[PATCH v2 0/3] clk: Implement a clock request API

[Maxime Ripard <maxime@cerno.tech>]

Hi,\r
\r
This is a follow-up of the discussion here:\r
https://lore.kernel.org/linux-clk/20210319150355.xzw7ikwdaga2dwhv@gilmour/\r
\r
This implements a mechanism to raise and lower clock rates based on consumer\r
workloads, with an example of such an implementation for the RaspberryPi4 HDMI\r
controller.\r
\r
There's a couple of things worth discussing:\r
\r
  - The name is in conflict with clk_request_rate, and even though it feels\r
    like the right name to me, we should probably avoid any confusion\r
\r
  - The code so far implements a policy of always going for the lowest rate\r
    possible. While we don't have an use-case for something else, this should\r
    maybe be made more flexible?\r
\r
Let me know what you think\r
Maxime\r
\r
Changes from v1:\r
  - Return NULL in clk_request_start if clk pointer is NULL\r
  - Test for clk_req pointer in clk_request_done\r
  - Add another user in vc4\r
  - Rebased on top of v5.15-rc1\r
\r
Dom Cobley (1):\r
  drm/vc4: hvs: Convert to the new clock request API\r
\r
Maxime Ripard (2):\r
  clk: Introduce a clock request API\r
  drm/vc4: hdmi: Convert to the new clock request API\r
\r
 drivers/clk/clk.c              | 126 +++++++++++++++++++++++++++++++++\r
 drivers/gpu/drm/vc4/vc4_hdmi.c |  15 ++--\r
 drivers/gpu/drm/vc4/vc4_hdmi.h |   3 +\r
 drivers/gpu/drm/vc4/vc4_kms.c  |   5 +-\r
 include/linux/clk.h            |   4 ++\r
 5 files changed, 146 insertions(+), 7 deletions(-)\r
\r
-- \r
2.31.1\r
\r