Refactor the dmaengine PCM library to allow the DMA channel to be requested
before opening a PCM substream. snd_dmaengine_pcm_open() now expects a DMA
channel instead of a filter function and filter parameter as its parameters.
snd_dmaengine_pcm_close() is updated to not release the DMA channel. This allows
a dmaengine based PCM driver to request its channels before the substream is
opened.

The patch also introduces two new functions, snd_dmaengine_pcm_open_request_chan()
and snd_dmaengine_pcm_close_release_chan(), which have the same signature and
behaviour of the old snd_dmaengine_pcm_{open,close}() and internally use the new
variants of these functions. All users of snd_dmaengine_pcm_{open,close}() are
updated to use snd_dmaengine_pcm_open_request_chan() and
snd_dmaengine_pcm_close_release_chan().
Signed-off-by: NLars-Peter Clausen <lars@metafoo.de>
Tested-by: NStephen Warren <swarren@nvidia.com>
Tested-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Use the common DAI DMA data struct for fsl/imx, this allows us to use the common
helper function to configure the DMA slave config based on the DAI DMA data.
Signed-off-by: NLars-Peter Clausen <lars@metafoo.de>
Tested-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Usually device_fc should be set to false for audio DMAs. Initialize it in a
common place so drivers don't have to do this manually.
Signed-off-by: NLars-Peter Clausen <lars@metafoo.de>
Tested-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: NNicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Currently the imx_dma_data struct, which gets passed to the dmaengine driver, is
allocated and constructed in the pcm driver from the data stored in the
dma_params struct. The dma_params struct gets passed to the pcm driver from the
dai driver. Instead of going this route of indirection embed the dma_data struct
directly into the dma_params struct and let the dai driver fill it in. This
allows us to simplify the imx-pcm-dma driver quite a bit, since it doesn't have
care about memory managing the imx_dma_data struct anymore.
Signed-off-by: NLars-Peter Clausen <lars@metafoo.de>
Tested-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Remove probe function from the init section.
Signed-off-by: NMarkus Pargmann <mpa@pengutronix.de>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: NMarkus Pargmann <mpa@pengutronix.de>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

Fix the following sparse warning:

sound/soc/fsl/imx-audmux.c:182:3: warning: symbol 'audmux_type' was not declared. Should it be static?
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

When building modules with CONFIG_SND_IMX_SOC=m in imx_v6_v7_defconfig,
we will see the following link error.

  LD [M]  sound/soc/fsl/snd-soc-fsl-ssi.o
  LD [M]  sound/soc/fsl/snd-soc-fsl-utils.o
  LD [M]  sound/soc/fsl/snd-soc-imx-ssi.o
  LD [M]  sound/soc/fsl/snd-soc-imx-audmux.o
  LD [M]  sound/soc/fsl/snd-soc-imx-pcm.o
sound/soc/fsl/imx-pcm-dma.o: In function `init_module':
imx-pcm-dma.c:(.init.text+0x0): multiple definition of `init_module'
sound/soc/fsl/imx-pcm-fiq.o:imx-pcm-fiq.c:(.init.text+0x0): first defined here
sound/soc/fsl/imx-pcm-dma.o: In function `cleanup_module':
imx-pcm-dma.c:(.exit.text+0x0): multiple definition of `cleanup_module'
sound/soc/fsl/imx-pcm-fiq.o:imx-pcm-fiq.c:(.exit.text+0x0): first defined here
make[4]: *** [sound/soc/fsl/snd-soc-imx-pcm.o] Error 1

The module snd-soc-imx-pcm is designed to link imx-pcm.o with
imx-pcm-dma.o or imx-pcm-fiq.o depending on if option SND_SOC_IMX_PCM_DMA
or SND_SOC_IMX_PCM_FIQ is enabled.  Both imx-pcm-dma and imx-pcm-fiq
register their own module_platform_driver.  However, these two options
are not mutually exclusive and can be enabled together.  And that's
why we see above multiple init_module definition error.

