com>
Signed-off-by: Samuel Holland <samuel@sholland.org>
Acked-by: Chen-Yu Tsai <wens@csie.org>
Link: https://lore.kernel.org/r/20200726025334.59931-2-samuel@sholland.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Fix cppcheck warnings:
sound/soc/intel/boards/bdw-rt5650.c:91:23: style: Local variable
'channels' shadows outer variable [shadowVariable]
sound/soc/intel/boards/bdw-rt5677.c:144:23: style: Local variable
'channels' shadows outer variable [shadowVariable]
sound/soc/intel/boards/broadwell.c:91:23: style: Local variable
'channels' shadows outer variable [shadowVariable]
This was fixed earlier in other machine drivers but keeps coming back
with copy/paste.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200813175839.59422-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Cppcheck reports the following warning:
sound/soc/sof/intel/hda-codec.c:191:1: style: Label 'error' is not
used. [unusedLabel]
This label is indeed only used conditionally, move it where it's
actually used.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200813175839.59422-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
On the A64, as tested using the PinePhone, the current code causes the
left/right channels to be swapped during I2S playback from the CPU on
AIF1, and breaks DSP_A communication with the modem on AIF2. Both of
these are fixed when LRCK is no longer inverted.
Trusting that the comment in the code is correct, the existing behavior
is kept for the A33.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Link: https://lore.kernel.org/r/20200726012557.38282-5-samuel@sholland.org
Signed-off-by: Mark Brown <broonie@kernel.org>
The sun8i-codec driver provides ALSA controls for enabling/disabling
each of the inputs to the AIF1 Slot 0 and DAC mixers. For two of these
inputs (ADC->DAC and AIF1 DA0->AIF1 AD0), the audio source is
implemented, so the mixer inputs can be used.
However, because the DAPM routes are missing, these mixer inputs only
work when both the source and the mixer happen to be part of other
active audio paths. Adding the appropriate routes makes these ALSA
controls function all of the time.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Link: https://lore.kernel.org/r/20200726012557.38282-4-samuel@sholland.org
Signed-off-by: Mark Brown <broonie@kernel.org>
The A33/A64 digital codec has 4 physical inputs and 4 physical outputs:
3 AIFs/DAIs and one ADC/DAC pair. Internal routing is accomplished by
a 4-channel mixer connected to each output.
The analog and digital sides of the ADC/DAC are in separate ASoC
components, so card-level DAPM routes (provided in the device tree) are
necessary to connect them together. Currently, these routes are wrong.
For AIF1 Playback, the correct topology is:
||<<============ sun8i-codec ===========>>||
|| ||
CPU DAI -> AIF1 DA0 -> DAC Mixer -> DAC (digital) -> DAC (analog)
|| ||
but the driver and device trees currently describe:
|| ||
CPU DAI -> AIF1 DA0 -------------------------------> DAC (analog)
|| \--> DAC Mixer -> ??? [dead end] ||
For AIF1 Capture, there is an additional problem, because the Mixer
route is backward. The topology should be:
|| ||
ADC (analog) -> ADC (digital) -> AIF1 AD0 Mixer -> AIF1 AD0 -> CPU DAI
|| ||
but the driver and device trees currently describe:
|| ||
ADC (analog) -> AIF1 AD0 ------------------------------------> CPU DAI
|| \--> ADC Mixer -> ??? [dead end] ||
The ADC/DAC are only powered because AIF1 AD0 (capture) has supply
routes from the ADC, and AIF1 DA0 (playback) has supply routes from the
DAC. However, neither set of supply routes matches the hardware
topology. Audio can be routed among AIF1/2/3 without using the ADC or
DAC at all; and audio can be routed from the ADC to the DAC without
using any AIFs (via the "ADC Digital DAC Playback Switch"). Because the
DAPM routes are wrong, both of these use cases are currently broken.
This commit adds the necessary widgets and routes to represent the real
hardware topology, with functionality equivalent to the current driver.
For the existing "allwinner,sun8i-a33-codec" compatible, widgets with
the old names are kept as wrappers around the new widgets, so existing
device trees will continue to work. For "allwinner,sun50i-a64-codec",
the old widgets can be omitted, because no device trees yet use that
compatible.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Link: https://lore.kernel.org/r/20200726012557.38282-3-samuel@sholland.org
Signed-off-by: Mark Brown <broonie@kernel.org>
The audio codecs in the A33 and A64 are both integrated variants of the
X-Powers AC100 codec. However, there are some differences between them
that merit having a separate compatible:
- The A64 has a second DRC block, not present in the AC100 or A33.
