Laurent Pinchart [Tue, 27 Jan 2015 13:58:53 +0000 (15:58 +0200)]
dmaengine: rcar-dmac: Fix oops due to unintialized list in error ISR
The error interrupt handler stops and reinitializes all channels. This
causes a crash for channels that have never been used, as their
descriptor lists are uninitialized. Fix it by initializing the
descriptor lists at probe time.
Lad, Prabhakar [Wed, 4 Feb 2015 13:03:27 +0000 (13:03 +0000)]
dmaenegine: edma: fix sparse warnings
this patch fixes following sparse warnings:
edma.c:537:32: warning: symbol 'edma_prep_dma_memcpy' was not declared. Should it be static?
edma.c:1070:6: warning: symbol 'edma_filter_fn' was not declared. Should it be static?
The desc variable is used uninitialized in the rcar_dmac_desc_get() and
rcar_dmac_xfer_chunk_get() functions if descriptors need to be
allocated. Fix it.
Sergei Shtylyov [Tue, 20 Jan 2015 21:13:21 +0000 (00:13 +0300)]
dmaengine: shdmac: extend PM methods
In order to make it possible to restore from hibernation not only in Linux but
also in e.g. U-Boot, we have to use sh_dmae_{suspend|resume}() for the {freeze|
thaw|restore}() PM methods. It's handy to achieve this with SIMPLE_DEV_PM_OPS()
macro; since that macro doesn't do anything when CONFIG_PM_SLEEP is undefined,
we don't need to #define sh_dmae_{suspend|resume} NULL anymore but we'll have to
enclose sh_dmae_{suspend|resume}() into the new #ifdef...
Based on original patch by Mikhail Ulyanov <mikhail.ulyanov@cogentembedded.com>.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Sergei Shtylyov [Tue, 20 Jan 2015 21:09:46 +0000 (00:09 +0300)]
dmaengine: shdmac: use SET_RUNTIME_PM_OPS()
Use SET_RUNTIME_PM_OPS() to initialize the runtime PM method pointers in the
'struct dev_pm_ops'; since that macro doesn't do anything if CONFIG_PM is
not defined, we have to move #ifdef up to also cover the runtime PM methods
in order to avoid compilation warnings.
Based on orignal patch by Mikhail Ulyanov <mikhail.ulyanov@cogentembedded.com>.
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
dmaengine: at_xdmac: allow muliple dwidths when doing slave transfers
When using FIFO, we need to support differents data width in a single
transfer. For example, serial device which usually uses 1-byte data
width will use 4-bytes data width when using the FIFO. If the transfer
size is not aligned on 4-bytes then the end of the transfer will be
performed with 1-byte data-width. For that reason,
at_xdmac_prep_slave_sg() now builds linked list descriptors using view 2
instead of view 1 so each of them can update the DWIDTH field into the
Channel Configuration Register.
This patch simplifies the channel configuration register management.
Relying on a "software snapshot" of the configuration is not safe and
too complex.
Multiple dwidths will be introduced for slave transfers. In this case,
it becomes quite difficult to have an accurate snapshot of the channel
configuration register in the way it is done. Using the channel
configuration available in the lli descriptor simplifies this stuff.
Andy Shevchenko [Tue, 13 Jan 2015 17:08:13 +0000 (19:08 +0200)]
dmaengine: dw: amend description of dma_dev field
The dma_dev field is widely used in filter functions to mach with a proper DMA
controller device. Thus it's not deprecated. The patch fixes the description of
that field. There is no functional change.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Add support for the IMG Multi-threaded DMA Controller (MDC) found on
certain IMG SoCs. Currently this driver supports the variant present
on the MIPS-based Pistachio SoC.
Andy Shevchenko [Mon, 19 Jan 2015 16:23:06 +0000 (18:23 +0200)]
dmaengine: dw: update MAINTAINERS file
This is a follow up to the previously done changes in the layout of the driver
files. We now have an additional file include/linux/dma/dw.h which is missed in
the MAINTAINERS data base.
Fixes: 3d588f83e4d6 (dmaengine: dw: split dma-dw.h to platform and private parts) Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
dmaengine: shdmac: fixup WARNING of slave caps retrieval
ecc19d17868be9c9f8f00ed928791533c420f3e0
(dmaengine: Add a warning for drivers not using the generic slave
caps retrieval) added WARN() for DMA_SLAVE.
Kernel will shows WARNING without this patch.
The driver doesn't support residue reporting at all.
residue_granularity should be set to DMA_RESIDUE_GRANULARITY_DESCRIPTOR.
