From: Laura Nao <laura.nao@collabora.com>
To: laura.nao@collabora.com, wenst@chromium.org
Cc: angelogioacchino.delregno@collabora.com, conor+dt@kernel.org,
devicetree@vger.kernel.org, guangjie.song@mediatek.com,
kernel@collabora.com, krzk+dt@kernel.org,
linux-arm-kernel@lists.infradead.org, linux-clk@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-mediatek@lists.infradead.org,
matthias.bgg@gmail.com, mturquette@baylibre.com,
netdev@vger.kernel.org, nfraprado@collabora.com,
p.zabel@pengutronix.de, richardcochran@gmail.com,
robh@kernel.org, sboyd@kernel.org
Subject: Re: [PATCH v4 02/27] clk: mediatek: clk-pll: Add ops for PLLs using set/clr regs and FENC
Date: Thu, 28 Aug 2025 11:09:58 +0200 [thread overview]
Message-ID: <20250828090958.29653-1-laura.nao@collabora.com> (raw)
In-Reply-To: <20250825123952.208448-1-laura.nao@collabora.com>
On 8/25/25 14:39, Laura Nao wrote:
> Hi Chen-Yu,
>
> On 8/15/25 05:18, Chen-Yu Tsai wrote:
>> On Tue, Aug 5, 2025 at 10:55 PM Laura Nao <laura.nao@collabora.com> wrote:
>>>
>>> MT8196 uses a combination of set/clr registers to control the PLL
>>> enable state, along with a FENC bit to check the preparation status.
>>> Add new set of PLL clock operations with support for set/clr enable and
>>> FENC status logic.
>>>
>>> Reviewed-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
>>> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
>>> Signed-off-by: Laura Nao <laura.nao@collabora.com>
>>> ---
>>> drivers/clk/mediatek/clk-pll.c | 42 +++++++++++++++++++++++++++++++++-
>>> drivers/clk/mediatek/clk-pll.h | 5 ++++
>>> 2 files changed, 46 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/clk/mediatek/clk-pll.c b/drivers/clk/mediatek/clk-pll.c
>>> index 49ca25dd5418..8f46de77f42d 100644
>>> --- a/drivers/clk/mediatek/clk-pll.c
>>> +++ b/drivers/clk/mediatek/clk-pll.c
>>> @@ -37,6 +37,13 @@ int mtk_pll_is_prepared(struct clk_hw *hw)
>>> return (readl(pll->en_addr) & BIT(pll->data->pll_en_bit)) != 0;
>>> }
>>>
>>> +static int mtk_pll_fenc_is_prepared(struct clk_hw *hw)
>>> +{
>>> + struct mtk_clk_pll *pll = to_mtk_clk_pll(hw);
>>> +
>>> + return readl(pll->fenc_addr) & pll->fenc_mask;
>>
>> Nits:
>>
>> I'd do a double-negate (!!) just to indicate that we only care about
>> true or false.
>>
>> Also, why do we need to store fenc_mask instead of just shifting the bit
>> here? Same goes for the register address. |pll| has the base address.
>> Why do we need to pre-calculate it?
>>
>> The code is OK; it just seems a bit wasteful on memory.
>>
>
> Thanks for the heads up - since these are only used here for now, I
> agree they’re not really adding value. I’ll drop fenc_mask and fenc_addr
> in the next revision.
>
On second thought, since pll->base_addr is initialized as base +
data->reg (i.e. at CON0 offset), and the FENC status register sits
before that in the register map, it can’t be derived directly from
pll->base_addr.
So we need to keep fenc_addr, but I’ll drop fenc_mask and shift the bit
inline instead.
