Re: Nvidia package installation problems
On Wed, 25 Dec 2013 21:09:35 + (UTC) Jarth Berilcosm wrote: Hello Jarth, >Yeah, i lost focus due to frustrations and wrote down an incorrect >pathname. Should have had some tea much sooner ;-) Tea? How very British of you. :-D >What i've figured so far is nomodeset is mostly required by older >hardware, this is an "NVIDIA Corporation G86M [Quadro FX 360M]" In which case, it may not have much effect with my GFX card, being a 8?00 >I wrote a guide on my experience. Yes, I read it. Quite interesting. >Right now, i hope I have not been deluding myself but at least it's a >stable delusion so far :-) :-) -- Regards _ / ) "The blindingly obvious is / _)radnever immediately apparent" I'll tell you something, I think that you should know Rich Kids - Rich Kids signature.asc Description: PGP signature
Re: Nvidia package installation problems
On Wed, 25 Dec 2013 08:46:21 +, Brad Rogers wrote: > On Wed, 25 Dec 2013 00:17:17 + (UTC) > Jarth Berilcosm wrote: > > Hello Jarth, > >>Honestly, i don't know what's going with Debian lately. Wheezy looked >>promising and has been quite a dissapointment when it comes to package >>quality. > > It's strange, isn't it; I've not had any problems at all. Admittedly, > I'm not using Wheezy (I am always on 'testing'), but even so maybe you'd > expect an issue or two. > >>Golden hint include nomodeset in /etc/default/grub on the line >>GRUB_CMD_LINE_DEFAULT="quiet splash nomodeset" > > No "nomodeset" here, and everything seems fine. Of course, I could set > it and see what changes/improvements it makes. > >>Make sure in /etc/modprobe.conf.d/nvidia.. there is a line stating >>blacklist nouveau > > Surely you mean '/etc/modprobe.d/' not '/etc/modprobe.conf.d/'? The > latter doesn't exist here. Hi Brad, Yeah, i lost focus due to frustrations and wrote down an incorrect pathname. Should have had some tea much sooner ;-) What i've figured so far is nomodeset is mostly required by older hardware, this is an "NVIDIA Corporation G86M [Quadro FX 360M]" graphics card which is quite old. Almost unsupported. My system logs were littered with segfault errors on gnome-shell and quite a few other programs. Since late last night there are no segfault errors anymore, it means I've done something right ;-) My system was left running for hours today with a few segfault sensitive applications running, no errors whatsoever, hurray. I wrote a guide on my experience. http://www.oxitech.info/helpdesk/nvidia.html Right now, i hope I have not been deluding myself but at least it's a stable delusion so far :-) -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l9fhif$cpp$1...@ger.gmane.org
Re: Nvidia package installation problems
On Tue, 24 Dec 2013 10:02:42 -0500 Jon N wrote: Hello Jon, >it got left over. The nvidia driver offered to register with it for >automatic updates. So I still get the main reason for changing. Good to hear your system is working properly again. A good catch on the Xorg nouveau call. I'd have not thought of it. >Thanks for all your help, You're welcome. -- Regards _ / ) "The blindingly obvious is / _)radnever immediately apparent" Never much liked playing there anyway Banned From The Roxy - Crass signature.asc Description: PGP signature
Re: Nvidia package installation problems
On Wed, 25 Dec 2013 00:17:17 + (UTC) Jarth Berilcosm wrote: Hello Jarth, >Honestly, i don't know what's going with Debian lately. Wheezy looked >promising and has been quite a dissapointment when it comes to package >quality. It's strange, isn't it; I've not had any problems at all. Admittedly, I'm not using Wheezy (I am always on 'testing'), but even so maybe you'd expect an issue or two. >Golden hint include nomodeset in /etc/default/grub on the line >GRUB_CMD_LINE_DEFAULT="quiet splash nomodeset" No "nomodeset" here, and everything seems fine. Of course, I could set it and see what changes/improvements it makes. >Make sure in /etc/modprobe.conf.d/nvidia.. there is a line stating >blacklist nouveau Surely you mean '/etc/modprobe.d/' not '/etc/modprobe.conf.d/'? The latter doesn't exist here. -- Regards _ / ) "The blindingly obvious is / _)radnever immediately apparent" I'll tell you something, I think that you should know Rich Kids - Rich Kids signature.asc Description: PGP signature
