linux.git/include/linux/clk.h, branch v7.3-rc2

clk: Add devm_clk_bulk_get_enable()

2026-07-28T15:25:30+00:00

devm_clk_bulk_get_optional_enable() gets, prepares and enables a set of
clocks with device-managed cleanup, but treats every clock as optional:
a missing clock is silently returned as NULL instead of failing.

Consumers that need a fixed set of mandatory clocks enabled for the
lifetime of the device currently have to open-code devm_clk_bulk_get()
followed by clk_bulk_prepare_enable(), which loses the managed disable on
unbind, or fall back to per-clock devm_clk_get_enabled() calls.

Add devm_clk_bulk_get_enable() as the non-optional counterpart. The
underlying __devm_clk_bulk_get_enable() helper already supports the
required (optional = false) path, so only export a thin wrapper for it.

Signed-off-by: Suraj Gupta 
Reviewed-by: Brian Masney 
Signed-off-by: Brian Masney

clk: document that clk_get_parent() returns NULL

2026-07-28T15:25:29+00:00

The documentation in the clk.h file says that clk_get_parent() returns
error pointers but it doesn't.  It's also not consistent with the
comments next to the clk_get_parent() implementation which say that
it returns NULL when the clk is NULL.  Update the comments so they
are consistent and accurate and say that it returns NULL.

Signed-off-by: Dan Carpenter 
Reviewed-by: Brian Masney 
Signed-off-by: Brian Masney

Merge tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux

2026-02-15T16:18:57+00:00

Pull clk updates from Stephen Boyd:
 "Not much changed in the clk framework this time except the clk.h
  consumer API moved the context saving APIs around to fix a build error
  in certain configurations.

  There was a change to the core framework for CLK_OPS_PARENT_ENABLE
  behavior during registration, but it wrecked existing drivers that
  didn't expect things to be turned off during clk registration so it
  got reverted.

  This cycle is really a large collection of new clk drivers, primarily
  for Qualcomm SoCs but also for Amlogic, SpacemiT, Google, and Aspeed.
  Another big change in here is support for automatic hardware clock
  gating on Samsung SoCs where the clks turn on and off when needed.
  Ideally more vendors move to this method for better power savings. The
  highlights are in the updates section below.

  Beyond all the new drivers we have a bunch of cleanups like converting
  drivers from divider_round_rate() to divider_determine_rate() and
  using scoped for each OF child loops. Otherwise it's the usual data
  fixes and plugging reference leaks, etc. that's all pretty ordinary
  but not critical enough to fix until the next release.

  New Drivers:
   - Qualcomm Kaanapali global, tcsr, rpmh, display, gpu, camera, and
     video clk controllers
   - Qualcomm SM8750 camera clk controllers
   - Qualcomm MSM8940 and SDM439 global clk controllers
   - Google GS101 Display Process Unit (DPU) clk controllers
   - SpacemiT K3 clk controllers
   - Amlogic t7 clk controllers
   - Aspeed AST2700 clk controllers

  Updates:
   - Convert clock dividers from round_rate() to determine_rate()
   - Fix sparse warnings, kernel-doc warnings, and plug leaked OF refs
   - Automatic hardware clk gating on Google GS101 SoCs
   - Amlogic s4 video clks
   - CAN-FD clks and resets on Renesas RZ/T2H, RZ/N2H, RZ/V2H, and
     RZ/V2N
   - Expanded Serial Peripheral Interface (xSPI) clocks and resets on
     Renesas RZ/T21H and RZ/N2H
   - DMAC, interrupt controller (ICU), SPI, and thermal (TSU) clocks and
     resets on Renesas RZ/V2N
   - More serial (RSCI) clocks and resets on Renesas RZ/V2H and RZ/V2N
   - CPU frequency scaling on T-HEAD TH1520"

* tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux: (165 commits)
  clk: aspeed: Add reset for HACE/VIDEO
  dt-bindings: clock: aspeed: Add VIDEO reset definition
  clk: aspeed: add AST2700 clock driver
  MAINTAINERS: Add entry for ASPEED clock drivers.
  clk: aspeed: Move the existing ASPEED clk drivers into aspeed subdirectory.
  Revert "clk: Respect CLK_OPS_PARENT_ENABLE during recalc"
  clk: Disable KUNIT_UML_PCI
  dt-bindings: clk: rs9: Fix DIF pattern match
  clk: rs9: Convert to DEFINE_SIMPLE_DEV_PM_OPS()
  clk: rs9: Reserve 8 struct clk_hw slots for for 9FGV0841
  clk: qcom: sm8750: Constify 'qcom_cc_desc' in SM8750 camcc
  clk: zynqmp: pll: Fix zynqmp_clk_divider_determine_rate kerneldoc
  clk: zynqmp: divider: Fix zynqmp_clk_divider_determine_rate kerneldoc
  clk: mediatek: Fix error handling in runtime PM setup
  clk: mediatek: don't select clk-mt8192 for all ARM64 builds
  clk: mediatek: Add mfg_eb as parent to mt8196 mfgpll clocks
  clk: mediatek: Refactor pllfh registration to pass device
  clk: mediatek: Pass device to clk_hw_register for PLLs
  clk: mediatek: Refactor pll registration to pass device
  clk: Respect CLK_OPS_PARENT_ENABLE during recalc
  ...

clk: Add devm_clk_bulk_get_optional_enable() helper

2026-01-22T02:57:07+00:00

Add a new managed clock framework helper function that combines getting
optional bulk clocks and enabling them in a single operation.

