[PATCH v4 0/4] drm/bridge: nwl-dsi: Various fixes

[PATCH v4 0/4] drm/bridge: nwl-dsi: Various fixes

[Esben Haabendal]

This series brings a few patches fixing horizontal sync timing, EOTP packet
behavior and hsync/vsync polarity selection when using DCSS endpoint, as
well as a small fix to properly handle if/when nwl_dsi_mode_set() fails.

The 3rd commit ("Correct auto-insert EOTP behavior") is taken directly from
NXP vendor tree, so I have kept Author unchanged, although I have modified
the commit message. Let me know if I should change Author for that.

Signed-off-by: Esben Haabendal <esben@geanix.com>
---
Changes in v4:
- Change calculation of hfp/hbp/hsa to use the raw clock rates directly to
  eliminate intermediate rounding errors.
- Drop patch 4 (will be included in follow-up patch series, together with
  a patch addressing the unblanced resource teardown and PM runtime
  underflow problems in the error path).
- Link to v3: https://patch.msgid.link/20260810-nwl-dsi-fixes-v3-0-256da9d672e1@geanix.com

Changes in v3:
- Added Fixes and Cc: stable tags.
- Split patch 1 into two, with change of endpoint mux control in a separate
  patch.
- Fixed mux_control_try_select() state argument.
- Link to v2: https://patch.msgid.link/20260803-nwl-dsi-fixes-v2-0-1809a0ac453f@geanix.com

Changes in v2:
- Added safeguard against small horizontal sync timings, protecting against
  division by zero and integer underflow.
- Added error handling of nwl_dsi_mode_set() call.
- Link to v1: https://patch.msgid.link/20260710-nwl-dsi-fixes-v1-0-927d918a86ad@geanix.com

To: Andrzej Hajda <andrzej.hajda@intel.com>
To: Neil Armstrong <neil.armstrong@linaro.org>
To: Robert Foss <rfoss@kernel.org>
To: Laurent Pinchart <Laurent.pinchart@ideasonboard.com>
To: Jonas Karlman <jonas@kwiboo.se>
To: Jernej Skrabec <jernej.skrabec@gmail.com>
To: Luca Ceresoli <luca.ceresoli@bootlin.com>
To: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
To: Maxime Ripard <mripard@kernel.org>
To: Thomas Zimmermann <tzimmermann@suse.de>
To: David Airlie <airlied@gmail.com>
To: Simona Vetter <simona@ffwll.ch>
To: Guido Günther <agx@sigxcpu.org>
To: Fabio Estevam <festevam@gmail.com>
To: Robert Chiras <robert.chiras@nxp.com>
To: Sam Ravnborg <sam@ravnborg.org>
Cc: dri-devel@lists.freedesktop.org
Cc: linux-kernel@vger.kernel.org

---
Esben Haabendal (3):
      drm/bridge: nwl-dsi: Cleanup endpoint mux control selection
      drm/bridge: nwl-dsi: Limit LCDIF specific sync override
      drm/bridge: nwl-dsi: Correct MIPI DSI horizontal sync timing

Robert Chiras (1):
      drm/bridge: nwl-dsi: Correct auto-insert EOTP behavior

 drivers/gpu/drm/bridge/nwl-dsi.c | 87 ++++++++++++++++++++++++++++++++++------
 1 file changed, 74 insertions(+), 13 deletions(-)
---
base-commit: cee9395acd8043be0644b25c34bfa86623f2b935
change-id: 20260710-nwl-dsi-fixes-c148b0adb685

Best regards,
--  
Esben Haabendal <esben@geanix.com>

[PATCH v4 1/4] drm/bridge: nwl-dsi: Cleanup endpoint mux control selection

[Esben Haabendal]

Using the endpoint define values makes it more clear that the argument to
mux_control_try_select() must be valid state integers, and not something
that looks like a boolean.

The endpoint value is kept, as it will be used in a following patch that
changes behavior slightly based on which endpoint is selected.