Re: Nvidia package installation problems
Hi, I've ran into the same issues on an old NVidia Quadro FX 360M. Honestly, i don't know what's going with Debian lately. Wheezy looked promising and has been quite a dissapointment when it comes to package quality. Golden hint include nomodeset in /etc/default/grub on the line GRUB_CMD_LINE_DEFAULT="quiet splash nomodeset" This will greatly improve the speed and stability of your system. Make sure in /etc/modprobe.conf.d/nvidia.. there is a line stating blacklist nouveau What i've done is to download the latest nvidia driver ( long life version ) and run the installer, make sure to remove ANY debian nvidia packages. So far i had to ditch gdm as login manager as it crashed and crashed, using lightdm now. J. On Tue, 24 Dec 2013 10:02:42 -0500, Jon N wrote: > On Mon, Dec 23, 2013 at 11:37 AM, Brad Rogers wrote: > > <---snip---> > >>>have any trouble doing that. But I don't want to have a package from >>>repository still installed at the same time as the nvidia download. >> >> There are problems with that as you can imagine. Debian packages can >> get confused by a previous nvidia installation. I don't know whether >> the same is true the other way round, but better safe than sorry. >> >> I include here a list of all nvidia specific packages I have installed, >> for the sake of comparison with your own list: >> >> glx-alternative-nvidia_0.4.1 libgl1-nvidia-glx_319.76-1 >> libnvidia-ml1_319.76-1 libvdpau1_0.7-1 libxnvctrl0_319.72-1 >> libxnvctrl0_319.72-1 libxnvctrl0_319.72-1 nvidia-driver_319.76-1 >> nvidia-kernel-dkms_319.76-1 nvidia-settings_319.72-1 >> nvidia-alternative_319.76-1 nvidia-xconfig_319.72-1 >> nvidia-installer-cleanup_20131102+1 * >> nvidia-kernel-common_20131102+1 nvidia-kernel-source_319.76-1 >> nvidia-support_20131102+1 xserver-xorg-video-nouveau_1.0.10-1 >> xserver-xorg-video-nvidia_319.76-1 >> >> * handy for ensuring the nvidia downloaded module stuff doesn't >> conflict with the Debian packages. > > Before I gave up I ran 'dpkg-reconfigure with each package name above > hoping that fix something, or at least generate an error message that > would give some clue. But, no luck. So I uninstalled it all and > rebooted. My desktop came up fine with (I think) the vesa driver. > > But, I often watch HD shows through MythTV, which didn't work well at > all with the vesa driver, nor with the nouveau driver. So, back to the > downloaded nvidia driver. Which had the exact same problem as the > nvidia package loaded through the repository, X would not load. Ouch, I > hadn't expected that. > > Poking around again I noticed a line in Xorg.0.log: > > "Kernel command line: BOOT_IMAGE+/vmlinux-3.11.-2-686-pae > root=/dev/mapper/MyVG-root_LV ro nomodeset nouveau.modeset=0" > > I did notice that earlier, but wasn't worried about nouveau being in > there because it's blacklisted. But with these continuing problems I > guess it's worth looking into. I edited it out in /etc/defaults/grub > and ran 'update-grub'. That was it! I am really puzzled how it got > there. It seem strange to me that uninstalling the downloaded nvidia > drivers (which is the first thing I did when this all started) would add > that. None of the 'grub.*' files in /etc/defaults was newer than 2012, > but it was in there, suggesting it was always in the command line. So > why was it a problem now??? As usual, I figure I must have missed > something, or done something dumb. But, at least it works! > > Oh, and one of my reasons for changing in the first place was to have > the driver automatically configured for each new kernel update that was > installed. Apparently I did not have 'dkms' installed before, but it > got left over. The nvidia driver offered to register with it for > automatic updates. So I still get the main reason for changing. > > Thanks for all your help, > Jon -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l9d86d$mar$1...@ger.gmane.org
Re: Nvidia package installation problems
