vc4.ko causes another misterious error

[Nicolas Saenz Julienne]

Hi Ryutaroh,

On Sun, 2021-03-21 at 04:31 +0900, Ryutaroh Matsumoto wrote:
> Hi, I sent the below to linux-rpi-kernel list on February 18,
> but it has remained unaccepted, so I resend this.
> 
> I found another mysterious error caused by vc4.ko.
> Only when vc4.ko is loaded AND the SD card slot is empty in my RPi4B 8GB,
> I observe the following harmless error message at every 10 seconds:
> 
> [  255.457584] mmc1: Timeout waiting for hardware cmd interrupt.
> [  255.460735] mmc1: sdhci: ============ SDHCI REGISTER DUMP ===========
> [  255.464244] mmc1: sdhci: Sys addr:  0x00000000 | Version:  0x00001002
> [  255.467752] mmc1: sdhci: Blk size:  0x00000000 | Blk cnt:  0x00000000
> [  255.471259] mmc1: sdhci: Argument:  0x00000000 | Trn mode: 0x00000000
> [  255.474765] mmc1: sdhci: Present:   0x1fff0001 | Host ctl: 0x00000001
> [  255.478271] mmc1: sdhci: Power:     0x0000000f | Blk gap:  0x00000080
> [  255.481777] mmc1: sdhci: Wake-up:   0x00000000 | Clock:    0x0000f447
> [  255.485282] mmc1: sdhci: Timeout:   0x00000000 | Int stat: 0x00000000
> [  255.488787] mmc1: sdhci: Int enab:  0x00ff1003 | Sig enab: 0x00ff1003
> [  255.492293] mmc1: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00000000
> [  255.495799] mmc1: sdhci: Caps:      0x45ee6432 | Caps_1:   0x0000a525
> [  255.499305] mmc1: sdhci: Cmd:       0x00000502 | Max curr: 0x00080008
> [  255.502809] mmc1: sdhci: Resp[0]:   0x00000000 | Resp[1]:  0x00000000
> [  255.506314] mmc1: sdhci: Resp[2]:   0x00000000 | Resp[3]:  0x00000000
> [  255.509819] mmc1: sdhci: Host ctl2: 0x00000000
> [  255.512239] mmc1: sdhci: ADMA Err:  0x00000000 | ADMA Ptr: 0x00000000
> [  255.515743] mmc1: sdhci: ============================================
> [  265.697585] mmc1: Timeout waiting for hardware cmd interrupt.
> [  265.700734] mmc1: sdhci: ============ SDHCI REGISTER DUMP ===========
> [  265.704244] mmc1: sdhci: Sys addr:  0x00000000 | Version:  0x00001002
> [  265.707752] mmc1: sdhci: Blk size:  0x00000000 | Blk cnt:  0x00000000
> [  265.711258] mmc1: sdhci: Argument:  0x00000000 | Trn mode: 0x00000000
> [  265.714764] mmc1: sdhci: Present:   0x1fff0001 | Host ctl: 0x00000001
> [  265.718270] mmc1: sdhci: Power:     0x0000000f | Blk gap:  0x00000080
> [  265.721775] mmc1: sdhci: Wake-up:   0x00000000 | Clock:    0x0000f447
> [  265.725279] mmc1: sdhci: Timeout:   0x00000000 | Int stat: 0x00000000
> [  265.728785] mmc1: sdhci: Int enab:  0x00ff1003 | Sig enab: 0x00ff1003
> [  265.732290] mmc1: sdhci: ACmd stat: 0x00000000 | Slot int: 0x00000000
> [  265.735795] mmc1: sdhci: Caps:      0x45ee6432 | Caps_1:   0x0000a525
> [  265.739300] mmc1: sdhci: Cmd:       0x00000502 | Max curr: 0x00080008
> [  265.742806] mmc1: sdhci: Resp[0]:   0x00000000 | Resp[1]:  0x00000000
> [  265.746311] mmc1: sdhci: Resp[2]:   0x00000000 | Resp[3]:  0x00000000
> [  265.749815] mmc1: sdhci: Host ctl2: 0x00000000
> [  265.752234] mmc1: sdhci: ADMA Err:  0x00000000 | ADMA Ptr: 0x00000000
> [  265.755739] mmc1: sdhci: ============================================
> 
> Kernel is plain upstream 5.10.17.
> As before, all WiFi frequencies remain unusable with vc4.ko.

Sorry for the late reply, I missed your mail as I assumed by the title it was
targeted at Maxime.

I've been looking at this issue for a while already. See here[1], it's not yet
ready for merging, I'm waiting for someone with access to proper documentation
to comment on it. But it does the trick.

Regards,
Nicolas

[1] https://lore.kernel.org/linux-mmc/20210322185816.27582-1-nsaenz@kernel.org/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: This is a digitally signed message part
URL: <http://lists.infradead.org/pipermail/linux-rpi-kernel/attachments/20210325/b00227c1/attachment.sig>