[gentoo-user] New monitor problems. < sighs >
Hi again, For those mot up to date. I built a new rig that has a Nvidia GT220 card in it. I bought a brand new monitor this morning, a LG W2253, and it worked one time. I had to reboot to move some things around and when I rebooted, the GUI doesn't come up. The BIOS screen shows up and I can see the services start up as well but when it switches to vt7, it just has a little blinking cursor at the top. This is a grep of the log: r...@smoker / # cat /media/disk/files/Xorg.0.log | grep EE [ 2081.047] Current Operating System: Linux fireball 2.6.36-gentoo-r4 #15 SMP PREEMPT Mon Dec 13 01:15:13 CST 2010 x86_64 (WW) warning, (EE) error, (NI) not implemented, (??) unknown. [ 2081.048] (II) Loading extension MIT-SCREEN-SAVER [ 2081.057] (EE) Failed to load module "dri" (module does not exist, 0) [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, 0) [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: Function not [ 2082.101] (EE) NVIDIA(0): implemented [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not exist, 0) [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA X driver not found) r...@smoker / # I have tried X -configure, Xorg -configure and nvidia-xconfig and none of them do any better. I also did a emerge -e all the way up to nvidia-driver as well then reconfiged everything. Still no joy. I'm not going to mention all the reinstalls of the nvidia drivers and different versions I tried. I attached my xorg.conf file too. I can send the whole xorg.log file if needed. Oh, I always get the dri load error. I also used eselect to switch to nvidia's opengl to, many times. Could it be something weird with the card maybe? This is monitor two that has issues with it. I have a ATI card handy. If someone knows of a idiot proof guide I could follow. You know, one that you used and are confident it will work if followed. I don't have much hair left right now. :-( Just for fun, I got a Mandriva install CD and booted it up and even installed it on a spare drive. After the install, it boots just like Gentoo does. I can see the BIOS, the services and such but when it switches to vt7, just a blinking cursor. This new rig is turning into a nightmare. No wonder I waited almost 8 years. lol Thoughts? Ideas? Suggestions, other than jumping in the creek. Dale :-) :-) # nvidia-xconfig: X configuration file generated by nvidia-xconfig # nvidia-xconfig: version 260.19.26 (buildmeis...@swio-display-x86-rhel47-07.nvidia.com) Mon Nov 29 01:13:31 PST 2010 Section "ServerLayout" Identifier "Layout0" Screen 0 "Screen0" 0 0 InputDevice"Keyboard0" "CoreKeyboard" InputDevice"Mouse0" "CorePointer" EndSection Section "Files" EndSection Section "InputDevice" # generated from data in "/etc/conf.d/gpm" Identifier "Mouse0" Driver "mouse" Option "Protocol" Option "Device" "/dev/input/mice" Option "Emulate3Buttons" "no" Option "ZAxisMapping" "4 5" EndSection Section "InputDevice" # generated from default Identifier "Keyboard0" Driver "kbd" EndSection Section "Monitor" Identifier "Monitor0" VendorName "Unknown" ModelName "Unknown" HorizSync 28.0 - 33.0 VertRefresh 43.0 - 72.0 Option "DPMS" EndSection Section "Device" Identifier "Device0" Driver "nvidia" VendorName "NVIDIA Corporation" EndSection Section "Screen" Identifier "Screen0" Device "Device0" Monitor"Monitor0" DefaultDepth24 SubSection "Display" Depth 24 EndSubSection EndSection
[gentoo-user] Strange boot problem - any ideas?
Hi, I have to power down one of my machines each day. Booting it the other day fails from time to time. On this (and all my machines) /usr is on an ext4 file system by its own. It looks as if mounting /usr fails sometimes (silently). The first unusual message is that it cannot find the file /usr/sbin/acpid . After that, most other actions fail as well. Once the machine has failed to boot it will fail every time afterwards unless I do the following: I boot by a rescue CD, change root and re-emerge sys-power/acpid (this package contains /usr/sbin/acpid). This single action has helped without any problem each time I had to try it. Of course, I've done many checks on the drive hosting the /usr partition - no errors at all (the drive is only a few months old and a good one (enterprise edition)). So, I suspect openrc. Might it be that this is a timing problem and openrc doesn't check if the partition is mounted? Booting has worked just flawlessly on that machine for more than a year. It looks as if a recent version of openrc (currently 0.6.8) has generated this problem. Thanks for any hint, Helmut.
