SAM9X60's PLL which is also part of SAMA7G5 is composed of 2 parts:
one fractional part and one divider. On SAMA7G5 the CPU PLL could be
changed at run-time to implement DVFS. The hardware clock tree on
SAMA7G5 for CPU PLL is as follows:
+---- div1 ----------------> cpuck
|
FRAC PLL ---> DIV PLL -+-> prescaler ---> div0 ---> mck0
The div1 block is not implemented in Linux; on prescaler block it has
been discovered a bug on some scenarios and will be removed from Linux
in next commits. Thus, the final clock tree that will be used in Linux
will be as follows:
+-----------> cpuck
|
FRAC PLL ---> DIV PLL -+-> div0 ---> mck0
It has been proposed in [1] to not introduce a new CPUFreq driver but
to overload the proper clock drivers with proper operation such that
cpufreq-dt to be used. To accomplish this DIV PLL and div0 implement
clock notifiers which applies safe dividers before FRAC PLL is changed.
The current commit treats only the DIV PLL by adding a notifier that
sets a safe divider on PRE_RATE_CHANGE events. The safe divider is
provided by initialization clock code (sama7g5.c). The div0 is treated
in next commits (to keep the changes as clean as possible).
[1] https://lore.kernel.org/lkml/20210105104426.4tmgc2l3vyicwedd@vireshk-i7/
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/20211011112719.3951784-12-claudiu.beznea@microchip.com
Acked-by: Nicolas Ferre <nicolas.ferre@microchip.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Some of the SAMA7G5 PLLs support multiple outputs (e.g. AUDIO PLL).
For these, split the PLL clock in two: fractional clock and
divider clock. In case PLLs supports multiple outputs (since these
outputs are dividers (with different settings) sharing the same
fractional part), it will register one fractional clock and multiple
divider clocks (dividers sharing the fractional clock).
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/1595403506-8209-17-git-send-email-claudiu.beznea@microchip.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Pass the ID of changeable parent at registration. This will allow
the scalability of this clock driver with regards to the changeable
parent ID for versions of this IP where changeable parent is not the
last one in the parents list (e.g. SAMA7G5). With this the clock flags
are set to zero in case we have no changeable parent. Also in
clk_generated_best_diff() the *best_diff variable is check against
tmp_diff variable using ">=" operator instead of ">" so that in case
the requested frequency could be obtained using fix parents + gck
dividers but the clock also supports changeable parent to be able
to force the usage of the changeable parent.
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/1595403506-8209-11-git-send-email-claudiu.beznea@microchip.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
According to datasheet the range of 600-1200MHz is for the
frequency generated by the fractional part of the PLL (namely
Fcorepllck according to datasheet). With this in mind the output
range of the PLL itself (fractional + div), taking into account
that the divider is 8 bits wide, is 600/256-1200Hz=2.3-1200MHz.
Fixes: a436c2a447 ("clk: at91: add sam9x60 PLL driver")
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Link: https://lore.kernel.org/r/1595403506-8209-6-git-send-email-claudiu.beznea@microchip.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
syscon_node_to_regmap() will make the created regmap get and enable the
first clock it can parse from the device tree. This clock is not needed to
access the registers and should not be enabled at that time.
Use device_node_to_regmap to resolve this as it looks up the regmap in
the same list but doesn't care about the clocks. This issue is detected
by lockdep when booting the sama5d3 with a device tree containing the
new clk bindings.
This fix already happened in 6956eb33ab ("clk: at91: fix possible
deadlock") for the drivers that had been migrated to the new clk binding
back then. This does the same for the new drivers as well.
Fixes: 01e2113de9 ("clk: at91: add sam9x60 pmc driver")
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.kernel.org/r/20200703073236.23923-1-a.fatoum@pengutronix.de
Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
The prescaler works as parent rate divided by (PRES + 1) (is_pres_direct == 1)
It does not work in the way of parent rate shifted to the right by (PRES + 1),
which means division by 2^(PRES + 1) (is_pres_direct == 0)
Thus is_pres_direct must be enabled for this SoC, to make the right computation.
This field was added in
commit 45b0668211 ("clk: at91: fix programmable clock for sama5d2")
SAM9X60 has the same field as SAMA5D2 in the PCK
Fixes: 01e2113de9 ("clk: at91: add sam9x60 pmc driver")
Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Link: https://lkml.kernel.org/r/1575977088-16781-1-git-send-email-eugen.hristev@microchip.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
It's just some static data that doesn't get changed after being used.
Mark it const everywhere.
Cc: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Add a driver for the PMC clocks of the sam9c60.
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
[sboyd@kernel.org: Staticize spinlock]
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
