Following the removal of the clk_regmap clock table from the s4-peripherals
clock controller driver, it appears some clocks are unused, which means
these are not exported or even registered.
In all likelihood, these clocks have not been tested. Remove the unused
clocks for now. These can added back later when they have been properly
tested.
Reviewed-by: Chuan Liu <chuan.liu@amlogic.com>
Link: https://lore.kernel.org/r/20250623-amlogic-clk-drop-clk-regmap-tables-v4-3-ff04918211cc@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Remove the clk_regmap tables that are used to keep track which clock
need to be initialised before being registered. The initialisation is now
done by the .init() operation of clk_regmap.
This rework saves a bit memory and makes maintenance a bit easier.
Link: https://lore.kernel.org/r/20250623-amlogic-clk-drop-clk-regmap-tables-v4-2-ff04918211cc@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Add clk_regmap_init() and use it with all clock types which derive from
clk_regmap. This helps initialise clk_regmap clocks without requiring
tables to keep track of the clock using this type.
The way it is done couples clk_regmap with the controllers, which is not
ideal. This is a temporary solution to get rid of the tables. The situation
will eventually be improved.
Link: https://lore.kernel.org/r/20250623-amlogic-clk-drop-clk-regmap-tables-v4-1-ff04918211cc@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Some Amlogic clock controller drivers have a dedicated headers file, some
do not. Over time, these headers have evolved and now only carry register
offset definitions. These offsets are only used by the related controller
and are not meant to be shared.
These headers are not serving any purpose now.
Start enforcing some consistency between the different Amlogic clock
drivers and move the register offset definitions to the related driver.
Link: https://lore.kernel.org/r/20250623-clk-meson-no-headers-v1-1-468161a7279e@baylibre.com
[jbrunet: checkpatch strict: removed extra blank line]
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Enabling the compile test should not cause automatic enabling of all
drivers.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/20250404-kconfig-defaults-clk-v1-1-4d2df5603332@linaro.org
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
SPICC is missing fclk_div2, which means fclk_div5 and fclk_div7 indexes
are wrong on this clock. This causes the spicc module to output sclk at
2.5x the expected rate when clock index 3 is picked.
Adding the missing fclk_div2 resolves this.
[jbrunet: amended commit description]
Fixes: a18c8e0b76 ("clk: meson: g12a: add support for the SPICC SCLK Source clocks")
Cc: stable@vger.kernel.org # 6.1
Signed-off-by: Da Xue <da@libre.computer>
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/20250512142617.2175291-1-da@libre.computer
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Fix a typo in MODULE_DESCRIPTION for a1 PLL driver, S4 should be A1.
Signed-off-by: Jian Hu <jian.hu@amlogic.com>
Reviewed-by: Dmitry Rokosov <ddrokosov@salutedevices.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20241231062552.2982266-1-jian.hu@amlogic.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The 32k clock reference a parent 'cts_slow_oscin' with a fixme note saying
that this clock should be provided by AO controller.
The HW probably has this clock but it does not exist at the moment in
any controller implementation. Furthermore, referencing clock by the global
name should be avoided whenever possible.
There is no reason to keep this hack around, at least for now.
Fixes: 14c735c8e3 ("clk: meson-gxbb: Add EE 32K Clock for CEC")
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20241220-amlogic-clk-gxbb-32k-fixes-v1-2-baca56ecf2db@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
gxbb_32k_clk_div sets CLK_DIVIDER_ROUND_CLOSEST in the init_data flag which
is incorrect. This is field is not where the divider flags belong.
Thankfully, CLK_DIVIDER_ROUND_CLOSEST maps to bit 4 which is an unused
clock flag, so there is no unintended consequence to this error.
Effectively, the clock has been used without CLK_DIVIDER_ROUND_CLOSEST
so far, so just drop it.
Fixes: 14c735c8e3 ("clk: meson-gxbb: Add EE 32K Clock for CEC")
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20241220-amlogic-clk-gxbb-32k-fixes-v1-1-baca56ecf2db@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Several clocks used by both g12a and g12b use the g12a cpu A clock hw
pointer as clock parent. This is incorrect on g12b since the parents of
cluster A cpu clock are different. Also the hw clock provided as parent to
these children is not even registered clock on g12b.