On Mon, Dec 23, 2013 at 11:37 AM, Brad Rogers wrote: <---snip---> >>have any trouble doing that. But I don't want to have a package from >>repository still installed at the same time as the nvidia download. > > There are problems with that as you can imagine. Debian packages can > get confused by a previous nvidia installation. I don't know whether > the same is true the other way round, but better safe than sorry. > > I include here a list of all nvidia specific packages I have installed, > for the sake of comparison with your own list: > > glx-alternative-nvidia_0.4.1 > libgl1-nvidia-glx_319.76-1 > libnvidia-ml1_319.76-1 > libvdpau1_0.7-1 > libxnvctrl0_319.72-1 > libxnvctrl0_319.72-1 > libxnvctrl0_319.72-1 > nvidia-driver_319.76-1 > nvidia-kernel-dkms_319.76-1 > nvidia-settings_319.72-1 > nvidia-alternative_319.76-1 > nvidia-xconfig_319.72-1 > nvidia-installer-cleanup_20131102+1 * > nvidia-kernel-common_20131102+1 > nvidia-kernel-source_319.76-1 > nvidia-support_20131102+1 > xserver-xorg-video-nouveau_1.0.10-1 > xserver-xorg-video-nvidia_319.76-1 > > * handy for ensuring the nvidia downloaded module stuff doesn't > conflict with the Debian packages. Before I gave up I ran 'dpkg-reconfigure with each package name above hoping that fix something, or at least generate an error message that would give some clue. But, no luck. So I uninstalled it all and rebooted. My desktop came up fine with (I think) the vesa driver. But, I often watch HD shows through MythTV, which didn't work well at all with the vesa driver, nor with the nouveau driver. So, back to the downloaded nvidia driver. Which had the exact same problem as the nvidia package loaded through the repository, X would not load. Ouch, I hadn't expected that. Poking around again I noticed a line in Xorg.0.log: "Kernel command line: BOOT_IMAGE+/vmlinux-3.11.-2-686-pae root=/dev/mapper/MyVG-root_LV ro nomodeset nouveau.modeset=0" I did notice that earlier, but wasn't worried about nouveau being in there because it's blacklisted. But with these continuing problems I guess it's worth looking into. I edited it out in /etc/defaults/grub and ran 'update-grub'. That was it! I am really puzzled how it got there. It seem strange to me that uninstalling the downloaded nvidia drivers (which is the first thing I did when this all started) would add that. None of the 'grub.*' files in /etc/defaults was newer than 2012, but it was in there, suggesting it was always in the command line. So why was it a problem now??? As usual, I figure I must have missed something, or done something dumb. But, at least it works! Oh, and one of my reasons for changing in the first place was to have the driver automatically configured for each new kernel update that was installed. Apparently I did not have 'dkms' installed before, but it got left over. The nvidia driver offered to register with it for automatic updates. So I still get the main reason for changing. Thanks for all your help, Jon -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CANcvmg2Mhrhw0EKbz6yDCd_Db0gdTcWT==nqpyv5pcu_c5o...@mail.gmail.com
Re: Nvidia package installation problems
On Mon, 23 Dec 2013 11:05:51 -0500 Jon N wrote: Hello Jon, >The card is a 8600 GT, which does use the latest drivers. in fact, I >uninstalled 319.76 that I had downloaded from nvida to install 319.76 Okay, now I know what GPU you've got. And yes, you're right of course, it uses the 319 drivers. >from repository. Duh! I'm starting to think I'm going to have to >back out and go back to nvidia's downloaded stuff. Hopefully I won't That would be a pity, but if you have to. >have any trouble doing that. But I don't want to have a package from >repository still installed at the same time as the nvidia download. There are problems with that as you can imagine. Debian packages can get confused by a previous nvidia installation. I don't know whether the same is true the other way round, but better safe than sorry. I include here a list of all nvidia specific packages I have installed, for the sake of comparison with your own list: glx-alternative-nvidia_0.4.1 libgl1-nvidia-glx_319.76-1 libnvidia-ml1_319.76-1 libvdpau1_0.7-1 libxnvctrl0_319.72-1 libxnvctrl0_319.72-1 libxnvctrl0_319.72-1 nvidia-driver_319.76-1 nvidia-kernel-dkms_319.76-1 nvidia-settings_319.72-1 nvidia-alternative_319.76-1 nvidia-xconfig_319.72-1 nvidia-installer-cleanup_20131102+1 * nvidia-kernel-common_20131102+1 nvidia-kernel-source_319.76-1 nvidia-support_20131102+1 xserver-xorg-video-nouveau_1.0.10-1 xserver-xorg-video-nvidia_319.76-1 * handy for ensuring the nvidia downloaded module stuff doesn't conflict with the Debian packages. -- Regards _ / ) "The blindingly obvious is / _)radnever immediately apparent" We're going to hell anyway, let's travel first class Saturday Night - Kaiser Chiefs signature.asc Description: PGP signature
