Arch Linux

zsugabubus

Member

Registered: 2019-07-29

Posts: 2

SD card internal clock never stabilized

Hi there,

Every time I insert an SD card I get the messages below. Internet didn't say much about it and everything worked fine after they first appeared, so I didn't care about it. I don't know exact dates (sorry) but I highly suspect that it is that commit that introduced the behavior: https://git.kernel.org/pub/scm/linux/ke … a43d1a9f31

mmc0: Internal clock never stabilised.
mmc0: sdhci: ============ SDHCI REGISTER DUMP ===========
mmc0: sdhci: Sys addr:  0x00000000 | Version:  0x0000c001
mmc0: sdhci: Blk size:  0x00000000 | Blk cnt:  0x00000000
mmc0: sdhci: Argument:  0x00000000 | Trn mode: 0x00000000
mmc0: sdhci: Present:   0x01ff0000 | Host ctl: 0x00000000
mmc0: sdhci: Power:     0x0000000f | Blk gap:  0x00000000
mmc0: sdhci: Wake-up:   0x00000000 | Clock:    0x00001003
mmc0: sdhci: Timeout:   0x00000000 | Int stat: 0x00000000
mmc0: sdhci: Int enab:  0x00ff0083 | Sig enab: 0x00ff0083
mmc0: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00000000
mmc0: sdhci: Caps:      0x21e608b2 | Caps_1:   0x00002077
mmc0: sdhci: Cmd:       0x00000000 | Max curr: 0x00000064
mmc0: sdhci: Resp[0]:   0x00000000 | Resp[1]:  0x00000000
mmc0: sdhci: Resp[2]:   0x00000000 | Resp[3]:  0x00000000
mmc0: sdhci: Host ctl2: 0x00000000
mmc0: sdhci: ============================================
mmc0: Internal clock never stabilised.
mmc0: sdhci: ============ SDHCI REGISTER DUMP ===========
mmc0: sdhci: Sys addr:  0xc30ee040 | Version:  0x0000c001
mmc0: sdhci: Blk size:  0x00007040 | Blk cnt:  0x00000001
mmc0: sdhci: Argument:  0x00fffff0 | Trn mode: 0x00000013
mmc0: sdhci: Present:   0x01ff0000 | Host ctl: 0x00000000
mmc0: sdhci: Power:     0x0000000f | Blk gap:  0x00000000
mmc0: sdhci: Wake-up:   0x00000000 | Clock:    0x00000003
mmc0: sdhci: Timeout:   0x0000000a | Int stat: 0x00000000
mmc0: sdhci: Int enab:  0x02ff008b | Sig enab: 0x02ff008b
mmc0: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00000000
mmc0: sdhci: Caps:      0x21e608b2 | Caps_1:   0x00002077
mmc0: sdhci: Cmd:       0x0000063a | Max curr: 0x00000064
mmc0: sdhci: Resp[0]:   0x00000900 | Resp[1]:  0xc8befbcf
mmc0: sdhci: Resp[2]:   0x325f5983 | Resp[3]:  0x00002600
mmc0: sdhci: Host ctl2: 0x00000000
mmc0: sdhci: ============================================
mmc0: new SD card at address e624

The issue, I'm faced now, is that I tried inserting a different SD card into my Dell laptop but I'm unable to write it. I mean I can't create files, can't format the device... etc. I can also reproduce the issue with another Dell Latitude laptop I have in the near.

At first I thought it's a hardware fault, but the same SD card on a different (Asus) laptop works fine without getting any "Internal clock never stabilised" and write errors. fsck confirms that the filesystem is clean.

mmc0: Internal clock never stabilised.
mmc0: sdhci: ============ SDHCI REGISTER DUMP ===========
mmc0: sdhci: Sys addr:  0x00000000 | Version:  0x0000c001
mmc0: sdhci: Blk size:  0x00000000 | Blk cnt:  0x00000000
mmc0: sdhci: Argument:  0x00000000 | Trn mode: 0x00000000
mmc0: sdhci: Present:   0x01ff0000 | Host ctl: 0x00000000
mmc0: sdhci: Power:     0x0000000f | Blk gap:  0x00000000
mmc0: sdhci: Wake-up:   0x00000000 | Clock:    0x00001003
mmc0: sdhci: Timeout:   0x00000000 | Int stat: 0x00000000
mmc0: sdhci: Int enab:  0x00ff0083 | Sig enab: 0x00ff0083
mmc0: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00000000
mmc0: sdhci: Caps:      0x21e608b2 | Caps_1:   0x00002077
mmc0: sdhci: Cmd:       0x00000000 | Max curr: 0x00000064
mmc0: sdhci: Resp[0]:   0x00000000 | Resp[1]:  0x00000000
mmc0: sdhci: Resp[2]:   0x00000000 | Resp[3]:  0x00000000
mmc0: sdhci: Host ctl2: 0x00000000
mmc0: sdhci: ============================================
mmc0: Internal clock never stabilised.
mmc0: sdhci: ============ SDHCI REGISTER DUMP ===========
mmc0: sdhci: Sys addr:  0x00000000 | Version:  0x0000c001
mmc0: sdhci: Blk size:  0x00000000 | Blk cnt:  0x00000000
mmc0: sdhci: Argument:  0xe6240000 | Trn mode: 0x00000000
mmc0: sdhci: Present:   0x01ff0000 | Host ctl: 0x00000000
mmc0: sdhci: Power:     0x0000000f | Blk gap:  0x00000000
mmc0: sdhci: Wake-up:   0x00000000 | Clock:    0x00000003
mmc0: sdhci: Timeout:   0x00000000 | Int stat: 0x00000000
mmc0: sdhci: Int enab:  0x00ff0083 | Sig enab: 0x00ff0083
mmc0: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00000000
mmc0: sdhci: Caps:      0x21e608b2 | Caps_1:   0x00002077
mmc0: sdhci: Cmd:       0x0000071a | Max curr: 0x00000064
mmc0: sdhci: Resp[0]:   0x00000700 | Resp[1]:  0x47804109
mmc0: sdhci: Resp[2]:   0x53553031 | Resp[3]:  0x00035344
mmc0: sdhci: Host ctl2: 0x00000000
mmc0: sdhci: ============================================
print_req_error: I/O error, dev mmcblk0, sector 137 flags 1
Buffer I/O error on dev mmcblk0, logical block 137, lost async page write
print_req_error: I/O error, dev mmcblk0, sector 380 flags 1
Buffer I/O error on dev mmcblk0, logical block 380, lost async page write
print_req_error: I/O error, dev mmcblk0, sector 488 flags 1
Buffer I/O error on dev mmcblk0, logical block 488, lost async page write

(Anyway, what does "Internal clock never stabilised" exacly mean?)

frostschutz

Member

Registered: 2013-11-15

Posts: 1,418

Re: SD card internal clock never stabilized

> I highly suspect that it is that commit that introduced the behavior

So does it work for you if you revert that commit? Nobody else can even try without the same hardware and problem...

If reverting the commit works then increasing the timeout a little might also work.

> Anyway, what does "Internal clock never stabilised" exacly mean?

not sure what it means for sd cards specifically, but in general, communication relies on protocol and protocol relies on timing and timing relies on clock speed

back in the day of dialup modems you could listen to negotiation, and devices today still work similar, you just don't hear it anymore ;-)

zsugabubus

Member

Registered: 2019-07-29

Posts: 2

Re: SD card internal clock never stabilized

So does it work for you if you revert that commit?

No, it's just my pure supposition. I couldn't try it out yet.

A little good news: From the register dump it seems that clock has been stabilized. Maybe we should just wait a bit longer?