i.MX6 Dual Lite fails to resume from suspend-to-ram mode

I.MX6
1

Hello,

I am encountering issues using iMX6, mounted on a SRMX6DLW00D01GE008V15IF module from SolidRun. This module is then mounted on a custom board.

Randomly, the i.MX won’t wake up well from a suspend-to-ram mode. It will stay in an “idle” mode as nothing can be done to interact with the system, except disconnect power and reconnect it to make it boot again.

Quick configuration info:

  • i.MX6 Dual Lite
  • SPL (not full U-boot, as i.MX6 Falcon Mode is used)
  • OS kernel 5.10

System logs with a succesfull resume from suspend to ram, then a failed one :

> DSM mode entry (OK):

[04:00:04.371645 57.746743] [ 290.857762] usb 2-1: USB disconnect, device number 14
[04:00:04.388191 0.016549] [ 290.863121] usb 2-1.1: USB disconnect, device number 16
[04:00:04.403335 0.015142] [ 290.885158] usb 2-1.2: USB disconnect, device number 17
[04:00:04.420400 0.017065] [ 290.897500] rndis_host 2-1.2:1.0 usb0: unregister 'rndis_host' usb-ci_hdrc.1-1.2, RNDIS device
[04:00:10.483855 6.063454] [ 296.970872] PM: suspend entry (deep)
[04:00:10.500545 0.016693] [ 296.974685] Filesystems sync: 0.000 seconds
[04:00:10.503274 0.002730] [ 296.979897] Freezing user space processes ... (elapsed 0.001 seconds) done.
[04:00:10.506772 0.003499] [ 296.988237] OOM killer disabled.
[04:00:10.515701 0.008928] [ 296.991505] Freezing remaining freezable tasks ... (elapsed 0.001 seconds) done.
[04:00:10.520343 0.004643] [ 297.000250] printk: Suspending console(s) (use no_console_suspend to debug)

> DSM mode exit, and re-entry (OK):

[04:03:58.957708 228.437359] [ 297.036285] fec 2188000.ethernet eth0: Link is Down
[04:03:58.962198 0.004495] [ 297.039063] Disabling non-boot CPUs ...
[04:03:58.963916 0.001719] [ 297.040658] Enabling non-boot CPUs ...
[04:03:58.973040 0.009124] [ 297.041422] CPU1 is up
[04:03:58.974646 0.001606] [ 297.108663] OOM killer enabled.
[04:03:58.976701 0.002054] [ 297.111836] Restarting tasks ... done.
[04:03:58.988202 0.011502] [ 297.125391] PM: suspend exit
[04:04:00.477769 1.489562] [ 298.608867] usb 2-1: new high-speed USB device number 18 using ci_hdrc
[04:04:00.605712 0.127943] [ 298.747155] hub 2-1:1.0: USB hub found
[04:04:00.620420 0.014709] [ 298.754354] hub 2-1:1.0: 4 ports detected
[04:04:00.892379 0.271961] [ 299.031810] usb 2-1.2: new high-speed USB device number 19 using ci_hdrc
[04:04:00.989769 0.097387] [ 299.135249] fec 2188000.ethernet eth0: Link is Up - 100Mbps/Full - flow control off
[04:04:01.565902 0.576132] [ 299.707733] usb 2-1.1: new full-speed USB device number 20 using ci_hdrc
[04:04:01.661784 0.095884] [ 299.796946] cdc_acm 2-1.1:1.0: ttyACM1: USB ACM device
[04:05:10.549342 0.003612] [ 368.679662] printk: Suspending console(s) (use no_console_suspend to debug)

> DSM exit failure (Not OK):

[04:11:07.262477 356.713130]
[04:11:07.263048 0.000573] U-Boot SPL 2021.01-00007-gadb0eb9497-dirty (Apr 28 2021 - 14:05:46 +0200)

Logs explanation :

This sequence shows the system going to deep sleep mode (= suspend-to-ram) and exiting it every few minutes, which is the normal lifecycle of the system.

The system enters deep sleep mode at 04:00. A first wake up from RTC is carried out at 04:03:58 as required by the rtcwake command I use (rtcwake -m mem).

