According to documentation, the DP PHY on x1e80100 has another clock called ref.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, when it comes to power management, this clock needs to be also disabled on suspend. So even though this change breaks the ABI, it is needed in order to make we disable this clock on runtime PM, when that is going to be enabled in the driver.
So rework the driver to allow different number of clocks, fix the dt-bindings schema and add the clock to the DT node as well.
Signed-off-by: Abel Vesa abel.vesa@linaro.org --- Changes in v3: - Use dev_err_probe() on clocks parsing failure. - Explain why the ABI break is necessary. - Drop the extra 'clk' suffix from the clock name. So ref instead of refclk. - Link to v2: https://lore.kernel.org/r/20250903-phy-qcom-edp-add-missing-refclk-v2-0-d88c...
Changes in v2: - Fix schema by adding the minItems, as suggested by Krzysztof. - Use devm_clk_bulk_get_all, as suggested by Konrad. - Rephrase the commit messages to reflect the flexible number of clocks. - Link to v1: https://lore.kernel.org/r/20250730-phy-qcom-edp-add-missing-refclk-v1-0-6f78...
--- Abel Vesa (3): dt-bindings: phy: qcom-edp: Add missing clock for X Elite phy: qcom: edp: Make the number of clocks flexible arm64: dts: qcom: Add missing TCSR ref clock to the DP PHYs
.../devicetree/bindings/phy/qcom,edp-phy.yaml | 28 +++++++++++++++++++++- arch/arm64/boot/dts/qcom/x1e80100.dtsi | 12 ++++++---- drivers/phy/qualcomm/phy-qcom-edp.c | 16 ++++++------- 3 files changed, 43 insertions(+), 13 deletions(-) --- base-commit: 65dd046ef55861190ecde44c6d9fcde54b9fb77d change-id: 20250730-phy-qcom-edp-add-missing-refclk-5ab82828f8e7
Best regards,
On X Elite platform, the eDP PHY uses one more clock called ref.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, even though this change breaks the ABI, it is needed in order to make the driver disables this clock along with the other ones, for a proper bring-down of the entire PHY.
So attach the this ref clock to the PHY.
Cc: stable@vger.kernel.org # v6.10 Fixes: 5d5607861350 ("dt-bindings: phy: qcom-edp: Add X1E80100 PHY compatibles") Signed-off-by: Abel Vesa abel.vesa@linaro.org --- .../devicetree/bindings/phy/qcom,edp-phy.yaml | 28 +++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-)
diff --git a/Documentation/devicetree/bindings/phy/qcom,edp-phy.yaml b/Documentation/devicetree/bindings/phy/qcom,edp-phy.yaml index eb97181cbb9579893b4ee26a39c3559ad87b2fba..bfc4d75f50ff9e31981fe602478f28320545e52b 100644 --- a/Documentation/devicetree/bindings/phy/qcom,edp-phy.yaml +++ b/Documentation/devicetree/bindings/phy/qcom,edp-phy.yaml @@ -37,12 +37,15 @@ properties: - description: PLL register block
clocks: - maxItems: 2 + minItems: 2 + maxItems: 3
clock-names: + minItems: 2 items: - const: aux - const: cfg_ahb + - const: ref
"#clock-cells": const: 1 @@ -64,6 +67,29 @@ required: - "#clock-cells" - "#phy-cells"
+allOf: + - if: + properties: + compatible: + enum: + - qcom,x1e80100-dp-phy + then: + properties: + clocks: + minItems: 3 + maxItems: 3 + clock-names: + minItems: 3 + maxItems: 3 + else: + properties: + clocks: + minItems: 2 + maxItems: 2 + clock-names: + minItems: 2 + maxItems: 2 + additionalProperties: false
examples:
On Tue, Sep 09, 2025 at 10:33:33AM +0300, Abel Vesa wrote:
On X Elite platform, the eDP PHY uses one more clock called ref.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, even though this change breaks the ABI, it is needed in order to make the driver disables this clock along with the other ones, for a proper bring-down of the entire PHY.
So attach the this ref clock to the PHY.
