Merge tag 'sunxi-clocks-for-3.19' of https://git.kernel.org/pub/scm/linux/kernel/git/mripard/linux into clk-next

Allwinner Clocks additions for 3.19

A few patches that should go through the clock tree, mostly fixes, cleanups,
and new clocks additions to start to support the A80.

A CLK_MUX_INDEX_BIT mux has one bit per parent, but the sun7i-a20-gmac-clk
has 2 bits selecting between 3 possible parents using values of 0, 1, 2,
which makes it a regular mux which should not have CLK_MUX_INDEX_BIT set in
its flag.

However we do not support parent 1 (an external clock), so use a table to
select parent 0 or 2, which are the 2 parents we support.

Note this has not been causing any issues sofar, because we start with a
parent setting of parent 0, and only ever re-parent to parent 2 (for which
we use an index of 1 as we skip parent 1) and with CLK_MUX_INDEX_BIT set
we write a value of 2 for index 1.

Tested on both a cubietruck (which uses rgmii mode) as well as a cs908
(an a31s board which uses mii mode).
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Maxime 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: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime 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: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime 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: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>

The apb2 clocks are actually the same as apb1 clocks on the other sunxi
platforms, hence compatible with "allwinner,sun4i-a10-apb1-clk".

Update the dtsi to use the new unified apb1 clk.
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>

With the new factors infrastructure in place, we can unify apb1 and
apb1_mux as a single clock now.
Signed-off-by: Emilio López <emilio@elopez.com.ar>
[wens@csie.org: Change apb1 node label to "apb1"; reword commit title]
Signed-off-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>

Use __initconst instead of __initdata for constant init data.
Signed-off-by: Bintian Wang <bintian.wang@huawei.com>
Acked-by: Haojian Zhuang <haojian.zhuang@linaro.org>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

If CLK_MUX_INDEX_BIT is set, then each bit turns on / off a single parent,
so theoretically multiple parents could be enabled at the same time, but in
practice only one bit should ever be 1. So to select parent 0, set
the register (*) to 0x01, to select parent 1 set it 0x02, parent 2, 0x04,
parent 3, 0x08, etc.

But the current code does:

                if (mux->flags & CLK_MUX_INDEX_BIT)
                        index = (1 << ffs(index));

Which means that:

For an input index of 0, ffs returns 0, so we set the register
to 0x01, ok.

For an input index of 1, ffs returns 1, so we set the register
to 0x02, ok.

For an input index of 2, ffs returns 2, so we set the register
to 0x04, ok.

For an input index of 3, ffs returns 1, so we set the register
to 0x02, not good!

The code should simply be:

                if (mux->flags & CLK_MUX_INDEX_BIT)
                        index = 1 << index;

Which always does the right thing, this commit fixes this.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Commit 8e33f91a ("clk: shmobile: clk-mstp: change to using
clock-indices") forgot to replace all occurrences of
"renesas,clock-indices" in the driver-specific binding documentation.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Acked-by: Simon Horman <horms+renesas@verge.net.au>
Acked-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

It's the same to the next statement, "ret = clk->parent". I think compiler will
optimize it, it's just not looking well.
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Merge tag 'v3.19-rockchip-clk1' of git://git.kernel.org/pub/scm/linux/kernel/git/mmind/linux-rockchip into clk-next-rockchip

- fixes for clock ordering/rate issues
- do not keep all clocks enabled anymore
- allow special pll rates for special cases

Have each pxa variant (pxa25x, pxa27x, pxa3xx) have its own device-tree
clock initializing function, to be able to register its own specific
core clocks.

Apply that change specifically to pxa27x.
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

As the clock descriptions are constant and only usefull at init time,
mark them as such by :
 - spliting clock description (desc) and clock private data (dynamic)
 - mark __initdata clock descriptions

This makes all the register and descriptions of the clocks to go after
kernel init phase.
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Trivial fix to check the A bit of CCCR for memory frequency
calculations, where the shift of the bit index was missing, triggering a
wrong calculation of memory frequency.
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Move pxa25x clock drivers from arch/arm/mach-pxa to driver/clk.
In the move :
 - convert to new clock framework legacy clocks
 - provide clocks as before for platform data based boards
 - provide clocks through devicetree with clk-pxa-dt

This is the preliminary step in the conversion. The remaining steps are
:
 - pxa3xx
 - once PXA is fully converted to device tree, if that happens,
   clk-pxa2* and clk-pxa3* should only hold the core clocks which cannot
   be described in devicetree.
Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr>
Tested-by: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Commit 79c6ab50 (clk: divider: add CLK_DIVIDER_READ_ONLY flag) in
v3.16 introduced the CLK_DIVIDER_READ_ONLY flag which caused the
recalc_rate() and round_rate() clock callbacks to be omitted.

However using this flag has the unfortunate side effect of causing the
clock recalculation code when a clock rate change is attempted to always
treat it as a pass-through clock, i.e. with a fixed divide of 1, which
may not be the case. Child clock rates are then recalculated using the
wrong parent rate.

Therefore instead of dropping the recalc_rate() and round_rate()
callbacks, alter clk_divider_bestdiv() to always report the current
divider as the best divider so that it is never altered.