Re: [gentoo-user] New monitor problems. < sighs >
On 12/15/10 09:40:48, Dale wrote: > Hi again, > > For those mot up to date. I built a new rig that has a Nvidia GT220 > card in it. I bought a brand new monitor this morning, a LG W2253, > and > it worked one time. I had to reboot to move some things around and > when > I rebooted, the GUI doesn't come up. The BIOS screen shows up and I > can > see the services start up as well but when it switches to vt7, it > just > > has a little blinking cursor at the top. Have you changed the kernel? The nvidia-drivers (as well as the ati- drivers) have to be re-emerged after (nearly) any change to the kernel. Just re-emerge it, unload and reload the corresponding kernel module and try again to startx. Helmut.
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, 15 Dec 2010 02:40:48 -0600, Dale wrote: > [ 2081.057] (EE) Failed to load module "dri" (module does not exist, 0) > [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, > 0) [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory > identifier: Function not > [ 2082.101] (EE) NVIDIA(0): implemented > [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not > exist, 0) > [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA > X driver not found) > r...@smoker / # You haven't loaded the glx module, try this in xorg.conf Section "Module" Load "glx" Disable "dri" Disable "dri2" EndSection Also, what does "eselect opengl list" show? -- Neil Bothwick Nostalgia isn't what it used to be. signature.asc Description: PGP signature
Re: [gentoo-user] New monitor problems. < sighs >
Helmut Jarausch wrote: On 12/15/10 09:40:48, Dale wrote: Hi again, For those mot up to date. I built a new rig that has a Nvidia GT220 card in it. I bought a brand new monitor this morning, a LG W2253, and it worked one time. I had to reboot to move some things around and when I rebooted, the GUI doesn't come up. The BIOS screen shows up and I can see the services start up as well but when it switches to vt7, it just has a little blinking cursor at the top. Have you changed the kernel? The nvidia-drivers (as well as the ati- drivers) have to be re-emerged after (nearly) any change to the kernel. Just re-emerge it, unload and reload the corresponding kernel module and try again to startx. Helmut. Just to make sure I understand you correctly. You mean: cd /usr/src/linux make all && make modules_install then cp the kernel to /boot ? That I didn't try. I knew if I updated the kernel I had to re-emerge the nvidia-drivers but I didn't know it was the other way around too. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 09:55:37 Helmut Jarausch wrote: > On 12/15/10 09:40:48, Dale wrote: > > Hi again, > > > > For those mot up to date. I built a new rig that has a Nvidia GT220 > > card in it. I bought a brand new monitor this morning, a LG W2253, > > and > > it worked one time. I had to reboot to move some things around and > > when > > I rebooted, the GUI doesn't come up. The BIOS screen shows up and I > > can > > see the services start up as well but when it switches to vt7, it > > just > > > > has a little blinking cursor at the top. > > Have you changed the kernel? The nvidia-drivers (as well as the ati- > drivers) have to be re-emerged after (nearly) any change to the kernel. > Just re-emerge it, unload and reload the corresponding kernel module > and try again to startx. > > Helmut. Also, when emerging the "nvidia-drivers", try making sure that "/usr/src/linux" points to the actual linux-sources you are booting from. (Not sure if necessary, but I always have that symlink to the current kernel) -- Joost