Cc: stable@vger.kernel.org # v6.10 Fixes: 5d5607861350 ("dt-bindings: phy: qcom-edp: Add X1E80100 PHY compatibles") Signed-off-by: Abel Vesa abel.vesa@linaro.org
.../devicetree/bindings/phy/qcom,edp-phy.yaml | 28 +++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-)
Reviewed-by: Krzysztof Kozlowski krzysztof.kozlowski@linaro.org
Best regards, Krzysztof
On X Elite, the DP PHY needs another clock called ref, while all other platforms do not.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, even though this change breaks the ABI, it is needed in order to make the driver disables this clock along with the other ones, for a proper bring-down of the entire PHY.
So in order to handle these clocks on different platforms, make the driver get all the clocks regardless of how many there are provided.
Cc: stable@vger.kernel.org # v6.10 Fixes: db83c107dc29 ("phy: qcom: edp: Add v6 specific ops and X1E80100 platform support") Signed-off-by: Abel Vesa abel.vesa@linaro.org --- drivers/phy/qualcomm/phy-qcom-edp.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/phy/qualcomm/phy-qcom-edp.c b/drivers/phy/qualcomm/phy-qcom-edp.c index f1b51018683d51df064f60440864c6031638670c..ca9bb9d70e29e1a132bd499fb9f74b5837acf45b 100644 --- a/drivers/phy/qualcomm/phy-qcom-edp.c +++ b/drivers/phy/qualcomm/phy-qcom-edp.c @@ -103,7 +103,9 @@ struct qcom_edp {
struct phy_configure_opts_dp dp_opts;
- struct clk_bulk_data clks[2]; + struct clk_bulk_data *clks; + int num_clks; + struct regulator_bulk_data supplies[2];
bool is_edp; @@ -218,7 +220,7 @@ static int qcom_edp_phy_init(struct phy *phy) if (ret) return ret;
- ret = clk_bulk_prepare_enable(ARRAY_SIZE(edp->clks), edp->clks); + ret = clk_bulk_prepare_enable(edp->num_clks, edp->clks); if (ret) goto out_disable_supplies;
@@ -885,7 +887,7 @@ static int qcom_edp_phy_exit(struct phy *phy) { struct qcom_edp *edp = phy_get_drvdata(phy);
- clk_bulk_disable_unprepare(ARRAY_SIZE(edp->clks), edp->clks); + clk_bulk_disable_unprepare(edp->num_clks, edp->clks); regulator_bulk_disable(ARRAY_SIZE(edp->supplies), edp->supplies);
return 0; @@ -1092,11 +1094,9 @@ static int qcom_edp_phy_probe(struct platform_device *pdev) if (IS_ERR(edp->pll)) return PTR_ERR(edp->pll);
- edp->clks[0].id = "aux"; - edp->clks[1].id = "cfg_ahb"; - ret = devm_clk_bulk_get(dev, ARRAY_SIZE(edp->clks), edp->clks); - if (ret) - return ret; + edp->num_clks = devm_clk_bulk_get_all(dev, &edp->clks); + if (edp->num_clks < 0) + return dev_err_probe(dev, edp->num_clks, "failed to parse clocks\n");
edp->supplies[0].supply = "vdda-phy"; edp->supplies[1].supply = "vdda-pll";
The DP PHYs on X1E80100 need the ref clock which is provided by the TCSR CC.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, even though this change breaks the ABI, it is needed in order to make the driver disables this clock along with the other ones, for a proper bring-down of the entire PHY.
So lets attach it to each of the DP PHYs in order to do that.
Cc: stable@vger.kernel.org # v6.9 Fixes: 1940c25eaa63 ("arm64: dts: qcom: x1e80100: Add display nodes") Signed-off-by: Abel Vesa abel.vesa@linaro.org --- arch/arm64/boot/dts/qcom/x1e80100.dtsi | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-)
diff --git a/arch/arm64/boot/dts/qcom/x1e80100.dtsi b/arch/arm64/boot/dts/qcom/x1e80100.dtsi index 737c5dbd1c808300041cc8897ca1f7450e16e019..551fa270e6ecbe6462ebc1736a36be65534d3ae0 100644 --- a/arch/arm64/boot/dts/qcom/x1e80100.dtsi +++ b/arch/arm64/boot/dts/qcom/x1e80100.dtsi @@ -5670,9 +5670,11 @@ mdss_dp2_phy: phy@aec2a00 { <0 0x0aec2000 0 0x1c8>;
clocks = <&dispcc DISP_CC_MDSS_DPTX2_AUX_CLK>, - <&dispcc DISP_CC_MDSS_AHB_CLK>; + <&dispcc DISP_CC_MDSS_AHB_CLK>, + <&tcsr TCSR_EDP_CLKREF_EN>; clock-names = "aux", - "cfg_ahb"; + "cfg_ahb", + "ref";
power-domains = <&rpmhpd RPMHPD_MX>;
@@ -5690,9 +5692,11 @@ mdss_dp3_phy: phy@aec5a00 { <0 0x0aec5000 0 0x1c8>;
clocks = <&dispcc DISP_CC_MDSS_DPTX3_AUX_CLK>, - <&dispcc DISP_CC_MDSS_AHB_CLK>; + <&dispcc DISP_CC_MDSS_AHB_CLK>, + <&tcsr TCSR_EDP_CLKREF_EN>; clock-names = "aux", - "cfg_ahb"; + "cfg_ahb", + "ref";
power-domains = <&rpmhpd RPMHPD_MX>;
On 9/9/2025 3:33 PM, Abel Vesa wrote:
According to documentation, the DP PHY on x1e80100 has another clock called ref.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, when it comes to power management, this clock needs to be also disabled on suspend. So even though this change breaks the ABI, it is needed in order to make we disable this clock on runtime PM, when that is going to be enabled in the driver.
