LS1027A - SD card boot

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

LS1027A - SD card boot

Jump to solution
898 Views
Mietto
Contributor I

Hello,

We have developed a custom board with the LS1027A, but we can't get it to boot with the SD Card.

The LS1027A is set to 1.8V at the EVDD pin, meaning the SDIO interface is always at 1.8V, unlike the NXP dev kit, which is set to 3.3V. Then, in our board, we have a 1.8V/3.3V level translator to interface with the SD card, but that should be transparent to the LS1027A.

We can see activity in and out from the SD card, but the boot stops at some point.

Is the LS1027A able to reliably boot from SD card when the interface is at 1.8V?

The question comes from the recommendation below, taken from AN12028.

ls1027a sd card.png

 Best regards.

0 Kudos
Reply
1 Solution
541 Views
yipingwang
NXP TechSupport
NXP TechSupport

Please refer design checklist AN12028, Table 23
For UHS mode, voltage switching is required from 3.3V to 1.8V
The SDIO expander you have used cannot support voltage switching as it supports two SD cards at a time and one may not be ready to switch to UHS.
There are voltage translators that can help to achieve the voltage switching.
LS1027A SDHC controller also supports voltage switching without requiring the in between voltage translator.
EVDD power rail can switch from 3.3V to 1.8V. You may refer NXP reference design for reference.

View solution in original post

0 Kudos
Reply
9 Replies
776 Views
yipingwang
NXP TechSupport
NXP TechSupport

SDHC specification requires all SD interfaces to come up at 3.3V at power and run at default speed.
When switching to UHS mode, voltage also should switch to 1.8V. The voltage translator on board should be capable of doing this voltage switching..
EVDD voltage rail can support both 3.3V and 1.8V. The way it has been designed is to switch to UHS mode without the need of an extra translator on board.
To answer your other question, it is possible to keep EVDD at 1.8V, add on board voltage translator and boot.
You need to share schematic snippet and more details to help us debug your system.
WHen you say boot problem, what fails? Is the RCW loading complete? Is the PBI complete?
Do you have a Code warrior tap? Can you connect it the board and dump the config chain?

0 Kudos
Reply
770 Views
Mietto
Contributor I

Hello,

Thanks for the reply, I'm sending additional details.

The current situation is: uBoot works from SD-card, Linux kernel boots from SD-Card, but then Linux can no longer access the same SD-card to continue booting.

I have attached a snippet from the schematic sheet with the level translator. The signals to the left named DL_EIO* are connected directly to the LS1027A pins. The fact that uBoot and Linux kernel boots indicates that the pins are correctly mapped.

Below I'm sharing the error log.

root@idl1027:~# cat /sys/kernel/debug/mmc0/ios
clock:          50000000 Hz
actual clock:   50000000 Hz
vdd:            22 (3.4 ~ 3.5 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)
root@idl1027:~# mmc0: Timeout waiting for hardware interrupt.
mmc0: sdhci: ============ SDHCI REGISTER DUMP ===========
mmc0: sdhci: Sys addr:  0x00000008 | Version:  0x00002202
mmc0: sdhci: Blk size:  0x00000200 | Blk cnt:  0x00000000
mmc0: sdhci: Argument:  0x00002000 | Trn mode: 0x0000002b
mmc0: sdhci: Present:   0x01f50008 | Host ctl: 0x00000032
mmc0: sdhci: Power:     0x00000002 | Blk gap:  0x00000000
mmc0: sdhci: Wake-up:   0x00000000 | Clock:    0x000000b8
mmc0: sdhci: Timeout:   0x0000000e | Int stat: 0x00000000
mmc0: sdhci: Int enab:  0x037f108f | Sig enab: 0x037f108b
mmc0: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00002202
mmc0: sdhci: Caps:      0x01fa0000 | Caps_1:   0x0000af00
mmc0: sdhci: Cmd:       0x0000193a | Max curr: 0x00000000
mmc0: sdhci: Resp[0]:   0x00000900 | Resp[1]:  0x01db9312
mmc0: sdhci: Resp[2]:   0x325b5900 | Resp[3]:  0x00000900
mmc0: sdhci: Host ctl2: 0x00000000
mmc0: sdhci: ADMA Err:  0x00000000 | ADMA Ptr: 0xa478a200
mmc0: sdhci: ============================================

 

0 Kudos
Reply
672 Views
yipingwang
NXP TechSupport
NXP TechSupport

From the schematic snippet, UHS speeds on SD cannot be supported on the board. Please ensure that the dts in use does not have any UHS mode configurations for SDHC controller.

0 Kudos
Reply
559 Views
Mietto
Contributor I
Hello,

Could you provide information about how to fix our design so that UHS is supported?

Best regards.
0 Kudos
Reply
556 Views
yipingwang
NXP TechSupport
NXP TechSupport

Discussing with the AE team.

0 Kudos
Reply
542 Views
yipingwang
NXP TechSupport
NXP TechSupport

Please refer design checklist AN12028, Table 23
For UHS mode, voltage switching is required from 3.3V to 1.8V
The SDIO expander you have used cannot support voltage switching as it supports two SD cards at a time and one may not be ready to switch to UHS.
There are voltage translators that can help to achieve the voltage switching.
LS1027A SDHC controller also supports voltage switching without requiring the in between voltage translator.
EVDD power rail can switch from 3.3V to 1.8V. You may refer NXP reference design for reference.

0 Kudos
Reply
668 Views
Mietto
Contributor I

Hello,

Thanks for the reply.

Could you provide more details as to why is this not possible?

And what should we do in our board to fix that problem?

Best regards.

0 Kudos
Reply
806 Views
Mietto
Contributor I

We still see an issue when the SDIO changes to high speed mode, it stops working at all.

Question remains: is it possible to keep EVDD at 1.8V at all times and properly boot from it?

0 Kudos
Reply
859 Views
yipingwang
NXP TechSupport
NXP TechSupport

Confirming with the AE team.

0 Kudos
Reply