It seems, for unknown reason, the system tries to boot at 04:11 instead of just resuming as it is required by rtcwake. And the boot does not seem to even finish for the system stays in this idle state and the last log line is related to U-boot SPL.

Tests we carried out :

  • We monitored power signals and initially detected a slight voltage drop on 3.2V SoM power input (-600mV). This drop was corrected, but the issue is still here.
  • We tried to alter peripherals power up/down to see if there was an influence. No influence.
  • We tried to replace suspend-to-ram mode by the power-off mode. Power-off mode performs well (no boot problem)
  • We tried to see if systemd was alive by blinking a LED whenever it is active. In the “idle” mode the system stays, no blinking.

All pieces of advice will be useful, as I tried a lot of workarounds and nothing seems to work.
I can provide further detailed information.

Thanks a lot,

Frank

2

Hi Aurélien, Frank,

Thanks for the details technical description of the issue.

Unfortunately we have no experience with suspend to RAM functionality of the kernel. This feature is not widely tested so bug might appear over time.

I guess that the SPL boot attempt you see is a result of the watchdog kicking in. You might want to try this patch:

http://lists.infradead.org/pipermail/linux-arm-kernel/2021-May/657491.html

See also the discussion on an earlier version of this patch for some background here:

https://lore.kernel.org/linux-arm-kernel/1620732032-7675-1-git-send-email-yibin.gong@nxp.com/t/#u

It is good to see you are using upstream kernel and U-Boot. Since this is the case I’d suggest you to reproduce the problem with latest mainline kernel, and report to upstream developers in the linux-arm-kernel mailing lists (linux-arm-kernel@lists.infradead.org). This list requires you to subscribe to post. see linux-arm-kernel Info Page .

Make sure to Cc the NXP Linux Team at linux-imx@nxp.com, and Fabio Estevam <festevam@gmail.com>. They might be able to help.

3

Please also add this option to your kernel commandline. This will then provide additional debug output to the console as the device goes in and out of suspend. I do agree that most likely this is caused by the watchdog kicking in on resume.

4

Hello @jnettlet and @YazanShhadySR,

Thank you for your answers. I have some news and more questions about the topic as we did a lot of tests since.

We tried with NXP kernel 5.4 from codeaurora. The problem apparently disappeared on our platform (which embed the SolidRun SOM): no more U-Boot and freeze when exiting suspend. This is geat news.

However, 2 other issues, which were initially very rare are now very common. Consequences are the same : randomly, the i.MX won’t wake up well from a suspend-to-ram mode. It will stay in an “idle” mode as nothing can be done to interact with the system, except disconnect power and reconnect it to make it boot again.