Fix the problem by reverting to the global namespace and let CCF pick
the appropriate, as it is already done for other clocks, such as
cpu_clk_trace_div.
Fixes: 25e682a02d ("clk: meson: g12a: migrate to the new parent description method")
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20241213-amlogic-clk-g12a-cpua-parent-fix-v1-1-d8c0f41865fe@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The bit index of the peripheral clock for mmc A is wrong
This was probably not a problem for mmc A as the peripheral is likely left
enabled by the bootloader.
No issues has been reported so far but it could be a problem, most likely
some form of conflict between the ethernet and mmc A clock, breaking
ethernet on init.
Use the value provided by the documentation for mmc A before this
becomes an actual problem.
Fixes: 085a4ea93d ("clk: meson: g12a: add peripheral clock controller")
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20241213-amlogic-clk-g12a-mmca-fix-v1-1-5af421f58b64@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
- Revert the attempt to make CLK_GET_RATE_NOCACHE flag work in
clk_set_rate() because it led to problems with the Qualcomm CPUFreq
driver
- Revert Amlogic reset driver back to the initial implementation. This
broke probe of the audio subsystem on axg based platforms and also
had compilation problems. We'll try again next time.
- Fix a clk frequency and fix array bounds runtime checks in the Airoha
EN7581 driver
-----BEGIN PGP SIGNATURE-----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=8Mf1
-----END PGP SIGNATURE-----
Merge tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Pull clk fixes from Stephen Boyd:
"Two reverts and two EN7581 driver fixes:
- Revert the attempt to make CLK_GET_RATE_NOCACHE flag work in
clk_set_rate() because it led to problems with the Qualcomm CPUFreq
driver
- Revert Amlogic reset driver back to the initial implementation.
This broke probe of the audio subsystem on axg based platforms and
also had compilation problems. We'll try again next time.
- Fix a clk frequency and fix array bounds runtime checks in the
Airoha EN7581 driver"
* tag 'clk-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
clk: en7523: Initialize num before accessing hws in en7523_register_clocks()
clk: en7523: Fix wrong BUS clock for EN7581
clk: amlogic: axg-audio: revert reset implementation
Revert "clk: Fix invalid execution of clk_set_rate"
The audio subsystem of axg based platform is not probing anymore.
This is due to the introduction of RESET_MESON_AUX and the config
not being enabled with the default arm64 defconfig.
This brought another discussion around proper decoupling between
the clock and reset part. While this discussion gets sorted out,
revert back to the initial implementation.
This reverts
* commit 681ed497d6 ("clk: amlogic: axg-audio: fix Kconfig dependency on RESET_MESON_AUX")
* commit 664988eb47 ("clk: amlogic: axg-audio: use the auxiliary reset driver")
Both are reverted with single change to avoid creating more compilation
problems.
Fixes: 681ed497d6 ("clk: amlogic: axg-audio: fix Kconfig dependency on RESET_MESON_AUX")
Cc: Arnd Bergmann <arnd@arndb.de>
Reported-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20241128-clk-audio-fix-rst-missing-v2-1-cf437d1a73da@baylibre.com
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
As in most cases, using 'imply' is wrong here and does not prevent
build failures since that code may not be visible to a built-in
clk driver:
axg-audio.c:(.text+0x15c): undefined reference to `devm_meson_rst_aux_register'
Replace the incorrt 'imply' with the necessary 'depends on'.
Fixes: 664988eb47 ("clk: amlogic: axg-audio: use the auxiliary reset driver")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20241111102932.3209861-1-arnd@kernel.org
Reviewed-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
Remove the implementation of the reset driver in axg audio
clock driver and migrate to the one provided by the reset framework
on the auxiliary bus
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20241008-meson-clk-rst-aux-v2-1-682ab9151f4f@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The spinlock in meson8b clock controller is now only protecting a call to
regmap_update_bits().
The regmap API already has its own locking system so this spinlock
is redundant. Remove it.