Cc: stable@vger.kernel.org
Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Esben Haabendal <esben@geanix.com>
---
 drivers/gpu/drm/bridge/nwl-dsi.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nwl-dsi.c
index 09992529f3d8..80e26a947ae6 100644
--- a/drivers/gpu/drm/bridge/nwl-dsi.c
+++ b/drivers/gpu/drm/bridge/nwl-dsi.c
@@ -110,6 +110,8 @@ struct nwl_dsi {
 	int error;
 
 	struct nwl_dsi_transfer *xfer;
+
+	unsigned int endpoint;
 };
 
 static const struct regmap_config nwl_dsi_regmap_config = {
@@ -1088,13 +1090,12 @@ static int nwl_dsi_parse_dt(struct nwl_dsi *dsi)
 static int nwl_dsi_select_input(struct nwl_dsi *dsi)
 {
 	struct device_node *remote;
-	u32 use_dcss = 1;
 	int ret;
 
 	remote = of_graph_get_remote_node(dsi->dev->of_node, 0,
 					  NWL_DSI_ENDPOINT_LCDIF);
 	if (remote) {
-		use_dcss = 0;
+		dsi->endpoint = NWL_DSI_ENDPOINT_LCDIF;
 	} else {
 		remote = of_graph_get_remote_node(dsi->dev->of_node, 0,
 						  NWL_DSI_ENDPOINT_DCSS);
@@ -1103,11 +1104,12 @@ static int nwl_dsi_select_input(struct nwl_dsi *dsi)
 				      "No valid input endpoint found\n");
 			return -EINVAL;
 		}
+		dsi->endpoint = NWL_DSI_ENDPOINT_DCSS;
 	}
 
 	DRM_DEV_INFO(dsi->dev, "Using %s as input source\n",
-		     (use_dcss) ? "DCSS" : "LCDIF");
-	ret = mux_control_try_select(dsi->mux, use_dcss);
+		     (dsi->endpoint == NWL_DSI_ENDPOINT_DCSS) ? "DCSS" : "LCDIF");
+	ret = mux_control_try_select(dsi->mux, dsi->endpoint);
 	if (ret < 0)
 		DRM_DEV_ERROR(dsi->dev, "Failed to select input: %d\n", ret);
 

-- 
2.55.0

[PATCH v4 2/4] drm/bridge: nwl-dsi: Limit LCDIF specific sync override

[Esben Haabendal]

When using DCSS with NWL, overriding the mode flags to enforce active high
sync is preventing the use of active low with downstream bridges, such as
ti-sn65dsi83, which will not see such mode flags set by the panel.

Fixes: 44cfc6233447 ("drm/bridge: Add NWL MIPI DSI host controller support")
Cc: stable@vger.kernel.org
Signed-off-by: Esben Haabendal <esben@geanix.com>
---
 drivers/gpu/drm/bridge/nwl-dsi.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nwl-dsi.c
index 80e26a947ae6..1aab46e1a168 100644
--- a/drivers/gpu/drm/bridge/nwl-dsi.c
+++ b/drivers/gpu/drm/bridge/nwl-dsi.c
@@ -823,10 +823,13 @@ static int nwl_dsi_bridge_atomic_check(struct drm_bridge *bridge,
 				       struct drm_connector_state *conn_state)
 {
 	struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
+	struct nwl_dsi *dsi = bridge_to_dsi(bridge);
 
 	/* At least LCDIF + NWL needs active high sync */
-	adjusted_mode->flags |= (DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
-	adjusted_mode->flags &= ~(DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC);
+	if (dsi->endpoint == NWL_DSI_ENDPOINT_LCDIF) {
+		adjusted_mode->flags |= (DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
+		adjusted_mode->flags &= ~(DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC);
+	}
 
 	/*
 	 * Do a full modeset if crtc_state->active is changed to be true.

