--- Begin Message ---
- To: Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>
- Subject: Re: [PATCH v3 28/65] clk: renesas: r9a06g032: Add a determine_rate hook
- From: Miquel Raynal <miquel.raynal@xxxxxxxxxxx>
- Date: Tue, 11 Apr 2023 15:09:30 +0200
- Cc: Maxime Ripard <maxime@xxxxxxxxxx>, Michael Turquette <mturquette@xxxxxxxxxxxx>, Stephen Boyd <sboyd@xxxxxxxxxx>, Andreas Färber <afaerber@xxxxxxx>, Manivannan Sadhasivam <mani@xxxxxxxxxx>, Nicolas Ferre <nicolas.ferre@xxxxxxxxxxxxx>, Alexandre Belloni <alexandre.belloni@xxxxxxxxxxx>, Claudiu Beznea <claudiu.beznea@xxxxxxxxxxxxx>, Max Filippov <jcmvbkbc@xxxxxxxxx>, Charles Keepax <ckeepax@xxxxxxxxxxxxxxxxxxxxx>, Richard Fitzgerald <rf@xxxxxxxxxxxxxxxxxxxxx>, Maxime Coquelin <mcoquelin.stm32@xxxxxxxxx>, Alexandre Torgue <alexandre.torgue@xxxxxxxxxxx>, Luca Ceresoli <luca.ceresoli@xxxxxxxxxxx>, David Lechner <david@xxxxxxxxxxxxxx>, Sekhar Nori <nsekhar@xxxxxx>, Abel Vesa <abelvesa@xxxxxxxxxx>, Shawn Guo <shawnguo@xxxxxxxxxx>, Sascha Hauer <s.hauer@xxxxxxxxxxxxxx>, Pengutronix Kernel Team <kernel@xxxxxxxxxxxxxx>, Fabio Estevam <festevam@xxxxxxxxx>, NXP Linux Team <linux-imx@xxxxxxx>, Matthias Brugger <matthias.bgg@xxxxxxxxx>, Geert Uytterhoeven <geert+renesas@xxxxxxxxx>, Dinh Nguyen <dinguyen@xxxxxxxxxx>, Peter De Schrijver <pdeschrijver@xxxxxxxxxx>, Prashant Gaikwad <pgaikwad@xxxxxxxxxx>, Thierry Reding <thierry.reding@xxxxxxxxx>, Jonathan Hunter <jonathanh@xxxxxxxxxx>, Ulf Hansson <ulf.hansson@xxxxxxxxxx>, Linus Walleij <linus.walleij@xxxxxxxxxx>, David Airlie <airlied@xxxxxxxxx>, Daniel Vetter <daniel@xxxxxxxx>, Vinod Koul <vkoul@xxxxxxxxxx>, Kishon Vijay Abraham I <kishon@xxxxxxxxxx>, Alessandro Zummo <a.zummo@xxxxxxxxxxxx>, Chen-Yu Tsai <wens@xxxxxxxx>, Jernej Skrabec <jernej.skrabec@xxxxxxxxx>, Samuel Holland <samuel@xxxxxxxxxxxx>, Liam Girdwood <lgirdwood@xxxxxxxxx>, Mark Brown <broonie@xxxxxxxxxx>, Takashi Iwai <tiwai@xxxxxxxx>, Paul Cercueil <paul@xxxxxxxxxxxxxxx>, Orson Zhai <orsonzhai@xxxxxxxxx>, Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>, Chunyan Zhang <zhang.lyra@xxxxxxxxx>, linux-clk@xxxxxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, AngeloGioacchino Del Regno <angelogioacchino.delregno@xxxxxxxxxxxxx>, linux-arm-kernel@xxxxxxxxxxxxxxxxxxx, linux-actions@xxxxxxxxxxxxxxxxxxx, patches@xxxxxxxxxxxxxxxxxxxxx, linux-stm32@xxxxxxxxxxxxxxxxxxxxxxxxxxxx, linux-mediatek@xxxxxxxxxxxxxxxxxxx, linux-renesas-soc@xxxxxxxxxxxxxxx, linux-tegra@xxxxxxxxxxxxxxx, dri-devel@xxxxxxxxxxxxxxxxxxxxx, linux-phy@xxxxxxxxxxxxxxxxxxx, linux-rtc@xxxxxxxxxxxxxxx, linux-sunxi@xxxxxxxxxxxxxxx, alsa-devel@xxxxxxxxxxxxxxxx, linux-mips@xxxxxxxxxxxxxxx, Gareth Williams <gareth.williams.jx@xxxxxxxxxxx>, Herve Codina <herve.codina@xxxxxxxxxxx>, Ralph Siemsen <ralph.siemsen@xxxxxxxxxx>
- In-reply-to: <CAMuHMdXUEOP_3zjf8nwDyHvZVG-D0AsBjnr=esKzejMMcEiLSQ@mail.gmail.com>
- Organization: Bootlin
- References: <20221018-clk-range-checks-fixes-v3-0-9a1358472d52@cerno.tech> <20221018-clk-range-checks-fixes-v3-28-9a1358472d52@cerno.tech> <CAMuHMdXUEOP_3zjf8nwDyHvZVG-D0AsBjnr=esKzejMMcEiLSQ@mail.gmail.com>
Hi Geert & Maxime,
geert@xxxxxxxxxxxxxx wrote on Tue, 11 Apr 2023 12:27:38 +0200:
> CC Gareth, Hervé, Miquel, Ralph
>
> On Tue, Apr 4, 2023 at 2:44 PM Maxime Ripard <maxime@xxxxxxxxxx> wrote:
> > The Renesas r9a06g032 bitselect clock implements a mux with a set_parent
> > hook, but doesn't provide a determine_rate implementation.
