borisfauret, your problem seems to be different, can you check if using kernel parameter
acpi_osi=! acpi_osi=“Windows 2009”
helps with suspend in your case?
Exactly the same problem here…
+1
lenovo thinkpad p50 nvidia quadro m2000m
When booting from live usb ubuntu 18.04, suspend works perfectly (Intel driver)
After installing nvidia driver nvidia-390 (and nvidia-340), suspend causes grey blank screen.
Hey,
I don’t have much idea how these things work, I too have the same problem with my laptop, Dell Vostro 5568, Ubuntu 16.04, Nvidia 940MX, CUDA 9.0 and nvidia driver version 396.
But after restarting the kernal(I used to work in jupyter notebook) the problem is solved, I can suspend my laptop normally.
:)
Thanks for the reply.
Actually, I’ve already tried various editions of the grub file, with no success. Anyway, I’ve just tried the one you mentionned.
Now, when I try to resume from suspend, the screen is not completely black… I can read
[ 95.479606] ERROR @wl_notify_san_status :
[ 97.069926] NVRM: Xid (PCI:0000:01:00): 79, GPU has fallen off the bus.
And then, I’m still stuck. I can only hardreset the laptop.
@borisfauret,
this looks more like a acpi bug. If you’re bold you could try using a 4.17-rc kernel, that has some fixes for those kind of problems.
Considering the worst perspective (being another new fresh install…), I’ve accepted the risk (kernel 4.17rc7)
It works and seems stable so far! Thanks a lot. :-)
Can other people confirm this?
So it was a ACPI bug and it is fixed in 4.17rc7 kernel? So when installing 4.17 when it comes out this bug will go away.
@tipi, borisfauret was suffering from a different suspend/resume bug, which was known so it is fixed. Yours is still unknown so it is unlikely it’ll be fixed with 4.17 as well. Of course you can try.
Desktop user here, using ubuntu 18.04 and Geforce GTX 950 with nvidia 390 driver and experiencing the same issue.
The default noveau works fine.
This problem is not specific to laptop. So I’m sure this can be easily reproduced.
Asus K501UX, NVIDIA GeForce GTX 950M, ubuntu 16.04 (4.13.0-38-generic), NVIDIA-Linux-x86_64-390.67. The same issue - black screen with no response after recovery from suspend.
Same issue, black unresponsive screen after resuming from suspend. Ubuntu Budgie 18.04, Asus r558u nvidia 940mx 4.15 kernel. Driver version 390 from Ubuntu repositories.
Edit: Attached bug reports obtained via ssh when the black screen was present on the laptop after resuming from suspend. Please reply if any more information is needed. There is also another crash report from when dialog popped up at next restart. Maybe that might contain some sensitive information so not attaching directly. Please reply if needed.
nvidia-bug-report.log.gz (92 KB)
dmesg.txt (69.5 KB)
acpidmp.txt (1.01 MB)
Xorg.0.log (26.2 KB)
Got the same problem with Ubuntu 16.04 and drivers 390.30, Linux kernel 4.15.0
Same problem, Laptop Asus-X541UV, Nvidia Geforce 920MX, all works perfectly with intel profile, but if I choose Nvidia profile suspend function won’t work. This happens with no difference in Ubuntu 16.04,16.10,17.04,17.10 and 18.04. This has been a known bug for many years but apparently no one cares in Nvidia Team. It’s a shame that some producers sell their laptops with OSs based on Linux but Nvidia drivers are full of high priority bugs.
I am experiencing the same issue since upgrading my desktop to 18.04 with kernel 4.15.0-23, running a GTX 660 with nvidia drivers 390. On previous versions like 17.10, 17.04 and before I have NEVER experienced this particular issue, suspend/resume worked just fine. In 18.04 suspending also works fine, but on resume the screen remains black. The system seems to be running but I cannot not get it to switch to a tty session. I have experimented a bit and when in this situation I connect using another machine via ssh to my system and kill the X server, it usually comes up with the logon screen. When I logon now and get to the gnome desktop, suspending and resuming suddenly works! However after rebooting the machine the old situation returns.