So rework the driver to allow different number of clocks, fix the dt-bindings schema and add the clock to the DT node as well.
Signed-off-by: Abel Vesa abel.vesa@linaro.org
Changes in v3:
- Use dev_err_probe() on clocks parsing failure.
- Explain why the ABI break is necessary.
- Drop the extra 'clk' suffix from the clock name. So ref instead of refclk.
- Link to v2: https://lore.kernel.org/r/20250903-phy-qcom-edp-add-missing-refclk-v2-0-d88c...
Changes in v2:
- Fix schema by adding the minItems, as suggested by Krzysztof.
- Use devm_clk_bulk_get_all, as suggested by Konrad.
- Rephrase the commit messages to reflect the flexible number of clocks.
- Link to v1: https://lore.kernel.org/r/20250730-phy-qcom-edp-add-missing-refclk-v1-0-6f78...
Abel Vesa (3): dt-bindings: phy: qcom-edp: Add missing clock for X Elite phy: qcom: edp: Make the number of clocks flexible arm64: dts: qcom: Add missing TCSR ref clock to the DP PHYs
.../devicetree/bindings/phy/qcom,edp-phy.yaml | 28 +++++++++++++++++++++- arch/arm64/boot/dts/qcom/x1e80100.dtsi | 12 ++++++---- drivers/phy/qualcomm/phy-qcom-edp.c | 16 ++++++------- 3 files changed, 43 insertions(+), 13 deletions(-)
base-commit: 65dd046ef55861190ecde44c6d9fcde54b9fb77d change-id: 20250730-phy-qcom-edp-add-missing-refclk-5ab82828f8e7
Best regards,
Hi,
I'm observing what looks like a related clock failure on sc8280xp when booting without a monitor connected to a DP-to-HDMI bridge on mdss0_dp2.
Do you think sc8280xp might require a similar fix, or could this be a different issue?
[ 0.390390] ------------[ cut here ]------------ [ 0.390398] disp0_cc_mdss_dptx2_link_clk_src: rcg didn't update its configuration. [ 0.390419] WARNING: CPU: 0 PID: 63 at drivers/clk/qcom/clk-rcg2.c:136 update_config+0xa4/0xb0 [ 0.390439] Modules linked in: [ 0.390448] CPU: 0 UID: 0 PID: 63 Comm: kworker/u32:1 Not tainted 6.16.3+ #45 PREEMPT(lazy) [ 0.390455] Hardware name: Qualcomm QRD, BIOS 6.0.250905.BOOT.MXF.1.1.c1-00167-MAKENA-1 09/ 5/2025 [ 0.390460] Workqueue: events_unbound deferred_probe_work_func [ 0.390476] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 0.390482] pc : update_config+0xa4/0xb0 [ 0.390492] lr : update_config+0xa4/0xb0 [ 0.390500] sp : ffff80008351b9e0 [ 0.390504] x29: ffff80008351b9e0 x28: 0000000000000000 x27: ffff0000850ec3c0 [ 0.390515] x26: ffff800081205320 x25: 0000000000000002 x24: 0000000000000000 [ 0.390523] x23: ffff8000812052a0 x22: ffff000080467800 x21: ffff800081207ef0 [ 0.390531] x20: ffff8000822ad6f0 x19: 0000000000000000 x18: ffffffffffc06b68 [ 0.390539] x17: 616c707369642e30 x16: 3030313065613a6d x15: ffff800081474230 [ 0.390547] x14: ffffffffff806b67 x13: 2e6e6f6974617275 x12: 6769666e6f632073 [ 0.390556] x11: 0000000000000058 x10: 0000000000000018 x9 : ffff8000814742b8 [ 0.390565] x8 : 0000000000afffa8 x7 : 0000000000000179 x6 : ffff800081f742b8 [ 0.390574] x5 : ffff800081f742b8 x4 : 0000000000000178 x3 : 00000000fffdffff [ 0.390582] x2 : ffff8000814741f8 x1 : ffff00008091cec0 x0 : 0000000100000000 [ 0.390591] Call trace: [ 0.390595] update_config+0xa4/0xb0 (P) [ 0.390606] clk_rcg2_set_parent+0x58/0x68 [ 0.390617] clk_core_set_parent_nolock+0xc4/0x1e0 [ 0.390630] clk_set_parent+0x40/0x144 [ 0.390638] of_clk_set_defaults+0x12c/0x520 [ 0.390645] platform_probe+0x38/0xdc [ 0.390652] really_probe+0xc0/0x390 [ 0.390657] __driver_probe_device+0x7c/0x150 [ 0.390663] driver_probe_device+0x40/0x120 [ 0.390667] __device_attach_driver+0xbc/0x168 [ 0.390673] bus_for_each_drv+0x74/0xc0 [ 0.390684] __device_attach+0x9c/0x1ac [ 0.390688] device_initial_probe+0x14/0x20 [ 0.390694] bus_probe_device+0x9c/0xa0 [ 0.390703] deferred_probe_work_func+0xa8/0xf8 [ 0.390713] process_one_work+0x150/0x2b0 [ 0.390725] worker_thread+0x2d0/0x3ec [ 0.390731] kthread+0x118/0x1e0 [ 0.390738] ret_from_fork+0x10/0x20 [ 0.390751] ---[ end trace 0000000000000000 ]--- [ 0.390760] clk: failed to reparent disp0_cc_mdss_dptx2_link_clk_src to aec2a00.phy::link_clk: -16 [ 0.401093] ------------[ cut here ]------------ [ 0.401096] disp0_cc_mdss_dptx2_pixel0_clk_src: rcg didn't update its configuration. [ 0.401112] WARNING: CPU: 0 PID: 63 at drivers/clk/qcom/clk-rcg2.c:136 update_config+0xa4/0xb0 [ 0.401126] Modules linked in: [ 0.401132] CPU: 0 UID: 0 PID: 63 Comm: kworker/u32:1 Tainted: G W 6.16.3+ #45 PREEMPT(lazy) [ 0.401141] Tainted: [W]=WARN [ 0.401144] Hardware name: Qualcomm QRD, BIOS 6.0.250905.BOOT.MXF.1.1.c1-00167-MAKENA-1 09/ 5/2025 [ 0.401147] Workqueue: events_unbound deferred_probe_work_func [ 0.401159] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 0.401164] pc : update_config+0xa4/0xb0 [ 0.401174] lr : update_config+0xa4/0xb0 [ 0.401182] sp : ffff80008351b9e0 [ 0.401185] x29: ffff80008351b9e0 x28: 00000000fffffff0 x27: ffff0000850ec3c0 [ 0.401194] x26: ffff800081205320 x25: 0000000000000002 x24: 0000000000000000 [ 0.401203] x23: ffff8000812052a0 x22: ffff000080467800 x21: ffff800081207ea0 [ 0.401211] x20: ffff8000822ad640 x19: 0000000000000000 x18: ffffffffffc07528 [ 0.401219] x17: 32636561206f7420 x16: 0001020ef3c08cb3 x15: ffff800081474230 [ 0.401227] x14: ffffffffff807527 x13: 2e6e6f6974617275 x12: 6769666e6f632073 [ 0.401235] x11: 0000000000000058 x10: 0000000000000018 x9 : ffff8000814742b8 [ 0.401243] x8 : 0000000000afffa8 x7 : 00000000000001a4 x6 : ffff800081f742b8 [ 0.401252] x5 : ffff800081f742b8 x4 : 00000000000001a3 x3 : 00000000fffdffff [ 0.401260] x2 : ffff8000814741f8 x1 : ffff00008091cec0 x0 : 0000000100000000 [ 0.401268] Call trace: [ 0.401271] update_config+0xa4/0xb0 (P) [ 0.401281] clk_rcg2_set_parent+0x58/0x68 [ 0.401291] clk_core_set_parent_nolock+0xc4/0x1e0 [ 0.401299] clk_set_parent+0x40/0x144 [ 0.401308] of_clk_set_defaults+0x12c/0x520 [ 0.401314] platform_probe+0x38/0xdc [ 0.401321] really_probe+0xc0/0x390 [ 0.401325] __driver_probe_device+0x7c/0x150 [ 0.401330] driver_probe_device+0x40/0x120 [ 0.401335] __device_attach_driver+0xbc/0x168 [ 0.401340] bus_for_each_drv+0x74/0xc0 [ 0.401349] __device_attach+0x9c/0x1ac [ 0.401353] device_initial_probe+0x14/0x20 [ 0.401358] bus_probe_device+0x9c/0xa0 [ 0.401367] deferred_probe_work_func+0xa8/0xf8 [ 0.401377] process_one_work+0x150/0x2b0 [ 0.401384] worker_thread+0x2d0/0x3ec [ 0.401390] kthread+0x118/0x1e0 [ 0.401395] ret_from_fork+0x10/0x20 [ 0.401405] ---[ end trace 0000000000000000 ]--- [ 0.401412] clk: failed to reparent disp0_cc_mdss_dptx2_pixel0_clk_src to aec2a00.phy::vco_div_clk: -16