- The A33 has some extra muxing options for AIF1/2/3 in the
AIF3_SGP_CTRL register, which are not present in the AC100 or A64.
- The A33 is missing registers providing jack detection functionality.
- The A33 is claimed to invert LRCK, but this is not seen on A64.
Since the driver will continue to work on the A64 using the A33
compatible, albeit without jack detection functionality and with
possibly inverted channels, as it does now, allow the A33 compatible
to be used as a fallback.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20200726012557.38282-2-samuel@sholland.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Fixes the register address for the timer3 entry on Arria10.
Fixes: 475dc86d08 ("arm: dts: socfpga: Add a base DTSI for Altera's Arria10 SOC")
Signed-off-by: Dinh Nguyen <dinguyen@kernel.org>
1f23816b8e ("virtio_blk: add discard and write zeroes support") starts
to support multi-range discard for virtio-blk. However, the virtio-blk
disk may report max discard segment as 1, at least that is exactly what
qemu is doing.
So far, block layer switches to normal request merge if max discard segment
limit is 1, and multiple bios can be merged to single segment. This way may
cause memory corruption in virtblk_setup_discard_write_zeroes().
Fix the issue by handling single max discard segment in straightforward
way.
Fixes: 1f23816b8e ("virtio_blk: add discard and write zeroes support")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Changpeng Liu <changpeng.liu@intel.com>
Cc: Daniel Verkamp <dverkamp@chromium.org>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When queue_max_discard_segments(q) is 1, blk_discard_mergable() will
return false for discard request, then normal request merge is applied.
However, only queue_max_segments() is checked, so max discard segment
limit isn't respected.
Check max discard segment limit in the request merge code for fixing
the issue.
Discard request failure of virtio_blk is fixed.
Fixes: 6984046608 ("block: fix the DISCARD request merge")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In case of block device backend, if the backend supports write zeros, the
loop device will set queue flag of QUEUE_FLAG_DISCARD. However,
limits.discard_granularity isn't setup, and this way is wrong,
see the following description in Documentation/ABI/testing/sysfs-block:
A discard_granularity of 0 means that the device does not support
discard functionality.
Especially 9b15d109a6 ("block: improve discard bio alignment in
__blkdev_issue_discard()") starts to take q->limits.discard_granularity
for computing max discard sectors. And zero discard granularity may cause
kernel oops, or fail discard request even though the loop queue claims
discard support via QUEUE_FLAG_DISCARD.
Fix the issue by setup discard granularity and alignment.
Fixes: c52abf5630 ("loop: Better discard support for block devices")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Coly Li <colyli@suse.de>
Cc: Hannes Reinecke <hare@suse.com>
Cc: Xiao Ni <xni@redhat.com>
Cc: Martin K. Petersen <martin.petersen@oracle.com>
Cc: Evan Green <evgreen@chromium.org>
Cc: Gwendal Grignou <gwendal@chromium.org>
Cc: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>
Cc: Andrzej Pietrasiewicz <andrzej.p@collabora.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Fix the tdmout inverter of the g12a and following SoC families.
This inverter is special and needs two bits to be the inverse of each other
for the inverter to operate properly.
Fixes: 075001385c ("clk: meson: axg-audio: add g12a support")
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20200729154359.1983085-4-jbrunet@baylibre.com
There are more differences than what we initially thought.
Let's keeps things clear and separate the axg and g12a regmap tables of the
audio clock controller.
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20200729154359.1983085-3-jbrunet@baylibre.com
Xiumei reported a bug with espintcp over IPv6 in transport mode,
because xfrm6_transport_finish expects to find IP6CB data (struct
inet6_skb_cb). Currently, espintcp zeroes the CB, but the relevant
part is actually preserved by previous layers (first set up by tcp,
then strparser only zeroes a small part of tcp_skb_tb), so we can just
relocate it to the start of skb->cb.
Fixes: e27cca96cd ("xfrm: add espintcp (RFC 8229)")
Reported-by: Xiumei Mu <xmu@redhat.com>
Signed-off-by: Sabrina Dubroca <sd@queasysnail.net>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
This is yet another simple but odd driver for the audio block of the g12a
and sm1 SoC families.