Special thanks to Laurent
Laurent Pinchart [Mon, 19 Jan 2015 11:54:27 +0000 (13:54 +0200)]
dmaengine: Add dma_get_slave_caps() inline stub when !CONFIG_DMA_ENGINE
Commit 0d5484b1c3db8a38 ("dmaengine: Move dma_get_slave_caps()
implementation to dmaengine.c") turned the inline dma_get_slave_caps()
function into an external function without adding an inline stub for the
cases where CONFIG_DMA_ENGINE isn't set. This breaks compilation of
drivers using the DMA engine API when CONFIG_DMA_ENGINE isn't set.
Arnd Bergmann [Tue, 13 Jan 2015 21:17:03 +0000 (22:17 +0100)]
dmaengine: coh901318: fix function return types build warnings
A recent patch that removed coh901318_control() replaced it
with a number of pointers to existing functions, but those
unfortunately have the wrong return type and need to be
changed to return an 'int' with an error value rather than
a 'void' to avoid these build warnings:
drivers/dma/coh901318.c:2697:32: warning: assignment from incompatible pointer type
base->dma_slave.device_config = coh901318_dma_set_runtimeconfig;
^
drivers/dma/coh901318.c:2698:31: warning: assignment from incompatible pointer type
base->dma_slave.device_pause = coh901318_pause;
^
drivers/dma/coh901318.c:2699:32: warning: assignment from incompatible pointer type
base->dma_slave.device_resume = coh901318_resume
The coh901318_base_init function has the correct return type
already, but needs to be marked 'static' to avoid a sparse
warning about a missing declaration.
After commit ecc19d17868be9c9f8f00ed928791533c420f3e0 ("dmaengine: Add
a warning for drivers not using the generic slave caps retrieval"),
the Tegra APB DMA driver causes this warning during boot:
WARNING: CPU: 0 PID: 1 at drivers/dma/dmaengine.c:830 dma_async_device_register+0x294/0x538()
this driver doesn't support generic slave capabilities reporting
Fix by setting the appropriate reporting structure fields that are
passed to dma_async_device_register().
Signed-off-by: Paul Walmsley <paul@pwsan.com> Tested-by: Thierry Reding <treding@nvidia.com> Acked-by: Thierry Reding <treding@nvidia.com> Acked-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Kevin Hao [Thu, 8 Jan 2015 10:38:17 +0000 (18:38 +0800)]
dmaengine: fsldma: remove the unused variable
Fix the following build warning:
drivers/dma/fsldma.c: In function 'fsl_dma_device_terminate_all':
drivers/dma/fsldma.c:947:6: warning: unused variable 'size' [-Wunused-variable]
Signed-off-by: Kevin Hao <haokexin@gmail.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
dmaengine: rcar-hpbdma: fixup WARNING of slave caps retrieval
ecc19d17868be9c9f8f00ed928791533c420f3e0
(dmaengine: Add a warning for drivers not using the generic slave
caps retrieval) added WARN() for DMA_SLAVE.
Kernel will shows WARNING without this patch.
Andy Shevchenko [Fri, 2 Jan 2015 14:17:24 +0000 (16:17 +0200)]
dmaengine: dw: provide DMA capabilities
The new DMAEngine requirement is to provide what the DMA controller can do,
such as directions, bus widths, and residue granularity. The patch sets those
properties for the DesignWare DMA controller driver.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
In a recent cleanup, the mmp_tdma_terminate_all function was
introduced but does not set a proper return value. Almost
no slave driver uses that return value, but if one does, the
result will be undefined, which the compiler warns about:
dma/mmp_tdma.c: In function 'mmp_tdma_terminate_all':
dma/mmp_tdma.c:474:1: warning: no return statement in function returning non-void [-Wreturn-type]
This changes the driver to return zero, like most other
drivers do.
Arnd Bergmann [Tue, 13 Jan 2015 13:25:19 +0000 (14:25 +0100)]
dmaengine: mmp-tdma: don't include mach/regs-icu.h
The mmp tdma driver does not actually require this header, and
we want to enable multiplatform support for MMP, which would
make it inaccessible and cause a build error.