On 25-09-19 19:06:36, Xilin Wu wrote:
On 9/9/2025 3:33 PM, Abel Vesa wrote:
According to documentation, the DP PHY on x1e80100 has another clock called ref.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, when it comes to power management, this clock needs to be also disabled on suspend. So even though this change breaks the ABI, it is needed in order to make we disable this clock on runtime PM, when that is going to be enabled in the driver.
So rework the driver to allow different number of clocks, fix the dt-bindings schema and add the clock to the DT node as well.
Signed-off-by: Abel Vesa abel.vesa@linaro.org
Changes in v3:
- Use dev_err_probe() on clocks parsing failure.
- Explain why the ABI break is necessary.
- Drop the extra 'clk' suffix from the clock name. So ref instead of refclk.
- Link to v2: https://lore.kernel.org/r/20250903-phy-qcom-edp-add-missing-refclk-v2-0-d88c...
Changes in v2:
- Fix schema by adding the minItems, as suggested by Krzysztof.
- Use devm_clk_bulk_get_all, as suggested by Konrad.
- Rephrase the commit messages to reflect the flexible number of clocks.
- Link to v1: https://lore.kernel.org/r/20250730-phy-qcom-edp-add-missing-refclk-v1-0-6f78...
Abel Vesa (3): dt-bindings: phy: qcom-edp: Add missing clock for X Elite phy: qcom: edp: Make the number of clocks flexible arm64: dts: qcom: Add missing TCSR ref clock to the DP PHYs
.../devicetree/bindings/phy/qcom,edp-phy.yaml | 28 +++++++++++++++++++++- arch/arm64/boot/dts/qcom/x1e80100.dtsi | 12 ++++++---- drivers/phy/qualcomm/phy-qcom-edp.c | 16 ++++++------- 3 files changed, 43 insertions(+), 13 deletions(-)
base-commit: 65dd046ef55861190ecde44c6d9fcde54b9fb77d change-id: 20250730-phy-qcom-edp-add-missing-refclk-5ab82828f8e7
Best regards,
Hi,
I'm observing what looks like a related clock failure on sc8280xp when booting without a monitor connected to a DP-to-HDMI bridge on mdss0_dp2.
Am I to understand that this is triggered by this patchset ?
I don't see how though.
Do you think sc8280xp might require a similar fix, or could this be a different issue?
There is no TCSR clock controller on sc8280xp, so it must be something else. My feeling is that this is probably triggered by the link clock source not being parented to the clock generated by the PHY, or PHY PLL isn't locked yet at that point, but I'm not sure.
I'm not able to reproduce this issue on my x13s.
On 9/19/2025 7:54 PM, Abel Vesa wrote:
On 25-09-19 19:06:36, Xilin Wu wrote:
On 9/9/2025 3:33 PM, Abel Vesa wrote:
According to documentation, the DP PHY on x1e80100 has another clock called ref.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, when it comes to power management, this clock needs to be also disabled on suspend. So even though this change breaks the ABI, it is needed in order to make we disable this clock on runtime PM, when that is going to be enabled in the driver.
