In the latest rawhide update, the nv driver (xorg-x11-drv-nv-2.1.6-6.fc9) kills the X server. The nouveau driver works.
Once X is running, it fails to start a window manager, repeatedly spewing this into the log file:
AUDIT: Wed Jan 23 09:01:06 2008: 2932 X: client 1 rejected from local host ( uid =0 gid=0 pid=2931 ) Auth name: XDM-AUTHORIZATION-1 ID: -1
Doesn't matter whether running xdm or gdm.
Dave
You have to downgrade. I was getting the same thing last night and downgrading fixed the problem. When using xorg-x11-drv-nv-2.1.6-6.fc9 I was getting this error starting the X Server. Rebuilding synaptics , xorg-x11-server , xorg-x11-server-utils, and xorg-x11-drivers-nv from src.rpm did not fix the problem
Error below: X.Org X Server 1.4.99.2 Release Date: 5 September 2007 X Protocol Version 11, Revision 0 Build Operating System: Linux 2.6.23.12-99.GAJ70.fc9 i686 Current Operating System: Linux localhost.localdomain 2.6.24-0.157.rc8.fc9 #1 SMP Thu Jan 17 14:32:47 EST 2008 i686 Build Date: 18 January 2008 05:04:33PM Build ID: xorg-x11-server 1.4.99.1-0.18.20080107.fc9 Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. Module Loader present Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (==) Log file: "/var/log/Xorg.0.log", Time: Tue Jan 22 21:58:17 2008 (==) Using config file: "/etc/X11/xorg.conf" (II) Module "i2c" already built-in (II) Module "ddc" already built-in
Backtrace: 0: X(xf86SigHandler+0x7a) [0x80c6236] 1: [0x131420] 2: /usr/lib/xorg/modules/drivers//nv_drv.so [0x51afe0] 3: X(xf86InitialCheckModeForDriver+0x1d9) [0x80ba756] 4: X(xf86ValidateModes+0xd9a) [0x80bc37d] 5: /usr/lib/xorg/modules/drivers//nv_drv.so [0x51e0d0] 6: X(InitOutput+0x927) [0x80a085c] 7: X(main+0x28d) [0x806b371] 8: /lib/libc.so.6(__libc_start_main+0xe6) [0x3474a6] 9: X(FontFileCompleteXLFD+0x1e1) [0x806a9c1]
Fatal server error: Caught signal 11. Server aborting
giving up.^M xinit: Connection refused (errno 111): unable to connect to X server^M xinit: No such process (errno 3): Server error.
David Mack wrote:
In the latest rawhide update, the nv driver (xorg-x11-drv-nv-2.1.6-6.fc9) kills the X server. The nouveau driver works.
Once X is running, it fails to start a window manager, repeatedly spewing this into the log file:
AUDIT: Wed Jan 23 09:01:06 2008: 2932 X: client 1 rejected from local host ( uid =0 gid=0 pid=2931 ) Auth name: XDM-AUTHORIZATION-1 ID: -1
Doesn't matter whether running xdm or gdm.
Dave
-- fedora-test-list mailing list fedora-test-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
gjohnson5 wrote: | You have to downgrade. I was getting the same thing last night and | downgrading fixed the problem. | When using xorg-x11-drv-nv-2.1.6-6.fc9 I was getting this error starting the | X Server. | Rebuilding synaptics , xorg-x11-server , xorg-x11-server-utils, and | xorg-x11-drivers-nv from src.rpm did not fix the problem
Actually you did *not* fix the problem. You repaired your machine.
Fixing the problem would have taken a bugzilla report so that the maintainer would be aware that there is a problem so he could correct it.
- --
~ David
On Wed, 2008-01-23 at 10:06 -0800, gjohnson5 wrote:
Backtrace: 0: X(xf86SigHandler+0x7a) [0x80c6236] 1: [0x131420] 2: /usr/lib/xorg/modules/drivers//nv_drv.so [0x51afe0] 3: X(xf86InitialCheckModeForDriver+0x1d9) [0x80ba756] 4: X(xf86ValidateModes+0xd9a) [0x80bc37d] 5: /usr/lib/xorg/modules/drivers//nv_drv.so [0x51e0d0] 6: X(InitOutput+0x927) [0x80a085c] 7: X(main+0x28d) [0x806b371] 8: /lib/libc.so.6(__libc_start_main+0xe6) [0x3474a6] 9: X(FontFileCompleteXLFD+0x1e1) [0x806a9c1]
Think I spotted the problem. nv 2.1.6-7.fc9 on its way to koji now, please test.
