Tegra124 introduces some small changes to the layout of some registers.
Modify the affected drivers to program those registers appropriately
based on which SoC they're running on.

Tegra124 also introduced some new modules on the AHUB configlink register
bus. These will require new entries in configlink_clocks[] in the AHUB
driver. However, supporting that change likely relies on switching Tegra
to the common reset framework, so I'll defer that change for now.
Based-on-work-by: NArun Shamanna Lakshmi <aruns@nvidia.com>
Based-on-work-by: NSonghee Baek <sbaek@nvidia.com>
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NMark Brown <broonie@linaro.org>

Now that all of the Tegra device trees have been updated to represent
the required audio clocks, remove the compatibility code from the Tegra
ASoC utility code, and always use clk_get() rather than clk_get_sys().
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Tegra114 requires different PLL rates. Modify the code to know about
this.

On Tegra114 only for now, use regular clk_get() rather than clk_get_sys()
to retrieve clocks. This assumes that the clocks will be represented in
device tree. We can assure that from the start of any Tegra114 audio
support. For older chips, I'll add the required clocks properties to the
device trees this kernel cycle, and switch this code to only support the
"new_clocks" path next cycle.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Tegra only supports, and always enables, device tree. Remove all runtime
checks for DT support from the driver.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

AC97 uses a fixed rate, unrelated to the sample rate. Add a function to
make the setup more trivial.
Signed-off-by: NLucas Stach <dev@lynxeye.de>
Acked-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Use clk_prepare/clk_unprepare as required by the generic clk framework.
Signed-off-by: NPrashant Gaikwad <pgaikwad@nvidia.com>
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Recent list discussions concluded that drivers should not be calling
of_have_populated_dt(), and hence of_have_populated_dt() should not be
exported. Use a different mechanism to detect DT vs. non-DT boot.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Tegra30 has some additional clocks that need to be manipulated, names
some clocks differently, runs PLLs at different base rates, etc. The
utility code needs to handle this.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Initialize the audio clock tree appropriately for some reasonable rate.
This makes sure the PLLs etc. are actually programmed to something
reasonable when the audio driver is loaded.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Lots of sound drivers were getting module.h via the implicit presence
of it in <linux/device.h> but we are going to clean that up.  So
fix up those users now.
Signed-off-by: NPaul Gortmaker <paul.gortmaker@windriver.com>

Only the clock programming code needs to know whether the clocks changed,
and that is encapsulated within tegra_asoc_utils_set_rate(). The machine
driver's call to snd_soc_dai_set_sysclk(codec_dai, ...) is safe
irrespective of whether the clocks changed.

(Applying Mark's TrimSlice review comments to the existing driver)
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Acked-by: NLiam Girdwood <lrg@ti.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The utilities will be required by every machine driver. Including the
utility object directly into every machine driver causes a build failure
if the modules are actually built into the kernel, since each will define
the symbols exported by the utility file. Solve this by moving the
utility object into a separate module.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Acked-by: NLiam Girdwood <lrg@slimlogic.co.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Instead, have the machine driver provide storage for the utility data
somehow.

For Harmony in particular, store this within struct tegra_harmony, itself
referenced by snd_soc_card's drvdata.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Acked-by: NLiam Girdwood <lrg@slimlogic.co.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

A recent discussion on linux-arm-kernel noted that the value returned by
clk_get_sys is an opaque token, and not strictly a pointer; it is
meaningful only to the clock API, clients should not dereference the value,
and the clock API must accept any non-IS_ERR value it returned.

Hence, only IS_ERR is appropriate to interpret the result, not
IS_ERR_OR_NULL.

I checked that clk_get_sys in both ASoC's for-next and Tegra's for-next
do behave as described; NULL is not returned in the case of error.
Signed-off-by: NStephen Warren <swarren@nvidia.com>
Acked-by: NLiam Girdwood <lrg@slimlogic.co.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Many portions of Tegra ASoC machine drivers will be similar or identical.
To avoid cut/paste, this file will act as a repository for all that common
code. For now, it solely includes code to reprogram the audio PLL for
44.1KHz- vs. 48KHz-based sample rates.
Signed-Off-By: NStephen Warren <swarren@nvidia.com>
Acked-by: NLiam Girdwood <lrg@slimlogic.co.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>