So rework the driver to allow different number of clocks, fix the dt-bindings schema and add the clock to the DT node as well.
Signed-off-by: Abel Vesa abel.vesa@linaro.org
Changes in v3:
- Use dev_err_probe() on clocks parsing failure.
- Explain why the ABI break is necessary.
- Drop the extra 'clk' suffix from the clock name. So ref instead of refclk.
- Link to v2: https://lore.kernel.org/r/20250903-phy-qcom-edp-add-missing-refclk-v2-0-d88c...
Changes in v2:
- Fix schema by adding the minItems, as suggested by Krzysztof.
- Use devm_clk_bulk_get_all, as suggested by Konrad.
- Rephrase the commit messages to reflect the flexible number of clocks.
- Link to v1: https://lore.kernel.org/r/20250730-phy-qcom-edp-add-missing-refclk-v1-0-6f78...
Abel Vesa (3): dt-bindings: phy: qcom-edp: Add missing clock for X Elite phy: qcom: edp: Make the number of clocks flexible arm64: dts: qcom: Add missing TCSR ref clock to the DP PHYs
.../devicetree/bindings/phy/qcom,edp-phy.yaml | 28 +++++++++++++++++++++- arch/arm64/boot/dts/qcom/x1e80100.dtsi | 12 ++++++---- drivers/phy/qualcomm/phy-qcom-edp.c | 16 ++++++------- 3 files changed, 43 insertions(+), 13 deletions(-)
base-commit: 65dd046ef55861190ecde44c6d9fcde54b9fb77d change-id: 20250730-phy-qcom-edp-add-missing-refclk-5ab82828f8e7
Best regards,
Hi,
I'm observing what looks like a related clock failure on sc8280xp when booting without a monitor connected to a DP-to-HDMI bridge on mdss0_dp2.
Am I to understand that this is triggered by this patchset ?
Sorry, it's not indeed. I just saw this patchset and wondered if it can fix the issue on sc8280xp. Just now I tried adding the missing GCC_EDP2_PHY_CLKREF_EN to DT and gcc driver, but it didn't fix the issue. :(
I don't see how though.
Do you think sc8280xp might require a similar fix, or could this be a different issue?
There is no TCSR clock controller on sc8280xp, so it must be something else. My feeling is that this is probably triggered by the link clock source not being parented to the clock generated by the PHY, or PHY PLL isn't locked yet at that point, but I'm not sure.
I'm not able to reproduce this issue on my x13s.
It only happens when mdss0_dp2 is not connected to a display during boot. I believe laptops usually use mdss0_dp3, and it's always connected.
I guess the Windows Dev Kit may have the same issue, since it also uses mdss0_dp2 as an external mini-DP port.
On Fri, Sep 19, 2025 at 07:06:36PM +0800, Xilin Wu wrote:
On 9/9/2025 3:33 PM, Abel Vesa wrote:
According to documentation, the DP PHY on x1e80100 has another clock called ref.
The current X Elite devices supported upstream work fine without this clock, because the boot firmware leaves this clock enabled. But we should not rely on that. Also, when it comes to power management, this clock needs to be also disabled on suspend. So even though this change breaks the ABI, it is needed in order to make we disable this clock on runtime PM, when that is going to be enabled in the driver.
So rework the driver to allow different number of clocks, fix the dt-bindings schema and add the clock to the DT node as well.
Signed-off-by: Abel Vesa abel.vesa@linaro.org
Changes in v3:
- Use dev_err_probe() on clocks parsing failure.
- Explain why the ABI break is necessary.
- Drop the extra 'clk' suffix from the clock name. So ref instead of refclk.
- Link to v2: https://lore.kernel.org/r/20250903-phy-qcom-edp-add-missing-refclk-v2-0-d88c...
Changes in v2:
- Fix schema by adding the minItems, as suggested by Krzysztof.
- Use devm_clk_bulk_get_all, as suggested by Konrad.
- Rephrase the commit messages to reflect the flexible number of clocks.
- Link to v1: https://lore.kernel.org/r/20250730-phy-qcom-edp-add-missing-refclk-v1-0-6f78...