- ajax
Am Mi 23.Januar 2008 schrieb Adam Jackson:
On Wed, 2008-01-23 at 10:06 -0800, gjohnson5 wrote:
Backtrace: 0: X(xf86SigHandler+0x7a) [0x80c6236] 1: [0x131420] 2: /usr/lib/xorg/modules/drivers//nv_drv.so [0x51afe0] 3: X(xf86InitialCheckModeForDriver+0x1d9) [0x80ba756] 4: X(xf86ValidateModes+0xd9a) [0x80bc37d] 5: /usr/lib/xorg/modules/drivers//nv_drv.so [0x51e0d0] 6: X(InitOutput+0x927) [0x80a085c] 7: X(main+0x28d) [0x806b371] 8: /lib/libc.so.6(__libc_start_main+0xe6) [0x3474a6] 9: X(FontFileCompleteXLFD+0x1e1) [0x806a9c1]
Think I spotted the problem. nv 2.1.6-7.fc9 on its way to koji now, please test.
This version works again (i386). (I've had the same issue in a livecd spin from today's rawhide).
Sebastian
On Wed, 2008-01-23 at 23:10 +0100, Sebastian Vahl wrote:
Am Mi 23.Januar 2008 schrieb Adam Jackson:
Think I spotted the problem. nv 2.1.6-7.fc9 on its way to koji now, please test.
This version works again (i386). (I've had the same issue in a livecd spin from today's rawhide).
Excellent. Thanks for the quick test, and I apologize for the interruption.
- ajax
No need for apologies. Thanks for working with us on this! Yes it is working for me as well
Adam Jackson-2 wrote:
On Wed, 2008-01-23 at 23:10 +0100, Sebastian Vahl wrote:
Am Mi 23.Januar 2008 schrieb Adam Jackson:
Think I spotted the problem. nv 2.1.6-7.fc9 on its way to koji now, please test.
This version works again (i386). (I've had the same issue in a livecd spin from today's rawhide).
Excellent. Thanks for the quick test, and I apologize for the interruption.
- ajax
-- fedora-test-list mailing list fedora-test-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
Yes, the nv driver is back up. However, X is still nonfunctional. I get the stippled gray background and an "X" cursor and nothing else:
AUDIT: Thu Jan 24 08:35:17 2008: 2846 X: client 1 rejected from local host ( uid =0 gid=0 pid=2845 ) Auth name: XDM-AUTHORIZATION-1 ID: -1
Dave
-----Original Message----- From: fedora-test-list-bounces@redhat.com [mailto:fedora-test-list-bounces@redhat.com] On Behalf Of gjohnson5 Sent: Wednesday, January 23, 2008 3:14 PM To: fedora-test-list@redhat.com Subject: Re: nv driver and gnome broken
No need for apologies. Thanks for working with us on this! Yes it is working for me as well
Adam Jackson-2 wrote:
On Wed, 2008-01-23 at 23:10 +0100, Sebastian Vahl wrote:
Am Mi 23.Januar 2008 schrieb Adam Jackson:
Think I spotted the problem. nv 2.1.6-7.fc9 on its way
to koji now,
please test.
This version works again (i386). (I've had the same issue
in a livecd
spin from today's rawhide).
Excellent. Thanks for the quick test, and I apologize for the interruption.
- ajax
-- fedora-test-list mailing list fedora-test-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
-- View this message in context: http://www.nabble.com/nv-driver-and-gnome-broken-tp15047077p15 055169.html Sent from the Fedora Test List mailing list archive at Nabble.com.
-- fedora-test-list mailing list fedora-test-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
On Thu, 2008-01-24 at 08:38 -0800, David Mack wrote:
Yes, the nv driver is back up. However, X is still nonfunctional. I get the stippled gray background and an "X" cursor and nothing else:
AUDIT: Thu Jan 24 08:35:17 2008: 2846 X: client 1 rejected from local host ( uid =0 gid=0 pid=2845 ) Auth name: XDM-AUTHORIZATION-1 ID: -1
What process is 2845 when this happens?
How are you starting X? What's the command line shown by 'ps auwx | grep X' ? If an authority file is listed on the command line, does it contain a valid cookie for that server, and is it the same one as your clients are using?
