Skip to contentSkip to conversation

MMC pinmux configuration fixes

PATCH Superseded Judith Mendez 3 participants 17 messages Jul 29 – Aug 3

Subsystems (2)

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

cover MMC pinmux configuration fixes

This series fixes MMC pin input/output and pullup/pulldown settings across AM62x, AM62AX, AM62LX and AM64x boards.

All patches have been tested except patch 5/8, so tested-bys are apreciated.

Judith Mendez (8):
arm64: dts: ti: k3-am62a: Fix MMC0_CLK register address
arm64: dts: ti: k3-am62a: Set MMC clock pins as OUTPUT
arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT
arm64: dts: ti: k3-am62l-evm: Remove unnecessary internal pullups for
MMC0
arm64: dts: ti: k3-am64: Enable IO expander interrupt
arm64: dts: ti: k3-am64: Let MMC1 clock loopback use hardware default
arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls
arm64: dts: ti: k3-am642-evm: Remove unnecessary internal pullups for
MMC1

 arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts      |  2 +-
 .../boot/dts/ti/k3-am62-pocketbeagle2.dts     |  2 +-
 .../arm64/boot/dts/ti/k3-am625-beagleplay.dts |  6 ++--
 .../arm64/boot/dts/ti/k3-am625-sk-common.dtsi |  2 +-
 arch/arm64/boot/dts/ti/k3-am62a7-sk.dts       |  5 ++--
 arch/arm64/boot/dts/ti/k3-am62l3-evm.dts      |  4 +--
 .../arm64/boot/dts/ti/k3-am62x-sk-common.dtsi |  2 +-
 arch/arm64/boot/dts/ti/k3-am642-evm.dts       | 19 +++++++++---
 arch/arm64/boot/dts/ti/k3-am642-sk.dts        | 29 +++++++++++++------
 9 files changed, 46 insertions(+), 25 deletions(-)
Signature · 1 line
2.54.0
#1 lore raw

PATCH 1/8 arm64: dts: ti: k3-am62a: Fix MMC0_CLK register address

MMC0_CLKLB and MMC0_CLK have incorrectly swapped register addresses as per datasheet [0]. Set the correct MMC0_CLK address and drop the unneeded MMC0_CLKLB pin configuration.

[0] https://www.ti.com/lit/gpn/am62a7 Rev. E: Table 5-1. Pin Attributes

Fixes e041ec6e86e7 ("arm64: dts: ti: k3-am62a7-sk: Enable eMMC support")
---
 arch/arm64/boot/dts/ti/k3-am62a7-sk.dts | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
arch/arm64/boot/dts/ti/k3-am62a7-sk.dts +1 -2
index 821a9705bb7d4..826fff38cfb77 100644
@@ -324,8 +324,7 @@ AM62AX_IOPAD(0x0b4, PIN_INPUT_PULLUP, 1) /* (K24) GPMC0_CSn3.I2C2_SDA */
 	main_mmc0_pins_default: main-mmc0-default-pins {
 		pinctrl-single,pins = <
 			AM62AX_IOPAD(0x220, PIN_INPUT, 0) /* (Y3) MMC0_CMD */
-			AM62AX_IOPAD(0x218, PIN_INPUT, 0) /* (AB1) MMC0_CLKLB */
-			AM62AX_IOPAD(0x21c, PIN_INPUT, 0) /* (AB1) MMC0_CLK */
+			AM62AX_IOPAD(0x218, PIN_INPUT, 0) /* (AB1) MMC0_CLK */
 			AM62AX_IOPAD(0x214, PIN_INPUT, 0) /* (AA2) MMC0_DAT0 */
 			AM62AX_IOPAD(0x210, PIN_INPUT_PULLUP, 0) /* (AA1) MMC0_DAT1 */
 			AM62AX_IOPAD(0x20c, PIN_INPUT_PULLUP, 0) /* (AA3) MMC0_DAT2 */
Signature · 1 line
2.54.0
#2 lore raw

PATCH 2/8 arm64: dts: ti: k3-am62a: Set MMC clock pins as OUTPUT

In reply to Judith Mendez

MMC clock pins must be configured as OUTPUT-only per TRM [0] with receiver disabled (RXACTIVE bit = 0). Change MMC CLK pins from PIN_INPUT to PIN_OUTPUT to match documentation.

[0] https://www.ti.com/lit/pdf/spruj16 Rev. C: Table 12-257. MMCSDi I/O Signals

---
 arch/arm64/boot/dts/ti/k3-am62a7-sk.dts | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/ti/k3-am62a7-sk.dts +2 -2
index 826fff38cfb77..9b120cd9b0c37 100644
@@ -324,7 +324,7 @@ AM62AX_IOPAD(0x0b4, PIN_INPUT_PULLUP, 1) /* (K24) GPMC0_CSn3.I2C2_SDA */
 	main_mmc0_pins_default: main-mmc0-default-pins {
 		pinctrl-single,pins = <
 			AM62AX_IOPAD(0x220, PIN_INPUT, 0) /* (Y3) MMC0_CMD */
-			AM62AX_IOPAD(0x218, PIN_INPUT, 0) /* (AB1) MMC0_CLK */
+			AM62AX_IOPAD(0x218, PIN_OUTPUT, 0) /* (AB1) MMC0_CLK */
 			AM62AX_IOPAD(0x214, PIN_INPUT, 0) /* (AA2) MMC0_DAT0 */
 			AM62AX_IOPAD(0x210, PIN_INPUT_PULLUP, 0) /* (AA1) MMC0_DAT1 */
 			AM62AX_IOPAD(0x20c, PIN_INPUT_PULLUP, 0) /* (AA3) MMC0_DAT2 */
@@ -340,7 +340,7 @@ AM62AX_IOPAD(0x1f8, PIN_INPUT_PULLUP, 0) /* (AC2) MMC0_DAT7 */
 	main_mmc1_pins_default: main-mmc1-default-pins {
 		pinctrl-single,pins = <
 			AM62AX_IOPAD(0x23c, PIN_INPUT, 0) /* (A21) MMC1_CMD */
-			AM62AX_IOPAD(0x234, PIN_INPUT, 0) /* (B22) MMC1_CLK */
+			AM62AX_IOPAD(0x234, PIN_OUTPUT, 0) /* (B22) MMC1_CLK */
 			AM62AX_IOPAD(0x230, PIN_INPUT, 0) /* (A22) MMC1_DAT0 */
 			AM62AX_IOPAD(0x22c, PIN_INPUT, 0) /* (B21) MMC1_DAT1 */
 			AM62AX_IOPAD(0x228, PIN_INPUT, 0) /* (C21) MMC1_DAT2 */
Signature · 1 line
2.54.0
#3 lore raw

PATCH 3/8 arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT

In reply to Judith Mendez

MMC clock pins must be configured as OUTPUT-only per TRM [0] with receiver disabled (RXACTIVE bit = 0). Change MMC CLK pins from PIN_INPUT to PIN_OUTPUT to match documentation.