For TDMOUT's sclk to be properly inverted, bit 29 of
AUDIO_CLK_TDMOUT_x_CTRL should be the inverse of bit 28.
IOW bit28 == !bit29 at all times
This setting is automatically applied on axg and the manual setting was
added on g12a.
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20200729154359.1983085-2-jbrunet@baylibre.com
SCHED_RESTART code path is relied to re-run queue for dispatch requests
in hctx->dispatch. Meantime the SCHED_RSTART flag is checked when adding
requests to hctx->dispatch.
memory barriers have to be used for ordering the following two pair of OPs:
1) adding requests to hctx->dispatch and checking SCHED_RESTART in
blk_mq_dispatch_rq_list()
2) clearing SCHED_RESTART and checking if there is request in hctx->dispatch
in blk_mq_sched_restart().
Without the added memory barrier, either:
1) blk_mq_sched_restart() may miss requests added to hctx->dispatch meantime
blk_mq_dispatch_rq_list() observes SCHED_RESTART, and not run queue in
dispatch side
or
2) blk_mq_dispatch_rq_list still sees SCHED_RESTART, and not run queue
in dispatch side, meantime checking if there is request in
hctx->dispatch from blk_mq_sched_restart() is missed.
IO hang in ltp/fs_fill test is reported by kernel test robot:
https://lkml.org/lkml/2020/7/26/77
Turns out it is caused by the above out-of-order OPs. And the IO hang
can't be observed any more after applying this patch.
Fixes: bd166ef183 ("blk-mq-sched: add framework for MQ capable IO schedulers")
Reported-by: kernel test robot <rong.a.chen@intel.com>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Jeffery <djeffery@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
As new function fsl_sai_dir_is_synced is included for checking if
stream is synced by the opposite stream, then replace the existing
synchronous checking with this new function.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Acked-by: Nicolin Chen <nicoleotsuka@gmail.com>
Link: https://lore.kernel.org/r/20200805063413.4610-4-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Tx synchronous with Rx: The RMR is the word mask register, it is used
to mask any word in the frame, it is not relating to clock generation,
So it is no need to be changed when Tx is going to be enabled.
Rx synchronous with Tx: The TMR is the word mask register, it is used
to mask any word in the frame, it is not relating to clock generation,
So it is no need to be changed when Rx is going to be enabled.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Nicolin Chen <nicoleotsuka@gmail.com>
Link: https://lore.kernel.org/r/20200805063413.4610-3-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Current code enables TCSR.TE and RCSR.RE together, and disable
TCSR.TE and RCSR.RE together in trigger(), which only supports
one operation mode：
1. Rx synchronous with Tx: TE is last enabled and first disabled
Other operation mode need to be considered also:
2. Tx synchronous with Rx: RE is last enabled and first disabled.
3. Asynchronous mode: Tx and Rx are independent.
So the enable TCSR.TE and RCSR.RE sequence and the disable
sequence need to be refined accordingly for #2 and #3.
There is slightly against what RM recommennds with this change.
For example in Rx synchronous with Tx mode, case "aplay 1.wav;
arecord 2.wav" enable TE before RE. But it should be safe to
do so, judging by years of testing results.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Nicolin Chen <nicoleotsuka@gmail.com>
Link: https://lore.kernel.org/r/20200805063413.4610-2-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Configure the SAI device node, configure audio clock
and pinctrl.
Enable the audio sound card, which use the SAI1 and
wm8960, and enable headphone detection.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Add header defining dai-id and mclk id for SC7180 lpass soc.
Signed-off-by: Ajit Pandey <ajitp@codeaurora.org>
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/1597402388-14112-9-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
platform_get_irq_byname() is used when there is list
of interrupts in the device node. As lpass-platform
has only one interrupt entry, use platform_get_irq()
instead.
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Link: https://lore.kernel.org/r/1597402388-14112-12-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
platform_get_resource_byname() is used when there
is list of reg entries. As lpass-cpu node has only
one reg entry, use platform_get_resource() instead.
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/1597402388-14112-11-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
i2sctl register value is set to 0 during hw_free(). This
impacts any ongoing concurrent session on the same i2s
port. As trigger() stop already resets enable bit to 0,
there is no need of explicit hw_free. Removing it to
fix the issue.