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/20240925-clk-meson8b-spinlock-v1-1-50b7f2d02d7d@baylibre.com
[jbrunet: amended to remove unused variable as reported by lkp]
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The register corresponding to MPLL does not share the same register
with other module drivers, so there is no concurrent access to the
register with other modules drivers. The spinlock defined in struct
meson_clk_mpll_data is no longer useful.
Signed-off-by: Chuan Liu <chuan.liu@amlogic.com>
Link: https://lore.kernel.org/r/20240925-mpll_spinlock-v2-1-8f9b73588ec1@amlogic.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The fractional denominator of C3's hifi_pll fractional multiplier is
fixed to 100000.
Fixes: 8a9a129dc5 ("clk: meson: c3: add support for the C3 SoC PLL clock")
Signed-off-by: Chuan Liu <chuan.liu@amlogic.com>
Link: https://lore.kernel.org/r/20240909-fix_clk-v3-2-a6d8f6333c04@amlogic.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Some PLLS with fractional multipliers have fractional denominators with
fixed values, instead of the previous "(1 << pll-> frc.width)".
Signed-off-by: Chuan Liu <chuan.liu@amlogic.com>
Link: https://lore.kernel.org/r/20240909-fix_clk-v3-1-a6d8f6333c04@amlogic.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The s4's hifi_pll supports a fractional frequency multiplier, but frac
parameters are not configured in the driver.
Signed-off-by: Chuan Liu <chuan.liu@amlogic.com>
Link: https://lore.kernel.org/r/20240906-fix_clk-v1-3-2977ef0d72e7@amlogic.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Symbols exported by the Amlogic clock modules are only meant to be used by
Amlogic clock controller drivers. Using a dedicated symbols namespace make
that clear and help clean the global namespace of symbols other modules do
no need.
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240719094228.3985595-1-jbrunet@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The register region of axg-audio tends to grow with the addition of
new supported SoC. Mapping slightly more has not been causing problem
so far but it is not viable to continue like this long term.
Setup the max register based on what is necessary on the related SoC.
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240719093934.3985139-3-jbrunet@baylibre.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
`clkc_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240703-clk-const-regmap-v1-6-7d15a0671d6f@gmail.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
`clkc_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240703-clk-const-regmap-v1-5-7d15a0671d6f@gmail.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
`clkc_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240703-clk-const-regmap-v1-4-7d15a0671d6f@gmail.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
`clkc_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240703-clk-const-regmap-v1-3-7d15a0671d6f@gmail.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
`a1_pll_regmap_cfg` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240703-clk-const-regmap-v1-2-7d15a0671d6f@gmail.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
`a1_periphs_regmap_cfg` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240703-clk-const-regmap-v1-1-7d15a0671d6f@gmail.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
When dealing with certain PLLs, it is necessary to avoid modifying them
if they have already been initialized by lower levels. For instance, in
the A1 SoC Family, the sys_pll is enabled as the parent for the cpuclk,
and it cannot be disabled during the initialization sequence. Therefore,
initialization phase must be skipped.
Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com>
Link: https://lore.kernel.org/r/20240515185103.20256-2-ddrokosov@salutedevices.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Add the C3 peripherals clock controller driver in the C3 SoC family.
[jbrunet: fix Kconfig select order and probe function name]
Co-developed-by: Chuan Liu <chuan.liu@amlogic.com>
Signed-off-by: Chuan Liu <chuan.liu@amlogic.com>
Signed-off-by: Xianwei Zhao <xianwei.zhao@amlogic.com>
Link: https://lore.kernel.org/r/20240522082727.3029656-6-xianwei.zhao@amlogic.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The fixed_pll_dco output frequency is not accurate,
add frac factor for fixed_pll_dco clk to fix it.
Fixes: 57b55c76aa ("clk: meson: S4: add support for Amlogic S4 SoC peripheral clock controller")
Signed-off-by: Xianwei Zhao <xianwei.zhao@amlogic.com>
Link: https://lore.kernel.org/r/20240603-s4_fixedpll-v1-1-2b2a98630841@amlogic.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Add MODULE_DEVICE_TABLE(), so modules could be properly autoloaded
based on the alias from of_device_id table. Clocks are considered core
components, so usually they are built-in, however these can be built and
used as modules on some generic kernel.
