irq_of_parse_and_map() returns 0 on error, not NO_IRQ.

Fix the following xtensa:allmodconfig build error.

sound/soc/fsl/fsl_ssi.c:705:26: error: 'NO_IRQ' undeclared (first use in this function)
make[4]: *** [sound/soc/fsl/fsl_ssi.o] Error 1

Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Grant Likely <grant.likely@linaro.org>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>
Signed-off-by: NMark Brown <broonie@linaro.org>

Driver core clears the driver data to NULL after device_release or on probe
failure, so just remove it from here.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NMark Brown <broonie@linaro.org>

We have to disable the ssi irq, as it is not safe for all platforms to
write back into the status register. It also runs into non-linefetch
aborts.
Signed-off-by: NMichael Grzeschik <m.grzeschik@pengutronix.de>
Signed-off-by: NMark Brown <broonie@linaro.org>

imx-pcm-fiq is checking for TE RE bits, so enable them only if
necessary.
Signed-off-by: NMichael Grzeschik <m.grzeschik@pengutronix.de>
Signed-off-by: NMark Brown <broonie@linaro.org>

The chip errata for the i.MX35, Rev.2 has the following errata:

ENGcm06222: SSI:Transmission does not take place in bit length early frame sync
	    configuration

The workaround states, that TX_EN and SSI_EN bits should be set in the same
register write. As the next errata in the document (ENGcm06532) says to always
write RX_EN and TX_EN in the same register write in network mode.

Therefore include the whole write to
	CCSR_SSI_SCR_TE and CCSR_SSI_SCR_RE
into the write to
	CCSR_SSI_SCR_SSIEN
Signed-off-by: NSteffen Trumtrar <s.trumtrar@pengutronix.de>
Signed-off-by: NMark Brown <broonie@linaro.org>

This patch adds ac97-slave support.

For ac97, the registers have to be setup earlier than for other ssi
modes because there is some communication with the external device
before streaming. So this patch introduces a fsl_ssi_setup function to
setup the registers for different ssi operation modes seperately.

This patch was tested with imx27-pca100.
Signed-off-by: NMarkus Pargmann <mpa@pengutronix.de>
Tested-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@linaro.org>

In fsl_ssi_remove() we need to remove the resources in the opposite order that
they were acquired in probe.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NMark Brown <broonie@linaro.org>

There may be some platforms using fsl-ssi that do not have a DMA driver
with generic DMA bindings. So this patch adds support for the generic
DMA bindings, while still accepting the old "fsl,dma-events" property if
"dmas" is not found.
Signed-off-by: NMarkus Pargmann <mpa@pengutronix.de>
Tested-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@linaro.org>

Add support for non-dma pcm for imx platforms with imx-pcm-fiq support.
Instead of imx-pcm-audio, in this case imx-pcm-fiq-audio device is added
and the SIER flags are set differently.

We need imx-pcm-fiq for some boards that use an incompatible codec.
imx-pcm-fiq handles those codecs differently and allows to operate with
them. DMA is not possible because some data sent by the codecs, e.g.
wm9712, is not in the datastream. Also some data is mixed up in the
fifos, so that we need to sort them out manually.
Signed-off-by: NMarkus Pargmann <mpa@pengutronix.de>
Tested-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@linaro.org>

Let CPU DAI drivers set SDMA periperal type directly to support more
dma types(SPDIF, ESAI) other than only two for SSI.
This will easily allow some non-SSI drivers to use the imx-pcm-dma
as well.
Signed-off-by: NNicolin Chen <b42378@freescale.com>
Acked-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@linaro.org>

Add MODULE_ALIAS, so that auto module loading can work.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Acked-by: NTimur Tavi <timur@tabi.org>
Signed-off-by: NMark Brown <broonie@linaro.org>

Use the ARM version by default as that's the more generally portable
one, it doesn't matter if they work well on random platforms when the
goal is only build coverage.
Signed-off-by: NMark Brown <broonie@linaro.org>
Acked-by: NTimur Tabi <timur@tabi.org>

clk_prepare_enable() may fail, so let's check its return value and propagate it
in the case of error.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NMark Brown <broonie@linaro.org>

Using devm_ functions can make the code cleaner and smaller.
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NMark Brown <broonie@linaro.org>

The code enabled SSIEN when triggered by SNDRV_PCM_TRIGGER_START,
so move the disable code to SNDRV_PCM_TRIGGER_STOP for symmetric.

This also allows us to use the SSI driver more flexible so that
it can support some use cases like "aplay S16_LE.wav S24_LE.wav"
which would call the driver in sequence like:
 startup()->hw_params(S16_LE)->trigger(START)->tirgger(STOP)->
 hw_params(S24_LE)->trigger(START)->tirgger(STOP)->shutdown()

If we disable SSIEN in shutdown(), the second hw_params() would
bypass the sample bits setting while using symmetric_rate.
Signed-off-by: NNicolin Chen <b42378@freescale.com>
Acked-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@linaro.org>

Rather than instantiating imx-pcm-audio to call imx_pcm_dma_init(),
fsl_ssi can just directly call it to save the use of imx-pcm-audio.
With this change, fsl_ssi becomes not only a cpu DAI but also a platform
device, so updates platform device setup in imx-sgtl5000 accordingly.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This allows us to access the DAI DMA data when we create the PCM. We'll use
this when converting imx to generic DMA engine PCM driver.
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>