You can check the contents of an authority file by doing:
% XAUTHORITY=/path/to/whatever xauth list
- ajax
-----Original Message----- From: fedora-test-list-bounces@redhat.com [mailto:fedora-test-list-bounces@redhat.com] On Behalf Of Adam Jackson Sent: Thursday, January 24, 2008 9:20 AM To: For testers of Fedora Core development releases Subject: RE: nv driver and gnome broken
On Thu, 2008-01-24 at 08:38 -0800, David Mack wrote:
Yes, the nv driver is back up. However, X is still
nonfunctional. I get
the stippled gray background and an "X" cursor and nothing else:
AUDIT: Thu Jan 24 08:35:17 2008: 2846 X: client 1 rejected
from local
host ( uid =0 gid=0 pid=2845 ) Auth name: XDM-AUTHORIZATION-1 ID: -1
What process is 2845 when this happens?
xinit
How are you starting X?
startx
How are you starting X? What's the command line shown by 'ps auwx | grep X' ? If an authority file is listed on the command line, does it contain a valid cookie for that server, and is it the same one as your clients are using?
AUDIT: Thu Jan 24 09:15:42 2008: 3198 X: client 1 rejected from local host ( uid=0 gid=0 pid=3197 ) Auth name: XDM-AUTHORIZATION-1 ID: -1
[root@garnet(Linux)]:~ 0# ps auwx | grep 3197 root 3197 0.0 0.0 2980 624 tty1 S+ 09:15 0:00 xinit /etc/X11/xinit/xinitrc -- -auth /root/.serverauth.3181
[root@garnet(Linux)]:~ 0# XAUTHORITY=/root/.serverauth.3181 xauth list garnet.leviatron.com/unix:0 MIT-MAGIC-COOKIE-1 f0f2afbe1232f3233a82486f515ce5c2 garnet.leviatron.com/unix:1 MIT-MAGIC-COOKIE-1 109fb1ed590f10613b71e8f3ac109dfe garnet.leviatron.com/unix:2 MIT-MAGIC-COOKIE-1 109fb1ed590f10613b71e8f3ac109dfe
[root@garnet(Linux)]:~ 0# ps auwx | grep X root 3197 0.0 0.0 2980 624 tty1 S+ 09:15 0:00 xinit /etc/X11/xinit/xinitrc -- -auth /root/.serverauth.3181 root 3198 0.8 0.5 13324 5788 tty7 S<s+ 09:15 0:00 X :0 -auth /root/.serverauth.3181
xinit tries to connect every 2 seconds for a couple of minutes, then gives up with
xinit: Resource temporarily unavailable (errno 11): unable to connect to X server
waiting for X server to shut down
xinit: server error
Dave
Nuking /root/.Xauthority before running startx fixed this problem.
Dave
-----Original Message----- From: fedora-test-list-bounces@redhat.com [mailto:fedora-test-list-bounces@redhat.com] On Behalf Of David Mack Sent: Thursday, January 24, 2008 9:33 AM To: For testers of Fedora Core development releases Subject: RE: nv driver and gnome broken
-----Original Message----- From: fedora-test-list-bounces@redhat.com [mailto:fedora-test-list-bounces@redhat.com] On Behalf Of
Adam Jackson
Sent: Thursday, January 24, 2008 9:20 AM To: For testers of Fedora Core development releases Subject: RE: nv driver and gnome broken
On Thu, 2008-01-24 at 08:38 -0800, David Mack wrote:
Yes, the nv driver is back up. However, X is still
nonfunctional. I get
the stippled gray background and an "X" cursor and nothing else:
AUDIT: Thu Jan 24 08:35:17 2008: 2846 X: client 1 rejected
from local
host ( uid =0 gid=0 pid=2845 ) Auth name: XDM-AUTHORIZATION-1 ID: -1
What process is 2845 when this happens?
xinit
How are you starting X?
startx
How are you starting X? What's the command line shown by 'ps auwx | grep X' ? If an authority file is listed on the command
line, does it
contain a valid cookie for that server, and is it the same
one as your
clients are using?