[0] https://www.ti.com/lit/pdf/spruiv7 Rev. C: Table 12-246. MMCSDi I/O Signals

---
 arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts         | 2 +-
 arch/arm64/boot/dts/ti/k3-am62-pocketbeagle2.dts | 2 +-
 arch/arm64/boot/dts/ti/k3-am625-beagleplay.dts   | 6 +++---
 arch/arm64/boot/dts/ti/k3-am625-sk-common.dtsi   | 2 +-
 arch/arm64/boot/dts/ti/k3-am62x-sk-common.dtsi   | 2 +-
 5 files changed, 7 insertions(+), 7 deletions(-)
arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts +1 -1
index dfd65c00f5b95..7fd508f4eddf1 100644
@@ -86,7 +86,7 @@ main_mmc0_pins_default: main-mmc0-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
 			AM62X_IOPAD(0x220, PIN_INPUT, 0) /* (V3) MMC0_CMD */
-			AM62X_IOPAD(0x218, PIN_INPUT, 0) /* (Y1) MMC0_CLK */
+			AM62X_IOPAD(0x218, PIN_OUTPUT, 0) /* (Y1) MMC0_CLK */
 			AM62X_IOPAD(0x214, PIN_INPUT, 0) /* (V2) MMC0_DAT0 */
 			AM62X_IOPAD(0x210, PIN_INPUT_PULLUP, 0) /* (V1) MMC0_DAT1 */
 			AM62X_IOPAD(0x20c, PIN_INPUT_PULLUP, 0) /* (W2) MMC0_DAT2 */
arch/arm64/boot/dts/ti/k3-am62-pocketbeagle2.dts +1 -1
index 7a4cffc27bda3..1c691629d4ad0 100644
@@ -235,7 +235,7 @@ AM62X_IOPAD(0x0020, PIN_OUTPUT, 3) /* (J25) OSPI0_D5.UART6_TXD */
 	main_mmc1_pins_default: main-mmc1-default-pins {
 		pinctrl-single,pins = <
 			AM62X_IOPAD(0x23c, PIN_INPUT, 0) /* (A21/C18) MMC1_CMD */
-			AM62X_IOPAD(0x234, PIN_INPUT, 0) /* (B22/A20) MMC1_CLK */
+			AM62X_IOPAD(0x234, PIN_OUTPUT, 0) /* (B22/A20) MMC1_CLK */
 			AM62X_IOPAD(0x230, PIN_INPUT, 0) /* (A22/A19) MMC1_DAT0 */
 			AM62X_IOPAD(0x22c, PIN_INPUT, 0) /* (B21/B19) MMC1_DAT1 */
 			AM62X_IOPAD(0x228, PIN_INPUT, 0) /* (C21/B20) MMC1_DAT2 */
arch/arm64/boot/dts/ti/k3-am625-beagleplay.dts +3 -3
index c468b9c5fc09a..898e4ca3fd87e 100644
@@ -323,7 +323,7 @@ emmc_pins_default: emmc-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
 			AM62X_IOPAD(0x0220, PIN_INPUT, 0) /* (Y3) MMC0_CMD */
-			AM62X_IOPAD(0x0218, PIN_INPUT, 0) /* (AB1) MMC0_CLK */
+			AM62X_IOPAD(0x0218, PIN_OUTPUT, 0) /* (AB1) MMC0_CLK */
 			AM62X_IOPAD(0x0214, PIN_INPUT, 0) /* (AA2) MMC0_DAT0 */
 			AM62X_IOPAD(0x0210, PIN_INPUT, 0) /* (AA1) MMC0_DAT1 */
 			AM62X_IOPAD(0x020c, PIN_INPUT, 0) /* (AA3) MMC0_DAT2 */
@@ -346,7 +346,7 @@ sd_pins_default: sd-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
 			AM62X_IOPAD(0x023c, PIN_INPUT, 0) /* (A21) MMC1_CMD */
-			AM62X_IOPAD(0x0234, PIN_INPUT, 0) /* (B22) MMC1_CLK */
+			AM62X_IOPAD(0x0234, PIN_OUTPUT, 0) /* (B22) MMC1_CLK */
 			AM62X_IOPAD(0x0230, PIN_INPUT, 0) /* (A22) MMC1_DAT0 */
 			AM62X_IOPAD(0x022c, PIN_INPUT, 0) /* (B21) MMC1_DAT1 */
 			AM62X_IOPAD(0x0228, PIN_INPUT, 0) /* (C21) MMC1_DAT2 */
@@ -358,7 +358,7 @@ AM62X_IOPAD(0x0240, PIN_INPUT, 7) /* (D17) MMC1_SDCD.GPIO1_48 */
 	wifi_pins_default: wifi-default-pins {
 		pinctrl-single,pins = <
 			AM62X_IOPAD(0x0120, PIN_INPUT, 0) /* (C24) MMC2_CMD */
-			AM62X_IOPAD(0x0118, PIN_INPUT, 0) /* (D25) MMC2_CLK */
+			AM62X_IOPAD(0x0118, PIN_OUTPUT, 0) /* (D25) MMC2_CLK */
 			AM62X_IOPAD(0x0114, PIN_INPUT, 0) /* (B24) MMC2_DAT0 */
 			AM62X_IOPAD(0x0110, PIN_INPUT, 0) /* (C25) MMC2_DAT1 */
 			AM62X_IOPAD(0x010c, PIN_INPUT, 0) /* (E23) MMC2_DAT2 */
arch/arm64/boot/dts/ti/k3-am625-sk-common.dtsi +1 -1
index 9c83626826453..085b067d96fbb 100644
@@ -98,7 +98,7 @@ main_mmc0_pins_default: main-mmc0-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
 			AM62X_IOPAD(0x220, PIN_INPUT, 0) /* (Y3) MMC0_CMD */
-			AM62X_IOPAD(0x218, PIN_INPUT, 0) /* (AB1) MMC0_CLK */
+			AM62X_IOPAD(0x218, PIN_OUTPUT, 0) /* (AB1) MMC0_CLK */
 			AM62X_IOPAD(0x214, PIN_INPUT, 0) /* (AA2) MMC0_DAT0 */
 			AM62X_IOPAD(0x210, PIN_INPUT_PULLUP, 0) /* (AA1) MMC0_DAT1 */
 			AM62X_IOPAD(0x20c, PIN_INPUT_PULLUP, 0) /* (AA3) MMC0_DAT2 */
arch/arm64/boot/dts/ti/k3-am62x-sk-common.dtsi +1 -1
index ab9e58c2d225e..d2f41e2ed6239 100644
@@ -194,7 +194,7 @@ main_mmc1_pins_default: main-mmc1-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
 			AM62X_IOPAD(0x23c, PIN_INPUT, 0) /* (A21/C18) MMC1_CMD */
-			AM62X_IOPAD(0x234, PIN_INPUT, 0) /* (B22/A20) MMC1_CLK */
+			AM62X_IOPAD(0x234, PIN_OUTPUT, 0) /* (B22/A20) MMC1_CLK */
 			AM62X_IOPAD(0x230, PIN_INPUT, 0) /* (A22/A19) MMC1_DAT0 */
 			AM62X_IOPAD(0x22c, PIN_INPUT, 0) /* (B21/B19) MMC1_DAT1 */
 			AM62X_IOPAD(0x228, PIN_INPUT, 0) /* (C21/B20) MMC1_DAT2 */
Signature · 1 line
2.54.0
#4 lore raw

