From: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
To: Anusha Arun Nandi <anusha.nandi@oss.qualcomm.com>,
david@ixit.cz, andersson@kernel.org, konradybcio@kernel.org,
robh@kernel.org, krzk+dt@kernel.org, conor+dt@kernel.org,
bod@kernel.org, vladimir.zapolskiy@linaro.org,
loic.poulain@oss.qualcomm.com, mchehab@kernel.org
Cc: linux-arm-msm@vger.kernel.org, devicetree@vger.kernel.org,
linux-kernel@vger.kernel.org,
gjorgji.rosikopulos@oss.qualcomm.com,
jigarkumar.zala@oss.qualcomm.com,
hariram.purushothaman@oss.qualcomm.com, akapatra@quicinc.com,
linux-media@vger.kernel.org,
krzysztof.kozlowski@oss.qualcomm.com, daniel@quora.org,
konrad.dybcio@oss.qualcomm.com, christopher.obbard@linaro.org,
alex.vinarskis@gmail.com
Subject: Re: [PATCH 1/3] phy: core: add generic MIPI C-PHY configuration API
Date: Sun, 18 Oct 2026 00:00:58 +0100 [thread overview]
Message-ID: <a6d8514b-269c-441d-b280-4895b5f21fac@linaro.org> (raw)
In-Reply-To: <20261017041733.3946816-2-anusha.nandi@oss.qualcomm.com>
On 17/10/2026 05:17, Anusha Arun Nandi wrote:
> Add PHY_MODE_MIPI_CPHY, struct phy_configure_opts_mipi_cphy and three
> helpers: phy_mipi_cphy_get_default_config(),
> phy_mipi_cphy_get_default_config_for_hsclk() and
> phy_mipi_cphy_config_validate(). These mirror the existing MIPI D-PHY
> generic configuration API (phy-core-mipi-dphy.c) so PHY providers can
> be configured for C-PHY via phy_configure().
>
> C-PHY encodes 16 bits per 7 symbols across a 3-wire trio; the default
> config helper derives the required per-trio symbol rate from the pixel
> clock, bits-per-pixel and lane (trio) count the same way the D-PHY
> helper derives its HS clock rate. The timing fields are filled in from
> the MIPI Alliance Specification for C-PHY (v1.0), Table 18 "Global
> Operation Timing Parameters"; unlike D-PHY, these are not data-rate
> dependent, so phy_mipi_cphy_get_default_config() fills the spec's
> minimum values directly rather than computing them from the symbol
> rate, and phy_mipi_cphy_config_validate() checks each field against
> its spec-defined bound.
>
> No existing phy consumer or provider selects GENERIC_PHY_MIPI_CPHY
> yet, so this commit has no functional effect on its own.
>
> Co-developed-by: Jigarkumar Zala <jigarkumar.zala@oss.qualcomm.com>
> Signed-off-by: Jigarkumar Zala <jigarkumar.zala@oss.qualcomm.com>
> Signed-off-by: Anusha Arun Nandi <anusha.nandi@oss.qualcomm.com>
> ---
> drivers/phy/Kconfig | 9 ++
> drivers/phy/Makefile | 1 +
> drivers/phy/phy-core-mipi-cphy.c | 131 ++++++++++++++++++++++
> include/linux/phy/phy-mipi-cphy.h | 177 ++++++++++++++++++++++++++++++
> include/linux/phy/phy.h | 5 +
> 5 files changed, 323 insertions(+)
> create mode 100644 drivers/phy/phy-core-mipi-cphy.c
> create mode 100644 include/linux/phy/phy-mipi-cphy.h
>
> diff --git a/drivers/phy/Kconfig b/drivers/phy/Kconfig
> index 19f3b7d12b7d..a6b0749a40cc 100644
> --- a/drivers/phy/Kconfig
> +++ b/drivers/phy/Kconfig
> @@ -46,6 +46,15 @@ config GENERIC_PHY_MIPI_DPHY
> Provides a number of helpers a core functions for MIPI D-PHY
> drivers to us.
>
> +config GENERIC_PHY_MIPI_CPHY
> + bool
> + select GENERIC_PHY
> + help
> + Generic MIPI C-PHY configuration helpers.
> +
> + Provides a number of helper core functions for MIPI C-PHY
> + drivers to use.
