| Age | Commit message (Collapse) | Author |
|---|
|
Add a new config CONFIG_CLK_AUTO_ID to support a unique clk id
for all the clock providers, managed by clk uclass, when the clock
reference arg[0] is the same.
When the CONFIG is activated, the clock id is limited to the lower
CLK_ID_SZ = 24 bits in default clock xlate function
and the sequence number + 1 of the clk provider device is
added for the 8 higher bits.
We use sequence number + 1 to avoid the "dummy" clock id = 0,
used for invalid clock when CCF is activated.
When this config is activated, the new function clk_get_id()
should be used to get back the internal reference to clock
for the each clock provider.
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>
Cc: Lukasz Majewski <lukma@denx.de>
Cc: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
|
|
The commit ac30d90f3367 ("clk: Ensure the parent clocks are enabled
while reparenting") add a call to clk_enable() for the parent clock.
For clock drivers that do not implement the enable() ops, like most
Rockchip clock drivers, this now cause the set_parent() ops to never
be called when CLK_CCF=n (default for Rockchip).
clk_enable() typically return -ENOSYS when the enable() ops is not
implemented by the clock driver, with CLK_CCF=y clk_enable() instead
return 0 when the enable() ops is unimplemented.
Change to ignore -ENOSYS from the newly introduced clk_enable() call to
fix this regression and restore the old behavior of set_parent() ops
being called regardless of if enable() ops is implemented or not.
Fixes: ac30d90f3367 ("clk: Ensure the parent clocks are enabled while reparenting")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Tested-by: Dang Huynh <danct12@riseup.net>
Acked-by: Miquel Raynal <miquel.raynal@bootlin.com>
|
|
It's currently possible to make the bootloader crash on calling
clk_set_rate caused by the loop in clk_clean_rate_cache.
The loop assume that every child of the clock node are also clock
device but this is not always the case. For example it's common for a
clock to bind to a reset device or also expose a syscon if the clock
register map is also used to apply special configuration.
In such case, on accessing a device as a clock, the bootloader crash. To
correctly handle this, check if the child device is actually a clock and
ignore otherwise.
Fixes: 6b7fd3128f71 ("clk: fix set_rate to clean up cached rates for the hierarchy")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Reviewed-by: Michael Trimarchi <michael@amarulasolutions.com>
|
|
Reparenting a clock C with a new parent P means that C will only
continue clocking if P is already clocking when the mux is updated. In
case the parent is currently disabled, failures (stalls) are likely to
happen.
This is exactly what happens on i.MX8 when enabling the video
pipeline. We tell LCDIF clocks to use the VIDEO PLL as input, while the
VIDEO PLL is currently off. This all happens as part of the
assigned-clocks handling procedure, where the reparenting happens before
the enable() calls. Enabling the parents as part of the reparenting
procedure seems sane and also matches the logic applied in other parts
of the CCM.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
|
|
Add clk_resolve_parent_clk() to resolve parent clock udevice name
based on clock-names DT property. This is used in SoC clock drivers
to look up the clock name in clock tables, which matches a clock
name in DT clock-names property, and convert it into udevice name
which is used by U-Boot clock framework to look up parent clock in
e.g. clk_register() using uclass_get_device_by_name(UCLASS_CLK,
parent_name, &parent);
Signed-off-by: Marek Vasut <marex@denx.de>
|
|
Sometimes clocks provided to a consumer might not have .set_rate
operation (like gate or mux clocks), but have CLK_SET_PARENT_RATE flag
set. In that case it's usually possible to find a parent up the tree
which is capable of setting the rate (div, pll, etc). Implement a simple
lookup procedure for such cases, to traverse the clock tree until
.set_rate capable parent is found, and use that parent to actually
change the rate. The search will stop once the first .set_rate capable
clock is found, which is usually enough to handle most cases.
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
|
|
Use the new symbol to refer to any 'SPL' build, including TPL and VPL
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
Print clk name in clk_enable and clk_disable. Make sense to know
what clock get disabled/enabled before a system crash or system
hang.
Signed-off-by: Michael Trimarchi <michael@amarulasolutions.com>
|
|
Fix a logical inversion of the printed text.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Reviewed-by: Sam Protsenko <semen.protsenko@linaro.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
|
|
Drop all duplicate newlines. No functional change.
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
|
|
As part of bringing the master branch back in to next, we need to allow
for all of these changes to exist here.
