Some determine_rate implementations are not returning an error
when they failed to adapt the rate according to the rate request.
Fix them so that they return an error instead of silently
returning 0.
Signed-off-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
CC: Jonathan Corbet <corbet@lwn.net>
CC: Tony Lindgren <tony@atomide.com>
CC: Ralf Baechle <ralf@linux-mips.org>
CC: "Emilio López" <emilio@elopez.com.ar>
CC: Maxime Ripard <maxime.ripard@free-electrons.com>
Cc: Tero Kristo <t-kristo@ti.com>
CC: Peter De Schrijver <pdeschrijver@nvidia.com>
CC: Prashant Gaikwad <pgaikwad@nvidia.com>
CC: Stephen Warren <swarren@wwwdotorg.org>
CC: Thierry Reding <thierry.reding@gmail.com>
CC: Alexandre Courbot <gnurou@gmail.com>
CC: linux-doc@vger.kernel.org
CC: linux-kernel@vger.kernel.org
CC: linux-arm-kernel@lists.infradead.org
CC: linux-omap@vger.kernel.org
CC: linux-mips@linux-mips.org
CC: linux-tegra@vger.kernel.org
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>

Clock rates are stored in an unsigned long field, but ->determine_rate()
(which returns a rounded rate from a requested one) returns a long
value (errors are reported using negative error codes), which can lead
to long overflow if the clock rate exceed 2Ghz.

Change ->determine_rate() prototype to return 0 or an error code, and pass
a pointer to a clk_rate_request structure containing the expected target
rate and the rate constraints imposed by clk users.

The clk_rate_request structure might be extended in the future to contain
other kind of constraints like the rounding policy, the maximum clock
inaccuracy or other things that are not yet supported by the CCF
(power consumption constraints ?).
Signed-off-by: NBoris Brezillon <boris.brezillon@free-electrons.com>
CC: Jonathan Corbet <corbet@lwn.net>
CC: Tony Lindgren <tony@atomide.com>
CC: Ralf Baechle <ralf@linux-mips.org>
CC: "Emilio López" <emilio@elopez.com.ar>
CC: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NTero Kristo <t-kristo@ti.com>
CC: Peter De Schrijver <pdeschrijver@nvidia.com>
CC: Prashant Gaikwad <pgaikwad@nvidia.com>
CC: Stephen Warren <swarren@wwwdotorg.org>
CC: Thierry Reding <thierry.reding@gmail.com>
CC: Alexandre Courbot <gnurou@gmail.com>
CC: linux-doc@vger.kernel.org
CC: linux-kernel@vger.kernel.org
CC: linux-arm-kernel@lists.infradead.org
CC: linux-omap@vger.kernel.org
CC: linux-mips@linux-mips.org
CC: linux-tegra@vger.kernel.org
[sboyd@codeaurora.org: Fix parent dereference problem in
__clk_determine_rate()]
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Tested-by: NRomain Perier <romain.perier@gmail.com>
Signed-off-by: NHeiko Stuebner <heiko@sntech.de>
[sboyd@codeaurora.org: Folded in fix from Heiko for fixed-rate
clocks without parents or a rate determining op]
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>

of_io_request_and map returns an error pointer, but the current code assumes
that on error the returned pointer will be NULL.

Obviously, that makes the check completely useless. Change the test to actually
check for the proper error code.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Cc: Mike Turquette <mturquette@linaro.org>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: linux-clk@vger.kernel.org

The pll6 has a /4 output that is used as an input to the ahb mux clock.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

The current sunxi clock driver has the base factor clock of divs clocks
as the last clock output of the clock node. This makes it rather difficult
to add new outputs, such as fixed dividers, which were previously unknown.

This patch makes the divs clocks data structure specify which output is
the factor clock, and updates all current divs clocks accordingly.

We can then add new outputs after the factor clocks, at least not breaking
backward compatibility with regards to the devicetree bindings.

Also replace kzalloc with kcalloc in sunxi_divs_clk_setup().
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

We want to reparent AHB clock to PLL6 on sun5i/sun7i using the assigned
clocks properties. AHB is a factor clock, while PLL6 is a divs clock.

Register divs clocks before factor clocks so reparenting works. This
is only needed because we do the reparenting on the clock provider.

The proper way to fix this is to split out all the old sunxi clocks
into separate CLK_OF_DECLARE statements, like we are doing for sun9i.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Now that the ahb clock on sun5i/sun7i is muxable, ahb is no longer
guaranteed to be a child of the cpu clock. Add the cpu clock to
the list of protected clocks so it doesn't get disabled.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

The AHB clock on sun5i and sun7i are muxable divider clocks.
Use a factors clock to support them.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

The USB clocks originally shared code with the gates clocks, but had
additional reset controllers. Move these to a separate file. This will
allow us to add new support for slightly different USB clocks, such as
on the A80, without affecting gates clocks, and also facilitate the
migration of gates clocks to a generic solution.