Case n°1 : error -84

[03:10:49.768217 1.105460] [ 8484.359835] PM: suspend entry (deep)
[03:10:49.773508 0.005295] [ 8484.363646] Filesystems sync: 0.000 seconds
[03:10:49.775322 0.001815] [ 8484.368561] Freezing user space processes … (elapsed 0.001 seconds) done.
[03:10:49.783931 0.008608] [ 8484.377170] OOM killer disabled.
[03:10:49.786007 0.002077] [ 8484.380455] Freezing remaining freezable tasks … (elapsed 0.001 seconds) done.
[03:10:49.790556 0.004549] [ 8484.389204] printk: Suspending console(s) (use no_console_suspend to debug)
[03:14:43.134265 233.343704] [ 8484.423579] fec 2188000.ethernet eth0: Link is Down
[03:14:43.138717 0.004456] [ 8484.427205] Disabling non-boot CPUs …
[03:14:43.140497 0.001781] [ 8484.428654] Enabling non-boot CPUs …
[03:14:43.149932 0.009434] [ 8484.429470] CPU1 is up
[03:14:43.151450 0.001519] [ 8484.461285] OOM killer enabled.
[03:14:43.153331 0.001881] [ 8484.464455] Restarting tasks … done.
[03:14:43.155550 0.002219] [ 8484.470775] PM: suspend exit
[03:14:43.534233 0.378677] [ 8484.845584] mmc1: tuning execution failed: -84
[03:14:43.538275 0.004047] [ 8484.850105] mmc1: error -84 doing runtime resume
[03:14:43.540420 0.002146] [ 8484.856967] EXT4-fs warning (device mmcblk1p1): ext4_end_bio:315: I/O error 10 writing to inode 134822 (offset 0 size 4096 starting block 627296)
[03:14:43.557243 0.016821] [ 8484.870143] Buffer I/O error on device mmcblk1p1, logical block 624223
[03:14:43.567174 0.009930] [ 8484.878993] EXT4-fs warning (device mmcblk1p1): ext4_end_bio:315: I/O error 10 writing to inode 134825 (offset 0 size 4096 starting block 722997)
[03:14:43.583654 0.016478] [ 8484.879466] mmc1: card 59b4 removed
[03:14:43.587762 0.004111] [ 8484.892161] Buffer I/O error on device mmcblk1p1, logical block 719924
[03:14:43.595249 0.007488] [ 8484.902267] EXT4-fs warning (device mmcblk1p1): ext4_end_bio:315: I/O error 10 writing to inode 8570 (offset 0 size 0 starting block 1790998)
[03:14:43.606834 0.011585] [ 8484.915050] Buffer I/O error on device mmcblk1p1, logical block 1787925
[03:14:43.610353 0.003520] [ 8484.921825] EXT4-fs warning (device mmcblk1p1): ext4_end_bio:315: I/O error 10 writing to inode 8518 (offset 0 size 0 starting block 1270168)
[03:14:43.618609 0.008255] [ 8484.934619] Buffer I/O error on device mmcblk1p1, logical block 1267095
[03:14:43.630478 0.011868] [ 8484.941356] EXT4-fs warning (device mmcblk1p1): ext4_end_bio:315: I/O error 10 writing to inode 8518 (offset 3768320 size 4096 starting block 1270169)
[03:14:43.646438 0.015960] [ 8484.954935] Buffer I/O error on device mmcblk1p1, logical block 1267096
[03:14:43.648368 0.001932] [ 8484.961674] EXT4-fs warning (device mmcblk1p1): ext4_end_bio:315: I/O error 10 writing to inode 8528 (offset 0 size 0 starting block 1191875)
[03:14:43.663168 0.014795] [ 8484.974461] Buffer I/O error on device mmcblk1p1, logical block 1188802
[03:14:43.665632 0.002468] [ 8484.981210] EXT4-fs warning (device mmcblk1p1): ext4_end_bio:315: I/O error 10 writing to inode 8528 (offset 3944448 size 4096 starting block 1191876)
[03:14:43.680338 0.014706] [ 8484.994778] Buffer I/O error on device mmcblk1p1, logical block 1188803
[03:14:43.682265 0.001928] [ 8485.001989] JBD2: Detected IO errors while flushing file data on mmcblk1p1-8
[03:14:43.696030 0.013763] [ 8485.010307] Aborting journal on device mmcblk1p1-8.
[03:14:43.697559 0.001531] [ 8485.015418] JBD2: Error -5 detected when updating journal superblock for mmcblk1p1-8.
[03:14:43.711773 0.014212] [ 8485.019025] EXT4-fs (mmcblk1p1): I/O error while writing superblock
[03:14:43.713703 0.001932] [ 8485.029683] EXT4-fs error (device mmcblk1p1): ext4_journal_check_start:61: Detected aborted journal
[03:14:43.727718 0.014005] [ 8485.038803] EXT4-fs (mmcblk1p1): Remounting filesystem read-only
[03:14:43.742761 0.015052] [ 8485.065542] EXT4-fs (mmcblk1p1): I/O error while writing superblock
[03:14:44.142696 0.399930] [ 8485.462667] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm systemd-udevd: reading directory lblock 0
[03:14:44.162843 0.020150] [ 8485.474523] EXT4-fs (mmcblk1p1): I/O error while writing superblock
[03:14:44.169231 0.006389] [ 8485.481023] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8098: comm sysnav-pr8011-m: reading directory lblock 0
[03:14:44.185245 0.016012] [ 8485.481665] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #5694: comm systemd-udevd: reading directory lblock 0
[03:14:44.195698 0.010455] [ 8485.502718] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #1907: comm kworker/u4:2: reading directory lblock 0
[03:14:44.206260 0.010563] [ 8485.516637] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm systemd-udevd: reading directory lblock 0
[03:14:44.210681 0.004417] [ 8485.528586] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm systemd-udevd: reading directory lblock 0
[03:14:44.224456 0.013780] [ 8485.541318] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #158: comm systemd-udevd: reading directory lblock 0
[03:14:44.239494 0.015036] [ 8485.553158] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #1907: comm systemd-udevd: reading directory lblock 0
[03:14:44.254443 0.014951] [ 8485.565565] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8098: comm systemd-udevd: reading directory lblock 0
[03:14:44.257817 0.003374] [ 8485.577924] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm systemd-udevd: reading directory lblock 0
[03:14:44.301742 0.043922] [ 8485.623240] mmc1: host does not support reading read-only switch, assuming write-enable