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>

Signed-off-by: NKuninori Morimoto <kuninori.morimoto.gx@renesas.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>

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>

This reverts commit 014e5b56 since
PowerPC doesn't use clkdev and hasn't implemented devm_clk_get() itself.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NRichard Zhao <richard.zhao@freescale.com>
Acked-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

For power saving, most IMX platform initilization code turns off
modules' clock, and expects driver turn on clock as needed.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The dma peripheral_type for SSI should be IMX_DMATYPE_SSI_SP if the SSI
is on SPBA bus.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Makes necessary changes on fsl_ssi to let it work with imx pcm and
machine drivers.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Provide different pair of accessors for accessing SSI registers on
PowerPC and ARM/IMX, so that fsl_ssi driver can be built on both
architectures.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NSascha Hauer <s.hauer@pengutronix.de>
Acked-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Check /compatible rather than /model to determine the machine name.
The p1022ds older device trees get a different /model from the new
ones, while /compatible is consistent there, so checking /compatible
will save the bother of detecting older p1022ds device trees.
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>
Acked-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Factor out some boilerplate code.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Acked-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Commit 1ee46ebd("ASoC: Make the DAI ops constant in the DAI structure")
introduced the possibility to have constant DAI ops structures, yet this is
barley used in both existing drivers and also new drivers being submitted,
although none of them modifies its DAI ops structure. The later is not
surprising since existing drivers are often used as templates for new drivers.
So this patch just constifies all existing snd_soc_dai_ops structs to eliminate
the issue altogether.

The patch was generated with the following coccinelle semantic patch:
// <smpl>
@@
identifier ops;
@@
-struct snd_soc_dai_ops ops =
+const struct snd_soc_dai_ops ops =
{ ... };
// </smpl>
Signed-off-by: NLars-Peter Clausen <lars@metafoo.de>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Commit 6992f533 ("sysfs: Use one lockdep class per sysfs attribute")
requires 'struct attribute' objects to be initialized with sysfs_attr_init().
Signed-off-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@kernel.org

The Freescale SSI audio controller supports "synchronous" and "asynchronous"
modes.  In synchronous mode, playback and capture use the same input clock,
so sample rates must be the same during simultaneous playback and capture.
Unfortunately, the code which supports asynchronous mode is just broken in
various ways.  In particular, it was constraining sample sizes as well as
the sample rate.

The fix also allows us to simplify the code by eliminating the 'asynchronous',
'playback', and 'capture' variables that were used to keep track of playback
and capture streams.

Unfortunately, it turns out that simulataneous playback and record does not
actually work on the only platform that supports asynchronous mode: the
Freescale P1022DS reference board.  If a second stream is started, the SSI
grinds to halt for both streams.  This is true even if the P1022 is configured
for synchronous mode, so it's likely a hardware problem that needs to be
worked around.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The PowerPC Freescale SSI driver is claiming the IRQ when the IRQ when
the device is opened, which means that the /proc/interrupts entry for
the SSI exists only during playback or capture.  This also meant that
the user won't know that the IRQ number is wrong until he tries to use
the device.  Instead, we should claim the IRQ when the device is probed.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This is a partial revert of 28f65c11 ("treewide: Convert uses of
struct resource to resource_size(ptr)") as the code is rewritten
in the sound tree and thus the change is obsolete.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Several fixes as well where the +1 was missing.

Done via coccinelle scripts like:

@@
struct resource *ptr;
@@

- ptr->end - ptr->start + 1
+ resource_size(ptr)

and some grep and typing.

Mostly uncompiled, no cross-compilers.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NJiri Kosina <jkosina@suse.cz>

Device tree integer properties are encoded in big-endian format, but some of
the Freescale ASoC drivers were assuming that the host is in big-endian format
as well.  Although this is true, it's better to use endian-safe accessors.

Also add a check for a failed ioremap() call in the SSI driver.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Get rid of users of of_platform_driver in drivers/sound.  The
of_platform_{,un}register_driver functions are going away, so the
users need to be converted to using the platform_bus_type directly.
Signed-off-by: NGrant Likely <grant.likely@secretlab.ca>

'struct of_device' no longer exists, and its functionality has been merged
into platform_device.  Update the MPC8610 HPCD audio drivers (fsl_ssi, fsl_dma,
and mpc8610_hpcd) accordingly.

Also add a #include for slab.h, which is now needed for kmalloc and kfree.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Add code that programs the DMA and SSI controllers differently based on the
FIFO depth of the SSI.

The SSI devices on the MPC8610 and the P1022 are identical in every way except
one: the transmit and receive FIFO depth.  On the MPC8610, the depth is eight.
On the P1022, it's fifteen.  The device tree nodes for the SSI include a
"fsl,fifo-depth" property that specifies the FIFO depth.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NLiam Girdwood <lrg@slimlogic.co.uk>

Add support for adding "status = disabled" to an SSI node to incidate that it
is not wired on the board.  This replaces the not-so-intuitive previous method
of omitting a codec-handle property.
Signed-off-by: NTimur Tabi <timur@freescale.com>
Acked-by: NKumar Gala <galak@kernel.crashing.org>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NLiam Girdwood <lrg@slimlogic.co.uk>