Reported-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Tom Rini <trini@konsulko.com>
|
|
When bringing in the series 'arm: dts: am62-beagleplay: Fix Beagleplay
Ethernet"' I failed to notice that b4 noticed it was based on next and
so took that as the base commit and merged that part of next to master.
This reverts commit c8ffd1356d42223cbb8c86280a083cc3c93e6426, reversing
changes made to 2ee6f3a5f7550de3599faef9704e166e5dcace35.
Reported-by: Jonas Karlman <jonas@kwiboo.se>
Signed-off-by: Tom Rini <trini@konsulko.com>
|
|
Remove <common.h> from this driver directory and when needed
add missing include files directly.
Signed-off-by: Tom Rini <trini@konsulko.com>
|
|
This function is a no-op. Remove it.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20231216193843.2463779-3-seanga2@gmail.com
|
|
Nothing uses this function. Remove it. Since clk_free no longer does
anything, just stub it out.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20231216193843.2463779-2-seanga2@gmail.com
|
|
When we call clk_get_rate(), we expect to get clock rate value as
ulong.
In that case we should not use log_ret() macro since it use internally
an int.
Otherwise we may return an invalid/truncated clock rate value.
Signed-off-by: Julien Masson <jmasson@baylibre.com>
Fixes: 5c5992cb90c ("clk: Add debugging for return values")
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Mattijs Korpershoek <mkorpershoek@baylibre.com>
Link: https://lore.kernel.org/r/87o7erv9p4.fsf@baylibre.com
|
|
assign clk_dev_ops(clkp->dev) to ops to ensure correct clk operations
are called on clocks.
This fixes the incorrect enable_count issue as described in [1].
[1]: https://lore.kernel.org/all/SEZPR06MB695927A6DEEEF8489A06897396A7A@SEZPR06MB6959.apcprd06.prod.outlook.com/
Signed-off-by: Yang Xiwen <forbidden405@outlook.com>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20231111-enable_count-v2-2-20e3728600b5@outlook.com
|
|
In clk_enable()/clk_disable() functions, when CCF is activated,
we must pass a private clk struct to enable()/disable() ops functions.
Otherwise, the use of a container_of() construction within these ops
should be banned. Because passing a non-private clk struct to
container_of() results in an out of range error.
At the moment, clk-mux, clk-fixed-factor, clk-gate and possibly other
clocks use container_of() in their enable()/disable() functions.
Therefore, for these functions to work correclty, private clk struct
must be passed.
Signed-off-by: Maksim Kiselev <bigunclemax@gmail.com>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20230905221649.3577929-1-bigunclemax@gmail.com
|
|
The second parameter for clk_release_all is used as an unsigned
(which makes sense) but the function prototype declares it as an int.
This causes warnings/error like such below:
include/clk.h:422:48: error: conversion to ‘int’ from ‘unsigned int’ may change the sign of the result [-Werror=sign-conversion]
422 | return clk_release_all(bulk->clks, bulk->count);
To fix this, changed the type of the count to `unsigned int`
Fixes: 82a8a669b4f7 ("clk: add clk_release_all()")
Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com>
Reviewed-by: Xavier Drudis Ferran <xdrudis@tinet.cat>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20230619104752.278500-1-eugen.hristev@collabora.com
|
|
In commit 5c5992cb90cf ("clk: Add debugging for return values"), a
temporary storage variable was added around the ops->get_rate() call
inside clk_get_rate(), so that the result could be passed through
log_ret.
This temporary variable was declared as an int, yet when we look in
struct clk_ops, we can see this needs to be a ulong:
ulong (*get_rate)(struct clk *clk);
This was resulting in a signed to unsigned casting error on our
builds, where a clock value of 0xABCDABCD was being incorrectly cast
to 0xFFFFFFFFABCDABCD.
Signed-off-by: Nathan Barrett-Morrison <nathan.morrison@timesys.com>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20230515195005.1961495-1-nathan.morrison@timesys.com
|
|
There is a chance that assigned-clock-rates is given and assigned-clocks
could be empty. Dont return error in that case, because the probe of the
corresponding driver will not be called at all if this fails.
Better to continue to look for it and return 0.
Signed-off-by: Ashok Reddy Soma <ashok.reddy.soma@amd.com>
Reviewed-by: Tom Rini <trini@konsulko.com>
Signed-off-by: Michal Simek <michal.simek@amd.com>
Link: https://lore.kernel.org/r/a9a9d853e0ac396cd9b3577cce26279a75765711.1693384296.git.michal.simek@amd.com
|
|
This allows devm_clock_get(dev, NULL) to work and get the first clock,
which is common in code ported from Linux.