Re: [gentoo-user] New monitor problems. < sighs >
2010/12/15 Neil Bothwick : > On Wed, 15 Dec 2010 02:40:48 -0600, Dale wrote: > >> [ 2081.057] (EE) Failed to load module "dri" (module does not exist, 0) >> [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, >> 0) [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory >> identifier: Function not >> [ 2082.101] (EE) NVIDIA(0): implemented >> [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not >> exist, 0) >> [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA >> X driver not found) >> r...@smoker / # > > You haven't loaded the glx module, try this in xorg.conf > > Section "Module" > Load "glx" > Disable "dri" > Disable "dri2" > EndSection > > Also, what does "eselect opengl list" show? > > > -- > Neil Bothwick > > Nostalgia isn't what it used to be. > Hi to everybody! This is my first post to this list. I think this guide help maybe? http://www.gentoo.org/doc/en/nvidia-guide.xml Vartsu
Re: [gentoo-user] New monitor problems. < sighs >
Neil Bothwick wrote: On Wed, 15 Dec 2010 02:40:48 -0600, Dale wrote: [ 2081.057] (EE) Failed to load module "dri" (module does not exist, 0) [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, 0) [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: Function not [ 2082.101] (EE) NVIDIA(0): implemented [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not exist, 0) [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA X driver not found) r...@smoker / # You haven't loaded the glx module, try this in xorg.conf Section "Module" Load "glx" Disable "dri" Disable "dri2" EndSection Also, what does "eselect opengl list" show? I reconfiged it so many times, I may have missed that the last time. I know it was in there the other times tho. I typed it all in a couple times. I'll give that a try again tho. Eselect shows nvidia. I always set it again to make sure tho. I ran into a problem with it switching on the old rig a while back. It also switches when I re-emerge the drivers too. I found that with google. It seems to work for everyone else. lol Thanks. Will post back what happens. I got a couple suggestions now. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 10:03:58 Dale wrote: > Helmut Jarausch wrote: > > On 12/15/10 09:40:48, Dale wrote: > >> Hi again, > >> > >> For those mot up to date. I built a new rig that has a Nvidia GT220 > >> card in it. I bought a brand new monitor this morning, a LG W2253, > >> and > >> it worked one time. I had to reboot to move some things around and > >> when > >> I rebooted, the GUI doesn't come up. The BIOS screen shows up and I > >> can > >> see the services start up as well but when it switches to vt7, it > >> just > >> > >> has a little blinking cursor at the top. > > > > Have you changed the kernel? The nvidia-drivers (as well as the ati- > > drivers) have to be re-emerged after (nearly) any change to the kernel. > > Just re-emerge it, unload and reload the corresponding kernel module > > and try again to startx. > > > > Helmut. > > Just to make sure I understand you correctly. You mean: > > cd /usr/src/linux > make all && make modules_install > then cp the kernel to /boot ? > > That I didn't try. I knew if I updated the kernel I had to re-emerge > the nvidia-drivers but I didn't know it was the other way around too. > > Dale > > :-) :-) Dale, Easy to test if the nvidia-driver is loaded correctly. What does "dmesg" say when loading the nvidia module? Also, is "nvidia" listed in the following file: /etc/modules.autoload.d/kernel-2.6 ? I don't have it there, for some reason it's "autoloaded", but it might help? -- Joost
[gentoo-user] Emerging package via NFS ?
