Bug 60674
| Summary: | linux 3.10.x RV740 (Radeon HD 4770) display problem | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | Y. G. (theYinYeti) (bugz) |
| Component: | Video(DRI - non Intel) | Assignee: | drivers_video-dri |
| Status: | NEW --- | ||
| Severity: | blocking | CC: | alexdeucher, nucrap, ondrej, systemdkiosk, szg00000, zxd123314 |
| Priority: | P1 | ||
| Hardware: | x86-64 | ||
| OS: | Linux | ||
| Kernel Version: | 3.10 | Subsystem: | |
| Regression: | No | Bisected commit-id: | |
| Attachments: |
Screenshot with kernel 3.10 (display BUG)
For comparison, same screenshot with kernel 3.9. possible fix |
||
Created attachment 107074 [details]
For comparison, same screenshot with kernel 3.9.
Just to be sure, is the only component you changed the kernel (i.e., you didn't change the mesa version as well)? If so, can you bisect the kernel? Same issue here. Very many Radeon HD 4770 users are experiencing it: https://bbs.archlinux.org/viewtopic.php?id=167248 This is probably a bug in the mesa or ddx driver due to some feature that requires a newer kernel and thus is only enabled on newer kernels. Can you bisect? Indeed downgrading to kernel 3.9.9 doesn't solve the issue for me.. will play with mesa now Ah, sorry! Downgrading to kernel 3.9.9 does solve the issue! SDDM (my login manager) still flickers but this seems to be Qt5 related. Really sorry, ignore my previous comment! I have taken the time, read some how-to's and bisected the kernel. I have managed narrow down the number responsible commits to 8. Unfortunately out of some reason Xorg sincerely broke on my system (with only two bisecting steps left -_-'): When trying to launch the X server e.g. via 'startx' the screen just gets black and after some long time it crashes with the warning: "No protocol specified". Although I can't explain this, it must have been the result of the corrupted kernel code while I was testing the bisected kernels. So this left me with a broken system, but I hope it will be of some help for solving the bug and won't fall on deaf ears. Here's the current output of 'git bisect visualize': https://gist.github.com/Nuc1eoN/6224207 best regards Philip Created attachment 107198 [details]
possible fix
This patch should fix it.
Sorry for the long time without an answer; I was on holidays. Just to be clear: the kernel (and associated packages for my setup) is the only component that changes; here’s the exact list: linux, linux-headers, spl-utils, spl, zfs-utils, zfs, virtualbox-host-modules. I don’t know how to “bissect” and all this stuf. However, I will try the patch suggested by Alex; thanks, by the way :) Hey Alex, I was able to resolve my Xorg issue (it was due to no disk space being left on the partition :s) and could fully test your patch. I can gladly confirm that it fully fixes the issue! Thanks for your great work at the Radeon open source drivers! regards, Philip Thank you guys for solving this one out! I can confirm the patch works; I applied it to an almost vanilla Arch kernel (package linux 3.10.7-1) and the driver now works with my HD 4770 beautifully, Gtk applications are drawn correctly. We encountered such visual issues on a different, older GPU: RV350 (Shivah) on AGP bus xorg-xserver 1.14.2-2 xf86-video-ati 1.7.2.0-1 qt4 4.8.5-1 kernel 3.10.6-2 32-bit Arch Linux, up-to-date Similar screenshots and reports, https://bbs.archlinux.org/viewtopic.php?id=167248 For us, switching the xorg.conf AccelMethod from EXA to glamor caused the mess. No problems occur under EXA. GTK+ and XUL apps suffer far more than Qt. KDE desktop and apps render OK. We get a random dead blank screen on likely hardware register or interrupt issues, but only in glamor mode. Boxes run stable EXA indefinitely. (In reply to Dave from comment #12) > RV350 (Shivah) on AGP bus [...] > For us, switching the xorg.conf AccelMethod from EXA to glamor caused the > mess. No problems occur under EXA. I doubt an RV350 can handle all of glamor's shaders, so you'll have to stick to EXA for that. Anyway, not the same problem this bug report is about. (In reply to Michel Dänzer from comment #13) Michel you could well be 100% right. I'll simply add, RV350 failures should emit X log outputs, not X crashes or worse, hardware/driver/kernel crashes. Code should fail gracefully. We see screenshots like others here, whatever our issue. People reporting use glamor. I got here googling for it. The patch offered struck me as specific to one Radeon GPU. Logic which needs patching for one GPU may need it for others. If not then such specificity needs more code comments, perhaps. I'm unaquainted with the code base. Thanks. (In reply to Dave from comment #14) > (In reply to Michel Dänzer from comment #13) > > Michel you could well be 100% right. While the symptoms may be similar, your issue is completely different, and not kernel related. Please open a new bug on https://bugs.freedesktop.org (for the ddx: product xorg, component Driver/Radeon). *** Bug 60741 has been marked as a duplicate of this bug. *** Please try to reproduce this bug with latest kernel image. Sorry, I changed of PC, so the hardware is now completely unrelated, which is why I let others do tests and all… On my new hardware, all is fine. |
Created attachment 107073 [details] Screenshot with kernel 3.10 (display BUG) Description: Hard to describe… It seems that fonts don’t get displayed. Seeing the screenshots, I’m not sure it is the only problem, though… (see icons for example) Linux console is OK. E17 is OK. Additional info: * OS: up-to-date Arch Linux * current (working kernel): 3.9.9 * tested non-working kernels: 3.10.2 and 3.10.3 Steps to reproduce: * upgrade kernel, * get this: [attached screenshot] As you can see, Enlightenment17 has no problem for some reason. However, all other applications I tried were displayed like the screenshot: Firefox, Thunderbird, Geany… and SLIM (the display manager)! On the login screen for example, I can enter my login name, but no text is shown on screen; it lets me login though.