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Link: https://lore.kernel.org/r/20230122000252.53642-1-samuel@sholland.org
|
|
In clk_clean_rate_cache, clk->rate should update the private clock
struct, in particular when CCF is activated, to save the cached
rate value.
When clk_get_parent_rate is called, the cached information
is read from pclk->rate, with pclk = clk_get_parent(clk).
As the cached is read from private clk data, the update should
be done also on it.
Fixes: 6b7fd3128f7 ("clk: fix set_rate to clean up cached rates for the hierarchy")
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20220620153717.v2.1.Ifa06360115ffa3f3307372e6cdd98ec16759d6ba@changeid
Link: https://lore.kernel.org/r/20220712142352.RESEND.v2.1.Ifa06360115ffa3f3307372e6cdd98ec16759d6ba@changeid/
|
|
All functions getting and setting clock rate use ulong for rate, only
clk_get_parent_rate is an exception. Change the return value to match
other clock rate funcrions.
Most users directly assign the rate to unsigned long anyway, and the few
users that use u64 (not s64) multiply the rate so they may need the
extra bits for the result in their use case.
Fixes: 4aa78300a0 ("dm: clk: Define clk_get_parent_rate() for clk operations")
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20220928103757.11870-1-msuchanek@suse.de
|
|
These functions are exactly the same as their "nodev" varients, except they
accept a device and not an ofnode. Rewrite them to just call the other
function.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20220227190113.1617498-1-seanga2@gmail.com
|
|
Most callers of this function do not check the return value, and it is
unclear what action they should take if it fails. If a function is freeing
multiple clocks, it should not stop just because the first one failed.
Since the callbacks can no longer fail, just convert the return type to
void.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Link: https://lore.kernel.org/r/20220115222504.617013-8-seanga2@gmail.com
|
|
When freeing a clock there is not much we can do if there is an error, and
most callers do not actually check the return value. Even e.g. checking to
make sure that clk->id is valid should have been done in request() in the
first place (unless someone is messing with the driver behind our back).
Just return void and don't bother returning an error.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Link: https://lore.kernel.org/r/20220115222504.617013-2-seanga2@gmail.com
|
|
It is safe to check if the uclass id on the device is UCLASS_CLK
before to call the clk_ functions, but today this comparison is
not done on the device used in API: clkp->dev->parent
but on the device himself: clkp->dev.
This patch corrects this behavior and tests if the parent device
is a clock device before to call the clock API, clk_enable or
clk_disable, on this device.
Fixes: 0520be0f67e3 ("clk: prograte clk enable/disable to parent")
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
|
|
The optional varients of clk_get_* functions are just simple wrappers.
Reduce code size a bit by inlining them. On platforms where it is not used
(most of them), it will not be compiled in any more. On platforms where
they are used, the inlined branch should not cause any significant growth.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Link: https://lore.kernel.org/r/20211222171114.3091780-3-seanga2@gmail.com
|
|
This normalizes the name of this accessor function to put "_optional" last.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Link: https://lore.kernel.org/r/20211222171114.3091780-2-seanga2@gmail.com
|
|
Somewhere along the way, someone misspelt "invalid" and it got copied
everywhere. Fix it.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Alexander Dahl <ada@thorsis.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
|
This reverts commit 92f1e9a4b31c0bf0f4f61ab823a6a88657323646.
The aforementioned patch causes massive breakage on all platforms which
have 'assigned-clock' DT property in their DT which references any clock
that are not supported by the platform clock driver. That can easily
happen either in SPL, or because the clock driver is reduced. Currently
it seems all iMX8M are affected and fail to boot altogether.
Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Peng Fan <peng.fan@oss.nxp.com>
Cc: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
|
Reorder include files in the U-Boot expected order:
the common.h header should always be first,
followed by other headers in order,
then headers with directories,
then local files.
It is a preliminary step for next patch.
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
|
|
This is actually a misnomer now, since the phandle info may contain
a driver_info index or a udevice index. Rename it to use the word
'phandle', which seems more accurate. Add a comment while we are here.
Also add a test for this function.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
Use this new Kconfig to simplify the compilation conditions where
appropriate.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
When the default clocks cannot be set, the clock is silently probed and
the error is ignored. This is incorrect, since having the clocks at the
correct speed may be important for operation of the system.
Fix it by checking the return code.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
Define LOG_CATEGORY for all uclass to allow filtering with
log command.
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
|
Remove the tab in clk_get_bulk to respect the coding rules.
