GNOME Bugzilla – Bug 663155
Gnome 3.2.x - FGLRX - gnome-shell : unusable Desktop
Last modified: 2014-04-26 14:40:09 UTC
Using FGLRX (ATi prop drivers) make the whole Gnome3 (gnome-shell) unusable. Please try to find a solution fast with ATi. As this is pissing me off bigtime. I can not understand why you Gnome devs and ATi devs are not cooperating better with eachother to release a WORKING desktop.
Please provide information on your hardware and driver version.
lspci: 00:00.0 Host bridge: Advanced Micro Devices [AMD] RS780 Host Bridge 00:01.0 PCI bridge: Advanced Micro Devices [AMD] RS780/RS880 PCI to PCI bridge (int gfx) 00:02.0 PCI bridge: Advanced Micro Devices [AMD] RS780 PCI to PCI bridge (ext gfx port 0) 00:09.0 PCI bridge: Advanced Micro Devices [AMD] RS780/RS880 PCI to PCI bridge (PCIE port 4) 00:11.0 SATA controller: ATI Technologies Inc SB7x0/SB8x0/SB9x0 SATA Controller [IDE mode] 00:12.0 USB Controller: ATI Technologies Inc SB7x0/SB8x0/SB9x0 USB OHCI0 Controller 00:12.1 USB Controller: ATI Technologies Inc SB7x0 USB OHCI1 Controller 00:12.2 USB Controller: ATI Technologies Inc SB7x0/SB8x0/SB9x0 USB EHCI Controller 00:13.0 USB Controller: ATI Technologies Inc SB7x0/SB8x0/SB9x0 USB OHCI0 Controller 00:13.1 USB Controller: ATI Technologies Inc SB7x0 USB OHCI1 Controller 00:13.2 USB Controller: ATI Technologies Inc SB7x0/SB8x0/SB9x0 USB EHCI Controller 00:14.0 SMBus: ATI Technologies Inc SBx00 SMBus Controller (rev 3a) 00:14.1 IDE interface: ATI Technologies Inc SB7x0/SB8x0/SB9x0 IDE Controller 00:14.2 Audio device: ATI Technologies Inc SBx00 Azalia (Intel HDA) 00:14.3 ISA bridge: ATI Technologies Inc SB7x0/SB8x0/SB9x0 LPC host controller 00:14.4 PCI bridge: ATI Technologies Inc SBx00 PCI to PCI Bridge 00:14.5 USB Controller: ATI Technologies Inc SB7x0/SB8x0/SB9x0 USB OHCI2 Controller 00:18.0 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor HyperTransport Configuration 00:18.1 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor Address Map 00:18.2 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor DRAM Controller 00:18.3 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor Miscellaneous Control 00:18.4 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor Link Control 01:05.0 VGA compatible controller: ATI Technologies Inc Radeon HD 3300 Graphics 01:05.1 Audio device: ATI Technologies Inc RS780 Azalia controller 02:00.0 VGA compatible controller: ATI Technologies Inc Barts PRO [ATI Radeon HD 6800 Series] 02:00.1 Audio device: ATI Technologies Inc Barts HDMI Audio [Radeon HD 6800 Series] 03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 02) ATi Drivers tested: 11.7 11.8 11.9 11.10 Kernel: 3.0.0 #1 SMP Sun Aug 28 10:49:31 UTC 2011 x86_64 AMD Phenom(tm) 9850 Quad-Core Processor AuthenticAMD GNU/Linux Tested on: Gnome 3.x Gnome 3.2.x Xorg: [ 20.810] X.Org X Server 1.10.4 Release Date: 2011-08-19 [ 20.811] X Protocol Version 11, Revision 0 [ 20.811] Build Operating System: Linux 3.0.0-sabayon x86_64 Gentoo [ 20.811] Current Operating System: Linux Cronos 3.0.0-fusion #1 SMP Sun Aug 28 10:49:31 UTC 2011 x86_64 [ 20.811] Kernel command line: BOOT_IMAGE=/boot/kernel-genkernel-x86_64-3.0.0-fusion ro init=/linuxrc splash=silent,theme:sabayon vga=791 console=tty1 quiet dokeymap keymap=us domdadm resume=swap:UUID=78e000dc-1$ [ 20.811] Build Date: 13 October 2011 05:54:43PM [ 20.811] [ 20.811] Current version of pixman: 0.22.0 [ 20.811] Before reporting problems, check http://wiki.x.org
Same issues on my laptop with Ati Mobility 4xxxx.