Re: Nvidia package installation problems
On Mon, Dec 23, 2013 at 10:20 AM, Brad Rogers wrote: > On Mon, 23 Dec 2013 09:35:52 -0500 > Jon N wrote: > > Hello Jon, > >>I did check that, but it looks OK (correct version installed). >> >>Thanks, > > No problem. I've just re-read your original message, and picked up on > some stuff I missed previously. Specifically, stuff about version > number mismatches (319 vs 304). Looks like you should be using > nvidia-legacy drivers. Search for nvidia-legacy-304. Sounds as though > your GFX card is not supported by latest (319) drivers. You'll need to > check what your GFX card is, and see which nvidia driver package > supports it. Either 304 or 173. The card is a 8600 GT, which does use the latest drivers. in fact, I uninstalled 319.76 that I had downloaded from nvida to install 319.76 from repository. Duh! I'm starting to think I'm going to have to back out and go back to nvidia's downloaded stuff. Hopefully I won't have any trouble doing that. But I don't want to have a package from repository still installed at the same time as the nvidia download. There could be problems with the repository version updating and over writing something form the nvidia download. Thanks, Jon -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CANcvmg3A+HR5GUkF+j+4Ab81hud9BS=50d6ku654zp7u2np...@mail.gmail.com
Re: Nvidia package installation problems
On Mon, 23 Dec 2013 09:35:52 -0500 Jon N wrote: Hello Jon, >I did check that, but it looks OK (correct version installed). > >Thanks, No problem. I've just re-read your original message, and picked up on some stuff I missed previously. Specifically, stuff about version number mismatches (319 vs 304). Looks like you should be using nvidia-legacy drivers. Search for nvidia-legacy-304. Sounds as though your GFX card is not supported by latest (319) drivers. You'll need to check what your GFX card is, and see which nvidia driver package supports it. Either 304 or 173. -- Regards _ / ) "The blindingly obvious is / _)radnever immediately apparent" Is she really going out with him? New Rose - The Damned signature.asc Description: PGP signature
Re: Nvidia package installation problems
On Mon, Dec 23, 2013 at 6:20 AM, Brad Rogers wrote: > On Sun, 22 Dec 2013 19:15:25 -0500 > Jon N wrote: > > Hello Jon, > >>I love the idea of not having to re-installed the downloaded Nvidia >>drivers every time the kernel updates (plus it seems to keep breaking) >>so I would like to get this to work. Any suggestions? > > As Andrei has said, make sure you have the relevant kernel headers > package installed. That's a show stopper. I did check that, but it looks OK (correct version installed). Thanks, Jon -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CANcvmg2fX6KmqYiTuv4_fTv_7Z7RzvQQ2mtYEEsvS48vGMx=o...@mail.gmail.com
Re: Nvidia package installation problems
On Mon, Dec 23, 2013 at 4:21 AM, Andrei POPESCU wrote: > On Du, 22 dec 13, 19:15:25, Jon N wrote: >> >> "NWRM: API mismatch: the client has the version 304.108, but this >> kernel module has the version 319.76. Please make sure that this >> kernel module and all NVIDIA driver components have the same version." >> >> That sounds simple enough, but I have searched all packages I can >> think make be related in Aptitude and can not find anything that >> mentions version 304.108. Either they all say 319.76, or a version >> like '20131102+1 (that's the 'nvidia-kernel-common among others). > > You have a leftover kernel module. See if nvidia-kernel-dkms can cleanup > and install a correct module for you. Make sure you have the matching > kernel headers for your kernel installed. > > Kind regards, > Andrei I ran 'dpkg-reconfigure dkms' (which exited after only a few seconds with no messages) and 'dpkg-reconfigure nvidia-kernel-dkms" which seemed to do a number of things. But basically it seemed to uninstall the nvidia module version.319.76 from kernel 3.11-2-686-pae (it said it was active on this kernel) and