> +
> config PHY_AIROHA_PCIE
> tristate "Airoha PCIe-PHY Driver"
> depends on ARCH_AIROHA || COMPILE_TEST
> diff --git a/drivers/phy/Makefile b/drivers/phy/Makefile
> index d7aa516bcc49..a94dc1b352e1 100644
> --- a/drivers/phy/Makefile
> +++ b/drivers/phy/Makefile
> @@ -7,6 +7,7 @@ obj-$(CONFIG_PHY_COMMON_PROPS) += phy-common-props.o
> obj-$(CONFIG_PHY_COMMON_PROPS_TEST) += phy-common-props-test.o
> obj-$(CONFIG_GENERIC_PHY) += phy-core.o
> obj-$(CONFIG_GENERIC_PHY_MIPI_DPHY) += phy-core-mipi-dphy.o
> +obj-$(CONFIG_GENERIC_PHY_MIPI_CPHY) += phy-core-mipi-cphy.o
> obj-$(CONFIG_PHY_AIROHA_PCIE) += phy-airoha-pcie.o
> obj-$(CONFIG_PHY_CAN_TRANSCEIVER) += phy-can-transceiver.o
> obj-$(CONFIG_PHY_ECONET_PCIE) += phy-econet-pcie.o
> diff --git a/drivers/phy/phy-core-mipi-cphy.c b/drivers/phy/phy-core-mipi-cphy.c
> new file mode 100644
> index 000000000000..303b9a02f096
> --- /dev/null
> +++ b/drivers/phy/phy-core-mipi-cphy.c
> @@ -0,0 +1,131 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
> + *
> + * Generic MIPI C-PHY configuration helpers.
> + */
> +#include <linux/phy/phy-mipi-cphy.h>
> +#include <linux/errno.h>
> +#include <linux/export.h>
> +#include <linux/kernel.h>
> +#include <linux/math64.h>
> +
> +/*
> + * C-PHY carries 16 bits per 7 symbols per trio (lane).
> + */
> +#define PHY_MIPI_CPHY_SYMBOLS_PER_GROUP 7
> +#define PHY_MIPI_CPHY_BITS_PER_GROUP 16
> +
> +/*
> + * Fill in the C-PHY global operation timing parameters based on the MIPI
> + * Alliance Specification for C-PHY (v1.0), Table 18 "Global Operation Timing
> + * Parameters". These parameters are not data rate dependent (see sections
> + * 6.14.3 and 6.14.4 of that specification), so unlike the D-PHY equivalent
> + * this helper does not need the symbol rate to compute them.
> + */
> +static int phy_mipi_cphy_calc_config(unsigned long pixel_clock,
> + unsigned int bpp, unsigned int lanes,
> + unsigned long long hs_clk_rate,
> + struct phy_configure_opts_mipi_cphy *cfg)
> +{
> + if (!cfg || !lanes)
> + return -EINVAL;
Lanes must be a multiple of three to be valid, at least for your three
submissions to be logically consistent, this statement must be so.
> +
> + if (!hs_clk_rate) {
> + if (!bpp)
> + return -EINVAL;
> +
> + /*
> + * Total payload bits/s = pixel_clock * bpp, spread over the
> + * active trios. C-PHY carries 16 bits per 7 symbols per trio,
> + * so the per-trio symbol rate is the bit rate scaled by 7/16.
> + */
> + hs_clk_rate = (unsigned long long)pixel_clock * bpp;
> + hs_clk_rate = DIV_ROUND_UP_ULL(hs_clk_rate *
> + PHY_MIPI_CPHY_SYMBOLS_PER_GROUP,
> + (unsigned long long)lanes *
> + PHY_MIPI_CPHY_BITS_PER_GROUP);
> + }
> +
> + cfg->lanes = lanes;
> + cfg->hs_clk_rate = hs_clk_rate;
> +
> + cfg->t3_prepare = 38000;
> + cfg->t3_term_en = 0;
> + cfg->t3_settle = 95000;
> + cfg->t3_hs_exit = 100000;
> + cfg->lpx = 50000;
> + cfg->init = 100;
Whatever about how DPHY does this in upstream Linux, these values should
be defines with names that mean something WRT the specification.
