Bug 85571
Summary: | wakeup from hibernate does not bring up networking (hard freeze, soft lockups and unusable system) | ||
---|---|---|---|
Product: | ACPI | Reporter: | itsme_410 (itsme_410) |
Component: | Config-Processors | Assignee: | Aaron Lu (aaron.lu) |
Status: | CLOSED INSUFFICIENT_DATA | ||
Severity: | blocking | CC: | aaron.lu, rui.zhang |
Priority: | P1 | ||
Hardware: | Intel | ||
OS: | Linux | ||
Kernel Version: | 3.17.0-1.vanilla.mainline.knurd.1.fc20.x86_64 | Subsystem: | |
Regression: | No | Bisected commit-id: | |
Attachments: |
output of dmesg
dmidecode output of lspic -v output of dmesg when booted with vanilla kernel output of dmidecode to reflect booting from mainline kernel output of lspci -v when booted with mainline kernel output of lsmod when booted with mainline kernel dmesg from 10/11/2014 (when the system crashed upon wakeup) |
Description
itsme_410@yahoo.com
2014-10-04 04:22:35 UTC
Created attachment 152361 [details]
output of dmesg
$dmesg yields the file:
Created attachment 152371 [details]
dmidecode
output of sudo dmidecode
Created attachment 152381 [details]
output of lspic -v
Used lspci -v to get the output in the attachment.
Also happens with the mainline 3.17.0 kernel. I get the following message from syslogd: Message from syslogd@machnie at Oct 10 07:50:47 ... kernel:[42283.515021] do_IRQ: 0.81 No irq handler for vector (irq -1) Message from syslogd@machine at Oct 10 07:50:47 ... kernel:do_IRQ: 0.81 No irq handler for vector (irq -1) Updating the submitted files to reflect this. Created attachment 153101 [details]
output of dmesg when booted with vanilla kernel
Created attachment 153111 [details]
output of dmidecode to reflect booting from mainline kernel
Created attachment 153121 [details]
output of lspci -v when booted with mainline kernel
Created attachment 153131 [details]
output of lsmod when booted with mainline kernel
Not completely sure about the component to be files under, so please reassign if needed. Happy to test and provide feedback. This issue only happens with my Dell Precision M3800, not with my Dell XPS 13. Does it happen on previous kernels? It happens at least from 3.16.2. (This is a new machine so I have not used kernels before those.) Occassionally, upon wakeup, the system crashes as it did this morning. I am providing the dmesg from this morning (10/11/2014). There is reference to a kernel bug. Thank you for looking into this! Created attachment 153241 [details]
dmesg from 10/11/2014 (when the system crashed upon wakeup)
Please follow this document: https://www.kernel.org/doc/Documentation/power/basic-pm-debugging.txt to do some debugging. Thanks! I am sorry for a naive question, but I justr tried 1. in that list (as root) five times and it came back up each time. dmesg was fine. (I can't try it as a mortal user -- am not allowed.) I wanted to mention that I hibernate using the recommended (in Fedora) systemctl hibernate as a mortal user. Am I supposed to do 1b even if 1a is not needed (because 1. worked)? The document is not clear to me. There is some bug in the kernel as per the dmesg I have submitted. But I am not sure what I should do next. Thanks again for looking into this! Please let me know what I should try next. Btw, I tried this: # echo reboot > /sys/power/disk # echo disk > /sys/power/state Aha! Got it to "work" aka crash. I had to do: # echo shutdown > /sys/power/disk # echo disk > /sys/power/state I will go into the 5 testing states now. Pay attention to the test modes of hibernation, they do not require shutdown or reboot the system. In short, please do test like this: # echo devices > /sys/power/pm_test # echo disk > state see if everything works well, if so, proceed to next level, i.e. platform. Thanks! I did the following: # echo freezer > /sys/power/pm_test # echo platform > /sys/power/disk # echo disk > /sys/power/state # echo devices > /sys/power/pm_test # echo platform > /sys/power/disk # echo disk > /sys/power/state # echo platform > /sys/power/pm_test # echo platform > /sys/power/disk # echo disk > /sys/power/state # echo processors > /sys/power/pm_test # echo platform > /sys/power/disk # echo disk > /sys/power/state # echo core > /sys/power/pm_test # echo platform > /sys/power/disk # echo disk > /sys/power/state It came up all five times. However, without the first of the three lines, I have a crash in the sense that the wakeup either crashes with garbage, or the mouse does not click -- two times that I tried. Do you mean if you did: # echo devices > /sys/power/pm_test # echo platform > /sys/power/disk # echo disk > /sys/power/state Then the system doesn't wakeup correctly? No, I am sorry for being confusing in my comment. I meant that if I only did: # echo platform > /sys/power/disk # echo disk > /sys/power/state the system almost always does not wakeup correctly. But if I did those five sets of three commands, the system came up fine. I repeated each of the five sets one more time, and still the system appeard to have come back up correctly. please check if the problem still exists if you build your kernel with CONFIG_I915_DRM cleared. ping... Bug closed as there is no response from the bug reporter since last Oct. please feel free to reopen it if the problem can be reproduced in the latest upstream kernel, and you can provide the information requested in comment #21. |