From: Christian Marangi <ansuelsmth@gmail.com>
To: Bjorn Andersson <bjorn.andersson@linaro.org>
Cc: Andy Gross <agross@kernel.org>,
Michael Turquette <mturquette@baylibre.com>,
Stephen Boyd <sboyd@kernel.org>, Rob Herring <robh+dt@kernel.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
linux-arm-msm@vger.kernel.org, linux-clk@vger.kernel.org,
linux-kernel@vger.kernel.org, devicetree@vger.kernel.org
Subject: Re: [PATCH v5 3/3] clk: qcom: lcc-ipq806x: convert to parent data
Date: Tue, 19 Jul 2022 14:23:02 +0200 [thread overview]
Message-ID: <62d6a229.1c69fb81.d5d0b.ac4a@mx.google.com> (raw)
In-Reply-To: <YtY2NWYq3Xbxu2pc@builder.lan>
On Mon, Jul 18, 2022 at 11:42:29PM -0500, Bjorn Andersson wrote:
> On Thu 07 Jul 19:03 CDT 2022, Christian Marangi wrote:
>
> > Convert lcc-ipq806x driver to parent_data API.
> >
> > Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
> > ---
> > v5:
> > - Fix the same compilation error (don't know what the hell happen
> > to my buildroot)
> > v4:
> > - Fix compilation error
> > v3:
> > - Inline pxo pll4 parent
> > - Change .name from pxo to pxo_board
> >
> > drivers/clk/qcom/lcc-ipq806x.c | 77 ++++++++++++++++++----------------
> > 1 file changed, 42 insertions(+), 35 deletions(-)
> >
> > diff --git a/drivers/clk/qcom/lcc-ipq806x.c b/drivers/clk/qcom/lcc-ipq806x.c
> > index ba90bebba597..72d6aea5be30 100644
> > --- a/drivers/clk/qcom/lcc-ipq806x.c
> > +++ b/drivers/clk/qcom/lcc-ipq806x.c
> > @@ -34,7 +34,9 @@ static struct clk_pll pll4 = {
> > .status_bit = 16,
> > .clkr.hw.init = &(struct clk_init_data){
> > .name = "pll4",
> > - .parent_names = (const char *[]){ "pxo" },
> > + .parent_data = &(const struct clk_parent_data) {
> > + .fw_name = "pxo", .name = "pxo_board",
>
> This changes the behavior from looking for the globally named "pxo" to
> look for the globally named "pxo_board", in the event that no
> clock-names of "pxo" was found (based on the .fw_name).
>
> So you probably want to keep this as .fw_name = "pxo", .name = "pxo".
>
Hi,
I will make this change but just for reference, I could be wrong by
Dimitry pointed out that the pattern is .fw_name pxo .name pxo_board.
The original patch had both set to pxo and it was asked to be changed.
> > + },
> > .num_parents = 1,
> > .ops = &clk_pll_ops,
> > },
> > @@ -64,9 +66,9 @@ static const struct parent_map lcc_pxo_pll4_map[] = {
> > { P_PLL4, 2 }
> > };
> >
> > -static const char * const lcc_pxo_pll4[] = {
> > - "pxo",
> > - "pll4_vote",
> > +static const struct clk_parent_data lcc_pxo_pll4[] = {
> > + { .fw_name = "pxo", .name = "pxo" },
> > + { .fw_name = "pll4_vote", .name = "pll4_vote" },
>
> This is a reference to a clock defined in this same driver, so you can
> use { .hw = &pll4_vote.clkr.hw } to avoid the lookup all together.
>
Eh... pll4_vote is defined in gcc (for some reason) the one we have here
is pll4.
I asked if this could be fixed in some way but it was said that it's
better to not complicate things too much.