Reference: http://ati.cchtml.com/show_bug.cgi?id=99 Reference2: http://ati.cchtml.com/show_bug.cgi?id=264 Reference3: http://ati.cchtml.com/show_bug.cgi?id=283
(In reply to comment #0) > I can not understand why you Gnome devs and ATi devs are not cooperating better > with eachother to release a WORKING desktop. GNOME devs cannot fix the proprietary ATI driver, obviously; there's no "cooperation" problem. The only thing GNOME can do is recommend users to file bugs, just like you did. Or to help fixing the free driver so that you can completely drop the proprietary one. ;-)
(In reply to comment #5) > (In reply to comment #0) > > I can not understand why you Gnome devs and ATi devs are not cooperating better > > with eachother to release a WORKING desktop. > GNOME devs cannot fix the proprietary ATI driver, obviously; there's no > "cooperation" problem. The only thing GNOME can do is recommend users to file > bugs, just like you did. Or to help fixing the free driver so that you can > completely drop the proprietary one. ;-) We can try to talk to AMD to let them know of all issues before release, of course. We'd love to do that, except we're not sure how to contact them. I posted a comment in the first bug link asking them for contact information, so we can be sure that this doesn't happen for 3.4. It also says that AMD fixed the problem internally and the fix should be in 11.10. I don't use an AMD graphics card, though, so I can't make sure of this.
They are hard to contact. But a few of them are on that bug mailing list where you posted the message. Hopefully they will contact you soon. The 11.10 drivers are on the latest GNOME useless.
(In reply to comment #5) > (In reply to comment #0) > > I can not understand why you Gnome devs and ATi devs are not cooperating better > > with eachother to release a WORKING desktop. > GNOME devs cannot fix the proprietary ATI driver, obviously; there's no > "cooperation" problem. The only thing GNOME can do is recommend users to file > bugs, just like you did. Or to help fixing the free driver so that you can > completely drop the proprietary one. ;-) I know you guys cant fix the closed source ATi prop drivers. I am aware of that :) But I had the feeling that ATi and GNOME are having zero communication about the bug that is stopping millions of ATi users to use the actual new GNOME product for a few months now. We are stuck on a forced Fail back and well yes some of us could use the opensource (free) driver. In my case not possible as it is not working on my hardware setup yet. I hope there is a fix anytime soon and that communication from a awesome GDE and 2 big Graphics Card production companies is a lot better. So that everyone can enjoy the Linux Desktop in a way you devs have intended it to be.
Hi, I've emailed John Bridgeman of AMD with a link to this thread and requesting that he could fix what someone referred as (and I quote): "They are hard to contact. But a few of them are on that bug mailing list where you posted the message." So lets see if that doesn't become an excuse to get this done. As for the open source driver, and speaking _ONLY_ for myself, here's the deal... When I buy a piece of hardware I expect the manufacturer provides decent support, and in the case of a graphics chipset, it doesn't work properly without software. Please leave the choice of what drivers to use for users, because for those who require decent openGL/3D support we can't rely on the open source drivers at the current state. Time maybe for leaving the "open source" ego out of this. NM
(In reply to comment #9) > Please leave the choice of what drivers to use for users, because for those who > require decent openGL/3D support we can't rely on the open source drivers at > the current state. Time maybe for leaving the "open source" ego out of this. > > NM It's not the open-source ego. If the closed-source drivers worked, but the open-source ones didn't, I'd recommend you used the closed-source ones for now. Besides that, I have as little power as you to try and coerce AMD to fixing their driver.
@Jasper St. Pierre: I can understand your point. Also I understand Nelson. But the point is, that many user are reliant the AMD driver (also Nvidia driver) because the installed 3D software works better then open source driver. Unfortunately, but I can understand that the companies do not want completely open the driver (Patent reasons, manufacturers competition). It makes me sad, that some open source developer are not working together with company of proprietary software/driver to solve any issue. If I would know the 3D programming in Gnome, I would like to solve the problem for the general public. I also have an open source spirit, but I also help companies that just have proprietary software. Is there a chance to defuse the 3D issue on the Gnome side? Thanks for reading. Regards, Sebastian (openSUSE Member)
More information about the crash of Gnome Shell: # gnome-shell --sync --replace Output: (gnome-shell:2383): Gdk-WARNING **: The program 'gnome-shell' received an X Window System error. This probably reflects a bug in the program. The error was 'GLXBadRenderRequest'. (Details: serial 181 error_code 153 request_code 137 minor_code 1) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get a meaningful backtrace from your debugger if you break on the gdk_x_error() function.)