Laura
> Best,
>
> Laura
>
>> Either way, this is
>>
>> Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
>>
>>> +}
>>> +
>>> static unsigned long __mtk_pll_recalc_rate(struct mtk_clk_pll *pll, u32 fin,
>>> u32 pcw, int postdiv)
>>> {
>>> @@ -274,6 +281,25 @@ void mtk_pll_unprepare(struct clk_hw *hw)
>>> writel(r, pll->pwr_addr);
>>> }
>>>
>>> +static int mtk_pll_prepare_setclr(struct clk_hw *hw)
>>> +{
>>> + struct mtk_clk_pll *pll = to_mtk_clk_pll(hw);
>>> +
>>> + writel(BIT(pll->data->pll_en_bit), pll->en_set_addr);
>>> +
>>> + /* Wait 20us after enable for the PLL to stabilize */
>>> + udelay(20);
>>> +
>>> + return 0;
>>> +}
>>> +
>>> +static void mtk_pll_unprepare_setclr(struct clk_hw *hw)
>>> +{
>>> + struct mtk_clk_pll *pll = to_mtk_clk_pll(hw);
>>> +
>>> + writel(BIT(pll->data->pll_en_bit), pll->en_clr_addr);
>>> +}
>>> +
>>> const struct clk_ops mtk_pll_ops = {
>>> .is_prepared = mtk_pll_is_prepared,
>>> .prepare = mtk_pll_prepare,
>>> @@ -283,6 +309,16 @@ const struct clk_ops mtk_pll_ops = {
>>> .set_rate = mtk_pll_set_rate,
>>> };
>>>
>>> +const struct clk_ops mtk_pll_fenc_clr_set_ops = {
>>> + .is_prepared = mtk_pll_fenc_is_prepared,
>>> + .prepare = mtk_pll_prepare_setclr,
>>> + .unprepare = mtk_pll_unprepare_setclr,
>>> + .recalc_rate = mtk_pll_recalc_rate,
>>> + .round_rate = mtk_pll_round_rate,
>>> + .set_rate = mtk_pll_set_rate,
>>> +};
>>> +EXPORT_SYMBOL_GPL(mtk_pll_fenc_clr_set_ops);
>>> +
>>> struct clk_hw *mtk_clk_register_pll_ops(struct mtk_clk_pll *pll,
>>> const struct mtk_pll_data *data,
>>> void __iomem *base,
>>> @@ -315,6 +351,9 @@ struct clk_hw *mtk_clk_register_pll_ops(struct mtk_clk_pll *pll,
>>> pll->hw.init = &init;
>>> pll->data = data;
>>>
>>> + pll->fenc_addr = base + data->fenc_sta_ofs;
>>> + pll->fenc_mask = BIT(data->fenc_sta_bit);
>>> +
>>> init.name = data->name;
>>> init.flags = (data->flags & PLL_AO) ? CLK_IS_CRITICAL : 0;
>>> init.ops = pll_ops;
>>> @@ -337,12 +376,13 @@ struct clk_hw *mtk_clk_register_pll(const struct mtk_pll_data *data,
>>> {
>>> struct mtk_clk_pll *pll;
>>> struct clk_hw *hw;
>>> + const struct clk_ops *pll_ops = data->ops ? data->ops : &mtk_pll_ops;
>>>
>>> pll = kzalloc(sizeof(*pll), GFP_KERNEL);
>>> if (!pll)
>>> return ERR_PTR(-ENOMEM);
>>>
>>> - hw = mtk_clk_register_pll_ops(pll, data, base, &mtk_pll_ops);
>>> + hw = mtk_clk_register_pll_ops(pll, data, base, pll_ops);
>>> if (IS_ERR(hw))
>>> kfree(pll);
>>>
>>> diff --git a/drivers/clk/mediatek/clk-pll.h b/drivers/clk/mediatek/clk-pll.h
>>> index c4d06bb11516..7fdc5267a2b5 100644
>>> --- a/drivers/clk/mediatek/clk-pll.h
>>> +++ b/drivers/clk/mediatek/clk-pll.h
>>> @@ -29,6 +29,7 @@ struct mtk_pll_data {
>>> u32 reg;
>>> u32 pwr_reg;
>>> u32 en_mask;
>>> + u32 fenc_sta_ofs;
>>> u32 pd_reg;
>>> u32 tuner_reg;
>>> u32 tuner_en_reg;
>>> @@ -51,6 +52,7 @@ struct mtk_pll_data {
>>> u32 en_clr_reg;
>>> u8 pll_en_bit; /* Assume 0, indicates BIT(0) by default */
>>> u8 pcw_chg_bit;
>>> + u8 fenc_sta_bit;
>>> };
>>>
>>> /*
>>> @@ -72,6 +74,8 @@ struct mtk_clk_pll {
>>> void __iomem *en_addr;
>>> void __iomem *en_set_addr;
>>> void __iomem *en_clr_addr;
>>> + void __iomem *fenc_addr;
>>> + u32 fenc_mask;
>>> const struct mtk_pll_data *data;
>>> };
>>>
>>> @@ -82,6 +86,7 @@ void mtk_clk_unregister_plls(const struct mtk_pll_data *plls, int num_plls,
>>> struct clk_hw_onecell_data *clk_data);
>>>
>>> extern const struct clk_ops mtk_pll_ops;
>>> +extern const struct clk_ops mtk_pll_fenc_clr_set_ops;
>>>
>>> static inline struct mtk_clk_pll *to_mtk_clk_pll(struct clk_hw *hw)
>>> {
>>> --
>>> 2.39.5
>>>
>
next prev parent reply other threads:[~2025-08-28 10:12 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-05 13:54 [PATCH v4 00/27] Add support for MT8196 clock controllers Laura Nao
2025-08-05 13:54 ` [PATCH v4 01/27] clk: mediatek: clk-pll: Add set/clr regs for shared PLL enable control Laura Nao