Instead of having both imx-pcm-dma and imx-pcm-fiq register their
own platform_driver, we should do only once in imx-pcm.c.  Using
platform_device_id to distinguish between imx-pcm-dma and imx-pcm-fiq,
we can run-time call imx-pcm-dma/fiq specific initialization in .probe
hook to have module snd-soc-imx-pcm work for both cases.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This reverts commit 25b8d314.

While the commit fixes multiple init_module definition error with
module build, it breaks build when both imx-pcm-fiq and imx-pcm-dma
are built in as below.

  LD      sound/soc/fsl/snd-soc-fsl-ssi.o
  LD      sound/soc/fsl/snd-soc-fsl-utils.o
  LD      sound/soc/fsl/snd-soc-imx-ssi.o
  LD      sound/soc/fsl/snd-soc-imx-audmux.o
  LD      sound/soc/fsl/snd-soc-imx-pcm-fiq.o
  LD      sound/soc/fsl/snd-soc-imx-pcm-dma.o
  LD      sound/soc/fsl/snd-soc-eukrea-tlv320.o
  LD      sound/soc/fsl/snd-soc-imx-sgtl5000.o
  LD      sound/soc/fsl/snd-soc-imx-mc13783.o
  LD      sound/soc/fsl/built-in.o
sound/soc/fsl/snd-soc-imx-pcm-dma.o: In function `imx_pcm_free':
imx-pcm.c:(.text+0x464): multiple definition of `imx_pcm_free'
sound/soc/fsl/snd-soc-imx-pcm-fiq.o:imx-pcm-fiq.c:(.text+0x1a8): first defined here
sound/soc/fsl/snd-soc-imx-pcm-dma.o: In function `snd_imx_pcm_mmap':
imx-pcm.c:(.text+0x35c): multiple definition of `snd_imx_pcm_mmap'
sound/soc/fsl/snd-soc-imx-pcm-fiq.o:imx-pcm-fiq.c:(.text+0xa0): first defined here
sound/soc/fsl/snd-soc-imx-pcm-dma.o: In function `imx_pcm_new':
imx-pcm.c:(.text+0x3dc): multiple definition of `imx_pcm_new'
sound/soc/fsl/snd-soc-imx-pcm-fiq.o:imx-pcm-fiq.c:(.text+0x120): first defined here
make[4]: *** [sound/soc/fsl/built-in.o] Error 1

Let's revert the commit and find a proper fix for multiple init_module
definition error later.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Convert all uses of devm_request_and_ioremap() to the newly introduced
devm_ioremap_resource() which provides more consistent error handling.

devm_ioremap_resource() provides its own error messages so all explicit
error messages can be removed from the failure code paths.
Signed-off-by: NThierry Reding <thierry.reding@avionic-design.de>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: Liam Girdwood <lrg@ti.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

With commit f2818d07 (ASoC: fsl: fix miscompilation of snd-soc-imx-pcm),
we will see the following build error when building modules with
CONFIG_SND_IMX_SOC=m in imx_v6_v7_defconfig.

  CC [M]  sound/soc/fsl/phycore-ac97.o
  LD [M]  sound/soc/fsl/snd-soc-fsl-ssi.o
  LD [M]  sound/soc/fsl/snd-soc-fsl-utils.o
  LD [M]  sound/soc/fsl/snd-soc-imx-ssi.o
  LD [M]  sound/soc/fsl/snd-soc-imx-audmux.o
  LD [M]  sound/soc/fsl/snd-soc-imx-pcm.o
sound/soc/fsl/imx-pcm-dma.o: In function `init_module':
imx-pcm-dma.c:(.init.text+0x0): multiple definition of `init_module'
sound/soc/fsl/imx-pcm-fiq.o:imx-pcm-fiq.c:(.init.text+0x0): first defined here
sound/soc/fsl/imx-pcm-dma.o: In function `cleanup_module':
imx-pcm-dma.c:(.exit.text+0x0): multiple definition of `cleanup_module'
sound/soc/fsl/imx-pcm-fiq.o:imx-pcm-fiq.c:(.exit.text+0x0): first defined here
make[4]: *** [sound/soc/fsl/snd-soc-imx-pcm.o] Error 1