Arnd Bergmann [Tue, 13 Jan 2015 13:23:13 +0000 (14:23 +0100)]
dmaengine: k3: fix duplicate function definition
Commit db08425ebd51f ("dmaengine: k3: Split device_control") introduced
two new helper functions, which unfortunately have the same names
as the existing suspend/resume functions, resulting in a build error
when CONFIG_PM_SLEEP is enabled:
drivers/dma/k3dma.c:823:12: error: conflicting types for 'k3_dma_resume'
static int k3_dma_resume(struct device *dev)
^
drivers/dma/k3dma.c:625:12: note: previous definition of 'k3_dma_resume' was here
static int k3_dma_resume(struct dma_chan *chan)
^
Signed-off-by: Arnd Bergmann <arnd@arndb.de> Fixes: db08425ebd51f ("dmaengine: k3: Split device_control") Reported-by: Mark Brown <broonie@kernel.org> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
dmaengine: k3: Fix duplicated function name and allmodconfig build
While splitting device control in db08425ebd51 ("dmaengine: k3:
Split device_control") new function with the same 'k3_dma_resume' name
was added, leading to build error:
drivers/dma/k3dma.c:823:12: error: conflicting types for ‘k3_dma_resume’
drivers/dma/k3dma.c:625:12: note: previous definition of ‘k3_dma_resume’ was here
Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
If the atomic DMA coherent pool is too small, disable use of hardware
descriptor lists instead of crashing the system:
ERROR: 256 KiB atomic DMA coherent pool is too small!
Please increase it with coherent_pool= kernel parameter!
Unable to handle kernel NULL pointer dereference at virtual address 00000004
Internal error: Oops: a07 [#1] PREEMPT SMP ARM
PC is at rcar_dmac_chan_reinit+0x3c/0x160
LR is at _raw_spin_lock_irqsave+0x18/0x5c
[<802132c0>] (rcar_dmac_chan_reinit) from [<80214818>] (rcar_dmac_isr_error+0x84/0xa0)
[<80214818>] (rcar_dmac_isr_error) from [<80060484>] (handle_irq_event_percpu+0x50/0x150)
[<80060484>] (handle_irq_event_percpu) from [<800605c0>] (handle_irq_event+0x3c/0x5c)
[<800605c0>] (handle_irq_event) from [<8006350c>] (handle_fasteoi_irq+0xb8/0x198)
[<8006350c>] (handle_fasteoi_irq) from [<8005fdb0>] (generic_handle_irq+0x20/0x30)
[<8005fdb0>] (generic_handle_irq) from [<8000fcd0>] (handle_IRQ+0x50/0xc4)
[<8000fcd0>] (handle_IRQ) from [<800092cc>] (gic_handle_irq+0x28/0x5c)
[<800092cc>] (gic_handle_irq) from [<80012700>] (__irq_svc+0x40/0x70)
Kernel panic - not syncing: Fatal exception in interrupt
Unlike DMA transfers descriptors that are preallocated and cached,
memory used to store hardware descriptors is allocated and freed with
the DMA coherent allocation API for every transfer. Besides degrading
performances, this creates a CMA stress test that seems to cause issues.
Running dmatest with the noverify option produces
dmaengine: rcar-dmac: Implement support for hardware descriptor lists
The DMAC supports hardware-based auto-configuration from descriptor
lists. This reduces the number of interrupts required for processing a
DMA transfer. Support that mode in the driver.
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Tested-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
The DMAC is a general purpose multi-channel DMA controller that supports
both slave and memcpy transfers.
The driver currently supports the DMAC found in the r8a7790 and r8a7791
SoCs. Support for compatible DMA controllers (such as the audio DMAC)
will be added later.
Feature-wise, automatic hardware handling of descriptors chains isn't
supported yet. LPAE support is implemented.
Vinod Koul [Mon, 8 Dec 2014 06:00:17 +0000 (11:30 +0530)]
dmaengine: tegra: fix incompatible pointer type warns
drivers/dma/tegra20-apb-dma.c:1428:37: warning: assignment from incompatible pointer type [enabled by default]
drivers/dma/ste_dma40.c: In function 'd40_terminate_all':
The function prototype expects return type 'int' whereas these where void
Vinod Koul [Mon, 8 Dec 2014 05:57:08 +0000 (11:27 +0530)]
dmaengine: ste_dma: fix incompatible pointer type warns
drivers/dma/ste_dma40.c:2627:3: warning: 'return' with a value, in function returning void [enabled by default]
drivers/dma/ste_dma40.c: In function 'd40_ops_init':
drivers/dma/ste_dma40.c:2869:28: warning: assignment from incompatible pointer type [enabled by default]
The function prototype expects return type 'int' whereas these where void
Vinod Koul [Mon, 8 Dec 2014 05:54:09 +0000 (11:24 +0530)]
dmaengine: mxs-dma: fix incompatible pointer type build warns
drivers/dma/mxs-dma.c: In function 'mxs_dma_probe':
drivers/dma/mxs-dma.c:848:35: warning: assignment from incompatible pointer type [enabled by default]
drivers/dma/mxs-dma.c:849:36: warning: assignment from incompatible pointer type [enabled by default]
The function prototype expects return type 'int' whereas these where void
Maxime Ripard [Mon, 17 Nov 2014 13:42:53 +0000 (14:42 +0100)]
dmaengine: Add a warning for drivers not using the generic slave caps retrieval
For the slave caps retrieval to be really useful, most drivers need to
implement it.