> + cfg->ta_get = 5 * cfg->lpx;
> + cfg->ta_go = 4 * cfg->lpx;
> + cfg->ta_sure = cfg->lpx;
> + cfg->wakeup = 1000;
> +
> + return 0;
> +}
> +
> +int phy_mipi_cphy_get_default_config(unsigned long pixel_clock,
> + unsigned int bpp, unsigned int lanes,
> + struct phy_configure_opts_mipi_cphy *cfg)
> +{
> + if (!bpp)
> + return -EINVAL;
> +
> + return phy_mipi_cphy_calc_config(pixel_clock, bpp, lanes, 0, cfg);
> +}
> +EXPORT_SYMBOL_GPL(phy_mipi_cphy_get_default_config);
> +
> +int phy_mipi_cphy_get_default_config_for_hsclk(unsigned long long hs_clk_rate,
> + unsigned int lanes,
> + struct phy_configure_opts_mipi_cphy *cfg)
> +{
> + if (!hs_clk_rate)
> + return -EINVAL;
> +
> + return phy_mipi_cphy_calc_config(0, 0, lanes, hs_clk_rate, cfg);
> +}
> +EXPORT_SYMBOL_GPL(phy_mipi_cphy_get_default_config_for_hsclk);
> +
> +/*
> + * Validate C-PHY configuration according to the MIPI Alliance
> + * Specification for C-PHY (v1.0), Table 18 "Global Operation Timing
> + * Parameters".
> + */
> +int phy_mipi_cphy_config_validate(struct phy_configure_opts_mipi_cphy *cfg)
> +{
> + if (!cfg || !cfg->lanes)
> + return -EINVAL;
> +
> + if (cfg->t3_prepare < 38000 || cfg->t3_prepare > 95000)
T3_PREPARE_MIN T3_PREPARE_MAX etc I'm stating a preference of course but
I think its better practice to define your constants with names that
mean things as frequently we end up reusing constants all over the place.
> + return -EINVAL;
> +
> + if (cfg->t3_term_en > 38000)
> + return -EINVAL;
> +
> + if (cfg->t3_settle < 95000 || cfg->t3_settle > 300000)
> + return -EINVAL;
> +
> + if (cfg->t3_hs_exit < 100000)
> + return -EINVAL;
> +
> + if (cfg->lpx < 50000)
> + return -EINVAL;
> +
> + if (cfg->init < 100)
> + return -EINVAL;
> +
> + if (cfg->ta_get != 5 * cfg->lpx)
> + return -EINVAL;
> +
> + if (cfg->ta_go != 4 * cfg->lpx)
> + return -EINVAL;
> +
> + if (cfg->ta_sure < cfg->lpx || cfg->ta_sure > (2 * cfg->lpx))
> + return -EINVAL;
> +
> + if (cfg->wakeup < 1000)
> + return -EINVAL;
> +
> + return 0;
> +}
> +EXPORT_SYMBOL_GPL(phy_mipi_cphy_config_validate);
> diff --git a/include/linux/phy/phy-mipi-cphy.h b/include/linux/phy/phy-mipi-cphy.h
> new file mode 100644
> index 000000000000..7627602eb57f
> --- /dev/null
> +++ b/include/linux/phy/phy-mipi-cphy.h
> @@ -0,0 +1,177 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +/*
> + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
> + */
> +#ifndef __PHY_MIPI_CPHY_H_
> +#define __PHY_MIPI_CPHY_H_
> +
> +#include <linux/types.h>
> +
> +#define PHY_MIPI_CPHY_MAX_DATA_LANES 4
> +
> +/**
> + * struct phy_configure_opts_mipi_cphy - MIPI C-PHY configuration set
> + *
> + * This structure is used to represent the configuration state of a
> + * MIPI C-PHY phy. The timing fields are named after their symbol in
> + * the MIPI Alliance Specification for C-PHY (v1.0), Table 18 "Global
> + * Operation Timing Parameters", and are all time values in picoseconds
> + * (microseconds for @init and @wakeup) that are not data rate dependent
> + * (see Table 18 and sections 6.14.3/6.14.4 of that specification).
> + */
> +struct phy_configure_opts_mipi_cphy {
> + /**
> + * @lanes:
> + *
> + * Number of active C-PHY trios (data lanes).
> + */
> + unsigned int lanes;
This statement and how you go about using the data aren't consistent.
In CAMSS you do
/* Set data trio lane positions and polarities (no clock lane) */
for (i = 0; i < num_lanes; i++) {
cphy_cfg->lane_positions[i] = lncfg->data[i].pos;
cphy_cfg->lane_polarities[i] = lncfg->data[i].pol;
}
where num_lanes == lncfg->num_data;
Let's assume dt has data-lanes = <1 2 3>; => num_data == 3
Then in qcom-mipi-csi2-core you do:
/* C-PHY has no clock lane; all entries map data trios. */
for (i = 0; i < csi2phy->stream_cfg.num_data_lanes; i++) {
if (data_lanes[i] < 1 || data_lanes[i] > CSI2_MAX_DATA_LANES) {
dev_err(dev, "Invalid lane %d\n", data_lanes[i]);
ret = -EINVAL;
goto out_put;
}
csi2phy->stream_cfg.lane_cfg.data[i].pos = data_lanes[i] - 1;
csi2phy->stream_cfg.lane_cfg.data[i].pol = lane_polarities[i];
}
This specifies which physical lanes will be in CPHY mode and what their
polarities will be
=>
lanes specifies the number of pads in a trio
data_lanes specifies which pads are in those trios
lane_polarities specifies the polarity of each of the given pads
To my knowledge the C-PHY specification doesn't mandate adjacent LNx as
required in a trio.