Instead of using bool for SND_SOC_IMX_PCM_FIQ and SND_SOC_IMX_PCM_DMA
to fix the original issue, we should completely remove SND_SOC_IMX_PCM
and have imx-pcm.o statically linked with imx-pcm-fiq.o or imx-pcm-dma.o.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The bestcomm dma hardware, and some of its users like the FEC ethernet
component, is used in different FreeScale parts, including non-powerpc
parts like the ColdFire MCF547x & MCF548x families.  Don't keep the
driver hidden in arch/powerpc where it is inaccessible for other arches.
.c files are moved to drivers/dma/bestcomm, while .h files are moved to
include/linux/fsl/bestcomm.  Makefiles, Kconfigs and #include directives
are updated for the new file locations.

Tested by recompiling for MPC5200 with all bestcomm users enabled.
Signed-off-by: NPhilippe De Muyter <phdm@macqel.be>
Signed-off-by: NAnatolij Gustschin <agust@denx.de>

CONFIG_HOTPLUG is going away as an option.  As result the __dev*
markings will be going away.

Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst,
and __devexit.
Signed-off-by: NBill Pemberton <wfp5p@virginia.edu>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Hi Mark,

thanks your insisting on a better description for the patch, I found a
more appropriate solution for the problem:

Compiling the SoC Audio driver for Freescale i.MX as a module
(CONFIG_SND_SOC_IMX_PCM=m) results in a non-functional sound driver
indicated by the error message:
| imx-sgtl5000 sound.1: platform imx-pcm-audio not registered
| imx-sgtl5000 sound.1: snd_soc_register_card failed (-517)
| platform sound.1: Driver imx-sgtl5000 requests probe deferral
instead of the message:
| imx-sgtl5000 sound.1:  sgtl5000 <-> 63fcc000.ssi mapping ok
that is to be expected upon loading the snd-soc-imx-pcm.ko module.

The build log reveals, that the file imx-pcm-dma.o (or imx-pcm-fiq.o
depending on the kernel configuration), which should be linked
together with imx-pcm.o into snd-imx-pcm.ko, is not being compiled in
this case.

The make rules for these files shows that the target object imx-pcm.o
is assigned to the variable snd-soc-imx-pcm-y while
imx-pcm-{dma,fiq}.o are added to to
snd-soc-imx-pcm-$(CONFIG_SND_SOC_IMX_PCM_DMA) and
snd-soc-imx-pcm-$(CONFIG_SND_SOC_IMX_PCM_FIQ) which resolve to
snd-soc-imx-pcm-m in this case.

According to Documentation/kbuild/modules.txt:
|When the module is built from multiple sources, an additional line is
|needed listing the files:
|
|        <module_name>-y := <src1>.o <src2>.o ...
Thus the type of the config variables CONFIG_SND_SOC_IMX_PCM_DMA and
CONFIG_SND_SOC_IMX_PCM_FIQ should be 'bool' instead of 'tristate' to
resolve to 'y' when selected.
Signed-off-by: NLothar Waßmann <LW@KARO-electronics.de>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Due to a broken make rule, sound/soc/fsl/imx-pcm-dma.c or
sound/soc/fsl/imx-pcm-fiq.c (whatever is selected via Kconfig) will
not be compiled into imx-pcm.o when building as module, i.e.:
CONFIG_SND_SOC_IMX_PCM=m
CONFIG_SND_SOC_IMX_PCM_DMA=m
resulting in a non-functional sound driver.