Abel Vesa (3): dt-bindings: phy: qcom-edp: Add missing clock for X Elite phy: qcom: edp: Make the number of clocks flexible arm64: dts: qcom: Add missing TCSR ref clock to the DP PHYs
.../devicetree/bindings/phy/qcom,edp-phy.yaml | 28 +++++++++++++++++++++- arch/arm64/boot/dts/qcom/x1e80100.dtsi | 12 ++++++---- drivers/phy/qualcomm/phy-qcom-edp.c | 16 ++++++------- 3 files changed, 43 insertions(+), 13 deletions(-)
base-commit: 65dd046ef55861190ecde44c6d9fcde54b9fb77d change-id: 20250730-phy-qcom-edp-add-missing-refclk-5ab82828f8e7
Best regards,
Hi,
I'm observing what looks like a related clock failure on sc8280xp when booting without a monitor connected to a DP-to-HDMI bridge on mdss0_dp2.
Do you think sc8280xp might require a similar fix, or could this be a different issue?
[ 0.390390] ------------[ cut here ]------------ [ 0.390398] disp0_cc_mdss_dptx2_link_clk_src: rcg didn't update its configuration. [ 0.390419] WARNING: CPU: 0 PID: 63 at drivers/clk/qcom/clk-rcg2.c:136 update_config+0xa4/0xb0 [ 0.390439] Modules linked in: [ 0.390448] CPU: 0 UID: 0 PID: 63 Comm: kworker/u32:1 Not tainted 6.16.3+ #45 PREEMPT(lazy) [ 0.390455] Hardware name: Qualcomm QRD, BIOS 6.0.250905.BOOT.MXF.1.1.c1-00167-MAKENA-1 09/ 5/2025 [ 0.390460] Workqueue: events_unbound deferred_probe_work_func [ 0.390476] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 0.390482] pc : update_config+0xa4/0xb0 [ 0.390492] lr : update_config+0xa4/0xb0 [ 0.390500] sp : ffff80008351b9e0 [ 0.390504] x29: ffff80008351b9e0 x28: 0000000000000000 x27: ffff0000850ec3c0 [ 0.390515] x26: ffff800081205320 x25: 0000000000000002 x24: 0000000000000000 [ 0.390523] x23: ffff8000812052a0 x22: ffff000080467800 x21: ffff800081207ef0 [ 0.390531] x20: ffff8000822ad6f0 x19: 0000000000000000 x18: ffffffffffc06b68 [ 0.390539] x17: 616c707369642e30 x16: 3030313065613a6d x15: ffff800081474230 [ 0.390547] x14: ffffffffff806b67 x13: 2e6e6f6974617275 x12: 6769666e6f632073 [ 0.390556] x11: 0000000000000058 x10: 0000000000000018 x9 : ffff8000814742b8 [ 0.390565] x8 : 0000000000afffa8 x7 : 0000000000000179 x6 : ffff800081f742b8 [ 0.390574] x5 : ffff800081f742b8 x4 : 0000000000000178 x3 : 00000000fffdffff [ 0.390582] x2 : ffff8000814741f8 x1 : ffff00008091cec0 x0 : 0000000100000000 [ 0.390591] Call trace: [ 0.390595] update_config+0xa4/0xb0 (P) [ 0.390606] clk_rcg2_set_parent+0x58/0x68 [ 0.390617] clk_core_set_parent_nolock+0xc4/0x1e0 [ 0.390630] clk_set_parent+0x40/0x144 [ 0.390638] of_clk_set_defaults+0x12c/0x520 [ 0.390645] platform_probe+0x38/0xdc [ 0.390652] really_probe+0xc0/0x390 [ 0.390657] __driver_probe_device+0x7c/0x150 [ 0.390663] driver_probe_device+0x40/0x120 [ 0.390667] __device_attach_driver+0xbc/0x168 [ 0.390673] bus_for_each_drv+0x74/0xc0 [ 0.390684] __device_attach+0x9c/0x1ac [ 0.390688] device_initial_probe+0x14/0x20 [ 0.390694] bus_probe_device+0x9c/0xa0 [ 0.390703] deferred_probe_work_func+0xa8/0xf8 [ 0.390713] process_one_work+0x150/0x2b0 [ 0.390725] worker_thread+0x2d0/0x3ec [ 0.390731] kthread+0x118/0x1e0 [ 0.390738] ret_from_fork+0x10/0x20 [ 0.390751] ---[ end trace 0000000000000000 ]--- [ 0.390760] clk: failed to reparent disp0_cc_mdss_dptx2_link_clk_src to aec2a00.phy::link_clk: -16 [ 0.401093] ------------[ cut here ]------------ [ 0.401096] disp0_cc_mdss_dptx2_pixel0_clk_src: rcg didn't update its configuration. [ 0.401112] WARNING: CPU: 0 PID: 63 at drivers/clk/qcom/clk-rcg2.c:136 update_config+0xa4/0xb0 [ 0.401126] Modules linked in: [ 0.401132] CPU: 0 UID: 0 PID: 63 Comm: kworker/u32:1 Tainted: G W 6.16.3+ #45 PREEMPT(lazy) [ 0.401141] Tainted: [W]=WARN [ 0.401144] Hardware name: Qualcomm QRD, BIOS 6.0.250905.BOOT.MXF.1.1.c1-00167-MAKENA-1 09/ 5/2025 [ 0.401147] Workqueue: events_unbound deferred_probe_work_func [ 0.401159] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 0.401164] pc : update_config+0xa4/0xb0 [ 0.401174] lr : update_config+0xa4/0xb0 [ 0.401182] sp : ffff80008351b9e0 [ 0.401185] x29: ffff80008351b9e0 x28: 00000000fffffff0 x27: ffff0000850ec3c0 [ 0.401194] x26: ffff800081205320 x25: 0000000000000002 x24: 0000000000000000 [ 0.401203] x23: ffff8000812052a0 x22: ffff000080467800 x21: ffff800081207ea0 [ 0.401211] x20: ffff8000822ad640 x19: 0000000000000000 x18: ffffffffffc07528 [ 0.401219] x17: 32636561206f7420 x16: 0001020ef3c08cb3 x15: ffff800081474230 [ 0.401227] x14: ffffffffff807527 x13: 2e6e6f6974617275 x12: 6769666e6f632073 [ 0.401235] x11: 0000000000000058 x10: 0000000000000018 x9 : ffff8000814742b8 [ 0.401243] x8 : 0000000000afffa8 x7 : 00000000000001a4 x6 : ffff800081f742b8 [ 0.401252] x5 : ffff800081f742b8 x4 : 00000000000001a3 x3 : 00000000fffdffff [ 0.401260] x2 : ffff8000814741f8 x1 : ffff00008091cec0 x0 : 0000000100000000 [ 0.401268] Call trace: [ 0.401271] update_config+0xa4/0xb0 (P) [ 0.401281] clk_rcg2_set_parent+0x58/0x68 [ 0.401291] clk_core_set_parent_nolock+0xc4/0x1e0 [ 0.401299] clk_set_parent+0x40/0x144 [ 0.401308] of_clk_set_defaults+0x12c/0x520 [ 0.401314] platform_probe+0x38/0xdc [ 0.401321] really_probe+0xc0/0x390 [ 0.401325] __driver_probe_device+0x7c/0x150 [ 0.401330] driver_probe_device+0x40/0x120 [ 0.401335] __device_attach_driver+0xbc/0x168 [ 0.401340] bus_for_each_drv+0x74/0xc0 [ 0.401349] __device_attach+0x9c/0x1ac [ 0.401353] device_initial_probe+0x14/0x20 [ 0.401358] bus_probe_device+0x9c/0xa0 [ 0.401367] deferred_probe_work_func+0xa8/0xf8 [ 0.401377] process_one_work+0x150/0x2b0 [ 0.401384] worker_thread+0x2d0/0x3ec [ 0.401390] kthread+0x118/0x1e0 [ 0.401395] ret_from_fork+0x10/0x20 [ 0.401405] ---[ end trace 0000000000000000 ]--- [ 0.401412] clk: failed to reparent disp0_cc_mdss_dptx2_pixel0_clk_src to aec2a00.phy::vco_div_clk: -16
The same happens on the X1E Devkit if there is nothing connected to the HDMI port. I believe you are looking for my patch series instead. :-)
https://lore.kernel.org/r/20250814-platform-delay-clk-defaults-v1-0-4aae5b33...
If it works for you, replying with a Tested-by there would be much appreciated. I'm still trying to convince folks that the approach of the series is the best way to move forward with this issue. :-)
Thanks, Stephan
linux-stable-mirror@lists.linaro.org
-
Abel Vesa -
Krzysztof Kozlowski -
Stephan Gerhold -
Xilin Wu