AUDIT: Thu Jan 24 09:15:42 2008: 3198 X: client 1 rejected from local host ( uid=0 gid=0 pid=3197 ) Auth name: XDM-AUTHORIZATION-1 ID: -1
0# ps auwx | grep 3197 root 3197 0.0 0.0 2980 624 tty1 S+ 09:15 0:00 xinit /etc/X11/xinit/xinitrc -- -auth /root/.serverauth.3181
0# XAUTHORITY=/root/.serverauth.3181 xauth list garnet.leviatron.com/unix:0 MIT-MAGIC-COOKIE-1 f0f2afbe1232f3233a82486f515ce5c2 garnet.leviatron.com/unix:1 MIT-MAGIC-COOKIE-1 109fb1ed590f10613b71e8f3ac109dfe garnet.leviatron.com/unix:2 MIT-MAGIC-COOKIE-1 109fb1ed590f10613b71e8f3ac109dfe
0# ps auwx | grep X root 3197 0.0 0.0 2980 624 tty1 S+ 09:15 0:00 xinit /etc/X11/xinit/xinitrc -- -auth /root/.serverauth.3181 root 3198 0.8 0.5 13324 5788 tty7 S<s+ 09:15 0:00 X :0 -auth /root/.serverauth.3181
xinit tries to connect every 2 seconds for a couple of minutes, then gives up with
xinit: Resource temporarily unavailable (errno 11): unable to connect to X server
waiting for X server to shut down
xinit: server error
Dave
-- fedora-test-list mailing list fedora-test-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
On Wed, 2008-01-23 at 09:09 -0800, David Mack wrote:
In the latest rawhide update, the nv driver (xorg-x11-drv-nv-2.1.6-6.fc9) kills the X server. The nouveau driver works.
Please, everybody, when reporting X failures, test whether X works standalone. Boot in runlevel 3 and just type X at a shell prompt.
Does nv 2.1.6-5 work?
- ajax
On Wed, 2008-01-23 at 09:09 -0800, David Mack wrote:
In the latest rawhide update, the nv driver (xorg-x11-drv-nv-2.1.6-6.fc9) kills the X server. The nouveau driver works.
I'm keen to test rawhide but since I've got a nvidia card I'm a little nervous about the past performance of the nv and nouveau drivers.
How are they working in the current rawhide (with the exception of this bug of course).
Does the nv driver resume after a suspend. Does nouveau support enough 3D to run compiz?
R.
On Thu, 2008-01-24 at 22:10 +1100, Rodd Clarkson wrote:
On Wed, 2008-01-23 at 09:09 -0800, David Mack wrote:
In the latest rawhide update, the nv driver (xorg-x11-drv-nv-2.1.6-6.fc9) kills the X server. The nouveau driver works.
I'm keen to test rawhide but since I've got a nvidia card I'm a little nervous about the past performance of the nv and nouveau drivers.
How are they working in the current rawhide (with the exception of this bug of course).
This bug affected nv only. nouveau was not affected.
Broadly speaking the nouveau driver is widely reported to have equal or better 2D performance than nv, occasionally even on par with nvidia.
Does the nv driver resume after a suspend.
Only if you're lucky. There are deep issues with the nv resume path; essentially VBE doesn't work reliably after suspend, so the only way to reinit the card is to execute the nvidia equivalent of what ATI calls ATOMBIOS. The nouveau driver contains a start towards implementing this, but it's not done. I don't know if anyone is owning that feature, it would certainly be a welcome area for contribution.
Does nouveau support enough 3D to run compiz?
Maybe. Certainly not in Fedora as shipped, we're not shipping the 3D part of the driver yet due to immaturity (of the driver, not of the developers).
- ajax
On Thu, 2008-01-24 at 09:47 -0500, Adam Jackson wrote:
On Thu, 2008-01-24 at 22:10 +1100, Rodd Clarkson wrote:
Does the nv driver resume after a suspend.
Only if you're lucky. There are deep issues with the nv resume path; essentially VBE doesn't work reliably after suspend, so the only way to reinit the card is to execute the nvidia equivalent of what ATI calls ATOMBIOS. The nouveau driver contains a start towards implementing this, but it's not done. I don't know if anyone is owning that feature, it would certainly be a welcome area for contribution.
I'm curious about this (and finally got around to asking).
The F8 vesa driver resumes properly after S2R and S2D with my nVidia Quadro NVS 140 M with the s3-mode quirk enabled. The latest F8 nv driver does not. Can you explain what's different?
The vesa driver won't turn the backlight off with this card, though.
Thanks.
-
Adam Jackson -
David -
David Mack -
gjohnson5 -
Matthew Saltzman -
Rodd Clarkson -
Sebastian Vahl