New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 720144 link

Starred by 1 user

Issue metadata

Status: Untriaged
Owner: ----
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 2
Type: Bug



Sign in to add a comment

complete freeze with messages: "mmc0: Timeout waiting for hardware interrupt"

Project Member Reported by semenzato@chromium.org, May 9 2017

Issue description

Chris's caroline got slow, and at the next suspend/resume cycle showed a black screen for about 20s after resume, then showed the lock screen but was totally unresponsive.

I did a triple alt-VolUpKey-X and then we sent this feedback report:

https://feedback.corp.google.com/#/Report/59811678283

The console-ramoops has lots of these, which seem to happen exactly every 10s:

[22479.883021] mmc0: Timeout waiting for hardware interrupt.
[22479.883056] sdhci: =========== REGISTER DUMP (mmc0)===========
[22479.883094] sdhci: Sys addr: 0x00000010 | Version:  0x00001002
[22479.883129] sdhci: Blk size: 0x00007200 | Blk cnt:  0x00000010
[22479.883164] sdhci: Argument: 0x00000000 | Trn mode: 0x00000023
[22479.883199] sdhci: Present:  0x1fff0000 | Host ctl: 0x00000019
[22479.883234] sdhci: Power:    0x0000000b | Blk gap:  0x00000080
[22479.883268] sdhci: Wake-up:  0x00000000 | Clock:    0x0000fa07
[22479.883302] sdhci: Timeout:  0x0000000e | Int stat: 0x00000000
[22479.883336] sdhci: Int enab: 0x02000008 | Sig enab: 0x02000008
[22479.883371] sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000
[22479.883405] sdhci: Caps:     0x546ec881 | Caps_1:   0x80000807
[22479.883440] sdhci: Cmd:      0x00000c1b | Max curr: 0x00000000
[22479.883472] sdhci: Host ctl2: 0x00000000
[22479.883503] sdhci: ADMA Err: 0x00000000 | ADMA Ptr: 0x0000000001402000
[22479.883527] sdhci: ===========================================
[22489.898997] mmc0: Timeout waiting for hardware interrupt.
[22489.899026] sdhci: =========== REGISTER DUMP (mmc0)===========
[22489.899059] sdhci: Sys addr: 0x00000010 | Version:  0x00001002
[22489.899090] sdhci: Blk size: 0x00007200 | Blk cnt:  0x00000010
[22489.899121] sdhci: Argument: 0x00010000 | Trn mode: 0x00000023
[22489.899151] sdhci: Present:  0x1fff0000 | Host ctl: 0x00000019
[22489.899181] sdhci: Power:    0x0000000b | Blk gap:  0x00000080
[22489.899211] sdhci: Wake-up:  0x00000000 | Clock:    0x0000fa07
[22489.899242] sdhci: Timeout:  0x0000000e | Int stat: 0x00000000
[22489.899272] sdhci: Int enab: 0x02000008 | Sig enab: 0x02000008
[22489.899303] sdhci: AC12 err: 0x00000000 | Slot int: 0x00000000
[22489.899333] sdhci: Caps:     0x546ec881 | Caps_1:   0x80000807
[22489.899363] sdhci: Cmd:      0x00000d1a | Max curr: 0x00000000
[22489.899391] sdhci: Host ctl2: 0x00000000
[22489.899419] sdhci: ADMA Err: 0x00000000 | ADMA Ptr: 0x0000000001402000
[22489.899439] sdhci: ===========================================
[22489.901667] mmcblk0: error -110 sending status command, retrying
[22499.915000] mmc0: Timeout waiting for hardware interrupt.

Almost all of the stacks in the ramoops, possibly all of them, show threads blocked in file system operations.

I've searched bugs for "Timeout waiting for hardware interrupt" and found this one, which unfortunately doesn't have much information.

https://bugs.chromium.org/p/chromium/issues/detail?id=644818
 
Check the firmware. This issue reminds me an ealier problem with Intel chipset: crbug/583014

I've seen a whole bunch of these on cyan (3.18 Braswell) coming out of resume.  They sometimes end with a hung tasks panic but I suspect a lot of users just hold down the power button to reset the system.

It is likely that these are not being counted as part of our stats so I filed bug 841442 to try to get a better handle on the problem.

Sign in to add a comment