The devm_clk_bulk_get_optional_enable() function simplifies the common
pattern where drivers need to get optional bulk clocks, prepare and enable
them, and have them automatically disabled/unprepared and freed when the
device is unbound.

This new API follows the established pattern of
devm_clk_bulk_get_all_enabled() and reduces boilerplate code in drivers
that manage multiple optional clocks.

Suggested-by: Andrew Lunn 
Signed-off-by: Suraj Gupta 
Reviewed-by: Brian Masney 
Acked-by: Stephen Boyd 
Link: https://patch.msgid.link/20260116192725.972966-2-suraj.gupta2@amd.com
Signed-off-by: Jakub Kicinski

clk: Annotate #else and #endif

2026-01-16T02:51:45+00:00

Annotate the #else and #endif keywords in large #ifdef/#else/#endif
sections, to improve readability.

Signed-off-by: Geert Uytterhoeven 
Signed-off-by: Stephen Boyd

clk: Merge prepare and unprepare sections

2026-01-16T02:51:44+00:00

 contains two consecutive #ifdef/#else/#endif sections
that check for CONFIG_HAVE_CLK_PREPARE: one for prepare-related
functionality, and a second for unprepare-related functionality.
Reduce #ifdef clutter by merging them.

Signed-off-by: Geert Uytterhoeven 
Signed-off-by: Stephen Boyd

clk: Move clk_{save,restore}_context() to COMMON_CLK section

2026-01-16T02:51:43+00:00

The clk_save_context() and clk_restore_context() helpers are only
implemented by the Common Clock Framework.  They are not available when
using legacy clock frameworks.  Dummy implementations are provided, but
only if no clock support is available at all.

Hence when CONFIG_HAVE_CLK=y, but CONFIG_COMMON_CLK is not enabled:

    m68k-linux-gnu-ld: drivers/net/phy/air_en8811h.o: in function `en8811h_resume':
    air_en8811h.c:(.text+0x83e): undefined reference to `clk_restore_context'
    m68k-linux-gnu-ld: drivers/net/phy/air_en8811h.o: in function `en8811h_suspend':
    air_en8811h.c:(.text+0x856): undefined reference to `clk_save_context'

Fix this by moving forward declarations and dummy implementions from the
HAVE_CLK to the COMMON_CLK section.

Fixes: 8b95d1ce3300c411 ("clk: Add functions to save/restore clock context en-masse")
Reported-by: kernel test robot 
Closes: https://lore.kernel.org/oe-kbuild-all/202511301553.eaEz1nEW-lkp@intel.com/
Signed-off-by: Geert Uytterhoeven 
Signed-off-by: Stephen Boyd

clk: Drop obsolete devm_clk_bulk_get_all_enable() helper

2024-12-23T20:53:52+00:00

Commit 265b07df758a ("clk: Provide managed helper to get and enable bulk
clocks") added devm_clk_bulk_get_all_enable() function, but missed to
return the number of clocks stored in the clk_bulk_data table referenced
by the clks argument.  Without knowing the number, it's not possible to
iterate these clocks when needed, hence the argument is useless and
could have been simply removed.

A new helper devm_clk_bulk_get_all_enabled() has been introduced, which
is consistent with devm_clk_bulk_get_all() in terms of the returned
value.

Drop the obsolete function since all users switched to the new helper.

Reviewed-by: AngeloGioacchino Del Regno 
Reviewed-by: Manivannan Sadhasivam 
Signed-off-by: Cristian Ciocaltea 
Link: https://lore.kernel.org/r/20241217-clk_bulk_ena_fix-v5-3-aafbbb245155@collabora.com
Signed-off-by: Stephen Boyd

clk: Provide devm_clk_bulk_get_all_enabled() helper

2024-10-22T20:55:22+00:00

Commit 265b07df758a ("clk: Provide managed helper to get and enable bulk
clocks") added devm_clk_bulk_get_all_enable() function, but missed to
return the number of clocks stored in the clk_bulk_data table referenced
by the clks argument.  Without knowing the number, it's not possible to
iterate these clocks when needed, hence the argument is useless and
could have been simply removed.

Introduce devm_clk_bulk_get_all_enabled() variant, which is consistent
with devm_clk_bulk_get_all() in terms of the returned value:

 > 0 if one or more clocks have been stored
 = 0 if there are no clocks
 < 0 if an error occurred

Moreover, the naming is consistent with devm_clk_get_enabled(), i.e. use
the past form of 'enable'.

To reduce code duplication and improve patch readability, make
devm_clk_bulk_get_all_enable() use the new helper, as suggested by
Stephen Boyd.

Reviewed-by: AngeloGioacchino Del Regno 
Reviewed-by: Manivannan Sadhasivam 
Signed-off-by: Cristian Ciocaltea 
Link: https://lore.kernel.org/r/20241019-clk_bulk_ena_fix-v4-1-57f108f64e70@collabora.com
Signed-off-by: Stephen Boyd

clk: provide devm_clk_get_optional_enabled_with_rate()

2024-09-05T20:51:46+00:00

There are clock users in the kernel that can't use
devm_clk_get_optional_enabled() as they need to set rate after getting
the clock and before enabling it. Provide a managed helper that wraps
these operations in the correct order.

Signed-off-by: Bartosz Golaszewski 
Link: https://lore.kernel.org/r/20240805-clk-new-helper-v2-1-e5fdd1e1d729@linaro.org
Signed-off-by: Stephen Boyd