This gives the error messages:
| imx-sgtl5000 sound.1: platform imx-pcm-audio not registered
| imx-sgtl5000 sound.1: snd_soc_register_card failed (-517)
| platform sound.1: Driver imx-sgtl5000 requests probe deferral
when loading the driver instead of what's to be expected:
| imx-sgtl5000 sound.1:  sgtl5000 <-> 63fcc000.ssi mapping ok
Signed-off-by: NLothar Waßmann <LW@KARO-electronics.de>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

set the 'onwer' field of the registered snd_soc_card object to prevent
removal of the module when its resources are in use.
Signed-off-by: NLothar Waßmann <LW@KARO-electronics.de>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

also set MODULE_AUTHOR and MODULE_DESCRIPTION
Signed-off-by: NLothar Waßmann <LW@KARO-electronics.de>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This config item has not carried much meaning for a while now and is
almost always enabled by default. As agreed during the Linux kernel
summit, remove it.
Signed-off-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This small reference boards has a Freescale P1022 dual-core PowerPC SOC
and a Wolfson Microelectronics WM8960 codec.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Assign each dai_link a unique name to avoid this run-time error.

[   18.978043] pcm030-audio-fabric sound.2:  wm9712-hifi <-> mpc5200-psc-ac97.0 mapping ok
[   19.003179] sysfs: cannot create duplicate filename '/devices/sound.2/AC97'
Signed-off-by: NEric Millbrandt <emillbrandt@dekaresearch.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The only mach/irqs.h user outside arch/arm/mach-imx is
sound/soc/fsl/imx-pcm-fiq.c, which refers to mxc_set_irq_fiq().
Move the declaration into include/linux/platform_data/asoc-imx-ssi.h,
so that we can remove mach/irqs.h includsion from imx-pcm-fiq.c.

Inside arch/arm/mach-imx, the only users to mach/irqs.h are avic.c
and tzic.c for referring to macro FIQ_START.  Let's move the macro
into irq-common.h and get rid of mach/irqs.h completely.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: alsa-devel@alsa-project.org

The inclusion of mach/hardware.h is not used by the driver at all.
Remove it.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Cc: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: alsa-devel@alsa-project.org

Convert eukrea-tlv320 to platform driver.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The mpc5200-psc-ac97 driver does not enumerate attached ac97 devices, so
register the device here.
Signed-off-by: NEric Millbrandt <emillbrandt@dekaresearch.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Convert pcm030-audio-fabric to use the new snd_soc_register_card api
instead of the older method of registering a separate platform device.
Create the dai_link to the mpc5200_psc_ac97 platform using the device tree.
Convert the pcm030-audio-fabric driver to a platform-driver and add a
remove function.
Signed-off-by: NEric Millbrandt <emillbrandt@dekaresearch.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

mpc52xx socs do not define FSL_SOC but need SND_POWERPC_SOC defined to build
ASoC drivers.
Signed-off-by: NEric Millbrandt <emillbrandt@dekaresearch.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Remove unreferenced header files.
Signed-off-by: NEric Millbrandt <emillbrandt@dekaresearch.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Add missing dai_driver information to avoid these runtime errors

[   16.433788] asoc: error - multiple DAI f0002c00.i2s registered with no name
[   16.453551] Failed to register DAI
[   16.461222] mpc5200-psc-i2s: probe of f0002c00.i2s failed with error -22
[   16.475242] asoc: error - multiple DAI f0002000.ac97 registered with no name
[   16.488087] mpc5200-psc-ac97 f0002000.ac97: Failed to register DAI
[   16.502222] mpc5200-psc-ac97: probe of f0002000.ac97 failed with error -22
Signed-off-by: NEric Millbrandt <emillbrandt@dekaresearch.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The mpc5200_psc_ac97 and mpc5200_psc_i2s modules rely on shared platform data
with mpc5200_dma.
Signed-off-by: NEric Millbrandt <emillbrandt@dekaresearch.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

When audmux_read_file is called, it means the driver is already
initialised successfully, so we don't need to check audmux_base.

