linux-phy.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Neil Armstrong <neil.armstrong@linaro.org>
To: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Cc: Andrzej Hajda <andrzej.hajda@intel.com>,
	Robert Foss <rfoss@kernel.org>,
	Laurent Pinchart <Laurent.pinchart@ideasonboard.com>,
	Jonas Karlman <jonas@kwiboo.se>,
	Jernej Skrabec <jernej.skrabec@gmail.com>,
	Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
	Maxime Ripard <mripard@kernel.org>,
	Thomas Zimmermann <tzimmermann@suse.de>,
	David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>,
	Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>, Vinod Koul <vkoul@kernel.org>,
	Kishon Vijay Abraham I <kishon@kernel.org>,
	Bjorn Andersson <andersson@kernel.org>,
	Konrad Dybcio <konradybcio@kernel.org>,
	dri-devel@lists.freedesktop.org, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-arm-msm@vger.kernel.org,
	linux-phy@lists.infradead.org
Subject: Re: [PATCH v3 4/5] phy: qcom: qmp-combo: get the USB3 & DisplayPort lanes mapping from DT
Date: Tue, 9 Sep 2025 09:21:33 +0200	[thread overview]
Message-ID: <b8156d1e-a4a5-46dd-8b78-4f2528a177c3@linaro.org> (raw)
In-Reply-To: <dfhvmeo3pnjubgiwvakpzhzgl6uwlopyrkwuqyhfqw6jfh4mty@t4lhuh4jh6af>

On 08/09/2025 23:24, Dmitry Baryshkov wrote:
> On Mon, Sep 08, 2025 at 03:04:21PM +0200, Neil Armstrong wrote:
>> The QMP USB3/DP Combo PHY hosts an USB3 phy and a DP PHY on top
>> of a combo glue to route either lanes to the 4 shared physical lanes.
>>
>> The routing of the lanes can be:
>> - 2 DP + 2 USB3
>> - 4 DP
>> - 2 USB3
>>
>> Get the lanes mapping from DT and stop registering the USB-C
>> muxes in favor of a static mode and orientation detemined
>> by the lanes mapping.
>>
>> This allows supporting boards with direct connection of USB3 and
>> DisplayPort lanes to the QMP Combo PHY lanes, not using the
>> USB-C Altmode feature.
>>
>> Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
>> ---
>>   drivers/phy/qualcomm/phy-qcom-qmp-combo.c | 132 ++++++++++++++++++++++++++++--
>>   1 file changed, 124 insertions(+), 8 deletions(-)
> 
> Looking at the patch... Would it be possible to make it more generic? I
> think some of the RockChips also have similar combo USB+DP PHY (and
> similar issues). I believe, Mediatek might also have the same issue.

Since they don't use the data-lanes property, it's hard to make it generic and I would
avoid migrating their DT to data-lanes just to solve HDMI on the Thinkpas T14s...

Let's solve this first, and code can be aggregated afterwards if the DT representation is correct.

> 
>> diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-combo.c b/drivers/phy/qualcomm/phy-qcom-qmp-combo.c
>> index 7b5af30f1d028c592500e723ecd27b54ed554709..f3f91a69dc8b81e049cd06f7ab4f04baf57776cd 100644
>> --- a/drivers/phy/qualcomm/phy-qcom-qmp-combo.c
>> +++ b/drivers/phy/qualcomm/phy-qcom-qmp-combo.c
>> @@ -13,6 +13,7 @@
>>   #include <linux/module.h>
>>   #include <linux/of.h>
>>   #include <linux/of_address.h>
>> +#include <linux/of_graph.h>
>>   #include <linux/phy/phy.h>
>>   #include <linux/platform_device.h>
>>   #include <linux/regulator/consumer.h>
>> @@ -1744,6 +1745,21 @@ static const u8 qmp_dp_v6_pre_emphasis_hbr_rbr[4][4] = {
>>   	{ 0x22, 0xff, 0xff, 0xff }
>>   };
>>   
>> +static const u32 usb3_data_lane_mapping[][2] = {
>> +	[TYPEC_ORIENTATION_NORMAL] = { 1, 0 },
>> +	[TYPEC_ORIENTATION_REVERSE] = { 2, 3 },
>> +};
>> +
>> +static const u32 dp_2_data_lanes_mapping[][2] = {
>> +	[TYPEC_ORIENTATION_NORMAL] = { 0, 1 },
>> +	[TYPEC_ORIENTATION_REVERSE] = { 3, 2 },
>> +};
>> +
>> +static const u32 dp_4_data_lanes_mapping[][4] = {
>> +	[TYPEC_ORIENTATION_NORMAL] = { 0, 1, 2, 3 },
>> +	[TYPEC_ORIENTATION_REVERSE] = { 3, 2, 1, 0 },
>> +};
>> +
>>   struct qmp_combo;
>>   
>>   struct qmp_combo_offsets {
>> @@ -4167,9 +4183,114 @@ static int qmp_combo_probe(struct platform_device *pdev)
>>   	if (ret)
>>   		goto err_node_put;
>>   
>> -	ret = qmp_combo_typec_register(qmp);
>> -	if (ret)
>> -		goto err_node_put;
>> +	qmp->qmpphy_mode = QMPPHY_MODE_USB3DP;
>> +
>> +	if (of_find_property(dev->of_node, "mode-switch", NULL) ||
>> +	    of_find_property(dev->of_node, "orientation-switch", NULL)) {
>> +		ret = qmp_combo_typec_register(qmp);
>> +		if (ret)
>> +			goto err_node_put;
>> +	} else {
> 
> At least this needs to be extracted to a function (or set of functions).