Re: [PATCH 3/8] arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT

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

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

Pre-existing issues:
- [High] This problem wasn't introduced by this patch, but the commit fails to apply the necessary AM62x SoC hardware constraint to several third-party device trees. While it correctly updates the TI SK and Beagle boards' MMC clock pins to PIN_OUTPUT to disable the receiver per the TRM (SPRUIV7), boards like the Toradex Verdin, PHYTEC phyCORE, TQ, and Variscite AM62x SOMs still incorrectly define MMC0/1/2 clock pins as PIN_INPUT or PIN_INPUT_PULLDOWN. As this is an underlying SoC requirement, all boards integrating the AM62x SoC must implement the correct pad configuration. --

--- Patch [3]: [PATCH 3/8] arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT --- commit 7601627985b20fa80186e314fd493c2b9a330fe1
Author: Judith Mendez <[email protected]>

arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT

This commit updates the MMC clock pin configuration on various AM62x boards to use PIN_OUTPUT. This matches the TRM requirement to disable the receiver for these pins.

arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts main_mmc0_pins_default: main-mmc0-default-pins { · line 89

Quoted 10 lines
arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts +1 -1
index dfd65c00f5b95..7fd508f4eddf1 100644
@@ -86,7 +86,7 @@ main_mmc0_pins_default: main-mmc0-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
 			AM62X_IOPAD(0x220, PIN_INPUT, 0) /* (V3) MMC0_CMD */
-			AM62X_IOPAD(0x218, PIN_INPUT, 0) /* (Y1) MMC0_CLK */
+			AM62X_IOPAD(0x218, PIN_OUTPUT, 0) /* (Y1) MMC0_CLK */

[Severity: High]
This isn't a bug introduced by this patch, but while this updates the TI SK and Beagle boards to use PIN_OUTPUT for MMC clock pins, it looks like several third-party AM62x boards still define these pins as PIN_INPUT or PIN_INPUT_PULLDOWN.

For example:

arch/arm64/boot/dts/ti/k3-am62-verdin.dtsi:pinctrl_sdhci0()

    AM62X_IOPAD(0x218, PIN_INPUT,        0) /* (AB1) MMC0_CLK  */

arch/arm64/boot/dts/ti/k3-am62-phycore-som.dtsi:main_mmc0_pins_default()

    AM62X_IOPAD(0x218, PIN_INPUT_PULLDOWN, 0) /* (AB1) MMC0_CLK */

arch/arm64/boot/dts/ti/k3-am62x-phyboard-lyra.dtsi:main_mmc1_pins_default()

    AM62X_IOPAD(0x234, PIN_INPUT_PULLDOWN, 0) /* (B22) MMC1_CLK */

arch/arm64/boot/dts/ti/k3-am62a-phycore-som.dtsi:main_mmc0_pins_default()

    AM62AX_IOPAD(0x218, PIN_INPUT_PULLDOWN, 0) /* (AB7) MMC0_CLK */

arch/arm64/boot/dts/ti/k3-am625-tqma62xx.dtsi:main_mmc0_pins()

    AM62X_IOPAD(0x218, PIN_INPUT, 0)

Since this is an underlying AM62x SoC hardware requirement from the TRM, should these other device trees also be updated to prevent potential signal integrity issues and intermittent SD/eMMC I/O failures?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=3
#5 lore raw

Re: [PATCH 3/8] arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT

Hi all,

Responding to sashiko review.

Quoted 45 lines · On 7/29/2026 3:08 PM, [email protected] wrote:

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

Pre-existing issues:
- [High] This problem wasn't introduced by this patch, but the commit fails to apply the necessary AM62x SoC hardware constraint to several third-party device trees. While it correctly updates the TI SK and Beagle boards' MMC clock pins to PIN_OUTPUT to disable the receiver per the TRM (SPRUIV7), boards like the Toradex Verdin, PHYTEC phyCORE, TQ, and Variscite AM62x SOMs still incorrectly define MMC0/1/2 clock pins as PIN_INPUT or PIN_INPUT_PULLDOWN. As this is an underlying SoC requirement, all boards integrating the AM62x SoC must implement the correct pad configuration. --

--- Patch [3]: [PATCH 3/8] arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT --- commit 7601627985b20fa80186e314fd493c2b9a330fe1
Author: Judith Mendez <[email protected]>

arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT

This commit updates the MMC clock pin configuration on various AM62x boards to use PIN_OUTPUT. This matches the TRM requirement to disable the receiver for these pins.

arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts main_mmc0_pins_default: main-mmc0-default-pins { · line 89

Quoted 10 lines
arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts +1 -1
index dfd65c00f5b95..7fd508f4eddf1 100644
@@ -86,7 +86,7 @@ main_mmc0_pins_default: main-mmc0-default-pins {
  		bootph-all;
  		pinctrl-single,pins = <
  			AM62X_IOPAD(0x220, PIN_INPUT, 0) /* (V3) MMC0_CMD */
-			AM62X_IOPAD(0x218, PIN_INPUT, 0) /* (Y1) MMC0_CLK */
+			AM62X_IOPAD(0x218, PIN_OUTPUT, 0) /* (Y1) MMC0_CLK */

[Severity: High]
This isn't a bug introduced by this patch, but while this updates the TI SK and Beagle boards to use PIN_OUTPUT for MMC clock pins, it looks like several third-party AM62x boards still define these pins as PIN_INPUT or PIN_INPUT_PULLDOWN.