[03:14:44.479202 0.177457] [ 8485.799751] mmc1: new ultra high speed SDR104 SDXC card at address 59b4
[03:14:44.498736 0.019536] [ 8485.809512] mmcblk1: mmc1:59b4 ED2S5 119 GiB
[03:14:44.509619 0.010886] [ 8485.818745] mmcblk1: p1
[03:14:44.750141 0.240517] [ 8486.066056] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:44.768298 0.018160] [ 8486.078438] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:44.781667 0.013369] [ 8486.090868] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:46.207127 1.425457] [ 8487.515586] fec 2188000.ethernet eth0: Link is Up - 100Mbps/Full - flow control off
[03:14:48.751008 2.543880] [ 8490.066028] EXT4-fs warning: 290 callbacks suppressed
[03:14:48.758837 0.007835] [ 8490.066043] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.770059 0.011222] [ 8490.083858] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.787349 0.017290] [ 8490.096472] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.798209 0.010860] [ 8490.108726] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.813594 0.015385] [ 8490.122140] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.817005 0.003411] [ 8490.134378] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.832828 0.015822] [ 8490.147012] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.847179 0.014352] [ 8490.159386] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.862323 0.015143] [ 8490.171781] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:48.868357 0.006034] [ 8490.184132] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:49.245687 0.377327] [ 8490.559277] EXT4-fs error: 623 callbacks suppressed
[03:14:49.251395 0.005712] [ 8490.559289] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #132307: comm redis-server: reading directory lblock 0
[03:14:49.758943 0.507544] [ 8491.066550] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm python3: reading directory lblock 0
[03:14:50.750966 0.992022] [ 8492.067228] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm python3: reading directory lblock 0
[03:14:50.877846 0.126881] [ 8492.196600] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:50.900332 0.022488] [ 8492.214039] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:50.918995 0.018662] [ 8492.229741] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:50.933549 0.014551] [ 8492.245426] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:50.945258 0.011716] [ 8492.261175] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:50.960112 0.014851] [ 8492.276780] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:50.976722 0.016612] [ 8492.292440] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:53.757771 2.781044] [ 8495.071244] EXT4-fs warning: 340 callbacks suppressed
[03:14:53.766294 0.008527] [ 8495.071258] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.776383 0.010089] [ 8495.088695] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.791296 0.014914] [ 8495.101173] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.799384 0.008087] [ 8495.113426] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.813276 0.013892] [ 8495.126699] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.827488 0.014212] [ 8495.138958] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.840785 0.013298] [ 8495.151337] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.854485 0.013699] [ 8495.163609] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.862683 0.008199] [ 8495.175954] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:53.876890 0.014205] [ 8495.188297] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:54.750019 0.873126] [ 8496.070112] EXT4-fs error: 222 callbacks suppressed
[03:14:54.767738 0.017723] [ 8496.070126] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm python3: reading directory lblock 0
[03:14:54.894596 0.126856] [ 8496.205677] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:54.909272 0.014677] [ 8496.222906] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:54.926467 0.017195] [ 8496.238667] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:54.942645 0.016177] [ 8496.254409] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:54.957815 0.015172] [ 8496.268656] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm systemd-journal: reading directory lblock 0
[03:14:54.973249 0.015434] [ 8496.281561] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm systemd-journal: reading directory lblock 0
[03:14:54.979630 0.006377] [ 8496.293561] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:54.993227 0.013601] [ 8496.305548] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm systemd-journal: reading directory lblock 0
[03:14:55.007274 0.014048] [ 8496.317532] EXT4-fs error (device mmcblk1p1): __ext4_find_entry:1532: inode #8100: comm sysnav-pr8011-r: reading directory lblock 0
[03:14:58.765976 3.758698] [ 8500.074233] EXT4-fs warning: 411 callbacks suppressed
[03:14:58.774299 0.008327] [ 8500.074247] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #155: lblock 0: comm bash: error -5 reading directory block
[03:14:58.781857 0.007559] [ 8500.078198] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #130615: lblock 0: comm python3: error -5 reading directory block
[03:14:58.787444 0.005585] [ 8500.089259] EXT4-fs warning (device mmcblk1p1): dx_probe:761: inode #155: lblock 0: comm bash: error -5 reading directory block
[03:14:58.802013 0.014570] [ 8500.092576] EXT4-fs warning (device mmcblk1p1): dx_probe:761:

And so on…

Case n°2 : Freeze at PM suspend exit

[11:38:02.865070 0.001584] [55577.217222] CPU1 is up
[11:38:02.866361 0.001287] [55577.249707] OOM killer enabled.
[11:38:02.875042 0.008682] [55577.252879] Restarting tasks ... done.
[11:38:02.889130 0.014085] [55577.268599] PM: suspend exit

And then nothing more.

Some explanations :

Case n°1 : the system declares error -84 then outputs continuously error messages, as you can read.

Case n°2 : the system seems to roll the suspend exit sequence normaly, but freezes after writing the line “PM: suspend exit”.

Frequency of appearance :

On our systems under tests (operating a deep sleep entry and exit every 3 minutes for prompting the issue as soon as possible), any of the two cases might occur (apparently in a random fashion).

We might have to wait 30 minutes or 60 hours as well for the system to crash (case 1 or 2). As I said, pretty random.

Strategy we would like to implement :

  • Before working on the source of the problem, we would like to implement a watchdog in the OS. This watchdog, by resuming early after deep sleep exit, will reboot the system in case n°2, and case n°1 too (in case n°1, we detect error-84 and prompt a kernel panic. Then, reboot is operated by the watchdog).
  • Once we ensure by these means that the system will not stay stuck when case 1 and case 2 crashes occur, we will tackle the issue source

Questions about our platform and the Hummingboard

  • We implemented watchdog and the kernel panic prompting. However the watchdog does not reboot with the SOM mounted on our platform. With the same SOM mounted on the hummingboard, the watchdog operates the reboot. The SOM implementation datasheet is fully respected on our platform. Is there any rule of thumb implemented in hummingboard for operating the SOM that is not specified in datasheets ?
  • We remarked some SD cards (where OS images are flashed) will enable the watchdog to boot on our platform, and some won’t. On the hummingboard, all SD seems to work. As the SOM implementation, the SD implementation datasheets are fully respected by our platform design. Do you know why we encounter such surprises ?
  • Regarding the SD card. We performed eye diagram drawing on one of the databus of the SD card. On our platform, the diagram is not very clean (lines are blurry). However, it is the same with Hummingboard. I’m affraid we may be at an edge of a malfunction which is without consequences on humminboard, but impacts our platform. What are your thoughts about it ?

I can provide more information if necessary.
I’m afraid this topic is not yet solved.

Thanks a lot for the help,

Frank

5

My guess is that the timing on your voltage regulator for the usdhc2 card needs to be tuned on your board. Early on in implementing UHS support we found that some card’s firmware would crash if they weren’t given a long enough delay.

Does the watchdog not trigger, or is the watchdog resetting the SOM and the device not booting because the SDHC card is not responding?

It is probably best if we have a call to discuss these details.