=> lanes % 3 should be invalid
=> for each trio lanes must increase by 3 also.
> +
> + /**
> + * @hs_clk_rate:
> + *
> + * High speed symbol rate in symbols/s derived from link rate.
> + *
> + * C-PHY encodes 16 bits per 7 symbols across a 3-wire trio.
> + * Consumers fill this from the sensor link frequency and hand
> + * it to phy_configure().
> + */
> + unsigned long hs_clk_rate;
> +
> + /**
> + * @lane_positions:
> + *
> + * Physical-to-logical trio (data lane) position mapping. Entry @i
> + * gives the physical lane position carrying logical trio @i. C-PHY
> + * has no clock lane, so all entries map data trios only.
> + */
> + unsigned char lane_positions[PHY_MIPI_CPHY_MAX_DATA_LANES];
> +
> + /**
> + * @lane_polarities:
> + *
> + * Polarity (0 = normal, 1 = swapped) of each logical trio. Entry
> + * @i corresponds to logical trio @i.
> + */
> + bool lane_polarities[PHY_MIPI_CPHY_MAX_DATA_LANES];
> +
> + /**
> + * @t3_prepare:
> + *
> + * Time, in picoseconds, that the transmitter drives the
> + * 3-wire LP-000 line state immediately before the HS_+x line
> + * state starting HS transmission.
> + *
> + * Minimum value: 38000 ps
> + * Maximum value: 95000 ps
> + */
> + unsigned int t3_prepare;
> +
> + /**
> + * @t3_term_en:
> + *
> + * Time, in picoseconds, for the slave to enable the HS line
> + * termination, starting from the time point when the A, B
> + * and C wires cross VIL_MAX.
> + *
> + * Maximum value: 38000 ps
> + */
> + unsigned int t3_term_en;
> +
> + /**
> + * @t3_settle:
> + *
> + * Time interval, in picoseconds, during which the HS receiver
> + * should ignore any HS transitions on the lane, starting from
> + * the beginning of @t3_prepare.
> + *
> + * Minimum value: 95000 ps
> + * Maximum value: 300000 ps
> + */
> + unsigned int t3_settle;
> +
> + /**
> + * @t3_hs_exit:
> + *
> + * Time, in picoseconds, that the transmitter drives LP-111
> + * following an HS burst.
> + *
> + * Minimum value: 100000 ps
> + */
> + unsigned int t3_hs_exit;
> +
> + /**
> + * @lpx:
> + *
> + * Transmitted length, in picoseconds, of any low-power state
> + * period.
> + *
> + * Minimum value: 50000 ps
> + */
> + unsigned int lpx;
> +
> + /**
> + * @init:
> + *
> + * Time, in microseconds, for the initialization period to
> + * complete.
> + *
> + * Minimum value: 100 us
> + */
> + unsigned int init;
> +
> + /**
> + * @ta_get:
> + *
> + * Time, in picoseconds, that the new transmitter drives the
> + * Bridge state (LP-000) after accepting control during a
> + * link turnaround.
> + *
> + * Value: 5 * @lpx
> + */
> + unsigned int ta_get;
> +
> + /**
> + * @ta_go:
> + *
> + * Time, in picoseconds, that the transmitter drives the
> + * Bridge state (LP-000) before releasing control during a
> + * link turnaround.
> + *
> + * Value: 4 * @lpx
> + */
> + unsigned int ta_go;
> +
> + /**
> + * @ta_sure:
> + *
> + * Time, in picoseconds, that the new transmitter waits after
> + * the LP-100 state before transmitting the Bridge state
> + * (LP-000) during a link turnaround.
> + *
> + * Minimum value: @lpx
> + * Maximum value: 2 * @lpx
> + */
> + unsigned int ta_sure;
> +
> + /**
> + * @wakeup:
> + *
> + * Time, in microseconds, that a transmitter drives a Mark-1
> + * state prior to a Stop state in order to initiate an exit
> + * from ULPS.
> + *
> + * Minimum value: 1000 us
> + */
> + unsigned int wakeup;
> +};
Should the following fields be included in this definition too ?