From what I've discovered, gnome-shell --sync doesn't work correctly. Try: MUTTER_SYNC=1 gnome-shell --replace
Hi Jasper, I try it in the tty: DISPLAY=":0" MUTTER_SYNC=1 gnome-shell --replace But the output is the same like above. Unfortunately I can not trigger a backtrace: DISPLAY=":0" MUTTER_SYNC=1 gdb --args gnome-shell --replace Any suggestions? Regards, Sebastian (openSUSE Member)
Anything I can do to help testing once there is a potential solution, I'll gladly help.
I think you need --synch (from --help-clutter).
@Vincent: I try it with --synch. No backtrace. # DISPLAY=":0" MUTTER_SYNC=1 gdb --args gnome-shell --synch --replace I also try these: # DISPLAY=":0" gdb --args gnome-shell --synch --replace # DISPLAY=":0" MUTTER_SYNC=1 gdb --args gnome-shell --sync --synch --replace I get only this output: (gnome-shell:5348): Gdk-WARNING **: The program 'gnome-shell' received an X Window System error. This probably reflects a bug in the program. The error was 'GLXBadRenderRequest'. (Details: serial 222 error_code 153 request_code 137 minor_code 1) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get a meaningful backtrace from your debugger if you break on the gdk_x_error() function.) [Inferior 1 (process 5348) exited with code 01] When I type "bt" in the prompt of gdb, I get: No stack. Any further suggestions? :-) Regards, Sebastian
As the message says, you need to break on gdk_x_error(). So maybe something like that will work: DISPLAY=":0" MUTTER_SYNC=1 gdb gnome-shell break gdk_x_error run --replace bt
@Milan: Thanks for the tips. I get more information: Breakpoint 1, gdk_x_error (xdisplay=0x63a680, error=0x7fffffffda70) at gdkmain-x11.c:277 277 gdkmain-x11.c: No such file or directory. in gdkmain-x11.c Command: # DISPLAY=":0" MUTTER_SYNC=1 gdb --args gnome-shell --sync --synch --replace Output: GNU gdb (GDB) SUSE (7.3-41.1.2) Copyright (C) 2011 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-suse-linux". For bug reporting instructions, please see: <http://www.gnu.org/software/gdb/bugs/>... Reading symbols from /usr/bin/gnome-shell...Reading symbols from /usr/lib/debug/usr/bin/gnome-shell.debug...done. done. (gdb) break gdk_x_error Function "gdk_x_error" not defined. Make breakpoint pending on future shared library load? (y or [n]) y Breakpoint 1 (gdk_x_error) pending. (gdb) run Starting program: /usr/bin/gnome-shell --sync --synch --replace [Thread debugging using libthread_db enabled] Breakpoint 1, gdk_x_error (xdisplay=0x63a680, error=0x7fffffffda70) at gdkmain-x11.c:277 277 gdkmain-x11.c: No such file or directory. in gdkmain-x11.c (gdb) continue Continuing. (gnome-shell:25383): Gdk-WARNING **: The program 'gnome-shell' received an X Window System error. This probably reflects a bug in the program. The error was 'GLXBadRenderRequest'. (Details: serial 222 error_code 153 request_code 137 minor_code 1) (Note to programmers: normally, X errors are reported asynchronously; that is, you will receive the error a while after causing it. To debug your program, run it with the --sync command line option to change this behavior. You can then get a meaningful backtrace from your debugger if you break on the gdk_x_error() function.) [Inferior 1 (process 25383) exited with code 01] (gdb) quit Does that help? Regards, Sebastian
I have forget to trigger a backtrace. :-/ Here the backtrace: Breakpoint 1, gdk_x_error (xdisplay=0x63a680, error=0x7fffffffda70) at gdkmain-x11.c:277 277 gdkmain-x11.c: No such file or directory. in gdkmain-x11.c (gdb) bt
+ Trace 229018
(In reply to comment #20) > I have forget to trigger a backtrace. :-/ > > > Here the backtrace: > > Breakpoint 1, gdk_x_error (xdisplay=0x63a680, error=0x7fffffffda70) at > gdkmain-x11.c:277 > 277 gdkmain-x11.c: No such file or directory. > in gdkmain-x11.c > (gdb) bt > Does running with COGL_DEBUG="disable-arbfp" work ?