This also cleans up the USB clocks code slightly, such as adding
newlines, getting rid of the unused clkdev call, using a simple
u32 instead of BITMAP for the clock masks, using BIT() macro to
declare the clock bitmasks, and using of_io_request_and_map() to
get the I/O address.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Adds a way for clock consumers to set maximum and minimum rates. This
can be used for thermal drivers to set minimum rates, or by misc.
drivers to set maximum rates to assure a minimum performance level.

Changes the signature of the determine_rate callback by adding the
parameters min_rate and max_rate.
Signed-off-by: NTomeu Vizoso <tomeu.vizoso@collabora.com>
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
[sboyd@codeaurora.org: set req_rate in __clk_init]
Signed-off-by: NMichael Turquette <mturquette@linaro.org>
[mturquette@linaro.org: min/max rate for sun6i_ahb1_clk_determine_rate
                        migrated clk-private.h changes to clk.c]

For all pll-s on sun6i n == 0 means use a multiplier of 1, rather then 0 as
it means on sun4i / sun5i / sun7i. n_start = 1 is already correctly set
for sun6i pll6, but was missing for pll1, this commit fixes this.

Cc: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Now that we don't have any user left for our custom phase function, we can
safely remove this hack from the code.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Reviewed-by: NChen-Yu Tsai <wens@csie.org>
Tested-by: NChen-Yu Tsai <wens@csie.org>

The cpu clock on sunxi machines is just a mux clock, which is normally
fed by the main PLL, but can be muxed to the main or low power oscillator.

Make the mux clock propagate rate changes to its parent, so we can
change the clock rate of the PLL, and thus actually implement rate
changing on the cpu clock.

This patch also removes the no reparenting limit.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

So far the A31s is 100% compatible with the A31, still lets do the same
as what we've done for the A13 / A10s and give it its own compatible string,
in case we need to differentiate later.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
[Maxime: Removed unusude CPU_OF_DECLARE_METHOD]
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Before this commit sunxi_factors_register uses of_iomap(node, 0) to get
the clk registers. The sun6i prcm has factor clocks, for which we want to
use sunxi_factors_register, but of_iomap(node, 0) does not work for the prcm
factor clocks, because the prcm uses the mfd framework, so the registers
are not part of the dt-node, instead they are added to the platform_device,
as platform_device resources.

This commit makes getting the registers the callers duty, so that
sunxi_factors_register can be used with mfd instantiated platform device too.

While at it also add error checking to the of_iomap calls.

This commit also drops the __init function from sunxi_factors_register since
platform driver probe functions are not __init.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

This patch unifies the sun6i AHB1 clock, originally supported
with separate mux and divider clks. It also adds support for
the pre-divider on the PLL6 input, thus allowing the clock to
be muxed to PLL6 with proper clock rate calculation.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

The ahb1_sdram clock gate is only used for accessing the sdram
controller's registers over the bus. It is not used for actually clock
the controller or the dram, hence it does not need to be protected.

This also gets rid of the problem when the protected ahb1_sdram gate
is prepared/enabled while it is still an orphan, and the operation
is not propagated to the correct parent.

This was confirmed on my A23 tablet and my A31 Hummingbird.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Some clock modules on the A31 use PLL6x2 as one of their inputs.
This patch changes the PLL6 implementation for A31 to a divs clock,
i.e. clock with multiple outputs that have different dividers.
The first output will be the normal PLL6 output, and the second
will be PLL6x2.

This patch fixes the PLL6 N factor in the clock driver, and removes
any /2 dividers in the PLL6 factors clock part. The N factor counts
from 1 to 32, mapping to values 0 to 31, as shown in the A31 manual.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Currently sunxi_divs_clk_setup assumes the number of child clocks
to be the same as the number of clock-output-names, and a maximum
of SUNXI_DIVS_MAX_QTY child clocks.

On sun6i, PLL6 only has 1 child clock, but the parent would be used
as well, thereby also having it's own clock-output-names entry. This
results in an extra bogus clock being registered.

This patch adds an entry for the number of child clocks alongside
the data structures for them.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

This driver does not match the hardware, which is actually compatible
to sun4i-a10-apb1-clk. Since we've switch to the correct one, drop
this driver.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

This commit unifies the APB1 mux with the APB1 clock, using the new
factors infrastructure.
Signed-off-by: NEmilio López <emilio@elopez.com.ar>
[wens@csie.org: Add mux mask bits]
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

This adds the gate clocks for AHB/APB busses on the A80 SoC.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Some of the factors-style clocks on the A80 have different widths
for the mux values in the registers.