- T3-PREBEGIN
- T3-PROGSEQ
- T3-PREEND
- T3-SYNC
- T3-POST
> +
> +int phy_mipi_cphy_get_default_config(unsigned long pixel_clock,
> + unsigned int bpp, unsigned int lanes,
> + struct phy_configure_opts_mipi_cphy *cfg);
> +int phy_mipi_cphy_get_default_config_for_hsclk(unsigned long long hs_clk_rate,
> + unsigned int lanes,
> + struct phy_configure_opts_mipi_cphy *cfg);
> +int phy_mipi_cphy_config_validate(struct phy_configure_opts_mipi_cphy *cfg);
> +
> +#endif /* __PHY_MIPI_CPHY_H_ */
> diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h
> index c480d6cb42c5..f12337834031 100644
> --- a/include/linux/phy/phy.h
> +++ b/include/linux/phy/phy.h
> @@ -19,6 +19,7 @@
> #include <linux/phy/phy-dp.h>
> #include <linux/phy/phy-hdmi.h>
> #include <linux/phy/phy-lvds.h>
> +#include <linux/phy/phy-mipi-cphy.h>
> #include <linux/phy/phy-mipi-dphy.h>
> #include <linux/phy/phy-thunderbolt.h>
>
> @@ -42,6 +43,7 @@ enum phy_mode {
> PHY_MODE_PCIE,
> PHY_MODE_ETHERNET,
> PHY_MODE_MIPI_DPHY,
> + PHY_MODE_MIPI_CPHY,
> PHY_MODE_SATA,
> PHY_MODE_LVDS,
> PHY_MODE_DP,
> @@ -69,6 +71,8 @@ union phy_notify {
> *
> * @mipi_dphy: Configuration set applicable for phys supporting
> * the MIPI_DPHY phy mode.
> + * @mipi_cphy: Configuration set applicable for phys supporting
> + * the MIPI_CPHY phy mode.
> * @dp: Configuration set applicable for phys supporting
> * the DisplayPort protocol.
> * @lvds: Configuration set applicable for phys supporting
> @@ -78,6 +82,7 @@ union phy_notify {
> */
> union phy_configure_opts {
> struct phy_configure_opts_mipi_dphy mipi_dphy;
> + struct phy_configure_opts_mipi_cphy mipi_cphy;
> struct phy_configure_opts_dp dp;
> struct phy_configure_opts_lvds lvds;
> struct phy_configure_opts_hdmi hdmi;
next prev parent reply other threads:[~2026-10-17 23:01 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-10-17 4:17 [PATCH 0/3] phy: core, qcom-mipi-csi2: add generic MIPI C-PHY support Anusha Arun Nandi
2026-10-17 4:17 ` [PATCH 1/3] phy: core: add generic MIPI C-PHY configuration API Anusha Arun Nandi
2026-10-17 23:00 ` Bryan O'Donoghue [this message]
2026-10-19 10:24 ` Bryan O'Donoghue
2026-10-19 11:24 ` Bryan O'Donoghue
2026-10-19 14:15 ` Loic Poulain
2026-10-17 4:17 ` [PATCH 2/3] phy: qcom-mipi-csi2: Add sa8775p C-PHY support Anusha Arun Nandi
2026-10-17 23:19 ` Bryan O'Donoghue
2026-10-17 4:17 ` [PATCH 3/3] phy: qcom-mipi-csi2: Add sc7280 " Anusha Arun Nandi
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=a6d8514b-269c-441d-b280-4895b5f21fac@linaro.org \
--to=bryan.odonoghue@linaro.org \
--cc=akapatra@quicinc.com \
--cc=alex.vinarskis@gmail.com \
--cc=andersson@kernel.org \
--cc=anusha.nandi@oss.qualcomm.com \
--cc=bod@kernel.org \
--cc=christopher.obbard@linaro.org \
--cc=conor+dt@kernel.org \
--cc=daniel@quora.org \
--cc=david@ixit.cz \
--cc=devicetree@vger.kernel.org \
--cc=gjorgji.rosikopulos@oss.qualcomm.com \
--cc=hariram.purushothaman@oss.qualcomm.com \
--cc=jigarkumar.zala@oss.qualcomm.com \
--cc=konrad.dybcio@oss.qualcomm.com \
--cc=konradybcio@kernel.org \
--cc=krzk+dt@kernel.org \
--cc=krzysztof.kozlowski@oss.qualcomm.com \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=loic.poulain@oss.qualcomm.com \
--cc=mchehab@kernel.org \
--cc=robh@kernel.org \
--cc=vladimir.zapolskiy@linaro.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
all inboxes | Powered by JetHome®