For example:

arch/arm64/boot/dts/ti/k3-am62-verdin.dtsi:pinctrl_sdhci0()

     AM62X_IOPAD(0x218, PIN_INPUT,        0) /* (AB1) MMC0_CLK  */

arch/arm64/boot/dts/ti/k3-am62-phycore-som.dtsi:main_mmc0_pins_default()

     AM62X_IOPAD(0x218, PIN_INPUT_PULLDOWN, 0) /* (AB1) MMC0_CLK */

arch/arm64/boot/dts/ti/k3-am62x-phyboard-lyra.dtsi:main_mmc1_pins_default()

     AM62X_IOPAD(0x234, PIN_INPUT_PULLDOWN, 0) /* (B22) MMC1_CLK */

arch/arm64/boot/dts/ti/k3-am62a-phycore-som.dtsi:main_mmc0_pins_default()

Quoted 8 lines
     AM62AX_IOPAD(0x218, PIN_INPUT_PULLDOWN, 0) /* (AB7) MMC0_CLK */

arch/arm64/boot/dts/ti/k3-am625-tqma62xx.dtsi:main_mmc0_pins()

     AM62X_IOPAD(0x218, PIN_INPUT, 0)

Since this is an underlying AM62x SoC hardware requirement from the TRM, should these other device trees also be updated to prevent potential signal integrity issues and intermittent SD/eMMC I/O failures?

As far as I know, this issue is more of a documentation fix and does not prevent potential signal integrity issues.

~ Judith

Quoted 1 line
#6 lore raw

PATCH 4/8 arm64: dts: ti: k3-am62l-evm: Remove unnecessary internal pullups for MMC0

In reply to Judith Mendez

AM62L EVM has external pullups on MMC0 CMD and D0 pins [0]. Remove internal pullups considering:
- Internal pullups are not required, without internal pullups, RDAT and RCMD values of 47K are within spec
- having both pullups (stronger pull) results in unnecessary power consumption

[0] https://www.ti.com/lit/zip/SPRCAL6

---
 arch/arm64/boot/dts/ti/k3-am62l3-evm.dts | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/ti/k3-am62l3-evm.dts +2 -2
index 84a0cfaaf196a..964387eb51788 100644
@@ -253,9 +253,9 @@ AM62LX_IOPAD(0x01e0, PIN_INPUT_PULLUP, 0) /* (D8) I2C2_SDA */
 
 	mmc0_pins_default: mmc0-default-pins {
 		pinctrl-single,pins = <
-			AM62LX_IOPAD(0x0214, PIN_INPUT_PULLUP, 0) /* (D2) MMC0_CMD */
+			AM62LX_IOPAD(0x0214, PIN_INPUT, 0) /* (D2) MMC0_CMD */
 			AM62LX_IOPAD(0x020c, PIN_OUTPUT, 0) /* (B2) MMC0_CLK */
-			AM62LX_IOPAD(0x0208, PIN_INPUT_PULLUP, 0) /* (D3) MMC0_DAT0 */
+			AM62LX_IOPAD(0x0208, PIN_INPUT, 0) /* (D3) MMC0_DAT0 */
 			AM62LX_IOPAD(0x0204, PIN_INPUT_PULLUP, 0) /* (D4) MMC0_DAT1 */
 			AM62LX_IOPAD(0x0200, PIN_INPUT_PULLUP, 0) /* (C1) MMC0_DAT2 */
 			AM62LX_IOPAD(0x01fc, PIN_INPUT_PULLUP, 0) /* (C2) MMC0_DAT3 */
Signature · 1 line
2.54.0
#7 lore raw

PATCH 5/8 arm64: dts: ti: k3-am64: Enable IO expander interrupt

In reply to Judith Mendez

On AM64 EVM and SK boards, MMC1_SDWP is used for IO expander interrupt, remove from MMC1 pinmux node and setup as exp1 IO expander interrupt.

---
 arch/arm64/boot/dts/ti/k3-am642-evm.dts | 14 +++++++++++++-
 arch/arm64/boot/dts/ti/k3-am642-sk.dts  | 14 +++++++++++++-
 2 files changed, 26 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/ti/k3-am642-evm.dts +13 -1
index bc4347cf71140..bb756c922252a 100644
@@ -259,11 +259,17 @@ AM64X_IOPAD(0x0284, PIN_INPUT_PULLUP, 0) /* (L21) MMC1_DAT1 */
 			AM64X_IOPAD(0x0280, PIN_INPUT_PULLUP, 0) /* (K19) MMC1_DAT2 */
 			AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */
 			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
-			AM64X_IOPAD(0x029c, PIN_INPUT, 0) /* (C20) MMC1_SDWP */
 			AM64X_IOPAD(0x0290, PIN_INPUT, 0) /* MMC1_CLKLB */
 		>;
 	};
 
+	main_gpio1_ioexp_intr_pins_default: main-gpio1-ioexp-intr-default-pins {
+		pinctrl-single,pins = <
+			AM64X_IOPAD(0x029c, PIN_INPUT, 7) /* (C20) MMC1_SDWP.GPIO1_78 */
+		>;
+		bootph-all;
+	};
+
 	main_uart1_pins_default: main-uart1-default-pins {
 		pinctrl-single,pins = <
 			AM64X_IOPAD(0x0248, PIN_INPUT, 0)		/* (D16) UART1_CTSn */
@@ -481,6 +487,12 @@ exp1: gpio@22 {
 		bootph-all;
 		compatible = "ti,tca6424";
 		reg = <0x22>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&main_gpio1_ioexp_intr_pins_default>;
+		interrupt-parent = <&main_gpio1>;
+		interrupts = <78 IRQ_TYPE_LEVEL_HIGH>;
+		interrupt-controller;
+		#interrupt-cells = <2>;
 		gpio-controller;
 		#gpio-cells = <2>;
 		gpio-line-names = "GPIO_eMMC_RSTn", "CAN_MUX_SEL",
arch/arm64/boot/dts/ti/k3-am642-sk.dts +13 -1
index d28a38c87f32a..3d7160400cef5 100644
@@ -198,7 +198,6 @@ &main_pmx0 {
 	main_mmc1_pins_default: main-mmc1-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
-			AM64X_IOPAD(0x029c, PIN_INPUT_PULLUP, 0) /* (C20) MMC1_SDWP */
 			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
 			AM64X_IOPAD(0x0294, PIN_INPUT_PULLUP, 0) /* (J19) MMC1_CMD */
 			AM64X_IOPAD(0x0290, PIN_INPUT, 0) /* (#N/A) MMC1_CLKLB */
@@ -210,6 +209,13 @@ AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */
 		>;
 	};
 
+	main_gpio1_ioexp_intr_pins_default: main-gpio1-ioexp-intr-default-pins {
+		pinctrl-single,pins = <
+			AM64X_IOPAD(0x029c, PIN_INPUT, 7) /* (C20) MMC1_SDWP.GPIO1_78 */
+		>;
+		bootph-all;
+	};
+
 	main_uart0_pins_default: main-uart0-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
@@ -383,6 +389,12 @@ exp1: gpio@70 {
 		bootph-all;
 		compatible = "nxp,pca9538";
 		reg = <0x70>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&main_gpio1_ioexp_intr_pins_default>;
+		interrupt-parent = <&main_gpio1>;
+		interrupts = <78 IRQ_TYPE_LEVEL_HIGH>;
+		interrupt-controller;
+		#interrupt-cells = <2>;
 		gpio-controller;
 		#gpio-cells = <2>;
 		gpio-line-names = "GPIO_CPSW2_RST", "GPIO_CPSW1_RST",
Signature · 1 line
2.54.0
#8 lore raw

Re: [PATCH 5/8] arm64: dts: ti: k3-am64: Enable IO expander interrupt

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider: - [High] The IO expander interrupt is incorrectly configured with IRQ_TYPE_LEVEL_HIGH polarity for an…

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] The IO expander interrupt is incorrectly configured with IRQ_TYPE_LEVEL_HIGH polarity for an active-low open-drain output.
- [Medium] The pad configuration for the SK board removes the internal pull-up, potentially leaving the open-drain interrupt line floating. --

--- Patch [5]: [PATCH 5/8] arm64: dts: ti: k3-am64: Enable IO expander interrupt --- commit 846c9ea4531cda7562459d539ef568783fb1e41c
Author: Judith Mendez <[email protected]>

arm64: dts: ti: k3-am64: Enable IO expander interrupt

On AM64 EVM and SK boards, MMC1_SDWP is used for IO expander interrupt, remove from MMC1 pinmux node and setup as exp1 IO expander interrupt.

Quoted 3 lines
arch/arm64/boot/dts/ti/k3-am642-evm.dts +0 -0

[ ... ]

arch/arm64/boot/dts/ti/k3-am642-evm.dts exp1: gpio@22 { · line 493

Quoted 8 lines
arch/arm64/boot/dts/ti/k3-am642-evm.dts +4 -0
@@ -481,6 +487,12 @@ exp1: gpio@22 {
 		bootph-all;
 		compatible = "ti,tca6424";
 		reg = <0x22>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&main_gpio1_ioexp_intr_pins_default>;
+		interrupt-parent = <&main_gpio1>;
+		interrupts = <78 IRQ_TYPE_LEVEL_HIGH>;

[Severity: High]
Will setting this interrupt to IRQ_TYPE_LEVEL_HIGH cause an interrupt storm?

Since the TCA6424 IO expander has an active-low open-drain interrupt output, an active-high configuration might trigger continuously when the line is idle, eventually causing the kernel to disable the interrupt.

Quoted 14 lines
arch/arm64/boot/dts/ti/k3-am642-evm.dts +2 -0
+		interrupt-controller;
+		#interrupt-cells = <2>;
 		gpio-controller;
 
arch/arm64/boot/dts/ti/k3-am642-sk.dts +0 -1
@@ -198,7 +198,6 @@ &main_pmx0 {
 	main_mmc1_pins_default: main-mmc1-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
-			AM64X_IOPAD(0x029c, PIN_INPUT_PULLUP, 0) /* (C20) MMC1_SDWP */
 			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
 			AM64X_IOPAD(0x0294, PIN_INPUT_PULLUP, 0) /* (J19) MMC1_CMD */

[ ... ]

arch/arm64/boot/dts/ti/k3-am642-sk.dts AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */ · line 214

Quoted 7 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +3 -0
@@ -210,6 +209,13 @@ AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */
 		>;
 	};
 
+	main_gpio1_ioexp_intr_pins_default: main-gpio1-ioexp-intr-default-pins {
+		pinctrl-single,pins = <
+			AM64X_IOPAD(0x029c, PIN_INPUT, 7) /* (C20) MMC1_SDWP.GPIO1_78 */

[Severity: Medium]
Does dropping the internal pull-up on this pad risk leaving the interrupt line floating?

The previous configuration for this pad on the SK board used PIN_INPUT_PULLUP. If the board lacks a dedicated external pull-up resistor for this open-drain interrupt, using PIN_INPUT might lead to spurious edge detections.

Quoted 3 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +3 -0
+		>;
+		bootph-all;
+	};

[ ... ]

arch/arm64/boot/dts/ti/k3-am642-sk.dts exp1: gpio@70 { · line 395

Quoted 8 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +4 -0
@@ -383,6 +389,12 @@ exp1: gpio@70 {
 		bootph-all;
 		compatible = "nxp,pca9538";
 		reg = <0x70>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&main_gpio1_ioexp_intr_pins_default>;
+		interrupt-parent = <&main_gpio1>;
+		interrupts = <78 IRQ_TYPE_LEVEL_HIGH>;

[Severity: High]
Could this also trigger an interrupt storm on the SK board?

Similar to the EVM board, the PCA9538 IO expander typically has an active-low open-drain interrupt. An active-high configuration will likely trigger unconditionally on boot.

Quoted 2 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +2 -0
+		interrupt-controller;
+		#interrupt-cells = <2>;
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=5
#9 lore raw

Re: [PATCH 5/8] arm64: dts: ti: k3-am64: Enable IO expander interrupt

Hi all,

Quoted 4 lines · On 7/29/2026 3:09 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [High] The IO expander interrupt is incorrectly configured with IRQ_TYPE_LEVEL_HIGH polarity for an active-low open-drain output.
- [Medium] The pad configuration for the SK board removes the internal pull-up, potentially leaving the open-drain interrupt line floating. --

This patch will be sent as a separate series, due to Vignesh's review, will test on am64x devices and fix before repining then.

~ Judith

Quoted 85 lines

--- Patch [5]: [PATCH 5/8] arm64: dts: ti: k3-am64: Enable IO expander interrupt --- commit 846c9ea4531cda7562459d539ef568783fb1e41c
Author: Judith Mendez <[email protected]>

arm64: dts: ti: k3-am64: Enable IO expander interrupt

On AM64 EVM and SK boards, MMC1_SDWP is used for IO expander interrupt, remove from MMC1 pinmux node and setup as exp1 IO expander interrupt.

Quoted 3 lines
arch/arm64/boot/dts/ti/k3-am642-evm.dts +0 -0

[ ... ]

arch/arm64/boot/dts/ti/k3-am642-evm.dts exp1: gpio@22 { · line 493

Quoted 8 lines
arch/arm64/boot/dts/ti/k3-am642-evm.dts +4 -0
@@ -481,6 +487,12 @@ exp1: gpio@22 {
  		bootph-all;
  		compatible = "ti,tca6424";
  		reg = <0x22>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&main_gpio1_ioexp_intr_pins_default>;
+		interrupt-parent = <&main_gpio1>;
+		interrupts = <78 IRQ_TYPE_LEVEL_HIGH>;

[Severity: High]
Will setting this interrupt to IRQ_TYPE_LEVEL_HIGH cause an interrupt storm?

Since the TCA6424 IO expander has an active-low open-drain interrupt output, an active-high configuration might trigger continuously when the line is idle, eventually causing the kernel to disable the interrupt.

Quoted 14 lines
arch/arm64/boot/dts/ti/k3-am642-evm.dts +2 -0
+		interrupt-controller;
+		#interrupt-cells = <2>;
  		gpio-controller;
 
arch/arm64/boot/dts/ti/k3-am642-sk.dts +0 -1
@@ -198,7 +198,6 @@ &main_pmx0 {
  	main_mmc1_pins_default: main-mmc1-default-pins {
  		bootph-all;
  		pinctrl-single,pins = <
-			AM64X_IOPAD(0x029c, PIN_INPUT_PULLUP, 0) /* (C20) MMC1_SDWP */
  			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
  			AM64X_IOPAD(0x0294, PIN_INPUT_PULLUP, 0) /* (J19) MMC1_CMD */

[ ... ]

arch/arm64/boot/dts/ti/k3-am642-sk.dts AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */ · line 214

Quoted 7 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +3 -0
@@ -210,6 +209,13 @@ AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */
  		>;
  	};
 
+	main_gpio1_ioexp_intr_pins_default: main-gpio1-ioexp-intr-default-pins {
+		pinctrl-single,pins = <
+			AM64X_IOPAD(0x029c, PIN_INPUT, 7) /* (C20) MMC1_SDWP.GPIO1_78 */

[Severity: Medium]
Does dropping the internal pull-up on this pad risk leaving the interrupt line floating?

The previous configuration for this pad on the SK board used PIN_INPUT_PULLUP. If the board lacks a dedicated external pull-up resistor for this open-drain interrupt, using PIN_INPUT might lead to spurious edge detections.

Quoted 3 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +3 -0
+		>;
+		bootph-all;
+	};

[ ... ]

arch/arm64/boot/dts/ti/k3-am642-sk.dts exp1: gpio@70 { · line 395

Quoted 8 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +4 -0
@@ -383,6 +389,12 @@ exp1: gpio@70 {
  		bootph-all;
  		compatible = "nxp,pca9538";
  		reg = <0x70>;
+		pinctrl-names = "default";
+		pinctrl-0 = <&main_gpio1_ioexp_intr_pins_default>;
+		interrupt-parent = <&main_gpio1>;
+		interrupts = <78 IRQ_TYPE_LEVEL_HIGH>;

[Severity: High]
Could this also trigger an interrupt storm on the SK board?

Similar to the EVM board, the PCA9538 IO expander typically has an active-low open-drain interrupt. An active-high configuration will likely trigger unconditionally on boot.

Quoted 2 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +2 -0
+		interrupt-controller;
+		#interrupt-cells = <2>;
#10 lore raw

PATCH 6/8 arm64: dts: ti: k3-am64: Let MMC1 clock loopback use hardware default

In reply to Judith Mendez

Clock loopback (CLKLB) should not be changed from its default state. Explicit pinmux config is unnecessary and overrides correct default. Drop CLKLB in pinmux node for MMC1.

---
 arch/arm64/boot/dts/ti/k3-am642-evm.dts | 1 -
 arch/arm64/boot/dts/ti/k3-am642-sk.dts  | 1 -
 2 files changed, 2 deletions(-)
arch/arm64/boot/dts/ti/k3-am642-evm.dts +0 -1
index bb756c922252a..f57f39f8579d5 100644
@@ -259,7 +259,6 @@ AM64X_IOPAD(0x0284, PIN_INPUT_PULLUP, 0) /* (L21) MMC1_DAT1 */
 			AM64X_IOPAD(0x0280, PIN_INPUT_PULLUP, 0) /* (K19) MMC1_DAT2 */
 			AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */
 			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
-			AM64X_IOPAD(0x0290, PIN_INPUT, 0) /* MMC1_CLKLB */
 		>;
 	};
 
arch/arm64/boot/dts/ti/k3-am642-sk.dts +0 -1
index 3d7160400cef5..7f2483968f4ee 100644
@@ -200,7 +200,6 @@ main_mmc1_pins_default: main-mmc1-default-pins {
 		pinctrl-single,pins = <
 			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
 			AM64X_IOPAD(0x0294, PIN_INPUT_PULLUP, 0) /* (J19) MMC1_CMD */
-			AM64X_IOPAD(0x0290, PIN_INPUT, 0) /* (#N/A) MMC1_CLKLB */
 			AM64X_IOPAD(0x028c, PIN_INPUT_PULLDOWN, 0) /* (L20) MMC1_CLK */
 			AM64X_IOPAD(0x0288, PIN_INPUT_PULLUP, 0) /* (K21) MMC1_DAT0 */
 			AM64X_IOPAD(0x0284, PIN_INPUT_PULLUP, 0) /* (L21) MMC1_DAT1 */
Signature · 1 line
2.54.0
#11 lore raw

PATCH 7/8 arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls

In reply to Judith Mendez

AM64 SK has external pullups on all MMC pins [0] and thus internal pullups are not required, drop them since:

Enabled, internal pullups on data pins further reduces the total pull resistance, violating the minimum RDAT and RCMD recommended spec values of 10K ohm.

Drop CD pullup since board has 1.1K external pullup resistor and internal pullup increases power consumption.