Signed-off-by: Patrick Delaunay <patrick.delaunay@foss.st.com>
Reviewed-by: Patrice Chotard <patrice.chotard@foss.st.com>
|
|
Since 291da96b8e ("clk: Allow clock defaults to be set during re-reloc
state for SPL only") it has been impossible to set clock defaults before
relocation. This is annoying on boards without SPL, since there is no way
to set clock defaults before U-Boot proper. In particular, the aisram rate
must be changed before relocation on the K210, since U-Boot will hang if we
try and change the rate while we are using aisram.
To get around this, extend the stage parameter to allow force setting
defaults, even if they would be otherwise postponed for later. A device
tree property was decided against because of the concerns in the original
commit thread about the overhead of repeatedly parsing the device tree.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
|
Clock rates are cached within the individual clock nodes, and right now
if one changes a clock rate somewhere in the middle of the tree, none
of its child clocks notice the change. To fix this, clear up all the
cached rates for us and our child clocks.
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tero Kristo <kristo@kernel.org>
|
|
If a clock provider is not ready for assigning default rates/parents
during its probe, it may return -EPROBE_DEFER directly from xlate.
Handle this special case properly by skipping the entry and adjusting the
return value to pass. The defaults will be handled properly in post probe
phase then.
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tero Kristo <kristo@kernel.org>
|
|
Bail out early if device returned for the parent clock is null.
This avoids warning prints like this when doing clk dump:
dev_get_uclass_priv: null device
Signed-off-by: Tero Kristo <t-kristo@ti.com>
Signed-off-by: Tero Kristo <kristo@kernel.org>
|
|
If the user/dev explicitly requests a clock be assigned a certain rate,
then we should warn them if we can't do it. This makes it clear if the
clock is running at the default rate.
Signed-off-by: Sean Anderson <seanga2@gmail.com>
|
|
This function finds a device by its driver_info index. With
of-platdata-inst we do not use driver_info, but instead instantiate
udevice records at build-time.
However the semantics of using the function are the same in each case:
the caller provides an index and gets back a device.
So rename the function to device_get_by_ofplat_idx(), so that it can be
used for both situations. The caller does not really need to worry about
the details.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
Move this out of the common header and include it only where needed. In
a number of cases this requires adding "struct udevice;" to avoid adding
another large header or in other cases replacing / adding missing header
files that had been pulled in, very indirectly. Finally, we have a few
cases where we did not need to include <asm/global_data.h> at all, so
remove that include.
Signed-off-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Tom Rini <trini@konsulko.com>
|
|
Use the log_msg_ret() mechanism to get error-return information when
clocks fail to probe, etc.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
It returns the rate which will be set if you ask clk_set_rate() to set
that rate. It provides a way to query exactly what rate you'll get if
you call clk_set_rate() with that same argument.
So essentially, clk_round_rate() and clk_set_rate() are equivalent
except the former does not modify the clock hardware in any way.
Signed-off-by: Dario Binacchi <dariobin@libero.it>
Reviewed-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Sean Anderson <seanga2@gmail.com>
|
|
We have two functions which do the same thing. Standardise on
dev_has_ofnode() since there is no such thing as an 'invalid' ofnode in
normal operation: it is either null or missing.
Also move the functions into one place.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
|
|
At present we use a 'node' pointer in the of-platadata phandle_n_arg
structs. This is a pointer to the struct driver_info for a particular
device, and we can use it to obtain the struct udevice pointer itself.
Since we don't know the struct udevice pointer until it is allocated in
memory, we have to fix up the phandle_n_arg.node at runtime. This is
annoying since it requires that SPL's data is writable and adds a small
amount of extra (generated) code in the dm_populate_phandle_data()
function.
Now that we can find a driver_info by its index, it is easier to put the
index in the phandle_n_arg structures.
Update dtoc to do this, add a new device_get_by_driver_info_idx() to look
up a device by drive_info index and update the tests to match.
Signed-off-by: Simon Glass <sjg@chromium.org>
|
|
The cell_count argument is required when cells_name is NULL.
This patch adds this parameter in live tree API
- of_count_phandle_with_args
- ofnode_count_phandle_with_args
- dev_count_phandle_with_args
This parameter solves issue when these API is used to count
the number of element of a cell without cell name. This parameter
allow to force the size cell.
For example:
count = dev_count_phandle_with_args(dev, "array", NULL, 3);
Signed-off-by: Patrick Delaunay <patrick.delaunay@st.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
|
