There are cases where there is no working codec on the soc-audio devices,
and snd_soc_suspend() will skip such device when suspending. Yet its
counterpart snd_soc_resume() does not check this, causing complaints
about spinlock lockup:

[  176.726087] BUG: spinlock lockup on CPU#0, kworker/0:2/1067, d8ab82a8
[  176.732539] [<80014a14>] (unwind_backtrace+0x0/0xec) from [<805b3fc8>] (dump_stack+0x20/0x24)
[  176.741082] [<805b3fc8>] (dump_stack+0x20/0x24) from [<80322208>] (do_raw_spin_lock+0x118/0x158)
[  176.749882] [<80322208>] (do_raw_spin_lock+0x118/0x158) from [<805b7874>] (_raw_spin_lock_irqsave+0x5c/0x68)
[  176.759723] [<805b7874>] (_raw_spin_lock_irqsave+0x5c/0x68) from [<8002a020>] (__wake_up+0x2c/0x5c)
[  176.768781] [<8002a020>] (__wake_up+0x2c/0x5c) from [<804a6de8>] (soc_resume_deferred+0x3c/0x2b0)
[  176.777666] [<804a6de8>] (soc_resume_deferred+0x3c/0x2b0) from [<8004ee20>] (process_one_work+0x2e8/0x50c)
[  176.787334] [<8004ee20>] (process_one_work+0x2e8/0x50c) from [<8004fd08>] (worker_thread+0x1c8/0x2e0)
[  176.796566] [<8004fd08>] (worker_thread+0x1c8/0x2e0) from [<80053ec8>] (kthread+0xa4/0xb0)
[  176.804843] [<80053ec8>] (kthread+0xa4/0xb0) from [<8000ea70>] (kernel_thread_exit+0x0/0x8)
Signed-off-by: NEric Miao <eric.miao@linaro.org>
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>

We don't want to clear the prefix while we're creating the DAPM controls
for the device as the prefix is applied during control creation.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NLiam Girdwood <lrg@ti.com>

This ensures they are available prior to the card late_probe().
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Specifically for the widgets added by machine driver late probe functions.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Ensure we only have one sync during the initial startup of the card by
making snd_soc_dapm_sync() a noop on non-instantiated cards. This avoids
any bounces due to things like jacks reporting their initial state on
partially initialised cards. The callers that don't also get called at
runtime should just be removed.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Handle the put_volsw/put_volsw_2r in one function.

To avoid build breakage in twl6040 keep the
snd_soc_put_volsw_2r as define, and map it snd_soc_put_volsw.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Handle the get_volsw/get_volsw_2r in one function.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Handle the info_volsw/info_volsw_2r in one function.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

In order to reduce the number of DAPM power checks we run keep a list of
widgets which have been changed since the last DAPM run and iterate over
that rather than the full widget list. Whenever we change the power state
for a widget we add all the source and sink widgets it has to the dirty
list, ensuring that all widgets in the path are checked.

This covers more widgets than we need to as some of the neighbour widgets
won't be connected but it's simpler as a first step. On one system I tried
this gave:

           Power    Path   Neighbour
Before:    207      1939   2461
After:     114      1066   1327

which seems useful.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The DAI init function may want to do something that needs the widgets to
be instantiated.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

For almost all machines the DAI format is a constant, always set to the
same thing. This means that not only should we normally set it on init
rather than in hw_params() (where it has been for historical reasons) we
should also allow users to configure this by setting a variable in the
dai_link structure. The combination of these two will make many machine
drivers even more data driven.

Implement a new dai_fmt field in the dai_link doing just that. Since 0 is
a valid value for many format flags and we need to be able to tell if the
field is actually set also add one to all the values used to configure
formats.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Commit 873bd4cb (ASoC: Don't set invalid name string to snd_card->driver
field) broke generation of a driver name for all ASoC cards relying on the
automatic generation of one. Fix this by using the old default with spaces
replaced by underscores.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NTakashi Iwai <tiwai@suse.de>

We don't use the step size so there's no need to work it out.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NLiam Girdwood <lrg@ti.com>

Similarly to PLLs/FLLs some modern CODECs provide selectable system clock
sources. When the clock is the clock for a DAI we do not usually need to
identify which clock is being configured so can use clk_id for the source
clock but with CODEC wide system clocks we will need to specify both the
clock being configured and the source.

Add a source argument to the CODEC driver set_sysclk() operation to
reflect this. As this operation is not as widely used as the DAI
set_sysclk() operation the change is not very invasive. We probably
ought to go and make the same alternation for DAIs at some point.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Use snd_soc_codec_readable_register instead of open-coding it.
Signed-off-by: NLars-Peter Clausen <lars@metafoo.de>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Change the default return value of snd_soc_codec_{readable,writable}_register to
true when no codec specific callback for this function is given. Otherwise all
registers of that codec will neither be readable nor writable, which is most
certainly not what we want.
Signed-off-by: NLars-Peter Clausen <lars@metafoo.de>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

GFP_ATOMIC is not needed here, use GFP_KERNEL instead.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

If devices can unconditionally support idle_bias_off let them flag it in
their driver structure.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NLiam Girdwood <lrg@ti.com>

Utterly trivial but it annoys me.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NLiam Girdwood <lrg@ti.com>

In preparation for Dynamic PCM (AKA DSP) support.

This adds a callback function to be called at the completion of a DAPM stream
event.