For me the read only clock was the system clock, which divided the PLL
rate by 2, from which both the UART and the SPI clocks were divided.
Initial setting of the UART rate set it correctly, but when the SPI
clock was set, the other child clocks were miscalculated. The UART clock
was recalculated using the PLL rate as the parent rate, resulting in a
UART new_rate of double what it should be, and a UART which spewed forth
garbage when the rate changes were propagated.
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Cc: Thomas Abraham <thomas.ab@samsung.com>
Cc: Tomasz Figa <t.figa@samsung.com>
Cc: Max Schwarz <max.schwarz@online.de>
Cc: <stable@vger.kernel.org> # v3.16+
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

There is a duplication in a clock name for apq8084 platform that causes
the following warning: "RBCPR_CLK_SRC" redefined

Resolve this by adding a MMSS_ prefix to this clock and making its name
coherent with msm8974 platform.

Fixes: 2b46cd23 ("clk: qcom: Add APQ8084 Multimedia Clock Controller (MMCC) support")
Signed-off-by: Georgi Djakov <gdjakov@mm-sol.com>
Reviewed-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

First check for rate == 0 in set_rate and round_rate to avoid div by zero.
Then, in order to get the closest rate, round all divisions to the closest
result instead of rounding them down.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

at91rm9200_clk_usb_set_rate might fail depending on the requested rate,
because the parent_rate / rate remainder is not necessarily zero.
Moreover, when rounding down the calculated rate we might alter the
divisor calculation and end up with an invalid divisor.

To solve those problems, accept a non zero remainder, and always round
division to the closest result.
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Reported-by: Andreas Henriksson <andreas.henriksson@endian.se>
Tested-by: Andreas Henriksson <andreas.henriksson@endian.se>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

According to rk3288 trm, the mux selector locate at bit[12:11]
of CRU_CLKSEL13_CON shows:
2'b00: select HOST0 USB pll clock (clk_otgphy1)
2'b01: select HOST1 USB pll clock (clk_otgphy2)
2'b10: select OTG USB pll clock   (clk_otgphy0)

The clock map is in Fig. 3-4 CRU Clock Architecture Diagram 3
- clk_otgphy0 -> USB PHY OTG
- clk_otgphy1 -> USB PHY host0
- clk_otgphy2 -> USB PHY host1
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>

According to rk3288 trm, the clk_usbphy480m_gate is located at
bit 14 of CRU_CLKGATE5_CON register.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>

Signed-off-by: Michael Turquette <mturquette@linaro.org>

Currently there is no driver owning these clocks and they have to stay
up for the system to function properly, so let's mark them as
CLK_IGNORE_UNUSED.

Without this patch we have trouble with suspend/resume and we have
trouble turning the eDP back on if it ever idles off.
Signed-off-by: Dmitry Torokhov <dtor@chromium.org>
Reviewed-by: Doug Anderson <dianders@chromium.org>
Tested-by: Doug Anderson <dianders@chromium.org>
Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>

Signed-off-by: Michael Turquette <mturquette@linaro.org>

Change the dtsi and dts file, soc initialization code to make
use of DT support clock.
So now in the code we do only need call of_clk_init to initialize
the clocks.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Add items in arch/arm/boot/dt/Makefile to compile the dtb
for mach-mmp.
Change the dts and dtsi file to use #include instead of \include\
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

It adds the DT support for mmp2 clock subsystem.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

It adds the DT support for pxa910 clock subsystem.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

It adds the DT support for pxa168 clock subsystem.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Some clock control regsiter has bit to reset the cotroller.
So before enable the clock, we need deassert the reset pin.
Make use of reset controller framework to export reset interface
for device drivers, then device driver can control the reset action.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

In order to support DT for mmp SOC clocks, it defines
some basic APIs which are shared by all mmp SOC clock
units.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Some SOCes have this kind of the gate clock
1. There are some bits to control the gate not only one bit.
2. It is not always that "1" is to enable while "0" is to disable
   when write register.

So we have to define the "mask", "enable_val", "disable_val" for
this kind of gate clock.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

The clock type mix is a kind of clock combines "div" and "mux".
This kind of clock can not allow to change div first then
mux or change mux first or div.
The reason is
1. Some clock has frequency change bit. Each time want to change
   the frequency, there are some operations based on this bit, and
   these operations are time-cost.
   Seperating div and mux change will make the process longer, and
   waste more time.
2. Seperting the div and mux may generate middle clock that the
   peripharals do not support. It may make the peripharals hang.

There are three kinds of this type of clock in all SOCes.
1. The clock has bit to trigger the frequency change.
2. Same as #1, but the operations for the bit is different
3. Do not have frequency change bit.

So this type of clock has implemented the callbacks
->determine_rate
->set_rate_and_parent
These callbacks can help to change the div and mux together.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

Move the definition of structure of mmp_clk_frac to
clk.h.
So device tree support can use this structure.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>

For the clk-frac, we need to make sure that the initial
clock rate is one item of the table.
If it is not, we use the first item in the table by default.
Signed-off-by: Chao Xie <chao.xie@marvell.com>
Acked-by: Haojian Zhuang <haojian.zhuang@gmail.com>
Signed-off-by: Michael Turquette <mturquette@linaro.org>