Add a .muxmask field to clk_factors_config to make it configurable.
Passing a bitmask instead of a width parameter will allow reuse
in case we support table-based muxes in the future.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Move the MBUS clock to the module clocks file. It's pretty trivial, but still
requires to enable the clocks to make sure it won't get disabled.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NHans de Goede <hdegoede@redhat.com>

Since we know have the ability to declare factors clock outside of clk-sunxi,
create a new mod0 driver to deal with the mod0 clocks.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NHans de Goede <hdegoede@redhat.com>

Even though the mbus clock is a regular module clock, given its nature, it
needs to be enabled all the time.

Introduce a new compatible, to differentiate it from the other module clocks.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NHans de Goede <hdegoede@redhat.com>

Until now, the factors clock probing was done directly by sunxi_init_clocks,
with the factors registration being called directly with the clocks data passed
as an argument.

This approch has shown its limits when we added more clocks, since we couldn't
really split code with such a logic in smaller files, and led to a huge file
having all the clocks.

Introduce an intermediate probing function, so that factor clocks will be able
to directly be called by CLK_OF_DECLARE, which will in turn ease the split into
several files.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NHans de Goede <hdegoede@redhat.com>

The sun4i-apb0 clock, as found on all platforms using it, is a
power-of-two-based divider clock, with a special divider of 2
for value 0.

This was causing the clock framework to incorrectly calculate
the clock rate for apb1 and related modules on sun6i and sun8i.
On sun[4/5/7]i, u-boot SPL configures the divider with value 1
for /2 divider, so no suprises there.

This patch adds a proper divider table for it, so the correct
clock rate can be calculated.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Acked-by: NEmilio López <emilio@elopez.com.ar>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

This commit adds __iomem thoughout the sunxi clock driver, in places
where it was ommited. This cleans most of the sparse warnings we
are getting here.
Signed-off-by: NEmilio López <emilio@elopez.com.ar>
Acked-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: NMike Turquette <mturquette@linaro.org>

The clock control unit on the A23 is similar to the one found on the A31.

The AHB1, APB1, APB2 gates on the A23 are almost identical to the ones
on the A31, but some outputs are missing.

The main CPU PLL (PLL1) however is like that on older Allwinner SoCs,
such as the A10 or A20, but the N factor starts from 1 instead of 0.

This patch adds support for PLL1 and all the basic clock muxes and gates.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Acked-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

A few of the clock modules have odd dividers, such as
the 2 lowest dividers being the same (2), or have the
same divider when the highest bit is set.

This patch adds support for optional divider tables,
so the clock framework will know about the odd values.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Acked-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

With sunxi_gates clocks registered with clkdev, we can use the
protected clocks list to enable the "ahb_sdram" clock, instead
of looking for it and adding CLK_IGNORE_UNUSED inline in the
clock setup code.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Acked-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

The new important clock protect code requires the clocks be
registered with clkdev. This was missing for sunxi_gates
type clocks.
Signed-off-by: NChen-Yu Tsai <wens@csie.org>
Acked-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>

Prevent the SDRAM controller from being gated by force-enabling it in the
machine code.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NMike Turquette <mturquette@linaro.org>
Signed-off-by: NEmilio López <emilio@elopez.com.ar>

Right now, AHB is an indirect child clock of the CPU clock. If that
happens to change, since the CPU clock has no other consumers declared
in Linux, it would be shut down, which is not really a good idea.

Prevent this by forcing it enabled.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NMike Turquette <mturquette@linaro.org>
Signed-off-by: NEmilio López <emilio@elopez.com.ar>

Since we start to have a lot of clocks to protect, some of them in a
few SoCs only, it becomes difficult to handle the clock protection
without having to add per machine exceptions.

Add per-SoC data to tell which clock to leave enabled.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NMike Turquette <mturquette@linaro.org>
Signed-off-by: NEmilio López <emilio@elopez.com.ar>

Since we have a folder of our own, we can actually make use of it by
splitting the huge clock file into several sub drivers.

The gmac clock is pretty easy to deal with, since it's pretty much
isolated and doesn't have any dependency on the other clocks.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NMike Turquette <mturquette@linaro.org>
Signed-off-by: NEmilio López <emilio@elopez.com.ar>

Since we have a folder of our own, we can actually make use of it by
splitting the huge clock file into several sub drivers.

The main oscillator is pretty easy to deal with, since it's pretty much
isolated.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NMike Turquette <mturquette@linaro.org>
Signed-off-by: NEmilio López <emilio@elopez.com.ar>

Callers of clk_put must disable the clock first. This also means that
as long as the clock is enabled the driver should hold a reference to
that clock. Hence, the call to clk_put here are bogus and should be
removed.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: NMike Turquette <mturquette@linaro.org>
Signed-off-by: NEmilio López <emilio@elopez.com.ar>

The A31 USB clock slightly differ from its older counterparts, mostly
because it has a different gate for each PHY, while the older one had
a single gate for all the phy.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Acked-by: NMike Turquette <mturquette@linaro.org>
Signed-off-by: NEmilio López <emilio@elopez.com.ar>