then load as new the exact same module. It even had an error message: "Error! Module version 319.76 for nvidia-current.ko is not newer than what is already found in kernel 3.11-2-686-pae (319.76). You may override by specifying --force" At the end it says: "DKMS: install completed" But, it did not fix the problem. I tried checking the kernel headers using Aptitude and as far as I can tell I have the following packages for linux-headers installed. These match the 2 kernels I have installed. BTW, I tried booting the older kernel and got the same results. linux-headers-3.11-2-686-pae linux-headers-3.11-2-common linux-headers-686-pae linux-headers-3.10-3-686-pae linux-headers-3.10-3-common That got me thinking again about versions. I searched for "304.*" (the 'mystery' version number that showed in some error messages) and found a whole bunch of 'libcuda' files, all version 304 or older. And a couple of symlinks setting the newest of those as 'libcuda.so' (or something similar). I deleted them all as these must have been leftovers from the downloaded nvidia binary drivers from nvidia. I didn't see that error message on the next boot, but still no X. Installing the latest libcuda (319.76) from Aptitude did help either. The only error message I can find now is from Xorg.0.log: "(EE) Failed to load /usr/lib/xorg/modules/drivers/nvidia_drv.so: /usr/lib/xorg/modules/drivers/nvidia_drv.so: undefined symbol: WindowTable" Two lines below that is: "(EE) Failed to load module "nvidia" (loader failed, 7)" I don't know if the number '7' is significant. I did start a search for the "... undefined symbol: WindowTable" message, but so far nothing really stood out as a likely cause. Thanks, Jon -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CANcvmg1Ev2YHt1y=wUr1gYhT5M+4x5ihC7sLKf=ctaurzmn...@mail.gmail.com
Re: Nvidia package installation problems
On Sun, 22 Dec 2013 19:15:25 -0500 Jon N wrote: Hello Jon, >I love the idea of not having to re-installed the downloaded Nvidia >drivers every time the kernel updates (plus it seems to keep breaking) >so I would like to get this to work. Any suggestions? As Andrei has said, make sure you have the relevant kernel headers package installed. That's a show stopper. -- Regards _ / ) "The blindingly obvious is / _)radnever immediately apparent" Just stop and take a second U & Ur Hand - P!nk signature.asc Description: PGP signature
Re: Nvidia package installation problems
On Du, 22 dec 13, 19:15:25, Jon N wrote: > > "NWRM: API mismatch: the client has the version 304.108, but this > kernel module has the version 319.76. Please make sure that this > kernel module and all NVIDIA driver components have the same version." > > That sounds simple enough, but I have searched all packages I can > think make be related in Aptitude and can not find anything that > mentions version 304.108. Either they all say 319.76, or a version > like '20131102+1 (that's the 'nvidia-kernel-common among others). You have a leftover kernel module. See if nvidia-kernel-dkms can cleanup and install a correct module for you. Make sure you have the matching kernel headers for your kernel installed. Kind regards, Andrei -- http://wiki.debian.org/FAQsFromDebianUser Offtopic discussions among Debian users and developers: http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic http://nuvreauspam.ro/gpg-transition.txt signature.asc Description: Digital signature
Re: Nvidia package installation problems
On Mon, Dec 23, 2013 at 12:41 AM, Jon N wrote: > On Sun, Dec 22, 2013 at 7:15 PM, Jon N wrote: <---snip---> > It's getting late so I'm going to have to look at this again in the > morning. I think I should be able to make a new blacklist file for > nouveau (hopefully you can't have it blacklisted too many times) and > copy the /etc/X11/xorg.conf to /root/xorg.conf.new to reference just > the nvidia driver. Anyone know where I can get a list of the order in > which, and locations of, config files for X? I would have sworn > /etc/X11 was first, but it's not the first time I would be wrong > either :-). Ok, still up. I discovered that I did not have an xorg.conf in /etc/X11 after all. So I put one there. Now nvidia is the only driver that seems to be loading in the Xorg.0.log file. But, if failed to load: "/usr/lib/xorg/modules/drivers/nvidia_drv.so: undefined symbol: WindowTable". So, I can look that up tomorrow, but if anyone is familiar with this error and would be interested in passing it along I will be very happy :-). Thanks, Jon P.S. I just did a quick google search on "nvidia_drv.so: undefined symbol: WindowTable" and really didn't see anything that looked useful, or even very recent. But it was only a quick check, I really have to get some sleep :-) zz. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/cancvmg2yv1qdqr-fsznra6ecqh4nl0t7ryyluxx8fgsyaaz...@mail.gmail.com