> >
> > This is a bit odd, since set_parent() is there to, as its name implies,
> > change the parent of a clock. However, the most likely candidate to
> > trigger that parent change is a call to clk_set_rate(), with
> > determine_rate() figuring out which parent is the best suited for a
> > given rate.
> >
> > The other trigger would be a call to clk_set_parent(), but it's far less
> > used, and it doesn't look like there's any obvious user for that clock.
> >
> > So, the set_parent hook is effectively unused, possibly because of an
> > oversight. However, it could also be an explicit decision by the
> > original author to avoid any reparenting but through an explicit call to
> > clk_set_parent().
> >
> > The latter case would be equivalent to setting the flag
> > CLK_SET_RATE_NO_REPARENT, together with setting our determine_rate hook
> > to __clk_mux_determine_rate(). Indeed, if no determine_rate
> > implementation is provided, clk_round_rate() (through
> > clk_core_round_rate_nolock()) will call itself on the parent if
> > CLK_SET_RATE_PARENT is set, and will not change the clock rate
> > otherwise. __clk_mux_determine_rate() has the exact same behavior when
> > CLK_SET_RATE_NO_REPARENT is set.
> >
> > And if it was an oversight, then we are at least explicit about our
> > behavior now and it can be further refined down the line.
> >
> > Signed-off-by: Maxime Ripard <maxime@xxxxxxxxxx>
>
> LGTM, so
> Reviewed-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
I searched for 'possible callers', I didn't find any places
where this would be used on the consumer side. However, downstream,
there is a rzn1-clock-bitselect.c clock driver which states:
+ * This clock provider handles the case of the RZN1 where you have peripherals
+ * that have two potential clock source and two gates, one for each of the
+ * clock source - the used clock source (for all sub clocks) is selected by a
+ * single bit.
+ * That single bit affects all sub-clocks, and therefore needs to change the
+ * active gate (and turn the others off) and force a recalculation of the rates.
I don't know how much of this file has been upstreamed (under a
different form) but this might very well be related to the fact that
reparenting in some cases would be a major issue and thus needs to be
avoided unless done on purpose (guessing?).
Maybe Ralph can comment, but for what I understand,
Reviewed-by: Miquel Raynal <miquel.raynal@xxxxxxxxxxx>
> But I do not have the hardware.
>
> > --- a/drivers/clk/renesas/r9a06g032-clocks.c
> > +++ b/drivers/clk/renesas/r9a06g032-clocks.c
> > @@ -1121,6 +1121,7 @@ static int r9a06g032_clk_mux_set_parent(struct clk_hw *hw, u8 index)
> > }
> >
> > static const struct clk_ops clk_bitselect_ops = {
> > + .determine_rate = __clk_mux_determine_rate,
> > .get_parent = r9a06g032_clk_mux_get_parent,
> > .set_parent = r9a06g032_clk_mux_set_parent,
> > };
> > @@ -1145,7 +1146,7 @@ r9a06g032_register_bitsel(struct r9a06g032_priv *clocks,
> >
> > init.name = desc->name;
> > init.ops = &clk_bitselect_ops;
> > - init.flags = CLK_SET_RATE_PARENT;
> > + init.flags = CLK_SET_RATE_PARENT | CLK_SET_RATE_NO_REPARENT;
> > init.parent_names = names;
> > init.num_parents = 2;
> >
>
> Gr{oetje,eeting}s,
>
> Geert
>
Thanks,
Miquèl
--- End Message ---