eMMC Data CRC Error in TK1

Hi All,

We are facing Data CRC failure issues during emmc read/write operations randomly in our custom tk1 setup. The following is the debug log of the failure,

[ 65.753667] mmc0: Data CRC error, intmask: 200000
[ 65.758371] sdhci: =========== REGISTER DUMP (mmc0)===========
[ 65.764194] sdhci: Sys addr: 0x00000000 | Version: 0x00000303
[ 65.770015] sdhci: Blk size: 0x00007200 | Blk cnt: 0x0000006f
[ 65.775837] sdhci: Argument: 0x007a6370 | Trn mode: 0x00000023
[ 65.781658] sdhci: Present: 0x01fb00f0 | Host ctl: 0x00000031
[ 65.787478] sdhci: Power: 0x0000000b | Blk gap: 0x00000000
[ 65.793299] sdhci: Wake-up: 0x00000000 | Clock: 0x00000007
[ 65.799120] sdhci: Timeout: 0x0000000e | Int stat: 0x00000000
[ 65.804942] sdhci: Int enab: 0x02ff000b | Sig enab: 0x02fc000b
[ 65.810761] sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000
[ 65.816582] sdhci: Caps: 0x376fd080 | Caps_1: 0x10002f77
[ 65.822403] sdhci: Cmd: 0x0000193a | Max curr: 0x00000000
[ 65.828221] sdhci: Host ctl2: 0x0000308b
[ 65.832135] sdhci: ADMA Err: 0x00000000 | ADMA Ptr: 0x9e040020
[ 65.837954] sdhci: ===========================================
[ 65.845952] mmcblk0: error -84 transferring data, sector 8020848, nr 8, cmd response 0x900, card status 0xc00
[ 65.856930] end_request: I/O error, dev mmcblk0, sector 8020848
[ 65.862929] Buffer I/O error on device mmcblk0p14, logical block 513134
[ 65.870865] EXT4-fs warning (device mmcblk0p14): ext4_end_bio:286: I/O error writing to inode 122403 (offset 450560 size 4096 starting block 1002607)
[ 65.884770] end_request: I/O error, dev mmcblk0, sector 6027800
[ 65.891050] Buffer I/O error on device mmcblk0p14, logical block 264003
[ 65.897789] Buffer I/O error on device mmcblk0p14, logical block 264004
[ 65.904745] Buffer I/O error on device mmcblk0p14, logical block 264005
[ 65.911485] Buffer I/O error on device mmcblk0p14, logical block 264006
[ 65.918241] Buffer I/O error on device mmcblk0p14, logical block 264007
[ 65.925194] Buffer I/O error on device mmcblk0p14, logical block 264008
[ 65.931934] Buffer I/O error on device mmcblk0p14, logical block 264009
[ 65.938867] Buffer I/O error on device mmcblk0p14, logical block 264010

Any pointers for fixing this issue would be helpful.

Thanks in advance.

Regards,
Sudar

I ran into the same/similar thing today… and I am uncertain as to what happened… and what needs to be done.

[62863.164173] sdhci: ================== REGISTER DUMP (mmc1)==================
[62863.171227] sdhci: Sys addr[0x000]: 0x00000000 | Version[0x0fe]:  0x00000303
[62863.178274] sdhci: Blk size[0x004]: 0x00007080 | Blk cnt[0x006]:  0x00000000
[62863.185320] sdhci: Argument[0x008]: 0x12003e00 | Trn mode[0x00c]: 0x00000000
[62863.192366] sdhci: Present[0x024]:  0x01fb00f0 | Host ctl[0x028]: 0x00000013
[62863.199412] sdhci: Power[0x029]:    0x0000000d | Blk gap[0x02a]:  0x00000000
[62863.206457] sdhci: Wake-up[0x02b]:  0x00000000 | Clock[0x02c]:    0x00000007
[62863.213501] sdhci: Timeout[0x02e]:  0x0000000e | Int stat[0x030]: 0x00000000
[62863.220546] sdhci: Int enab[0x034]: 0x02ff000b | Sig enab[0x038]: 0x02fc000b
[62863.227591] sdhci: AC12 err[0x03c]: 0x00000000 | Slot int[0x0fc]: 0x00000000
[62863.234636] sdhci: Caps[0x040]:     0x376cd08c | Caps_1[0x044]:   0x10006f73
[62863.241681] sdhci: Cmd[0x00e]:      0x0000341a | Max curr[0x048]: 0x00000000
[62863.248722] sdhci: Host ctl2[0x03e]: 0x0000300b
[62863.253254] sdhci: ADMA Err[0x054]: 0x00000000 | ADMA Ptr[0x058]: 0xfdd00010
[62863.260300] mmc1: tuning_window[0]: 0x0
[62863.260310] mmc1: tuning_window[1]: 0x0
[62863.260320] mmc1: tuning_window[2]: 0x0
[62863.260329] mmc1: tuning_window[3]: 0x0
[62863.260339] mmc1: tuning_window[4]: 0x0
[62863.260349] mmc1: tuning_window[5]: 0x0
[62863.260358] mmc1: tuning_window[6]: 0x0
[62863.260367] mmc1: tuning_window[7]: 0x0
[62863.260373] sdhci: Tap value: 45 | Trim value: 8
[62863.260379] sdhci: SDMMC Interrupt status: 0x00040000
[62863.260382] sdhci: =========================================================
[62863.267421] mmc1: Command CRC error, intmask: 60001 Interface clock = 204000000Hz
[599081.035698] sock: Can't find a process for this sock.

I haven’t yet run a fsck… or done any further troubleshooting… but this raises a few eyebrows. Could this indicate flashrot… or perhaps some data-corruption?

hello TheCompWiz,

here’s Tk1 custom board issue for your reference.
[url]Custom Tegra K1 board MMC problem - Jetson TK1 - NVIDIA Developer Forums

could you check if sending the board ID resolve your issue.
thanks

Hi TheCompWiz,

Have you clarified and resolved the problem?
Any result can be shared?

Thanks