Yes I was not sure about how to split, and overall I'm not supeer happy about the design.

> 
>> +		enum typec_orientation usb3_orientation = TYPEC_ORIENTATION_NONE;
>> +		enum typec_orientation dp_orientation = TYPEC_ORIENTATION_NONE;
>> +		struct device_node *usb3_ep, *dp_ep;
>> +		u32 data_lanes[4];
>> +		int count, i;
>> +
>> +		usb3_ep = of_graph_get_endpoint_by_regs(dev->of_node, 0, 0);
>> +		dp_ep = of_graph_get_endpoint_by_regs(dev->of_node, 0, 1);
>> +
>> +		if (usb3_ep) {
>> +			ret = of_property_count_u32_elems(usb3_ep, "data-lanes");
>> +			if (ret == -EINVAL)
>> +				/* Property isn't here, ignore property */
> 
> In all thsese error cases we are leaking a ref count on usb3_ep and
> dp_ep.

It would be much simpler to solve in a separate function.

> 
>> +				goto usb3_mapping_done;
>> +			if (ret < 0)
>> +				goto err_node_put;
>> +
>> +			count = ret;
>> +			if (count != 2)
>> +				/* Property size is invalid, ignore property */
>> +				goto usb3_mapping_done;
>> +
>> +			ret = of_property_read_u32_array(usb3_ep, "data-lanes", data_lanes, count);
>> +			if (ret)
>> +				goto err_node_put;
>> +
>> +			for (i = TYPEC_ORIENTATION_NORMAL; i <= TYPEC_ORIENTATION_REVERSE; i++)
>> +				if (!memcmp(data_lanes, usb3_data_lane_mapping[i], sizeof(u32) * 2))
>> +					break;
>> +
>> +			if (i >= TYPEC_ORIENTATION_REVERSE)
>> +				/* Property value is invalid, ignore property */
>> +				goto usb3_mapping_done;
>> +
>> +			usb3_orientation = i;
>> +		}
>> +
>> +usb3_mapping_done:
>> +		of_node_put(usb3_ep);
>> +
>> +		if (dp_ep) {
>> +			ret = of_property_count_u32_elems(dp_ep, "data-lanes");
>> +			if (ret == -EINVAL)
>> +				/* Property isn't here, ignore property */
>> +				goto dp_mapping_done;
>> +			if (ret < 0)
>> +				goto err_node_put;
>> +
>> +			count = ret;
>> +			if (count != 2 && count != 4)
>> +				/* Property size is invalid, ignore property */
>> +				goto dp_mapping_done;
>> +
>> +			ret = of_property_read_u32_array(dp_ep, "data-lanes", data_lanes, count);
>> +
>> +			if (ret)
>> +				goto err_node_put;
>> +
>> +			for (i = TYPEC_ORIENTATION_NORMAL; i <= TYPEC_ORIENTATION_REVERSE; i++) {
>> +				switch (count) {
>> +				case 2:
>> +					ret = memcmp(data_lanes, dp_2_data_lanes_mapping[i],
>> +						      sizeof(u32) * count);
>> +					break;
>> +				case 4:
>> +					ret = memcmp(data_lanes, dp_4_data_lanes_mapping[i],
>> +						     sizeof(u32) * count);
>> +					break;
>> +				}
>> +
>> +				if (!ret)
>> +					break;
>> +			}
>> +
>> +			if (i >= TYPEC_ORIENTATION_REVERSE)
>> +				/* Property value is invalid, ignore property */
>> +				goto dp_mapping_done;
>> +
>> +			dp_orientation = i;
>> +		}
>> +
>> +dp_mapping_done:
>> +		of_node_put(dp_ep);
>> +
>> +		if (dp_orientation == TYPEC_ORIENTATION_NONE &&
>> +		    usb3_orientation != TYPEC_ORIENTATION_NONE) {
>> +			qmp->qmpphy_mode = QMPPHY_MODE_USB3_ONLY;
>> +			qmp->orientation = usb3_orientation;
>> +		} else if (usb3_orientation == TYPEC_ORIENTATION_NONE &&
>> +			 dp_orientation != TYPEC_ORIENTATION_NONE) {
>> +			qmp->qmpphy_mode = QMPPHY_MODE_DP_ONLY;
>> +			qmp->orientation = dp_orientation;
>> +		} else if (dp_orientation != TYPEC_ORIENTATION_NONE &&
>> +			 dp_orientation == usb3_orientation) {
>> +			qmp->qmpphy_mode = QMPPHY_MODE_USB3DP;
>> +			qmp->orientation = dp_orientation;
>> +		} else {
>> +			dev_warn(dev, "unable to determine orientation & mode from data-lanes");
>> +		}
>> +	}
>>   
>>   	ret = drm_aux_bridge_register(dev);
>>   	if (ret)
>> @@ -4189,11 +4310,6 @@ static int qmp_combo_probe(struct platform_device *pdev)
>>   	if (ret)
>>   		goto err_node_put;
>>   
>> -	/*
>> -	 * The hw default is USB3_ONLY, but USB3+DP mode lets us more easily
>> -	 * check both sub-blocks' init tables for blunders at probe time.
>> -	 */
>> -	qmp->qmpphy_mode = QMPPHY_MODE_USB3DP;
>>   
>>   	qmp->usb_phy = devm_phy_create(dev, usb_np, &qmp_combo_usb_phy_ops);
>>   	if (IS_ERR(qmp->usb_phy)) {
>>
>> -- 
>> 2.34.1
>>
> 