> > };
> >
> > static struct freq_tbl clk_tbl_aif_mi2s[] = {
> > @@ -131,18 +133,14 @@ static struct clk_rcg mi2s_osr_src = {
> > .enable_mask = BIT(9),
> > .hw.init = &(struct clk_init_data){
> > .name = "mi2s_osr_src",
> > - .parent_names = lcc_pxo_pll4,
> > - .num_parents = 2,
> > + .parent_data = lcc_pxo_pll4,
> > + .num_parents = ARRAY_SIZE(lcc_pxo_pll4),
> > .ops = &clk_rcg_ops,
> > .flags = CLK_SET_RATE_GATE,
> > },
> > },
> > };
> >
> > -static const char * const lcc_mi2s_parents[] = {
> > - "mi2s_osr_src",
> > -};
> > -
> > static struct clk_branch mi2s_osr_clk = {
> > .halt_reg = 0x50,
> > .halt_bit = 1,
> > @@ -152,7 +150,9 @@ static struct clk_branch mi2s_osr_clk = {
> > .enable_mask = BIT(17),
> > .hw.init = &(struct clk_init_data){
> > .name = "mi2s_osr_clk",
> > - .parent_names = lcc_mi2s_parents,
> > + .parent_hws = (const struct clk_hw*[]){
> > + &mi2s_osr_src.clkr.hw,
> > + },
> > .num_parents = 1,
> > .ops = &clk_branch_ops,
> > .flags = CLK_SET_RATE_PARENT,
> > @@ -167,7 +167,9 @@ static struct clk_regmap_div mi2s_div_clk = {
> > .clkr = {
> > .hw.init = &(struct clk_init_data){
> > .name = "mi2s_div_clk",
> > - .parent_names = lcc_mi2s_parents,
> > + .parent_hws = (const struct clk_hw*[]){
>
> It would be wonderful if you could keep a space between ) and { in
> these.
>
You mean only here or in the entire patch? I assume the latter.
> > + &mi2s_osr_src.clkr.hw,
> > + },
> > .num_parents = 1,
> > .ops = &clk_regmap_div_ops,
> > },
> > @@ -183,7 +185,9 @@ static struct clk_branch mi2s_bit_div_clk = {
> > .enable_mask = BIT(15),
> > .hw.init = &(struct clk_init_data){
> > .name = "mi2s_bit_div_clk",
> > - .parent_names = (const char *[]){ "mi2s_div_clk" },
> > + .parent_hws = (const struct clk_hw*[]){
> > + &mi2s_div_clk.clkr.hw,
> > + },
> > .num_parents = 1,
> > .ops = &clk_branch_ops,
> > .flags = CLK_SET_RATE_PARENT,
> > @@ -191,6 +195,10 @@ static struct clk_branch mi2s_bit_div_clk = {
> > },
> > };
> >
> > +static const struct clk_parent_data lcc_mi2s_bit_div_codec_clk[] = {
> > + { .hw = &mi2s_bit_div_clk.clkr.hw, },
> > + { .fw_name = "mi2s_codec_clk", .name = "mi2s_codec_clk" },
>
> Is mi2s_codec_clk and external clock? I don't see it documented in the
> DT binding. And if we're introducing new clock-names, perhaps we could
> skip the _clk suffix - because obviously it's a clock :)
>
> Regards,
> Bjorn
>
I also didn't find where is mi2s_codec_clk... but yes I will change the
fw_name with the clock with _clk stripped.
Will send v6 with the other question clarified.
> > +};
> >
> > static struct clk_regmap_mux mi2s_bit_clk = {
> > .reg = 0x48,
> > @@ -199,11 +207,8 @@ static struct clk_regmap_mux mi2s_bit_clk = {
> > .clkr = {
> > .hw.init = &(struct clk_init_data){
> > .name = "mi2s_bit_clk",
> > - .parent_names = (const char *[]){
> > - "mi2s_bit_div_clk",
> > - "mi2s_codec_clk",
> > - },
> > - .num_parents = 2,
> > + .parent_data = lcc_mi2s_bit_div_codec_clk,
> > + .num_parents = ARRAY_SIZE(lcc_mi2s_bit_div_codec_clk),
> > .ops = &clk_regmap_mux_closest_ops,
> > .flags = CLK_SET_RATE_PARENT,
> > },
> > @@ -245,8 +250,8 @@ static struct clk_rcg pcm_src = {
> > .enable_mask = BIT(9),
> > .hw.init = &(struct clk_init_data){
> > .name = "pcm_src",
> > - .parent_names = lcc_pxo_pll4,
> > - .num_parents = 2,
> > + .parent_data = lcc_pxo_pll4,
> > + .num_parents = ARRAY_SIZE(lcc_pxo_pll4),
> > .ops = &clk_rcg_ops,
> > .flags = CLK_SET_RATE_GATE,
> > },
> > @@ -262,7 +267,9 @@ static struct clk_branch pcm_clk_out = {