Hi, is it possible to emerge packages to a $ROOT directory mounted via NFS ? The setup is - machine A is equipped with a Quad core CPU - machine B is equipped with an N330 Atom-CPU - machine A is doing the system update on a local chroot-environment for machine B and generates binary packages. These packages are installed on machine B using the binary package feature of portage. I expected that the above setup would give an performance improvement over letting machine B do the portage update itself. However a trial run did not show significant improvement that justifies the effort. Machine B still needs a reasonable amount of time to fetch unpack and install the packages. An alternative way might be to mount machine B's / directory via NFS and change make.conf's $ROOT variable to that mount point. Does that sound as a reasonable approach ? Regards, Thomas
Re: [gentoo-user] New monitor problems. < sighs >
Dale [10-12-15 09:48]: > Hi again, > > For those mot up to date. I built a new rig that has a Nvidia GT220 > card in it. I bought a brand new monitor this morning, a LG W2253, and > it worked one time. I had to reboot to move some things around and > when I rebooted, the GUI doesn't come up. The BIOS screen shows up and > I can see the services start up as well but when it switches to vt7, it > just has a little blinking cursor at the top. > > This is a grep of the log: > > r...@smoker / # cat /media/disk/files/Xorg.0.log | grep EE > [ 2081.047] Current Operating System: Linux fireball 2.6.36-gentoo-r4 > #15 SMP PREEMPT Mon Dec 13 01:15:13 CST 2010 x86_64 > (WW) warning, (EE) error, (NI) not implemented, (??) unknown. > [ 2081.048] (II) Loading extension MIT-SCREEN-SAVER > [ 2081.057] (EE) Failed to load module "dri" (module does not exist, > 0) > [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, > 0) > [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory > identifier: Function not > [ 2082.101] (EE) NVIDIA(0): implemented > [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not > exist, 0) > [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA > X driver not found) > r...@smoker / # > > I have tried X -configure, Xorg -configure and nvidia-xconfig and none > of them do any better. I also did a emerge -e all the way up to > nvidia-driver as well then reconfiged everything. Still no joy. I'm > not going to mention all the reinstalls of the nvidia drivers and > different versions I tried. > > I attached my xorg.conf file too. I can send the whole xorg.log file > if needed. Oh, I always get the dri load error. I also used eselect > to switch to nvidia's opengl to, many times. > > Could it be something weird with the card maybe? This is monitor two > that has issues with it. > > I have a ATI card handy. If someone knows of a idiot proof guide I > could follow. You know, one that you used and are confident it will > work if followed. I don't have much hair left right now. :-( > > Just for fun, I got a Mandriva install CD and booted it up and even > installed it on a spare drive. After the install, it boots just like > Gentoo does. I can see the BIOS, the services and such but when it > switches to vt7, just a blinking cursor. > > This new rig is turning into a nightmare. No wonder I waited almost 8 > years. lol > > Thoughts? Ideas? Suggestions, other than jumping in the creek. > > Dale > > :-) :-) Hi Dale, me again ... 8) Dont hesitate...I think everything is alright. First of all: In grub add a "2" at the end of the kernel command line. This will boot you into runlevel 2 (console) without trying to switch to X. This way you will have a "clean" machine with no confused nvidia driver hanging a around and a polling X waiting to something to connect to. Next is: Make yourself a cup of tea (or coffee, if you prefer) cause the next steps may take some compilation time... As root: 1.) Add "hal" to the USE flags of /etc/make.conf. Do a emerge --color=n -p -v --newuse --update --deep world to find those packages, which need to be updated/recompiled due to the changes to the USE flags. This morning there was an update to the nvidia driver also. I does not harm to execute the emerge command, even if you have hal already installed and configured. 2.) cd to /etc mkdir hal and extract the fdi rulesset, I will send you in a personal mail soon. If you have already a rulesset installed DONT overwrite it with mine. May be you can use mine for comparison reasons, if something goes wrong. The rulesset works for me -- but that is no reason, that it have to work for you. BE WARNED :) Folder layout: /etc/hal/fdi/.. 