-- 
2.55.0

[PATCH v4 3/4] drm/bridge: nwl-dsi: Correct auto-insert EOTP behavior

[Esben Haabendal]

From: Robert Chiras <robert.chiras@nxp.com>

In order to respect the DSI protocol, make sure that auto-insert EOTP is
enabled according to the NO_EOT_PACKET flag instead of the
CLOCK_NON_CONTINUOUS flag.

Fixes: 44cfc6233447 ("drm/bridge: Add NWL MIPI DSI host controller support")
Cc: stable@vger.kernel.org
Signed-off-by: Esben Haabendal <esben@geanix.com>
Cc: Robert Chiras <robert.chiras@nxp.com>
Cc: Laurentiu Palcu <laurentiu.palcu@oss.nxp.com>
---
 drivers/gpu/drm/bridge/nwl-dsi.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nwl-dsi.c
index 1aab46e1a168..f3f4334136c0 100644
--- a/drivers/gpu/drm/bridge/nwl-dsi.c
+++ b/drivers/gpu/drm/bridge/nwl-dsi.c
@@ -222,13 +222,15 @@ static int nwl_dsi_config_host(struct nwl_dsi *dsi)
 	DRM_DEV_DEBUG_DRIVER(dsi->dev, "DSI Lanes %d\n", dsi->lanes);
 	nwl_dsi_write(dsi, NWL_DSI_CFG_NUM_LANES, dsi->lanes - 1);
 
-	if (dsi->dsi_mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS) {
+	if (dsi->dsi_mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS)
 		nwl_dsi_write(dsi, NWL_DSI_CFG_NONCONTINUOUS_CLK, 0x01);
-		nwl_dsi_write(dsi, NWL_DSI_CFG_AUTOINSERT_EOTP, 0x01);
-	} else {
+	else
 		nwl_dsi_write(dsi, NWL_DSI_CFG_NONCONTINUOUS_CLK, 0x00);
+
+	if (dsi->dsi_mode_flags & MIPI_DSI_MODE_NO_EOT_PACKET)
 		nwl_dsi_write(dsi, NWL_DSI_CFG_AUTOINSERT_EOTP, 0x00);
-	}
+	else
+		nwl_dsi_write(dsi, NWL_DSI_CFG_AUTOINSERT_EOTP, 0x01);
 
 	/* values in byte clock cycles */
 	cycles = ui2bc(cfg->clk_pre);

-- 
2.55.0

[PATCH v4 4/4] drm/bridge: nwl-dsi: Correct MIPI DSI horizontal sync timing

[Esben Haabendal]

The NWL MIPI Host controller registers specifies the horizontal front
porch, sync pulse, and back porch in DSI packet payload size in bytes (ie.
not in pixel clocks).

The calculation for this is (mostly) described in section 13.6.3.5.1.2 of
the i.MX 8M Dual/8M QuadLite/8M Quad Applications Processors Reference
Manual (rev. 3.1). The formula shown there does not take packet header size
into account though.

The formula implemented here converts the hfp, hsa, and hbp to DSI packet
bytes and then subtracts the number of packet header bytes.

It is worth noting that these values only needs to match approximately,
according to the reference manual.

Based on commit in NXP vendor tree
commit f2a61699749d ("LF-7981-4: drm/bridge: nwl-dsi: Correct MIPI DSI horizontal sync timing")

Signed-off-by: Esben Haabendal <esben@geanix.com>
Cc: Robert Chiras <robert.chiras@nxp.com>
Cc: Oliver F. Brown <oliver.brown@oss.nxp.com>
---
 drivers/gpu/drm/bridge/nwl-dsi.c | 60 ++++++++++++++++++++++++++++++++++++++--
 1 file changed, 57 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nwl-dsi.c
index f3f4334136c0..eb604733eea0 100644
--- a/drivers/gpu/drm/bridge/nwl-dsi.c
+++ b/drivers/gpu/drm/bridge/nwl-dsi.c
@@ -22,6 +22,7 @@
 #include <linux/reset.h>
 #include <linux/sys_soc.h>
 #include <linux/time64.h>
+#include <linux/math64.h>
 