2025-08-15 3:03 ` Chen-Yu Tsai
2025-08-05 13:54 ` [PATCH v4 02/27] clk: mediatek: clk-pll: Add ops for PLLs using set/clr regs and FENC Laura Nao
2025-08-15 3:18 ` Chen-Yu Tsai
2025-08-25 12:39 ` Laura Nao
2025-08-28 9:09 ` Laura Nao [this message]
2025-08-05 13:54 ` [PATCH v4 03/27] clk: mediatek: clk-mux: Add ops for mux gates with set/clr/upd " Laura Nao
2025-08-15 3:23 ` Chen-Yu Tsai
2025-08-25 12:42 ` Laura Nao
2025-08-05 13:54 ` [PATCH v4 04/27] clk: mediatek: clk-mtk: Introduce mtk_clk_get_hwv_regmap() Laura Nao
2025-08-15 3:25 ` Chen-Yu Tsai
2025-08-05 13:54 ` [PATCH v4 05/27] clk: mediatek: clk-mux: Add ops for mux gates with HW voter and FENC Laura Nao
2025-08-15 3:31 ` Chen-Yu Tsai
2025-08-25 12:45 ` Laura Nao
2025-08-05 13:54 ` [PATCH v4 06/27] clk: mediatek: clk-gate: Refactor mtk_clk_register_gate to use mtk_gate struct Laura Nao
2025-08-15 3:42 ` Chen-Yu Tsai
2025-08-25 12:49 ` Laura Nao
2025-08-05 13:54 ` [PATCH v4 07/27] clk: mediatek: clk-gate: Add ops for gates with HW voter Laura Nao
2025-08-15 3:37 ` Chen-Yu Tsai
2025-08-25 12:51 ` Laura Nao
2025-08-25 14:50 ` Chen-Yu Tsai
2025-08-26 8:36 ` Laura Nao
2025-08-05 13:54 ` [PATCH v4 08/27] clk: mediatek: clk-mtk: Add MUX_DIV_GATE macro Laura Nao
2025-08-15 3:43 ` Chen-Yu Tsai
2025-08-05 13:54 ` [PATCH v4 09/27] dt-bindings: clock: mediatek: Describe MT8196 clock controllers Laura Nao
2025-08-07 6:58 ` Krzysztof Kozlowski
2025-08-05 13:54 ` [PATCH v4 10/27] clk: mediatek: Add MT8196 apmixedsys clock support Laura Nao
2025-08-05 13:54 ` [PATCH v4 11/27] clk: mediatek: Add MT8196 topckgen " Laura Nao
2025-08-05 13:54 ` [PATCH v4 12/27] clk: mediatek: Add MT8196 topckgen2 " Laura Nao
2025-08-05 13:54 ` [PATCH v4 13/27] clk: mediatek: Add MT8196 vlpckgen " Laura Nao
2025-08-25 13:12 ` Laura Nao
2025-08-05 13:54 ` [PATCH v4 14/27] clk: mediatek: Add MT8196 peripheral " Laura Nao
2025-08-05 13:54 ` [PATCH v4 15/27] clk: mediatek: Add MT8196 ufssys " Laura Nao
2025-08-15 3:50 ` Chen-Yu Tsai
2025-08-25 12:54 ` Laura Nao
2025-08-05 13:54 ` [PATCH v4 16/27] clk: mediatek: Add MT8196 pextpsys " Laura Nao
2025-08-15 3:53 ` Chen-Yu Tsai
2025-08-05 13:54 ` [PATCH v4 17/27] clk: mediatek: Add MT8196 I2C " Laura Nao
2025-08-15 6:13 ` Chen-Yu Tsai
2025-08-05 13:54 ` [PATCH v4 18/27] clk: mediatek: Add MT8196 mcu " Laura Nao
2025-08-05 13:54 ` [PATCH v4 19/27] clk: mediatek: Add MT8196 mdpsys " Laura Nao
2025-08-15 7:16 ` Chen-Yu Tsai
2025-08-05 13:54 ` [PATCH v4 20/27] clk: mediatek: Add MT8196 mfg " Laura Nao
2025-08-05 13:54 ` [PATCH v4 21/27] clk: mediatek: Add MT8196 disp0 " Laura Nao
2025-08-05 13:54 ` [PATCH v4 22/27] clk: mediatek: Add MT8196 disp1 " Laura Nao
2025-08-05 13:54 ` [PATCH v4 23/27] clk: mediatek: Add MT8196 disp-ao " Laura Nao
2025-08-05 13:54 ` [PATCH v4 24/27] clk: mediatek: Add MT8196 ovl0 " Laura Nao
2025-08-05 13:54 ` [PATCH v4 25/27] clk: mediatek: Add MT8196 ovl1 " Laura Nao
2025-08-05 13:54 ` [PATCH v4 26/27] clk: mediatek: Add MT8196 vdecsys " Laura Nao
2025-08-05 13:54 ` [PATCH v4 27/27] clk: mediatek: Add MT8196 vencsys " Laura Nao
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=20250828090958.29653-1-laura.nao@collabora.com \
--to=laura.nao@collabora.com \
--cc=angelogioacchino.delregno@collabora.com \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=guangjie.song@mediatek.com \
--cc=kernel@collabora.com \
--cc=krzk+dt@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-clk@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mediatek@lists.infradead.org \
--cc=matthias.bgg@gmail.com \
--cc=mturquette@baylibre.com \
--cc=netdev@vger.kernel.org \
--cc=nfraprado@collabora.com \
--cc=p.zabel@pengutronix.de \
--cc=richardcochran@gmail.com \
--cc=robh@kernel.org \
--cc=sboyd@kernel.org \
--cc=wenst@chromium.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).