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Link: https://lore.kernel.org/r/20240410155406.224128-1-krzk@kernel.org
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
In order to setup the DSI clock, let's make the unused VCLK2 clock path
configuration via CCF.
The nocache option is removed from following clocks:
- vclk2_sel
- vclk2_input
- vclk2_div
- vclk2
- vclk_div1
- vclk2_div2_en
- vclk2_div4_en
- vclk2_div6_en
- vclk2_div12_en
- vclk2_div2
- vclk2_div4
- vclk2_div6
- vclk2_div12
- cts_encl_sel
vclk2 and vclk2_div uses the newly introduced vclk regmap driver
to handle the enable and reset bits.
In order to set a rate on cts_encl via the vclk2 clock path,
the NO_REPARENT flag is set on cts_encl_sel & vclk2_sel in order
to keep CCF from selection a parent.
The parents of cts_encl_sel & vclk2_sel are expected to be defined
in DT or manually set by the display driver at some point.
The following clock scheme is to be used for DSI:
xtal
\_ gp0_pll_dco
\_ gp0_pll
|- vclk2_sel
| \_ vclk2_input
| \_ vclk2_div
| \_ vclk2
| \_ vclk2_div1
| \_ cts_encl_sel
| \_ cts_encl -> to VPU LCD Encoder
|- mipi_dsi_pxclk_sel
\_ mipi_dsi_pxclk_div
\_ mipi_dsi_pxclk -> to DSI controller
The mipi_dsi_pxclk_div is set as bypass with a single /1 entry in div_table
in order to use the same GP0 for mipi_dsi_pxclk and vclk2_input.
The SET_RATE_PARENT is only set on the mipi_dsi_pxclk_sel clock so the
DSI bitclock is the reference base clock to calculate the vclk2_div value
when pixel clock is set on the cts_encl endpoint.
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240403-amlogic-v6-4-upstream-dsi-ccf-vim3-v12-3-99ecdfdc87fc@linaro.org
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
The VCLK and VCLK_DIV clocks have supplementary bits.
The VCLK gate has a "SOFT RESET" bit to toggle after the whole
VCLK sub-tree rate has been set, this is implemented in
the gate enable callback.
The VCLK_DIV clocks as enable and reset bits used to disable
and reset the divider, associated with CLK_SET_RATE_GATE it ensures
the rate is set while the divider is disabled and in reset mode.
The VCLK_DIV enable bit isn't implemented as a gate since it's part
of the divider logic and vendor does this exact sequence to ensure
the divider is correctly set.
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20240403-amlogic-v6-4-upstream-dsi-ccf-vim3-v12-2-99ecdfdc87fc@linaro.org
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
In most meson systems, multiple PLLs are present, making it difficult to
identify the specific PLL that fails to lock. To address this issue,
print out the name of the PLL that cannot be locked.
Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com>
Link: https://lore.kernel.org/r/20240328195733.30572-1-ddrokosov@salutedevices.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
When the max_register value is not set, the regmap debugfs 'registers'
file does not display the entire range of the regmap.
Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com>
Link: https://lore.kernel.org/r/20240320155512.3544-5-ddrokosov@salutedevices.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
When the max_register value is not set, the regmap debugfs 'registers'
file does not display the entire range of the regmap.
Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com>
Link: https://lore.kernel.org/r/20240320155512.3544-4-ddrokosov@salutedevices.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
When the max_register value is not set, the regmap debugfs 'registers'
file does not display the entire range of the regmap.
Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com>
Link: https://lore.kernel.org/r/20240320155512.3544-3-ddrokosov@salutedevices.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
When the max_register value is not set, the regmap debugfs 'registers'
file does not display the entire range of the regmap.
Signed-off-by: Dmitry Rokosov <ddrokosov@salutedevices.com>
Link: https://lore.kernel.org/r/20240320155512.3544-2-ddrokosov@salutedevices.com
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