Re: Nvidia package installation problems
On Sun, Dec 22, 2013 at 7:15 PM, Jon N wrote: > I have been getting pretty good with breaking my system lately. Today > I got it in mind to stop downloading the Nvidia binary driver and > installing it myself and switch to using the version available as a > Debian package. I first uninstalled the downloaded version and > rebooted to the normal desktop OK. I tried using lspci to see just > what driver it was currently using but either it didn't say or I > didn't understand the output. But the vesa driver was installed and > it seemed unlikely it was using the driver I just uninstalled, so i > went forward thinking I had the old drivers removed. I'm not sure if > that was necessary, but cleaning out the old before installing the new > seemed to be a good idea. > > I wasn't sure exactly what to select (in Synaptic) for the new > drivers, and unfortunately don't remember exactly everything that may > have been selected as a dependency. But I know I at least the > following were installed: > xserver-xorg-video-nvidia' > nvidia-kernel-dkms > dkms > > And I'm pretty sure these were installed > nvidia-vdpau-driver > nvidia-support > nvidia-kernel-commen > nvidia-installer-cleanup > nvidia-alternative > nvidia-kernel-686-pae (it does match my kernel, and making the kernel > module ends without error) > > There are also other packages related to glx, and probably more I > don't remember. > > But, I can't get the xserver to start. I can log into virtual > terminals OK. At the end of the boot sequence I get the following > message: > > "NWRM: API mismatch: the client has the version 304.108, but this > kernel module has the version 319.76. Please make sure that this > kernel module and all NVIDIA driver components have the same version." > > That sounds simple enough, but I have searched all packages I can > think make be related in Aptitude and can not find anything that > mentions version 304.108. Either they all say 319.76, or a version > like '20131102+1 (that's the 'nvidia-kernel-common among others). > > I just noticed that 'glx-alternative-nvidia has a version of '0.4.1', > but even if that's the problem I can't uninstall that without it > taking most of the packages listed above. > > I checked my 'sources.list' and all the '...debian.org' lines are set > to jessie. I'm not sure if any of the packages related to my problem > could have come from other repositories, but I do have the following: > > www.deb-multimedia.org/ testing non-free main > packages.mate-desktop.org/repo/debian/ jessie main > ftp.mowgli.ch/pub/debian/ wheezy unofficial > Also apt.last.fm, download.skype.com and dowload.webmin.com, but these > seem pretty unlikely sources of wrong version Nvidia drivers to me > (then again, if I knew what I was doing I woudn't be writing to you > :-)). > > I love the idea of not having to re-installed the downloaded Nvidia > drivers every time the kernel updates (plus it seems to keep breaking) > so I would like to get this to work. Any suggestions? > > Thanks, > Jon Hmmm, I've been having a little luck trying to track down the source of my own problem. I checked the Xorg.0.log and noticed it seemed to be trying to load about every dirver that made sense at all for my Nvidia 8600 GT card. They are (not necessarily in this order): nv, nvidia, fbdev, vesa and nouveau. And I know that I had nouveau blacklisted before. Heck, I had to, the downloaded nvidia drivers wouldn't work if it wasn't. I did notice that the contents of /etc/modprode.d had changed with 2 new symlinks (nvidia-blacklists-nouveau.conf and nvidia.conf) since trying to install the nvidia driver package that both point