Il 17/02/23 05:09, Chen-Yu Tsai ha scritto:

On Tue, Feb 14, 2023 at 9:42 PM AngeloGioacchino Del Regno
[off-list ref] wrote:

quoted

Propagate struct device for divider clocks registered through clk-mtk
helpers to be able to get runtime PM support for MTK clocks.

Sidenote: I think we still need to call pm_runtime_enable() (or some other
function) on the underlying device before any clk_hw_register() calls are
made for the clk core to pick up runtime PM support?

The clk_core checks...

	if (dev && pm_runtime_enabled(dev))
		core->rpm_enabled = true;

...at __clk_register() time, and there's no way to enable it *later*, which
is something that may or may not be good for us.

We could call pm_runtime_enable() on the clock driver itself (so, in simple_probe)
but that's something I didn't want to do right now because this needs a rather big
amount of testing on multiple SoCs.

But yes I agree with you on that we still need to *flip the switch* to actually
enable rpm support. These commits are meant to make us able to do that in one line,
instead of one line per file :-P

quoted

Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>

Thanks! :-)

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel