RE: [PATCH v2] arm64: dts: rockchip: Fix SD card init on rk3399-nanopi4

Posted by Christian Loehle 3 years, 9 months ago

I only briefly skimmed the discussion, but does this mean that a soft-reset (CMD0) of a UHS (post-voltage-switch) will not work?
(As the card/spec requires a power-cycle by the host which will not come, right?)
Can you try this real quick? I can give you a mmc-utils snippet if you have trouble issuing one.
If that does indeed not work I think the general approach is to disable uhs in the dts or at least document that.
Regards,
Christian

-----Original Message-----
From: Christian Kohlschütter <christian@kohlschutter.com> 
Sent: Donnerstag, 14. Juli 2022 18:27
To: Robin Murphy <robin.murphy@arm.com>; Markus Reichl <m.reichl@fivetechno.de>; Heiko Stübner <heiko@sntech.de>; linux-arm-kernel@lists.infradead.org; linux-rockchip@lists.infradead.org; linux-kernel@vger.kernel.org; Linux MMC List <linux-mmc@vger.kernel.org>
Subject: [PATCH v2] arm64: dts: rockchip: Fix SD card init on rk3399-nanopi4

mmc/SD-card initialization may fail on NanoPi r4s with
"mmc1: problem reading SD Status register" /
"mmc1: error -110 whilst initialising SD card"

Moreover, rebooting would also sometimes hang.

This is caused by the gpio entry for the vcc3v0-sd regulator;
even though it appears to be the correct GPIO pin, the presence
of the binding causes these errors.

Fix the regulator to drop the gpio binding and add a comment
to prevent accidental reintroduction of that entry.

Signed-off-by: Christian Kohlschütter <christian@kohlschutter.com>
---
 arch/arm64/boot/dts/rockchip/rk3399-nanopi4.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm64/boot/dts/rockchip/rk3399-nanopi4.dtsi b/arch/arm64/boot/dts/rockchip/rk3399-nanopi4.dtsi
index 8c0ff6c96e03..d5f8a62e01be 100644
--- a/arch/arm64/boot/dts/rockchip/rk3399-nanopi4.dtsi
+++ b/arch/arm64/boot/dts/rockchip/rk3399-nanopi4.dtsi
@@ -67,7 +67,7 @@ vcc1v8_s3: vcc1v8-s3 {
 	vcc3v0_sd: vcc3v0-sd {
 		compatible = "regulator-fixed";
 		enable-active-high;
-		gpio = <&gpio0 RK_PA1 GPIO_ACTIVE_HIGH>;
+		// gpio = <&gpio0 RK_PA1 GPIO_ACTIVE_HIGH>; // breaks SDHC card support
 		pinctrl-names = "default";
 		pinctrl-0 = <&sdmmc0_pwr_h>;
 		regulator-always-on;
-- 
2.36.1



Hyperstone GmbH | Reichenaustr. 39a  | 78467 Konstanz
Managing Director: Dr. Jan Peter Berns.
Commercial register of local courts: Freiburg HRB381782

Re: [PATCH v2] arm64: dts: rockchip: Fix SD card init on rk3399-nanopi4

Posted by Christian Kohlschütter 3 years, 9 months ago

> Am 14.07.2022 um 18:44 schrieb Christian Loehle <CLoehle@hyperstone.com>:
> 
> I only briefly skimmed the discussion, but does this mean that a soft-reset (CMD0) of a UHS (post-voltage-switch) will not work?
> (As the card/spec requires a power-cycle by the host which will not come, right?)
> Can you try this real quick? I can give you a mmc-utils snippet if you have trouble issuing one.
> If that does indeed not work I think the general approach is to disable uhs in the dts or at least document that.
> Regards,
> Christian

I tried disabling UHS in the DTS, but that would still cause mmc detection issues.

"rmmod dw_mmc_rockchip" followed by "modprobe dw_mmc_rockchip" still detects the card:
> [ 4481.141764] mmc1: card 0001 removed
> [ 4488.133398] dwmmc_rockchip fe320000.mmc: IDMAC supports 32-bit address mode.
> [ 4488.133462] dwmmc_rockchip fe320000.mmc: Using internal DMA controller.
> [ 4488.133484] dwmmc_rockchip fe320000.mmc: Version ID is 270a
> [ 4488.133541] dwmmc_rockchip fe320000.mmc: DW MMC controller at irq 32,32 bit host data width,256 deep fifo
> [ 4488.134320] dwmmc_rockchip fe320000.mmc: Got CD GPIO
> [ 4488.147329] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 4488.218364] mmc_host mmc1: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
> [ 4488.678181] dwmmc_rockchip fe320000.mmc: Successfully tuned phase to 214
> [ 4488.678239] mmc1: new ultra high speed SDR104 SDHC card at address 0001
> [ 4488.680315] mmcblk1: mmc1:0001 ASTC 14.6 GiB 
> [ 4488.684871]  mmcblk1: p1 p2

Ejecting/re-inserting the card also works:
> [ 4607.521119] mmc1: card 0001 removed
> [ 4608.517343] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 4608.632987] mmc_host mmc1: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
> [ 4609.065445] dwmmc_rockchip fe320000.mmc: Successfully tuned phase to 213
> [ 4609.065535] mmc1: new ultra high speed SDR104 SDHC card at address 0001
> [ 4609.067942] mmcblk1: mmc1:0001 ASTC 14.6 GiB 
> [ 4609.073521]  mmcblk1: p1 p2

and so is changing the clock back and forth:
> echo 400000 > /sys/kernel/debug/mmc1/clock; echo 150000000 > kernel/debug/mmc1/clock; fdisk -l /dev/mmcblk1

> [ 4817.829078] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 4820.063457] mmc_host mmc1: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
> [ 4835.305419] dwmmc_rockchip fe320000.mmc: Successfully tuned phase to 213
> [ 4836.346928]  mmcblk1: p1 p2

Swapping with a "highspeed" (non-UHS) card also seems to work
> [ 5733.702083] mmc1: card 0001 removed
> [ 5738.858439] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 5739.378487] mmc_host mmc1: Bus speed (slot 0) = 50000000Hz (slot req 50000000Hz, actual 50000000HZ div = 0)
> [ 5739.378627] mmc1: new high speed SD card at address 21bb
> [ 5739.380491] mmcblk1: mmc1:21bb APPSD 480 MiB 
> [ 5739.382795] debugfs: Directory 'mmcblk1' with parent 'block' already present!
> [ 5739.385096]  mmcblk1: p1
> [ 5774.386536] FAT-fs (mmcblk1p1): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive!
> [ 5795.486365] mmc1: card 21bb removed
> [ 5801.302688] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 5801.447128] mmc_host mmc1: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
> [ 5801.880374] dwmmc_rockchip fe320000.mmc: Successfully tuned phase to 211
> [ 5801.880440] mmc1: new ultra high speed SDR104 SDHC card at address 0001
> [ 5801.882253] mmcblk1: mmc1:0001 ASTC 14.6 GiB 
> [ 5801.884145] debugfs: Directory 'mmcblk1' with parent 'block' already present!
> [ 5801.886558]  mmcblk1: p1 p2
> 

Some debug output: cat /sys/kernel/debug/mmc1/ios
UHC
> clock:		150000000 Hz
> actual clock:	148500000 Hz
> vdd:		18 (3.0 ~ 3.1 V)
> bus mode:	2 (push-pull)
> chip select:	0 (don't care)
> power mode:	2 (on)
> bus width:	2 (4 bits)
> timing spec:	6 (sd uhs SDR104)
> signal voltage:	1 (1.80 V)
> driver type:	0 (driver type B)
non-UHC
> cat /sys/kernel/debug/mmc1/ios
> clock:		50000000 Hz
> vdd:		18 (3.0 ~ 3.1 V)
> bus mode:	2 (push-pull)
> chip select:	0 (don't care)
> power mode:	2 (on)
> bus width:	2 (4 bits)
> timing spec:	2 (sd high-speed)
> signal voltage:	0 (3.30 V)
> driver type:	0 (driver type B)
> 

How do I make sure I specifically send the soft-reset command? I'm happy to help but I'm really a novice here.

Best,
Christian

RE: [PATCH v2] arm64: dts: rockchip: Fix SD card init on rk3399-nanopi4

Posted by Christian Loehle 3 years, 9 months ago

Please try my softreset patch for mmc-utils.

If some UHS cards have trouble coming up again then we have a problem on that hardware.

-----Original Message-----
From: Christian Kohlschütter <christian@kohlschutter.com> 
Sent: Donnerstag, 14. Juli 2022 19:21
To: Christian Loehle <CLoehle@hyperstone.com>
Cc: Robin Murphy <robin.murphy@arm.com>; Markus Reichl <m.reichl@fivetechno.de>; Heiko Stübner <heiko@sntech.de>; linux-arm-kernel@lists.infradead.org; linux-rockchip@lists.infradead.org; linux-kernel@vger.kernel.org; Linux MMC List <linux-mmc@vger.kernel.org>
Subject: Re: [PATCH v2] arm64: dts: rockchip: Fix SD card init on rk3399-nanopi4

> Am 14.07.2022 um 18:44 schrieb Christian Loehle <CLoehle@hyperstone.com>:
> 
> I only briefly skimmed the discussion, but does this mean that a soft-reset (CMD0) of a UHS (post-voltage-switch) will not work?
> (As the card/spec requires a power-cycle by the host which will not come, right?)
> Can you try this real quick? I can give you a mmc-utils snippet if you have trouble issuing one.
> If that does indeed not work I think the general approach is to disable uhs in the dts or at least document that.
> Regards,
> Christian

I tried disabling UHS in the DTS, but that would still cause mmc detection issues.

"rmmod dw_mmc_rockchip" followed by "modprobe dw_mmc_rockchip" still detects the card:
> [ 4481.141764] mmc1: card 0001 removed
> [ 4488.133398] dwmmc_rockchip fe320000.mmc: IDMAC supports 32-bit address mode.
> [ 4488.133462] dwmmc_rockchip fe320000.mmc: Using internal DMA controller.
> [ 4488.133484] dwmmc_rockchip fe320000.mmc: Version ID is 270a
> [ 4488.133541] dwmmc_rockchip fe320000.mmc: DW MMC controller at irq 32,32 bit host data width,256 deep fifo
> [ 4488.134320] dwmmc_rockchip fe320000.mmc: Got CD GPIO
> [ 4488.147329] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 4488.218364] mmc_host mmc1: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
> [ 4488.678181] dwmmc_rockchip fe320000.mmc: Successfully tuned phase to 214
> [ 4488.678239] mmc1: new ultra high speed SDR104 SDHC card at address 0001
> [ 4488.680315] mmcblk1: mmc1:0001 ASTC 14.6 GiB 
> [ 4488.684871]  mmcblk1: p1 p2

Ejecting/re-inserting the card also works:
> [ 4607.521119] mmc1: card 0001 removed
> [ 4608.517343] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 4608.632987] mmc_host mmc1: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
> [ 4609.065445] dwmmc_rockchip fe320000.mmc: Successfully tuned phase to 213
> [ 4609.065535] mmc1: new ultra high speed SDR104 SDHC card at address 0001
> [ 4609.067942] mmcblk1: mmc1:0001 ASTC 14.6 GiB 
> [ 4609.073521]  mmcblk1: p1 p2

and so is changing the clock back and forth:
> echo 400000 > /sys/kernel/debug/mmc1/clock; echo 150000000 > kernel/debug/mmc1/clock; fdisk -l /dev/mmcblk1

> [ 4817.829078] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 4820.063457] mmc_host mmc1: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
> [ 4835.305419] dwmmc_rockchip fe320000.mmc: Successfully tuned phase to 213
> [ 4836.346928]  mmcblk1: p1 p2

Swapping with a "highspeed" (non-UHS) card also seems to work
> [ 5733.702083] mmc1: card 0001 removed
> [ 5738.858439] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 5739.378487] mmc_host mmc1: Bus speed (slot 0) = 50000000Hz (slot req 50000000Hz, actual 50000000HZ div = 0)
> [ 5739.378627] mmc1: new high speed SD card at address 21bb
> [ 5739.380491] mmcblk1: mmc1:21bb APPSD 480 MiB 
> [ 5739.382795] debugfs: Directory 'mmcblk1' with parent 'block' already present!
> [ 5739.385096]  mmcblk1: p1
> [ 5774.386536] FAT-fs (mmcblk1p1): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive!
> [ 5795.486365] mmc1: card 21bb removed
> [ 5801.302688] mmc_host mmc1: Bus speed (slot 0) = 400000Hz (slot req 400000Hz, actual 400000HZ div = 0)
> [ 5801.447128] mmc_host mmc1: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
> [ 5801.880374] dwmmc_rockchip fe320000.mmc: Successfully tuned phase to 211
> [ 5801.880440] mmc1: new ultra high speed SDR104 SDHC card at address 0001
> [ 5801.882253] mmcblk1: mmc1:0001 ASTC 14.6 GiB 
> [ 5801.884145] debugfs: Directory 'mmcblk1' with parent 'block' already present!
> [ 5801.886558]  mmcblk1: p1 p2
> 

Some debug output: cat /sys/kernel/debug/mmc1/ios
UHC
> clock:		150000000 Hz
> actual clock:	148500000 Hz
> vdd:		18 (3.0 ~ 3.1 V)
> bus mode:	2 (push-pull)
> chip select:	0 (don't care)
> power mode:	2 (on)
> bus width:	2 (4 bits)
> timing spec:	6 (sd uhs SDR104)
> signal voltage:	1 (1.80 V)
> driver type:	0 (driver type B)
non-UHC
> cat /sys/kernel/debug/mmc1/ios
> clock:		50000000 Hz
> vdd:		18 (3.0 ~ 3.1 V)
> bus mode:	2 (push-pull)
> chip select:	0 (don't care)
> power mode:	2 (on)
> bus width:	2 (4 bits)
> timing spec:	2 (sd high-speed)
> signal voltage:	0 (3.30 V)
> driver type:	0 (driver type B)
> 

How do I make sure I specifically send the soft-reset command? I'm happy to help but I'm really a novice here.

Best,
Christian

Hyperstone GmbH | Reichenaustr. 39a  | 78467 Konstanz
Managing Director: Dr. Jan Peter Berns.
Commercial register of local courts: Freiburg HRB381782