3.) Strip you xorg.conf. I will you send mine with the hal/fdi-ruleset. I have a different monitori/mouse and a different nvidia card. But these differences should be hidden behind hal and the nvidia driver. Before installing it, check it against your hardware. Again: YOU HAVE BEEN WARNED ! :) 4.) Add input_devices_evdev video_cards_nvidia to the USE flags of xorg-drivers. Add hal to the USE-flags of your xorg-server. 5.) Check your kernel config file for CONFIG_INPUT_EVDEV=y If not set, configured it and recompile the kernel. Reboot. 5.) Recompile xorg-drivers, xorg-servers, xf86-input-evdev xf86-input-keyboard xf86-input-mouse, nvidia-drivers 6.) Reboot 7.) Submit telinit 3 as root from the commandline. Should work I cannot make responsible for any consequences directly or indirectly resulting from the usage of these instructions Good luck! Best regards, mcc
Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: On Wednesday 15 December 2010 09:55:37 Helmut Jarausch wrote: On 12/15/10 09:40:48, Dale wrote: Hi again, For those mot up to date. I built a new rig that has a Nvidia GT220 card in it. I bought a brand new monitor this morning, a LG W2253, and it worked one time. I had to reboot to move some things around and when I rebooted, the GUI doesn't come up. The BIOS screen shows up and I can see the services start up as well but when it switches to vt7, it just has a little blinking cursor at the top. Have you changed the kernel? The nvidia-drivers (as well as the ati- drivers) have to be re-emerged after (nearly) any change to the kernel. Just re-emerge it, unload and reload the corresponding kernel module and try again to startx. Helmut. Also, when emerging the "nvidia-drivers", try making sure that "/usr/src/linux" points to the actual linux-sources you are booting from. (Not sure if necessary, but I always have that symlink to the current kernel) -- Joost That I know is important. Since this is a new build, it only has one kernel installed and it is pointing to the right one. Good call tho. That would cause problems for sure and could be easily forgotten. Thanks. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
Teemu Vartiainen wrote: 2010/12/15 Neil Bothwick: On Wed, 15 Dec 2010 02:40:48 -0600, Dale wrote: [ 2081.057] (EE) Failed to load module "dri" (module does not exist, 0) [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, 0) [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: Function not [ 2082.101] (EE) NVIDIA(0): implemented [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not exist, 0) [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA X driver not found) r...@smoker / # You haven't loaded the glx module, try this in xorg.conf Section "Module" Load "glx" Disable "dri" Disable "dri2" EndSection Also, what does "eselect opengl list" show? -- Neil Bothwick Nostalgia isn't what it used to be. Hi to everybody! This is my first post to this list. I think this guide help maybe? http://www.gentoo.org/doc/en/nvidia-guide.xml Vartsu I already found it. I was in links when I used it too. That is nasty. ;-) Links works in a pinch but I have to be in a real pinch. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: On Wednesday 15 December 2010 10:03:58 Dale wrote: Helmut Jarausch wrote: On 12/15/10 09:40:48, Dale wrote: Hi again, For those mot up to date. I built a new rig that has a Nvidia GT220 card in it. I bought a brand new monitor this morning, a LG W2253, and it worked one time. I had to reboot to move some things around and when I rebooted, the GUI doesn't come up. The BIOS screen shows up and I can see the services start up as well but when it switches to vt7, it just has a little blinking cursor at the top. Have you changed the kernel? The nvidia-drivers (as well as the ati- drivers) have to be re-emerged after (nearly) any change to the kernel. Just re-emerge it, unload and reload the corresponding kernel module and try again to startx. Helmut. Just to make sure I understand you correctly. You mean: cd /usr/src/linux make all&& make modules_install then cp the kernel to /boot ? That I didn't try. I knew if I updated the kernel I had to re-emerge the nvidia-drivers but I didn't know it was the other way around too. Dale :-) :-) Dale, Easy to test if the nvidia-driver is loaded correctly. What does "dmesg" say when loading the nvidia module? Also, is "nvidia" listed in the following file: /etc/modules.autoload.d/kernel-2.6 ? I don't have it there, for some reason it's "autoloaded", but it might help? -- Joost I confirmed with lsmod and even rmmod and modprobed it back in a few times. Still no joy. I really think it is either the card or the xorg.conf file. Thing is, when Mandriva failed to work, that makes me wonder about the card. It was using the generic "nv" driver and it still wouldn't work. Got to go take care of some medical issues. Back in a bit. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