to files in /etc/alternatives. I'm not sure if they are blacklisting it or not at this point. But, I'm not sure if that is the problem, since nouveau doesn't load either. I was also surprised that /etc/X11/xorg.conf apparently isn't being looked at, it's using /root/xorg.conf.new, and /usr/share/X11/xorg.conf.d. /root/xorg.conf.new has sections for each of the drivers mentioned above, so I guess that's why it's trying to load them. But I renamed it and the behavior remains. Apparently, that lack of an xorg.conf file of some type just causes X to try a bunch of stuff. But nvidia isn't one of them (I'll take a guess that because it's closed source it's not included in some sort of list of default drivers to try). It's getting late so I'm going to have to look at this again in the morning. I think I should be able to make a new blacklist file for nouveau (hopefully you can't have it blacklisted too many times) and copy the /etc/X11/xorg.conf to /root/xorg.conf.new to reference just the nvidia driver. Anyone know where I can get a list of the order in which, and locations of, config files for X? I would have sworn /etc/X11 was first, but it's not the first time I would be wrong either :-). Thanks, Jon -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Tro
Nvidia package installation problems
I have been getting pretty good with breaking my system lately. Today I got it in mind to stop downloading the Nvidia binary driver and installing it myself and switch to using the version available as a Debian package. I first uninstalled the downloaded version and rebooted to the normal desktop OK. I tried using lspci to see just what driver it was currently using but either it didn't say or I didn't understand the output. But the vesa driver was installed and it seemed unlikely it was using the driver I just uninstalled, so i went forward thinking I had the old drivers removed. I'm not sure if that was necessary, but cleaning out the old before installing the new seemed to be a good idea. I wasn't sure exactly what to select (in Synaptic) for the new drivers, and unfortunately don't remember exactly everything that may have been selected as a dependency. But I know I at least the following were installed: xserver-xorg-video-nvidia' nvidia-kernel-dkms dkms And I'm pretty sure these were installed nvidia-vdpau-driver nvidia-support nvidia-kernel-commen nvidia-installer-cleanup nvidia-alternative nvidia-kernel-686-pae (it does match my kernel, and making the kernel module ends without error) There are also other packages related to glx, and probably more I don't remember. But, I can't get the xserver to start. I can log into virtual terminals OK. At the end of the boot sequence I get the following message: "NWRM: API mismatch: the client has the version 304.108, but this kernel module has the version 319.76. Please make sure that this kernel module and all NVIDIA driver components have the same version." That sounds simple enough, but I have searched all packages I can think make be related in Aptitude and can not find anything that mentions version 304.108. Either they all say 319.76, or a version like '20131102+1 (that's the 'nvidia-kernel-common among others). I just noticed that 'glx-alternative-nvidia has a version of '0.4.1', but even if that's the problem I can't uninstall that without it taking most of the packages listed above. I checked my 'sources.list' and all the '...debian.org' lines are set to jessie. I'm not sure if any of the packages related to my problem could have come from other repositories, but I do have the following: www.deb-multimedia.org/ testing non-free main packages.mate-desktop.org/repo/debian/ jessie main ftp.mowgli.ch/pub/debian/ wheezy unofficial Also apt.last.fm, download.skype.com and dowload.webmin.com, but these seem pretty unlikely sources of wrong version Nvidia drivers to me (then again, if I knew what I was doing I woudn't be writing to you :-)). I love the idea of not having to re-installed the downloaded Nvidia drivers every time the kernel updates (plus it seems to keep breaking) so I would like to get this to work. Any suggestions? Thanks, Jon -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/cancvmg0wsxfactszc3zbabbfaw6xenmw83case2fjifr6fz...@mail.gmail.com