-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy

  reply	other threads:[~2025-09-09  8:50 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-08 13:04 [PATCH v3 0/5] arm64: qcom: x1e78100-lenovo-thinkpad-t14s: add support for HDMI output Neil Armstrong
2025-09-08 13:04 ` [PATCH v3 1/5] dt-bindings: display: bridge: simple: document the Realtek RTD2171 DP-to-HDMI bridge Neil Armstrong
2025-09-08 13:04 ` [PATCH v3 2/5] drm/bridge: simple: add " Neil Armstrong
2025-09-08 13:35   ` Maud Spierings
2025-09-08 13:45     ` Dmitry Baryshkov
2025-09-08 14:07       ` neil.armstrong
2025-09-08 13:04 ` [PATCH v3 3/5] dt-bindings: phy: qcom,sc8280xp-qmp-usb43dp-phy: Document lanes mapping when not using in USB-C complex Neil Armstrong
2025-09-08 21:14   ` Dmitry Baryshkov
2025-09-09  7:14     ` Neil Armstrong
2025-09-09 11:16       ` Dmitry Baryshkov
2025-09-09 11:21         ` Konrad Dybcio
2025-09-09 13:58           ` Dmitry Baryshkov
2025-09-12 12:12           ` Dmitry Baryshkov
2025-09-09 12:27         ` Neil Armstrong
2025-09-08 13:04 ` [PATCH v3 4/5] phy: qcom: qmp-combo: get the USB3 & DisplayPort lanes mapping from DT Neil Armstrong
2025-09-08 21:24   ` Dmitry Baryshkov
2025-09-09  7:21     ` Neil Armstrong [this message]
2025-09-09 11:24       ` Dmitry Baryshkov
2025-09-08 13:04 ` [PATCH v3 5/5] arm64: dts: qcom: x1e78100-lenovo-thinkpad-t14s: add HDMI nodes Neil Armstrong
2025-09-08 21:28 ` (subset) [PATCH v3 0/5] arm64: qcom: x1e78100-lenovo-thinkpad-t14s: add support for HDMI output Dmitry Baryshkov

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=b8156d1e-a4a5-46dd-8b78-4f2528a177c3@linaro.org \
    --to=neil.armstrong@linaro.org \
    --cc=Laurent.pinchart@ideasonboard.com \
    --cc=airlied@gmail.com \
    --cc=andersson@kernel.org \
    --cc=andrzej.hajda@intel.com \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=dmitry.baryshkov@oss.qualcomm.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=jernej.skrabec@gmail.com \
    --cc=jonas@kwiboo.se \
    --cc=kishon@kernel.org \
    --cc=konradybcio@kernel.org \
    --cc=krzk+dt@kernel.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-phy@lists.infradead.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=mripard@kernel.org \
    --cc=rfoss@kernel.org \
    --cc=robh@kernel.org \
    --cc=simona@ffwll.ch \
    --cc=tzimmermann@suse.de \
    --cc=vkoul@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).