> r...@smoker / # cat /media/disk/files/Xorg.0.log | grep EE > [ 2081.047] Current Operating System: Linux fireball 2.6.36-gentoo-r4 #15 > SMP PREEMPT Mon Dec 13 01:15:13 CST 2010 x86_64 >(WW) warning, (EE) error, (NI) not implemented, (??) unknown. > [ 2081.048] (II) Loading extension MIT-SCREEN-SAVER > [ 2081.057] (EE) Failed to load module "dri" (module does not exist, 0) > [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, 0) > [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: > Function not > [ 2082.101] (EE) NVIDIA(0): implemented > [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not > exist, 0) > [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA X > driver not found) > r...@smoker / # > Can you locate libdri.so, libdri2.so and libglx.so on your file system? On my ATI system they're in /usr/lib64/opengl/xorg-x11/extensions/ so to let xorg know where to look; Section "Files" ModulePath "/usr/lib64/opengl/xorg-x11/extensions/" EndSection
Re: [gentoo-user] New monitor problems. < sighs >
On 12/15/10 10:03:58, Dale wrote: > Helmut Jarausch wrote: > > On 12/15/10 09:40:48, Dale wrote: > > > >> Hi again, > >> > >> For those mot up to date. I built a new rig that has a Nvidia > GT220 > >> card in it. I bought a brand new monitor this morning, a LG > W2253, > >> and > >> it worked one time. I had to reboot to move some things around > and > >> when > >> I rebooted, the GUI doesn't come up. The BIOS screen shows up and > I > >> can > >> see the services start up as well but when it switches to vt7, it > >> just > >> > >> has a little blinking cursor at the top. > >> > > Have you changed the kernel? The nvidia-drivers (as well as the > ati- > > drivers) have to be re-emerged after (nearly) any change to the > kernel. > > Just re-emerge it, unload and reload the corresponding kernel > module > > and try again to startx. > > > > Helmut. > > > > > > Just to make sure I understand you correctly. You mean: > > cd /usr/src/linux > make all && make modules_install > then cp the kernel to /boot ? > > That I didn't try. I knew if I updated the kernel I had to re-emerge > the nvidia-drivers but I didn't know it was the other way around too. > No, just emerge -1 x11-drivers/nvidia-drivers (while the /usr/src/linux symlink is OK as others have mentioned already) Helmut. (By the way, my x11-base/xorg-server-1.9.2.902 doesn't use hal anymore)
Re: [gentoo-user] Strange boot problem - any ideas?
Helmut Jarausch [10-12-15 10:00]: > Hi, > > I have to power down one of my machines each day. > Booting it the other day fails from time to time. > On this (and all my machines) /usr is on an ext4 file system by its > own. It looks as if mounting /usr fails sometimes (silently). > The first unusual message is that it cannot find the file > /usr/sbin/acpid . After that, most other actions fail as well. > > Once the machine has failed to boot it will fail every time afterwards > unless I do the following: > I boot by a rescue CD, change root and re-emerge sys-power/acpid (this > package contains /usr/sbin/acpid). > This single action has helped without any problem each time I had to > try it. > > Of course, I've done many checks on the drive hosting the /usr > partition - no errors at all (the drive is only a few months old and a > good one (enterprise edition)). > > So, I suspect openrc. Might it be that this is a timing problem and > openrc doesn't check if the partition is mounted? Booting has worked > just flawlessly on that machine for more than a year. It looks as if a > recent version of openrc (currently 0.6.8) has generated this problem. > > Thanks for any hint, > Helmut. > > Hi Helmut, it would be interesting to know, whether /usr/sbin/acpid is really missing, when the boot fails. If it is not missing, the boot fails due to a mount problem and not the re-installing of acpid as such cures the problem, but what this installation is doing else. Or in other words: Any installation, which put someting into /usr cures the problem. If the mount of /usr is the problem, which arises from a timing problem than there is something done in parallel which should be done in sequence (and was previously done in sequence). I dont know openrc, but may be a setting in its configuration file regarding this aspect will cure the problem? Good luck! Best regards, mcc
Re: [gentoo-user] New monitor problems. < sighs >
On 12/15/10 10:26:16, Dale wrote: > > Got to go take care of some medical issues. Back in a bit. > I'd try to boot by SystemRescueCD and let it start X11 to rule out some hardware problems. (Version 1.6.4 is using xorg-server-1.9.2) Helmut.