It has also happened that the system completely froze on killing the Xserver.
I can confirm that upgrading to e.g. kernel 4.17.2 fixes the issue, suspend and resume works fine there but unfortunately I did suffer a complete system lockup with that kernel rather quickly. Will try it some more and also earlier 4.17 versions.
You can add mine to the list. I’ve been looking into this problem for a while now to no avail. It is a right pain in the butt - we really need a fix for this. Thanks.
At night I close the lid on my laptop which suspends it. On resume from suspend in the morning to start my day I get a blank screen and the backlight on the keyboard keys does not show either. Nor can I get into command line mode with ctrl-alt-f1. I need to physically power down with the on/off button and restart my laptop.
ASUS laptop G751JY Intel Core i7-4860HQ 2.4GHz x8
32 GB Ram
Nvidia GeForce GTX 980M / PCIe / SSE2
64bit OS Ubuntu 14.04 with latest updates installed
Running Unity Desktop
kernel version 3.13.0-151-generic
using nVIDIA BINARY driver version 384.130 from nvidia-384 (proprietary, tested)
I speculate that dbus-pm is not compatible to nvidia proprietary driver
try the following
sudo nano /etc/default/acpi-support
modify the line as follow
#SUSPEND_METHODS="dbus-pm dbus-hal pm-utils"
SUSPEND_METHODS="dbus-hal pm-utils"
then
sudo /etc/init.d/acpid restart
Tried your suggestion but it does not fix the problem for me.
I was going to say this was a necropost, but apparently people are still having issues with this, which sucks, but I have a workaround that is working for me.
I am running ArchLinux with the latest NVIDIA Drivers (396.24-1)
For me the issue only appeared if I powered off the display (TV on HDMI) after putting the computer into suspend mode, and then powered it on after waking the computer (by a keypress). Apparently the graphics card / xorg / xrandr didn’t know the display was connected, so I got a “No signal” message on screen.
The workaround I have found is a little script to check xrandr outputs to see if they are connected, and then restart the output using xrandr when they are:
#!path/to/node/node
const exec = require('child_process').exec;
const parse = require('xrandr-parse');
var xrandr = function(){
this.interval = setInterval(function() {
exec('xrandr', function(err, stdout){
var output = parse(stdout);
console.log(stdout);
console.log(output);
console.log(output["HDMI-0"].connected);
if(output["HDMI-0"].connected)
{
exec('xrandr --output HDMI-0 --mode 3840x2160', function(err,stdout){console.log(stdout);});
}
}.bind(this));
}.bind(this), 1000);
}
var read = new xrandr();
It’s a little rough, as I just put it together just now, and will need some modification in order to work for others, but it is better than searching endless forums and finding nothing, and hoping a driver update will fix it (as I have been doing for months). So far it does use about 15-20mb of ram (probably due to node, or a short interval of running) and a negligible amount of cpu (less than 1%) but it does fix the issue. I have my Window Manager (i3) launch it on startup in the background.
Requirements: xrandr-parse - npm
npm i -G xrandr-parse
Hi!
My laptop: Asus-X556UQ
Video: NVidia GeForce 940 MX
I was using Antegros OS and this problem it was appering when resume from suspend, the only solution: hardreset…
kernel: [ 79.856794] nouveau 0000:01:00.0: fifo: SCHED_ERROR 20 []
I’ve just install the last version of Ubuntu (18.04) and the problem is still there… (in just two days my log file have 141506 lines haha) and now is accompanied whit the following:
nouveau 0000:01:00.0: fifo: PBDMA0: 00024000 [GPPTR PBPTR] ch 3 [007faae000 systemd-logind[871]] subc 0 mthd 0000 data 00000000
I hope that you could understand my poor english… thanks for the help!
Hi ,
am on archlinux too ,
where to save the script and what to name it ?