 #include <drm/drm_atomic_state_helper.h>
 #include <drm/drm_bridge.h>
@@ -264,6 +265,9 @@ static int nwl_dsi_config_dpi(struct nwl_dsi *dsi)
 	bool burst_mode;
 	int hfront_porch, hback_porch, vfront_porch, vback_porch;
 	int hsync_len, vsync_len;
+	int hfp, hbp, hsa;
+	u64 hs_clk_rate, pclk_rate;
+	int h_blank, pkt_hdr_len, pkt_len;
 
 	hfront_porch = dsi->mode.hsync_start - dsi->mode.hdisplay;
 	hsync_len = dsi->mode.hsync_end - dsi->mode.hsync_start;
@@ -317,9 +321,59 @@ static int nwl_dsi_config_dpi(struct nwl_dsi *dsi)
 			      dsi->mode.hdisplay);
 	}
 
-	nwl_dsi_write(dsi, NWL_DSI_HFP, hfront_porch);
-	nwl_dsi_write(dsi, NWL_DSI_HBP, hback_porch);
-	nwl_dsi_write(dsi, NWL_DSI_HSA, hsync_len);
+	hs_clk_rate = dsi->phy_cfg.mipi_dphy.hs_clk_rate;
+	pclk_rate = (u64)dsi->mode.clock * 1000;
+	DRM_DEV_DEBUG_DRIVER(dsi->dev, "hs_clk_rate: %llu Hz\n", hs_clk_rate);
+
+	/*
+	 * Calculate the bytes needed, according to the RM formula:
+	 * Time of DPI event = time to transmit x number of bytes on the DSI interface
+	 * dpi_event_size * dpi_pclk_period = dsi_bytes * 8 * hs_bit_period / num_lanes
+	 * ===>
+	 * dsi_bytes = dpi_event_size * dpi_pclk_period * num_lanes / (8 * hs_bit_period)
+	 * ===> (substituting period = 1 / rate)
+	 * dsi_bytes = dpi_event_size * num_lanes * hs_clk_rate / (8 * dpi_pclk_rate)
+	 */
+	hfp = DIV64_U64_ROUND_UP((u64)hfront_porch * dsi->lanes * hs_clk_rate, 8 * pclk_rate);
+	hbp = DIV64_U64_ROUND_UP((u64)hback_porch * dsi->lanes * hs_clk_rate, 8 * pclk_rate);
+	hsa = DIV64_U64_ROUND_UP((u64)hsync_len * dsi->lanes * hs_clk_rate, 8 * pclk_rate);
+
+	/* Make sure horizontal blankins are even numbers */
+	hfp = roundup(hfp, 2);
+	hbp = roundup(hbp, 2);
+	hsa = roundup(hsa, 2);
+
+	/*
+	 * We need to subtract the packet header length: 32
+	 * In order to make sure we don't get negative values,
+	 * subtract a proportional value to the total length of the
+	 * horizontal blanking duration.
+	 */
+	h_blank = hfp + hbp + hsa;
+
+	if (h_blank) {
+		pkt_len = min(hfp, roundup(((hfp * 100 / h_blank) * 32) / 100, 2));
+		pkt_hdr_len = pkt_len;
+		hfp -= pkt_len;
+
+		pkt_len = min(hbp, roundup(((hbp * 100 / h_blank) * 32) / 100, 2));
+		pkt_hdr_len += pkt_len;
+		hbp -= pkt_len;
+
+		pkt_len = min(hsa, 32 - pkt_hdr_len);
+		pkt_hdr_len += pkt_len;
+		hsa -= pkt_len;
+		if (pkt_hdr_len != 32)
+			dev_warn(dsi->dev,
+				 "Unable to subtract packet header length fully: %d left\n",
+				 32 - pkt_hdr_len);
+	}
+
+	DRM_DEV_DEBUG_DRIVER(dsi->dev, "Register values: hfp=%d hbp=%d hsa=%d\n", hfp, hbp, hsa);
+
+	nwl_dsi_write(dsi, NWL_DSI_HFP, hfp);
+	nwl_dsi_write(dsi, NWL_DSI_HBP, hbp);
+	nwl_dsi_write(dsi, NWL_DSI_HSA, hsa);
 
 	nwl_dsi_write(dsi, NWL_DSI_ENABLE_MULT_PKTS, 0x0);
 	nwl_dsi_write(dsi, NWL_DSI_BLLP_MODE, 0x1);