Fixes: 80beab8e1d ("ASoC: qcom: Add LPASS CPU DAI driver")
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/1597402388-14112-7-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
I2SCTL and DMACTL registers has different bits alignment for newer
LPASS variants of SC7180 soc. Use REG_FIELD_ID() to define the
reg_fields in platform specific file and removed shifts and mask
macros for such registers from header file.
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Link: https://lore.kernel.org/r/1597402388-14112-6-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
lpass_pcm_data is never freed. Free it in close
ops to avoid memory leak.
Fixes: 022d00ee0b ("ASoC: lpass-platform: Fix broken pcm data usage")
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/1597402388-14112-5-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
We are allocating dma memory for component->dev but trying to mmap
such memory for substream->pcm->card->dev. Replace device argument
in mmap with component->dev to fix this.
Signed-off-by: Ajit Pandey <ajitp@codeaurora.org>
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/1597402388-14112-4-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Ahbix clock is optional clock and not needed for all platforms.
Move it to lpass-apq8016/ipq806x as it is not needed for sc7180.
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/1597402388-14112-3-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
LPASS variants have their own soc specific clocks that needs to be
enabled for MI2S audio support. Added a common variable in drvdata to
initialize such clocks using bulk clk api. Such clock names is
defined in variants specific data and needs to fetched during init.
Signed-off-by: Ajit Pandey <ajitp@codeaurora.org>
Signed-off-by: Rohit kumar <rohitkr@codeaurora.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/1597402388-14112-2-git-send-email-rohitkr@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Commit ("03fd42d458 powerpc/fixmap: Fix FIX_EARLY_DEBUG_BASE when
page size is 256k") reworked the setup of the early debug area and
mistakenly replaced 128 * 1024 by SZ_128.
Change to SZ_128K to restore the original 128 kbytes size of the area.
Fixes: 03fd42d458 ("powerpc/fixmap: Fix FIX_EARLY_DEBUG_BASE when page size is 256k")
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/996184974d674ff984643778cf1cdd7fe58cc065.1597644194.git.christophe.leroy@csgroup.eu
IS_ENABLED() instead of #ifdef still requires variable declaration.
In this specific case, default_uamor is declared in asm/pkeys.h which
is only included if PPC_MEM_KEYS is enabled.
arch/powerpc/mm/book3s64/hash_utils.c: In function ‘hash__early_init_mmu_secondary’:
arch/powerpc/mm/book3s64/hash_utils.c:1119:21: error: ‘default_uamor’ undeclared (first use in this function)
1119 | mtspr(SPRN_UAMOR, default_uamor);
| ^~~~~~~~~~~~~
Fixes: 6553fb799f ("powerpc/pkeys: Fix boot failures with Nemo board (A-EON AmigaOne X1000)")
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20200817103301.158836-1-aneesh.kumar@linux.ibm.com
Add device trees for TQMa6S, TQMa6Q, and TQMa6QP embedded modules.
The A and B SoM variants are for hardware revisions that differ in
how the I2C devices are connected. For details, see [1].
This is a combination of the patches
"arm: dt: imx6qdl: add tqma6[qdl] som on mba6 mainboard",
"arm: dt: tqma6: add spi with spi nor flash on SOM",
"arm: dt: add basic support for tqma6qp on mba6", and
"arm: dt: imx6qdl-tqma6: use generic jedec,spi-nor"
from the TQMa6x BSP, with the MBa6 specific parts removed.
[1] https://support.tq-group.com/en/arm/tqma6x/linux/ptxdist/overview
Signed-off-by: Markus Niebel <Markus.Niebel@tq-group.com>
[bst@pengutronix.de: remove unnecessary container node in the iomuxc
node]
Signed-off-by: Bastian Krause <bst@pengutronix.de>
[p.zabel@pengutronix.de: merged patches from TQMa6x BSP REV.0114,
separated MBa6 DTs, fixed checkpatch and dtbs_check warnings,
added no-sd(io) properties to eMMC, added SPDX license identifiers
and commit message]
Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Disable Ethernet switch reset GPIO with ZII platforms that have it
enabled. HW switch reset results in a reset of the copper PHYs
inside of the switch. We want to avoid this reset of the copper PHYs
in the switch as this results in unnecessary broader network disruption on
a soft reboot of the application processor.