[0] https://www.ti.com/lit/zip/SPRR460

Fixes 744545ffec14f ("arm64: dts: ti: k3-am642-sk: Fix mmc1 pinmux")
---
 arch/arm64/boot/dts/ti/k3-am642-sk.dts | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
arch/arm64/boot/dts/ti/k3-am642-sk.dts +7 -7
index 7f2483968f4ee..fde4f80524301 100644
@@ -198,13 +198,13 @@ &main_pmx0 {
 	main_mmc1_pins_default: main-mmc1-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
-			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
-			AM64X_IOPAD(0x0294, PIN_INPUT_PULLUP, 0) /* (J19) MMC1_CMD */
-			AM64X_IOPAD(0x028c, PIN_INPUT_PULLDOWN, 0) /* (L20) MMC1_CLK */
-			AM64X_IOPAD(0x0288, PIN_INPUT_PULLUP, 0) /* (K21) MMC1_DAT0 */
-			AM64X_IOPAD(0x0284, PIN_INPUT_PULLUP, 0) /* (L21) MMC1_DAT1 */
-			AM64X_IOPAD(0x0280, PIN_INPUT_PULLUP, 0) /* (K19) MMC1_DAT2 */
-			AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */
+			AM64X_IOPAD(0x0298, PIN_INPUT, 0) /* (D19) MMC1_SDCD */
+			AM64X_IOPAD(0x0294, PIN_INPUT, 0) /* (J19) MMC1_CMD */
+			AM64X_IOPAD(0x028c, PIN_INPUT, 0) /* (L20) MMC1_CLK */
+			AM64X_IOPAD(0x0288, PIN_INPUT, 0) /* (K21) MMC1_DAT0 */
+			AM64X_IOPAD(0x0284, PIN_INPUT, 0) /* (L21) MMC1_DAT1 */
+			AM64X_IOPAD(0x0280, PIN_INPUT, 0) /* (K19) MMC1_DAT2 */
+			AM64X_IOPAD(0x027c, PIN_INPUT, 0) /* (K18) MMC1_DAT3 */
 		>;
 	};
 
Signature · 1 line
2.54.0
#12 lore raw

Re: [PATCH 7/8] arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] Unintended removal of pulldown and incorrect receiver state for `MMC1_CLK`. --

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Unintended removal of pulldown and incorrect receiver state for MMC1_CLK. --

--- Patch [7]: [PATCH 7/8] arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls --- commit 3a83d1bd44a1ef684b5bd29a93a2127e82509201
Author: Judith Mendez <[email protected]>

arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls

This commit drops internal pullups on MMC data and CMD pins to prevent violating minimum pull resistance specifications, as external pullups are already present on the board. It also removes the CD pullup to reduce power consumption.

arch/arm64/boot/dts/ti/k3-am642-sk.dts &main_pmx0 { · line 203

Quoted 13 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +3 -3
@@ -198,13 +198,13 @@ &main_pmx0 {
 	main_mmc1_pins_default: main-mmc1-default-pins {
 		bootph-all;
 		pinctrl-single,pins = <
-			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
-			AM64X_IOPAD(0x0294, PIN_INPUT_PULLUP, 0) /* (J19) MMC1_CMD */
-			AM64X_IOPAD(0x028c, PIN_INPUT_PULLDOWN, 0) /* (L20) MMC1_CLK */
+			AM64X_IOPAD(0x0298, PIN_INPUT, 0) /* (D19) MMC1_SDCD */
+			AM64X_IOPAD(0x0294, PIN_INPUT, 0) /* (J19) MMC1_CMD */
+			AM64X_IOPAD(0x028c, PIN_INPUT, 0) /* (L20) MMC1_CLK */

[Severity: Medium]
The commit message mentions removing internal pullups because of external pullups on the board, but MMC1_CLK in main_mmc1_pins_default was configured with PIN_INPUT_PULLDOWN. Does this blanket change unintentionally leave the clock pin floating when the MMC controller tri-states the clock line (for example, during idle or suspend)?

Additionally, keeping this as PIN_INPUT leaves the receiver enabled. If the clock line floats without a pulldown, could the floating voltage cause the input receiver to rapidly toggle and increase power consumption? Should this pin be configured as OUTPUT-only instead?

[ ... ]

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=7
#13 lore raw

Re: [PATCH 7/8] arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls

Hi all,

Addressing Sashiko review.

Quoted 26 lines · On 7/29/2026 3:13 PM, [email protected] wrote:

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Unintended removal of pulldown and incorrect receiver state for MMC1_CLK. --

--- Patch [7]: [PATCH 7/8] arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls --- commit 3a83d1bd44a1ef684b5bd29a93a2127e82509201
Author: Judith Mendez <[email protected]>

arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls

This commit drops internal pullups on MMC data and CMD pins to prevent violating minimum pull resistance specifications, as external pullups are already present on the board. It also removes the CD pullup to reduce power consumption.

Quoted 12 lines
arch/arm64/boot/dts/ti/k3-am642-sk.dts +2 -3
@@ -198,13 +198,13 @@ &main_pmx0 {
  	main_mmc1_pins_default: main-mmc1-default-pins {
  		bootph-all;
  		pinctrl-single,pins = <
-			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
-			AM64X_IOPAD(0x0294, PIN_INPUT_PULLUP, 0) /* (J19) MMC1_CMD */
-			AM64X_IOPAD(0x028c, PIN_INPUT_PULLDOWN, 0) /* (L20) MMC1_CLK */
+			AM64X_IOPAD(0x0298, PIN_INPUT, 0) /* (D19) MMC1_SDCD */
+			AM64X_IOPAD(0x0294, PIN_INPUT, 0) /* (J19) MMC1_CMD */
Quoted 8 lines

arch/arm64/boot/dts/ti/k3-am642-sk.dts &main_pmx0 { · line 203

Quoted 1 line
arch/arm64/boot/dts/ti/k3-am642-sk.dts +1 -0
+			AM64X_IOPAD(0x028c, PIN_INPUT, 0) /* (L20) MMC1_CLK */

[Severity: Medium]
The commit message mentions removing internal pullups because of external pullups on the board, but MMC1_CLK in main_mmc1_pins_default was configured with PIN_INPUT_PULLDOWN. Does this blanket change unintentionally leave the clock pin floating when the MMC controller tri-states the clock line (for example, during idle or suspend)?

MMC1_CLK has a 49.9K external pulldown, so there should be no issue dropping the internal pulldown.

Quoted 5 lines

Additionally, keeping this as PIN_INPUT leaves the receiver enabled. If the clock line floats without a pulldown, could the floating voltage cause the input receiver to rapidly toggle and increase power consumption? Should this pin be configured as OUTPUT-only instead?

I think I missed one patch for setting MMC1_CLK pin as output for am64x, will check and add patch in respin.

~ Judith

#14 lore raw

PATCH 8/8 arm64: dts: ti: k3-am642-evm: Remove unnecessary internal pullups for MMC1

In reply to Judith Mendez

AM64 EVM has external pullups on MMC1 CLK and MMC1 CD pins [0], remove internal pullups since both enabled results in unnecessary power consumption.

[0] https://www.ti.com/lit/zip/SPRR462

---
 arch/arm64/boot/dts/ti/k3-am642-evm.dts | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/ti/k3-am642-evm.dts +2 -2
index f57f39f8579d5..37e3dd862a43f 100644
@@ -253,12 +253,12 @@ &main_pmx0 {
 	main_mmc1_pins_default: main-mmc1-default-pins {
 		pinctrl-single,pins = <
 			AM64X_IOPAD(0x0294, PIN_INPUT_PULLUP, 0) /* (J19) MMC1_CMD */
-			AM64X_IOPAD(0x028c, PIN_INPUT_PULLDOWN, 0) /* (L20) MMC1_CLK */
+			AM64X_IOPAD(0x028c, PIN_INPUT, 0) /* (L20) MMC1_CLK */
 			AM64X_IOPAD(0x0288, PIN_INPUT_PULLUP, 0) /* (K21) MMC1_DAT0 */
 			AM64X_IOPAD(0x0284, PIN_INPUT_PULLUP, 0) /* (L21) MMC1_DAT1 */
 			AM64X_IOPAD(0x0280, PIN_INPUT_PULLUP, 0) /* (K19) MMC1_DAT2 */
 			AM64X_IOPAD(0x027c, PIN_INPUT_PULLUP, 0) /* (K18) MMC1_DAT3 */
-			AM64X_IOPAD(0x0298, PIN_INPUT_PULLUP, 0) /* (D19) MMC1_SDCD */
+			AM64X_IOPAD(0x0298, PIN_INPUT, 0) /* (D19) MMC1_SDCD */
 		>;
 	};
 
Signature · 1 line
2.54.0
#15 lore raw
In reply to Judith Mendez

Hi Judith

Quoted 6 lines · Judith Mendez

This series fixes MMC pin input/output and pullup/pulldown settings across AM62x, AM62AX, AM62LX and AM64x boards.

All patches have been tested except patch 5/8, so tested-bys are apreciated.

Quoted 12 lines

Judith Mendez (8):
arm64: dts: ti: k3-am62a: Fix MMC0_CLK register address
arm64: dts: ti: k3-am62a: Set MMC clock pins as OUTPUT
arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT
arm64: dts: ti: k3-am62l-evm: Remove unnecessary internal pullups for
MMC0
arm64: dts: ti: k3-am64: Enable IO expander interrupt
arm64: dts: ti: k3-am64: Let MMC1 clock loopback use hardware default
arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls
arm64: dts: ti: k3-am642-evm: Remove unnecessary internal pullups for
MMC1

Apart from 5/8, these patches seems like bug fixes. Please repost them with appropriate Fixes tag. Its better to separate out 5/8 as standalone patch.

Quoted 11 lines
 arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts      |  2 +-
 .../boot/dts/ti/k3-am62-pocketbeagle2.dts     |  2 +-
 .../arm64/boot/dts/ti/k3-am625-beagleplay.dts |  6 ++--
 .../arm64/boot/dts/ti/k3-am625-sk-common.dtsi |  2 +-
 arch/arm64/boot/dts/ti/k3-am62a7-sk.dts       |  5 ++--
 arch/arm64/boot/dts/ti/k3-am62l3-evm.dts      |  4 +--
 .../arm64/boot/dts/ti/k3-am62x-sk-common.dtsi |  2 +-
 arch/arm64/boot/dts/ti/k3-am642-evm.dts       | 19 +++++++++---
 arch/arm64/boot/dts/ti/k3-am642-sk.dts        | 29 +++++++++++++------
 9 files changed, 46 insertions(+), 25 deletions(-)
Signature · 3 lines
Regards
Vignesh
https://ti.com/opensource
#16 lore raw

Hi Vignesh,

Quoted 22 lines · Vignesh Raghavendra

Hi Judith

Quoted 6 lines · Judith Mendez

This series fixes MMC pin input/output and pullup/pulldown settings across AM62x, AM62AX, AM62LX and AM64x boards.

All patches have been tested except patch 5/8, so tested-bys are apreciated.

Quoted 12 lines

Judith Mendez (8):
arm64: dts: ti: k3-am62a: Fix MMC0_CLK register address
arm64: dts: ti: k3-am62a: Set MMC clock pins as OUTPUT
arm64: dts: ti: k3-am62: Set MMC clock pins as OUTPUT
arm64: dts: ti: k3-am62l-evm: Remove unnecessary internal pullups for
MMC0
arm64: dts: ti: k3-am64: Enable IO expander interrupt
arm64: dts: ti: k3-am64: Let MMC1 clock loopback use hardware default
arm64: dts: ti: k3-am64-sk: Disable MMC1 internal pulls
arm64: dts: ti: k3-am642-evm: Remove unnecessary internal pullups for
MMC1

Quoted 3 lines

Apart from 5/8, these patches seems like bug fixes. Please repost them with appropriate Fixes tag. Its better to separate out 5/8 as standalone patch.

Since they were not functional fixes and were more of documentation fixes, I did not include the fixes tag, but sure, will go ahead and re-spin with fixes tag.

~ Judith

Quoted 13 lines
Quoted 12 lines
  arch/arm64/boot/dts/ti/k3-am62-lp-sk.dts      |  2 +-
  .../boot/dts/ti/k3-am62-pocketbeagle2.dts     |  2 +-
  .../arm64/boot/dts/ti/k3-am625-beagleplay.dts |  6 ++--
  .../arm64/boot/dts/ti/k3-am625-sk-common.dtsi |  2 +-
  arch/arm64/boot/dts/ti/k3-am62a7-sk.dts       |  5 ++--
  arch/arm64/boot/dts/ti/k3-am62l3-evm.dts      |  4 +--
  .../arm64/boot/dts/ti/k3-am62x-sk-common.dtsi |  2 +-
  arch/arm64/boot/dts/ti/k3-am642-evm.dts       | 19 +++++++++---
  arch/arm64/boot/dts/ti/k3-am642-sk.dts        | 29 +++++++++++++------
  9 files changed, 46 insertions(+), 25 deletions(-)