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 10:26:16 Dale wrote: > J. Roeleveld wrote: > > On Wednesday 15 December 2010 10:03:58 Dale wrote: > >> Helmut Jarausch wrote: > >>> On 12/15/10 09:40:48, Dale wrote: > Hi again, > > For those mot up to date. I built a new rig that has a Nvidia GT220 > card in it. I bought a brand new monitor this morning, a LG W2253, > and > it worked one time. I had to reboot to move some things around and > when > I rebooted, the GUI doesn't come up. The BIOS screen shows up and I > can > see the services start up as well but when it switches to vt7, it > just > > has a little blinking cursor at the top. > >>> > >>> Have you changed the kernel? The nvidia-drivers (as well as the ati- > >>> drivers) have to be re-emerged after (nearly) any change to the kernel. > >>> Just re-emerge it, unload and reload the corresponding kernel module > >>> and try again to startx. > >>> > >>> Helmut. > >> > >> Just to make sure I understand you correctly. You mean: > >> > >> cd /usr/src/linux > >> make all&& make modules_install > >> then cp the kernel to /boot ? > >> > >> That I didn't try. I knew if I updated the kernel I had to re-emerge > >> the nvidia-drivers but I didn't know it was the other way around too. > >> > >> Dale > >> > >> :-) :-) > > > > Dale, > > > > Easy to test if the nvidia-driver is loaded correctly. > > What does "dmesg" say when loading the nvidia module? > > > > Also, is "nvidia" listed in the following file: > > /etc/modules.autoload.d/kernel-2.6 ? > > > > I don't have it there, for some reason it's "autoloaded", but it might > > help? > > > > -- > > Joost > > I confirmed with lsmod and even rmmod and modprobed it back in a few > times. Still no joy. I really think it is either the card or the > xorg.conf file. Thing is, when Mandriva failed to work, that makes me > wonder about the card. It was using the generic "nv" driver and it > still wouldn't work. > > Got to go take care of some medical issues. Back in a bit. > > Dale > > :-) :-) Dale, Rather then rebooting, did you shut down the whole machine, eg. power off and unplug power and then start it again? I have, in the past, had issues where a card wasn't being "reset" properly during a reboot and only a cold start would lead to correct behaviour. -- Joost Roeleveld
Re: [gentoo-user] Strange boot problem - any ideas?
On 12/15/10 10:32:34, meino.cra...@gmx.de wrote: > Helmut Jarausch [10-12-15 10:00]: > > Hi, > > > > I have to power down one of my machines each day. > > Booting it the other day fails from time to time. > > On this (and all my machines) /usr is on an ext4 file system by > its > > > own. It looks as if mounting /usr fails sometimes (silently). > > The first unusual message is that it cannot find the file > > /usr/sbin/acpid . After that, most other actions fail as well. > > > > Once the machine has failed to boot it will fail every time > afterwards > > unless I do the following: > > I boot by a rescue CD, change root and re-emerge sys-power/acpid > (this > > package contains /usr/sbin/acpid). > > This single action has helped without any problem each time I had > to > > > try it. > > > > Of course, I've done many checks on the drive hosting the /usr > > partition - no errors at all (the drive is only a few months old > and > a > > good one (enterprise edition)). > > > > So, I suspect openrc. Might it be that this is a timing problem and > > openrc doesn't check if the partition is mounted? Booting has > worked > > > just flawlessly on that machine for more than a year. It looks as > if > a > > recent version of openrc (currently 0.6.8) has generated this > problem. > > > > Thanks for any hint, > > Helmut. > > > > > > Hi Helmut, > > it would be interesting to know, whether /usr/sbin/acpid is really > missing, when the boot fails. The system allows me to login as root - though / is still mounted read only. And /usr isn't mounted. I can remount the root partition read/ write and then mount /usr. Now the file /usr/sbin/acpid is there (of course). Just reboot(ing) after that is not enough. Yes, I should try to write something to the /usr partition instead of emerging something and see if it helps. > > If it is not missing, the boot fails due to a mount problem and not > the re-installing of acpid as such cures the problem, but what this > installation is doing else. > Or in other words: Any installation, which put someting into /usr > cures the problem. > > If the mount of /usr is the problem, which arises from a timing > problem than there is something done in parallel which should be done > in sequence (and was previously done in sequence). > > I dont know openrc, but may be a setting in its configuration file > regarding this aspect will cure the problem? I'll search for such a configuration problem, otherwise I'll create a bug report. > > Good luck! Probably, I do need that. Many thanks, Helmut.