It also fix smatch warning:
sound/soc/fsl/imx-audmux.c:78 audmux_read_file() warn: possible memory leak of 'buf'
Signed-off-by: NRichard Zhao <richard.zhao@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

snd_imx_open should return error code returned by snd_dmaengine_pcm_open.
Signed-off-by: NRichard Zhao <richard.zhao@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

It fixed smatch warning:
sound/soc/fsl/imx-pcm-dma.c:112 snd_imx_open() error: potential null dereference 'dma_data'.  (kzalloc returns null)
Signed-off-by: NRichard Zhao <richard.zhao@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Remove a call to dma_unmap_single() from the PowerPC ASoC DMA driver.  The
buffer is allocated and not actually mapped, so the unmap call doesn't
make sense.  It was probably left over from some early version of the driver.

This bug was unnoticed for so long because the DMA mapping functions normally
don't do anything on PowerPC.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Use snd_soc_register_card() instead of platform_device_alloc("soc-audio")
to register the sound card from the machine drivers.  The use of
platform_device_alloc is deprecated.

Although several other drivers still use platform_device_alloc(), the
Freescale drivers were not using it to pass driver data.  Instead of fixing
the driver data usage, it's better to replace the deprecated code.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Rather than including mach/iomux-mx27.h to define gpio numbers and set
up the pins, the patch moves all these into machine code and has the
gpio numbers passed to driver via platform_data.  As the result, we
can remove the mach/iomux-mx27.h inclusion from driver.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NJavier Martin <javier.martin@vista-silicon.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Platform data for device drivers should be defined in
include/linux/platform_data/*.h, not in the architecture
and platform specific directories.

This moves such data out of the imx include directories
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Acked-by: NNicolas Pitre <nico@linaro.org>
Acked-by: NMauro Carvalho Chehab <mchehab@redhat.com>
Acked-by: NSascha Hauer <kernel@pengutronix.de>
Acked-by: NVinod Koul <vinod.koul@linux.intel.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Dan Williams <djbw@fb.com>
Cc: "Ben Dooks (embedded platforms)" <ben-linux@fluff.org>
Cc: "Wolfram Sang (embedded platforms)" <w.sang@pengutronix.de>
Cc: Chris Ball <cjb@laptop.org>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: Alan Cox <alan@linux.intel.com>
Cc: Felipe Balbi <balbi@ti.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Cc: Liam Girdwood <lrg@ti.com>
Cc: Javier Martin <javier.martin@vista-silicon.com>
Cc: Shawn Guo <shawn.guo@linaro.org>
Cc: Guennadi Liakhovetski <g.liakhovetski@gmx.de>

Revert 0865a75d(ASoC: imx-ssi: Remove mono support).

The bug this patch is meant to solve doesn't occur in Visstrim_M10 boards.
Furthermore, after applying this patch sound in Visstrim_M10 is played
at slower rates.
Signed-off-by: NJavier Martin <javier.martin@vista-silicon.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Initialize ret on the second call to imx_audmux_v2_configure_port so that
the subsequent test checks that result and not the previous one.

A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
(
if@p1 (\(ret < 0\|ret != 0\))
 { ... return ret; }
|
ret@p1 = 0
)
... when != ret = e1
    when != &ret
*if(...)
{
  ... when != ret = e2
      when forall
 return ret;
}

// </smpl>
Signed-off-by: NJulia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Playing a mono track results in incorrect playback rate, ie, the audio
is played at a faster rate.

Remove mono support in the driver by setting 'channes_min' to dual-channel
and this allows mono tracks to be played correctly.
Reported-by: NGaëtan Carlier <gcembed@gmail.com>
Tested-by: NGaëtan Carlier <gcembed@gmail.com>
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Saves anyone wondering what happened if they run into this error.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>