-- 
2.55.0

Re: [PATCH v4 1/4] drm/bridge: nwl-dsi: Cleanup endpoint mux control selection

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Dangling pointer to stack variable left in dsi->xfer after transfer completes or times out.
--

commit dea46a0fff2442d283e886a1f263495c782e6706
Author: Esben Haabendal <esben@geanix.com>

drm/bridge: nwl-dsi: Cleanup endpoint mux control selection

This commit cleans up the endpoint mux control selection by storing the
endpoint value in the nwl_dsi struct instead of using a boolean flag. It uses
the proper endpoint define values for mux_control_try_select().

Cc: stable@vger.kernel.org

> diff --git a/drivers/gpu/drm/bridge/nwl-dsi.c b/drivers/gpu/drm/bridge/nwl-dsi.c
> index 09992529f3d85..80e26a947ae69 100644
> --- a/drivers/gpu/drm/bridge/nwl-dsi.c
> +++ b/drivers/gpu/drm/bridge/nwl-dsi.c
> @@ -110,6 +110,8 @@ struct nwl_dsi {
>  	int error;
>  
>  	struct nwl_dsi_transfer *xfer;
[Severity: High]
This isn't a bug introduced by this patch, but looking at how dsi->xfer is
handled, could it leave a dangling pointer to a stack variable?

In nwl_dsi_host_transfer(), dsi->xfer is assigned a pointer to a local stack
variable:

drivers/gpu/drm/bridge/nwl-dsi.c:nwl_dsi_host_transfer() {
	struct nwl_dsi_transfer xfer;
	...
	dsi->xfer = &xfer;
	...
	if (!wait_for_completion_timeout(&xfer.completed,
					 NWL_DSI_MIPI_FIFO_TIMEOUT)) {
		DRM_DEV_ERROR(dsi_host->dev, "[%02X] DSI transfer timed out\n",
			      xfer.cmd);
		ret = -ETIMEDOUT;
	} else {
		ret = xfer.status;
	}

	clk_disable_unprepare(dsi->rx_esc_clk);

	return ret;
}

When this function returns (either on success or timeout), dsi->xfer is not
cleared. If a delayed or spurious hardware interrupt fires afterwards,
nwl_dsi_irq_handler() calls nwl_dsi_finish_transmission():

drivers/gpu/drm/bridge/nwl-dsi.c:nwl_dsi_finish_transmission() {
	struct nwl_dsi_transfer *xfer = dsi->xfer;
	bool end_packet = false;

	if (!xfer)
		return;

	if (xfer->direction == DSI_PACKET_SEND &&
	    status & NWL_DSI_TX_PKT_DONE) {
		xfer->status = xfer->tx_len;
	...
}

Does this cause the interrupt handler to dereference and write to the popped
stack frame of nwl_dsi_host_transfer(), potentially corrupting the memory of
whatever thread is currently executing?

> +
> +	unsigned int endpoint;
>  };

[ ... ]

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260831-nwl-dsi-fixes-v4-0-6a2a1cb23910@geanix.com?part=1

Re: [PATCH v4 1/4] drm/bridge: nwl-dsi: Cleanup endpoint mux control selection

[Esben Haabendal]

<sashiko-bot@kernel.org> writes:

> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
>
> Pre-existing issues:
> - [High] Dangling pointer to stack variable left in dsi->xfer after transfer completes or times out.

I will add a fix for this pre-existing issue to the next patch series I
have lined up for the nwl-dsi.c driver.

So let's try and get this one merged :)

/Esben