Summary: | AMD graphics hardware hangs with an homogeneous coloured screen or blank screen, and with chirp coming from the graphics card | ||
---|---|---|---|
Product: | Mesa | Reporter: | Alberto Salvia Novella <es20490446e> |
Component: | EGL | Assignee: | mesa-dev |
Status: | RESOLVED INVALID | QA Contact: | mesa-dev |
Severity: | critical | ||
Priority: | highest | ||
Version: | unspecified | ||
Hardware: | x86-64 (AMD64) | ||
OS: | Linux (All) | ||
URL: | https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/881526 | ||
Whiteboard: | |||
i915 platform: | i915 features: | ||
Attachments: |
xorg.0.log
dmesg.0.log |
Description
Alberto Salvia Novella
2015-01-17 15:13:31 UTC
You are using fglrx. Is this an issue with the open source drivers? Yes, I'm not using fgrlx at all. It happens using both the libre and the proprietary drivers. Because this normally happens after first boot, but not in posterior ones, I greatly suspect this is a hardware bug. Probably this is happening because of the GPU passing from a cold state to a warm state too fast, under graphic demanding operations as watching videos are. And Windows users won't be experiencing this as the GPU doesn't stay in a maximum power state all the time. So this should be a bug in the Linux kernel, Direct Rendering Manager's, Dinamic Power Management feature. You seem to be using a very old version of Ubuntu and in particular the Linux kernel. Can you try newer versions? Current versions of the kernel radeon driver support DPM. It also happens to me using Ubuntu 14.10 with xorg 7.7. (In reply to Alberto Salvia Novella from comment #6) > It also happens to me using Ubuntu 14.10 with xorg 7.7. Please attach the full /var/log/Xorg.0.log and output of dmesg corresponding to that here. Created attachment 112508 [details]
xorg.0.log
Created attachment 112509 [details]
dmesg.0.log
Both fglrx and radeon support dynamic power management so this does not likely have anything to do with power management. It looks like a plain old GPU hang. I'd suggest updating your mesa stack in the case of the open source driver. Does fglrx depend on Mesa? How can I figure out if this is a bug in Mesa? (In reply to Alberto Salvia Novella from comment #11) > Does fglrx depend on Mesa? No > How can I figure out if this is a bug in Mesa? Make sure you have completely removed fglrx (there are many guides on how to do this in Ubuntu just google it) and then try using this ppa with updated drivers https://launchpad.net/~oibaf/+archive/ubuntu/graphics-drivers So why is this happening also with the proprietary driver? And about upgrading using the PPA, I have just done it. If in three weeks the problem doesn't reproduce, I will report it as gone in git version. I can confirm upgrading from the PPA didn't solve the problem :) Graphics hanged again just after finishing the latest comment. The open and closed drivers currently don't play nice together. Hopefully one day that will change. But to me it looks like you haven't fully removed fglrx and this is causing you problems. The Synaptic package manager says there isn't any fglrx package installed, and following instructions at <http://askubuntu.com/questions/78675/how-do-i-remove-the-fglrx-drivers-after-ive-installed-them-by-hand> says the same. Wait, comment 2 says further steps are needed... I'm testing now. I will test this for a while. I can still hang graphics by doing this: 1. Let the computer cool down for several hours. 2. Preferably in the morning, when ambiance is cool, switch on the computer. 3. Using the HTML5 YouTube player, play a video. It looks like if the computer has been on for a while, is harder to make it to hang. Other symptom is sound hangs too. And sometimes, when graphics hang, what appear in screen is a pattern of stripes of near the same colour. I confirm it also happens with the Adobe Flash player when reproducing videos in YouTube. https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/881526/comments/56 User reported this to be an overheating issue |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.