With the HW GPIO removed, the switch driver still performs a soft reset of
the switch core which has been shown to sufficiently meet our needs with
other ZII platforms that do not have the HW switch reset GPIO defined.
Signed-off-by: Chris Healy <cphealy@gmail.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Add the reset-gpio property, and enable PCIe on iMX6QP SABREAUTO board.
Signed-off-by: Richard Zhu <hongxing.zhu@nxp.com>
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
The (new?) style of clk registration uses clk_hw based APIs so that we
can more easily see the difference between clk providers and clk
consumers. Use the clk_hw based APIs to do this and migrate to devm for
the clkdev creation so that we can reduce the amount of code.
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Cc: Shuming Fan <shumingf@realtek.com>
Link: https://lore.kernel.org/r/20200804000531.920688-4-swboyd@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>
The __clk_get_name() API is deprecated. Use clk_hw_get_name() or
proper registration techniques to avoid it.
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Cc: Shuming Fan <shumingf@realtek.com>
Link: https://lore.kernel.org/r/20200804000531.920688-3-swboyd@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>
I see a spew of "sysclk/dai not set correctly" whenever I cat
/sys/kernel/debug/clk/clk_summary on my device. This is because the
master pointer isn't set yet in this driver. A user isn't going to be
able to do much if this check is failing so this error message isn't
really an error, it's more of a kernel debug message. Lower the priority
to dev_dbg() so that it isn't so noisy.
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Cc: Shuming Fan <shumingf@realtek.com>
Link: https://lore.kernel.org/r/20200804000531.920688-2-swboyd@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>
Currently dwc3 doesn't handle usb_request->zero for SG requests. This
change checks and prepares extra TRBs for the ZLP for SG requests.
Cc: <stable@vger.kernel.org> # v4.5+
Fixes: 04c03d10e5 ("usb: dwc3: gadget: handle request->zero")
Signed-off-by: Thinh Nguyen <thinhn@synopsys.com>
Signed-off-by: Felipe Balbi <balbi@kernel.org>
The usb_request->zero doesn't apply for isoc. Also, if we prepare a
0-length (ZLP) TRB for the OUT direction, we need to prepare an extra
TRB to pad up to the MPS alignment. Use the same bounce buffer for the
ZLP TRB and the extra pad TRB.
Cc: <stable@vger.kernel.org> # v4.5+
Fixes: d6e5a549cc ("usb: dwc3: simplify ZLP handling")
Fixes: 04c03d10e5 ("usb: dwc3: gadget: handle request->zero")
Signed-off-by: Thinh Nguyen <thinhn@synopsys.com>
Signed-off-by: Felipe Balbi <balbi@kernel.org>
The SG list may be set up with entry size more than the requested
length. Check the usb_request->length and make sure that we don't setup
the TRBs to send/receive more than requested. This case may occur when
the SG entry is allocated up to a certain minimum size, but the request
length is less than that. It can also occur when the request is reused
for a different request length.
Cc: <stable@vger.kernel.org> # v4.18+
Fixes: a31e63b608 ("usb: dwc3: gadget: Correct handling of scattergather lists")
Signed-off-by: Thinh Nguyen <thinhn@synopsys.com>
Signed-off-by: Felipe Balbi <balbi@kernel.org>
The code modifies rdev, but locks c_rdev instead. Remove the lock
as this is held together by regulator_list_mutex taken in the caller.
Fixes: f9503385b1 ("regulator: core: Mutually resolve regulators coupling")
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Reviewed-by: Dmitry Osipenko <digetx@gmail.com>
Link: https://lore.kernel.org/r/25eb81cefb37a646f3e44eaaf1d8ae8881cfde52.1597195321.git.mirq-linux@rere.qmqm.pl
Signed-off-by: Mark Brown <broonie@kernel.org>
Since only regulator_ena_gpio_request() allocates rdev->ena_pin, and it
guarantees that same gpiod gets same pin structure, it is enough to
compare just the pointers. Also we know there can be only one matching
entry on the list. Rework the code take advantage of the facts.
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Link: https://lore.kernel.org/r/3ff002c7aa3bd774491af4291a9df23541fcf892.1597195321.git.mirq-linux@rere.qmqm.pl
Signed-off-by: Mark Brown <broonie@kernel.org>