> > .enable_mask = BIT(11),
> > .hw.init = &(struct clk_init_data){
> > .name = "pcm_clk_out",
> > - .parent_names = (const char *[]){ "pcm_src" },
> > + .parent_hws = (const struct clk_hw*[]){
> > + &pcm_src.clkr.hw,
> > + },
> > .num_parents = 1,
> > .ops = &clk_branch_ops,
> > .flags = CLK_SET_RATE_PARENT,
> > @@ -270,6 +277,11 @@ static struct clk_branch pcm_clk_out = {
> > },
> > };
> >
> > +static const struct clk_parent_data lcc_pcm_clk_out_codec_clk[] = {
> > + { .hw = &pcm_clk_out.clkr.hw, },
> > + { .fw_name = "pcm_codec_clk", .name = "pcm_codec_clk" },
> > +};
> > +
> > static struct clk_regmap_mux pcm_clk = {
> > .reg = 0x54,
> > .shift = 10,
> > @@ -277,11 +289,8 @@ static struct clk_regmap_mux pcm_clk = {
> > .clkr = {
> > .hw.init = &(struct clk_init_data){
> > .name = "pcm_clk",
> > - .parent_names = (const char *[]){
> > - "pcm_clk_out",
> > - "pcm_codec_clk",
> > - },
> > - .num_parents = 2,
> > + .parent_data = lcc_pcm_clk_out_codec_clk,
> > + .num_parents = ARRAY_SIZE(lcc_pcm_clk_out_codec_clk),
> > .ops = &clk_regmap_mux_closest_ops,
> > .flags = CLK_SET_RATE_PARENT,
> > },
> > @@ -325,18 +334,14 @@ static struct clk_rcg spdif_src = {
> > .enable_mask = BIT(9),
> > .hw.init = &(struct clk_init_data){
> > .name = "spdif_src",
> > - .parent_names = lcc_pxo_pll4,
> > - .num_parents = 2,
> > + .parent_data = lcc_pxo_pll4,
> > + .num_parents = ARRAY_SIZE(lcc_pxo_pll4),
> > .ops = &clk_rcg_ops,
> > .flags = CLK_SET_RATE_GATE,
> > },
> > },
> > };
> >
> > -static const char * const lcc_spdif_parents[] = {
> > - "spdif_src",
> > -};
> > -
> > static struct clk_branch spdif_clk = {
> > .halt_reg = 0xd4,
> > .halt_bit = 1,
> > @@ -346,7 +351,9 @@ static struct clk_branch spdif_clk = {
> > .enable_mask = BIT(12),
> > .hw.init = &(struct clk_init_data){
> > .name = "spdif_clk",
> > - .parent_names = lcc_spdif_parents,
> > + .parent_hws = (const struct clk_hw*[]){
> > + &spdif_src.clkr.hw,
> > + },
> > .num_parents = 1,
> > .ops = &clk_branch_ops,
> > .flags = CLK_SET_RATE_PARENT,
> > @@ -384,8 +391,8 @@ static struct clk_rcg ahbix_clk = {
> > .enable_mask = BIT(11),
> > .hw.init = &(struct clk_init_data){
> > .name = "ahbix",
> > - .parent_names = lcc_pxo_pll4,
> > - .num_parents = 2,
> > + .parent_data = lcc_pxo_pll4,
> > + .num_parents = ARRAY_SIZE(lcc_pxo_pll4),
> > .ops = &clk_rcg_lcc_ops,
> > },
> > },
> > --
> > 2.36.1
> >
--
Ansuel
next prev parent reply other threads:[~2022-07-19 12:58 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-08 0:03 [PATCH v5 1/3] dt-bindings: clock: add pcm reset for ipq806x lcc Christian Marangi
2022-07-08 0:03 ` [PATCH v5 2/3] clk: qcom: lcc-ipq806x: add reset definition Christian Marangi
2022-07-08 0:03 ` [PATCH v5 3/3] clk: qcom: lcc-ipq806x: convert to parent data Christian Marangi
2022-07-19 4:42 ` Bjorn Andersson
2022-07-19 12:23 ` Christian Marangi [this message]
2022-07-19 15:23 ` Dmitry Baryshkov
2022-07-19 21:41 ` Bjorn Andersson
2022-07-19 13:18 ` Marijn Suijten
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=62d6a229.1c69fb81.d5d0b.ac4a@mx.google.com \
--to=ansuelsmth@gmail.com \
--cc=agross@kernel.org \
--cc=bjorn.andersson@linaro.org \
--cc=devicetree@vger.kernel.org \
--cc=krzysztof.kozlowski+dt@linaro.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-clk@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mturquette@baylibre.com \
--cc=robh+dt@kernel.org \
--cc=sboyd@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).