(In reply to comment #21) > > Does running with COGL_DEBUG="disable-arbfp" work ? Yes, the Gnome Shell works evidently. But ... When I start any application, the desktop will be corrupted. The window of the application is not visible. I see only a shadow of the windows. Gnome Shell is unfortunately not still operable. I have got this output from gnome-shell: ... (gnome-shell:6715): Cogl-WARNING **: ./cogl-framebuffer.c:924: Failed to create an OpenGL framebuffer (gnome-shell:6715): St-CRITICAL **: setup_framebuffers: assertion `priv->old_offscreen != COGL_INVALID_HANDLE' failed (gnome-shell:6715): Cogl-WARNING **: ./cogl-framebuffer.c:924: Failed to create an OpenGL framebuffer ...
I am also badly affected by this. Is there anything that we can provide to track that issue down?
(In reply to comment #23) > I am also badly affected by this. Is there anything that we can provide to > track that issue down? fglrx's source code ;) No seriously there is nothing we (GNOME) can do here ... you'd have to talk to AMD.
That's what some of us are doing. Although it seems that some of the changes from 3.0 to 3.2 broke with that driver and in case anyone of the shell developers can point out what is broken in the driver it would be helpful to know.
First of all, let me state that the issues are actually due to bugs in AMD's fglrx driver. Please try the latest driver (currently 12.1) and give brief feedback. I am glad to say that in recent 2-3 months some major issues have been addressed by AMD, others still remain. I say that judging by the comments people give and by my own experience. Graphical corruptions and slanted graphics like http://ati.cchtml.com/show_bug.cgi?id=99 http://ati.cchtml.com/show_bug.cgi?id=339 should have been fixed in many cases, according to comments. Please try the latest driver and try filing bugs in that inofficial AMD bug reporting site. Look for already existing reports first. Known issues remain, like a segmentation fault http://ati.cchtml.com/show_bug.cgi?id=408 that was also reported in GNOME bugzilla, and general poor performance of the Shell. For bad performance with gnome shell, as a workaround try adding export CLUTTER_VBLANK=none to /etc/environment and restart X. It might help.
Exactly the same problem in: -- https://bugzilla.gnome.org/show_bug.cgi?id=668800 (openSUSE 12.1+fglrx 12.1) -- https://bugzilla.gnome.org/show_bug.cgi?id=668930 (Linux mint 12+fglrx 12.1) -- https://bugzilla.redhat.com/show_bug.cgi?id=702257 (Fedora 15,16+fglrx 12.1) -- https://bugzilla.gnome.org/show_bug.cgi?id=663155#c4 (GNOME 3.2.X+FGLRX 12.1) -- http://ati.cchtml.com/show_bug.cgi?id=99 -- http://ati.cchtml.com/show_bug.cgi?id=264 -- http://ati.cchtml.com/show_bug.cgi?id=283 -- http://ati.cchtml.com/show_bug.cgi?id=339 P.s. We must send all of our bugs on ati bagtracker
Alex, please don't paste the same links in all Catalyst bugs you know of. These are not always the same problem. For example, the first series of bugs is about a crash, while the others are about image corruption.
I received this sugestion from the ati-developers on the ati bugzilla: """ One mechanism which seems to work is to use the beta mailing list for catalyst, which goes straight to few Linux devs at amd. if you could ask them to subscribe, this is preferable. """ So far I have not found out where that list is though ...
Hello, As Christian stated in comment 26, adn there was not activity for a while, could you confirm the issue is fixed (even if it's not caused by Shell). Anyway use of closed driver is always discouraged it is impossible to track down the problem and we have to rely on the good-will of the developer. So I suggest you to use the open-source driver. Regards.
Closing this bug report as no further information has been provided. Please feel free to reopen this bug if you can provide the information asked for. Thanks!