This can be used by DSP components to perform calculations based on DAPM graphs
after completion of stream events.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

In one comment, cpu_dai was mentioned although codec_dai was used in the
code. Also, fix the name for the card dai list which has no seperation
into card_dai and codec_dai.
Signed-off-by: NWolfram Sang <w.sang@pengutronix.de>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Since quite a few drivers are not managing to flag the cache as needing
to be resynced after suspend and it's a reasonable thing to do flag the
cache as needing sync automatically when suspending.

The expectation is that systems will mainly only keep the CODEC powered
when doing audio through the CODEC so we won't actually suspend the
device anyway; drivers which want to can override this behaviour when
they resume.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NLiam Girdwood <lrg@ti.com>
Cc: stable@kernel.org

Trace platform IO just like CODEC IO.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Allow platform probe to register platform kcontrols and DAPM just like
the CODEC probe().
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Allow platform driver widgets to perform any IO required for DAPM.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

In preparation for Dynamic PCM (AKA DSP) support.

Allow platform drivers to register kcontrols.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The snd_card->driver field contains a driver name string, and in
general it shouldn't contain space or special letters.  The commit
2b39535b changed the string copy from
card->name, but the long name string may contain such letters, thus
it may still lead to a segfault.

A temporary fix is not to copy the long name string but just keep it
empty as the earlier version did.
Reported-by: NKuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

In preparation for ASoC Dynamic PCM (AKA DSP) support.

Allow platform driver to perform IO. Intended for platform DAPM.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

In preparation for ASoC dynamic PCM support (AKA ASoC DSP)

Platform will also support DAPM so separate out the probe function
to simplify the code (just like the codec probe).
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

In preparation for the new ASoC Dynamic PCM support (AKA DSP support).

The new ASoC Dynamic PCM core allows DAIs to be dynamically re-routed
at runtime between the PCM device end (or Frontend - FE) and the physical DAI
(Backend - BE) using regular kcontrols (just like a hardware CODEC routes
audio in the analog domain). The Dynamic PCM core therefore must be
able to call PCM operations for both the Frontend and Backend(s) DAIs at
the same time.

Currently we have a global pcm_mutex that is used to serialise
the ASoC PCM operations. This patch removes the global mutex
and adds a mutex per RTD allowing the PCM operations to be reentrant and
allow control of more than one DAI at at time. e.g. a frontend PCM hw_params()
could configure multiple backend DAI hw_params() with similar or different
hw parameters at the same time.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

In preparation for Dynamic PCM support (AKA DSP support).

There will be future patches that add support to allow PCMs to be dynamically
routed to multiple DAIs at startup and also during stream runtime. This patch
moves the ASoC core PCM operaitions into a new file called soc-pcm.c.  This will
in simplify the ASoC core features into distinct files.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Some ASoC components depend on other ASoC components to provide clocks and
power resources in order to probe() and vice versa for remove().

Allow components to be ordered so that components can be probed() and removed()
in sequences that conform to their dependencies.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Currently pcm_new() passes in 3 arguments :- card, pcm and DAI.

Refactor this to only pass in 1 argument (i.e. the rtd) since struct rtd contains
card, pcm and DAI along with other members too that are useful too.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Avoids issues if someone does a read followed by restore and doesn't mask
out only the bits being updated.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NLiam Girdwood <lrg@ti.com>

For cards that have two or more DAIs, snd_soc_resume's loop over all
DAIs ends up calling schedule_work(deferred_resume_work) once per DAI.
Since this is the same work item each time, the 2nd and subsequent
calls return 0 (work item already queued), and trigger the dev_err
message below stating that a work item may have been lost.

Solve this by adjusting the loop to simply calculate whether to run the
resume work immediately or defer it, and then call schedule work (or not)
one time based on that.

Note: This has not been tested in mainline, but only in chromeos-2.6.38;
mainline doesn't support suspend/resume on Tegra, nor does the mainline
Tegra ASoC driver contain multiple DAIs. It has been compile-checked in
mainline.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Currently CODEC and platform drivers have their module reference count
incremented soc_probe_dai_link() whilst CPU DAI drivers have their reference
count incremented in soc_bind_dai_link().

CPU DAIs should have their reference count incremented in soc_probe_dai_link()
just like the CODEC and platform drivers.
Signed-off-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Commit 22de71ba ("ASoC: core - allow ASoC more flexible machine name")
writes "(null)" to driver name string in struct snd_card if card->driver_name
is NULL. This causes segmentation faults with some user space ALSA utilities
like aplay and arecord.

Fix this by using the card->name if no driver name is specified.
Signed-off-by: NJarkko Nikula <jhnikula@gmail.com>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Enabling DEBUG_STRICT_USER_COPY_CHECKS causes the following
warning:

In file included from arch/x86/include/asm/uaccess.h:573,
                 from include/linux/poll.h:14,
                 from include/sound/pcm.h:29,
                 from include/sound/ac97_codec.h:31,
                 from sound/soc/soc-core.c:34:
In function 'copy_from_user',
    inlined from 'codec_reg_write_file' at
    sound/soc/soc-core.c:252:
arch/x86/include/asm/uaccess_64.h:65:
warning: call to 'copy_from_user_overflow' declared with
attribute warning: copy_from_user() buffer size is not provably
correct

presumably due to buf_size being signed causing GCC to fail to
see that buf_size can't become negative.
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>