Hence, we need to be slightly more aggressive, and trigger a warning at
registration time for drivers that don't fill their caps infos in order to
encourage them to implement it.
Maxime Ripard [Mon, 17 Nov 2014 13:42:43 +0000 (14:42 +0100)]
dmaengine: rapidio: tsi721: Rename device_control
Rename the device_control callback of the TXX9 DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:42 +0000 (14:42 +0100)]
dmaengine: txx9: Rename device_control
Rename the device_control callback of the TXX9 DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:41 +0000 (14:42 +0100)]
dmaengine: td: Rename device_control
Rename the device_control callback of the Timberdal DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:40 +0000 (14:42 +0100)]
dmaengine: pch-dma: Rename device_control
Rename the device_control callback of the Intel PCH DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:39 +0000 (14:42 +0100)]
dmaengine: mv_xor: Remove device_control
The Marvell XOR engine doesn't allow any operations that use to be defined in
device_control, it shouldn't need to be defined. Since it's going to be
deprecated, remove it altogether.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:38 +0000 (14:42 +0100)]
dmaengine: xilinx: Split device_control
Split the device_control callback of the Xilinx VDMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Maxime Ripard [Mon, 17 Nov 2014 13:42:37 +0000 (14:42 +0100)]
dmaengine: tegra20: Split device_control
Split the device_control callback of the NVidia Tegra20 APB DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:36 +0000 (14:42 +0100)]
dmaengine: d40: Split device_control
Split the device_control callback of the ST-Ericsson DMA 40 driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Maxime Ripard [Mon, 17 Nov 2014 13:42:35 +0000 (14:42 +0100)]
dmaengine: sun6i: Split device_control
Split the device_control callback of the Allwinner A31 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:34 +0000 (14:42 +0100)]
dmaengine: sirf: Split device_control
Split the device_control callback of the SiRF Prima 2 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:33 +0000 (14:42 +0100)]
dmaengine: sh: Split device_control
Split the device_control callback of the Super-H DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Maxime Ripard [Mon, 17 Nov 2014 13:42:32 +0000 (14:42 +0100)]
dmaengine: sa11x0: Split device_control
Split the device_control callback of the SA-11x0 DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:31 +0000 (14:42 +0100)]
dmaengine: s3c24xx: Split device_control
Split the device_control callback of the Samsung S3C24xxx DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:30 +0000 (14:42 +0100)]
dmaengine: bam-dma: Split device_control
Split the device_control callback of the Qualcomm BAM DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:29 +0000 (14:42 +0100)]
dmaengine: pl330: Split device_control
Split the device_control callback of the AMBA PL330 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:28 +0000 (14:42 +0100)]
dmaengine: omap: Split device_control
Split the device_control callback of the TI OMAP DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:27 +0000 (14:42 +0100)]
dmaengine: nbpfaxi: Split device_control
Split the device_control callback of the NBPF AXI DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:26 +0000 (14:42 +0100)]
dmaengine: mxs: Split device_control
Split the device_control callback of the Freescale MXS DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:25 +0000 (14:42 +0100)]
dmaengine: mpc512x: Split device_control
Split the device_control callback of the Freescale MPC512x DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:24 +0000 (14:42 +0100)]
dmaengine: fsl-dma: Split device_control
Split the device_control callback of the Freescale Elo DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
While we're at it, remove the useless prep_sg callback.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:23 +0000 (14:42 +0100)]
dmaengine: moxart: Split device_control
Split the device_control callback of the Moxart DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:22 +0000 (14:42 +0100)]
dmaengine: mmp-tdma: Split device_control
Split the device_control callback of the Marvell MMP TDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Maxime Ripard [Mon, 17 Nov 2014 13:42:21 +0000 (14:42 +0100)]
dmaengine: mmp-pdma: Split device_control
Split the device_control callback of the Marvell MMP PDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com> Signed-off-by: Vinod Koul <vinod.koul@intel.com>