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, 15 Dec 2010 10:17:01 +0100, meino.cra...@gmx.de wrote: > First of all: In grub add a "2" at the end of the kernel command line. > This will boot you into runlevel 2 (console) without trying to switch > to X. This way you will have a "clean" machine with no confused nvidia > driver hanging a around and a polling X waiting to something to > connect to. Gentoo doesn't use runlevels like that. The correct way to boot without X in Gentoo is to add "gentoo=nox" to the kernel options. -- Neil Bothwick Windows Error #09: Game Over. Exiting Windows. signature.asc Description: PGP signature
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, 15 Dec 2010 10:17:01 +0100, meino.cra...@gmx.de wrote: > 1.) Add "hal" to the USE flags of /etc/make.conf. This is going to be fun ;-) -- Neil Bothwick Few women admit their age. Few men act theirs. signature.asc Description: PGP signature
[gentoo-user] Re: Can't move Chromium windows between screens
Everyone here running kde+xinerama is able to drag Chromium through screens without kwin bar on top, when Chromium is maximized/full screen? Hard to believe i'm the one suffering this pain :'( PD: Issue just happens with window maximized/full screen. 2010/11/16 Pau Peris : > Hi, after updating Chromium to chromium-7.0.517.44 from portage i > can't move it between different screens, if i go to Preferencies -> > Personal -> Theme and enable "Use borders and title bar" (so it shows > kwin) i can move chromium windows between screens as i normally do > with any other app. Before the update i was using chromum 7.x from > portage and had no issues, does any one know how can i check what's > happening? > > Thanks in advanced! > > PD: I'm a kde user, xinerama enabled everywhere. >
Re: [gentoo-user] New monitor problems. < sighs >
> I confirmed with lsmod and even rmmod and modprobed it back in a few times. > Still no joy. I really think it is either the card or the xorg.conf file. > Thing is, when Mandriva failed to work, that makes me wonder about the > card. It was using the generic "nv" driver and it still wouldn't work. > > IMO its a waste of time to worry about any of this other stuff - the EE's in the log file are ERRORs, directly pointing at serious issues with missing modules. So either the modules dont exist, or xorg cant find them. There's no evidence of a problem with the nvidia module or the hardware.
Re: [gentoo-user] Emerging package via NFS ?
On Wed, Dec 15, 2010 at 10:15:12AM +0100, Thomas Drueke wrote:
> Hi,
>
> is it possible to emerge packages to a $ROOT directory mounted via NFS ?
>
> The setup is
> - machine A is equipped with a Quad core CPU
> - machine B is equipped with an N330 Atom-CPU
> - machine A is doing the system update on a local chroot-environment
> for machine B and generates binary packages. These packages are
> installed on machine B using the binary package feature of portage.
>
> I expected that the above setup would give an performance improvement
> over letting machine B do the portage update itself. However a trial run
> did not show significant improvement that justifies the effort. Machine
> B still needs a reasonable amount of time to fetch unpack and install
> the packages.
>
> An alternative way might be to mount machine B's / directory via NFS
> and change make.conf's $ROOT variable to that mount point.
>
> Does that sound as a reasonable approach ?
I had a very old machine, that was really slow. Compiles could be
offloaded by distcc, but even the ./configure-s and portage stuff
(checking, upacking, ...) was reaaly slow...
So I just used to export / through nfs, mounted it on a fast amd64 and
basically did (other is the slow machine)
mount other:/ /mnt/other
mount -t proc proc /mnt/other/proc
mount --bind /dev /mnt/other/dev
mkdir /tmp/other
mount --bind /tmp/other /mnt/other/var/tmp/portage
mkdir /home/gentoo-other
mount --bind /home/gentoo-other /mnt/other/home/gentoo
linux32 chroot /mnt/other /bin/bash
emerge.
For the last mkdir/mount, I have DISTDIR=/home/gentoo/distfiles and
PKGDIR=/home/gentoo/packages in make.conf, you can do that with the
standart /usr/portage/{distfiles,packages}
This way most of the compile is done "localy" on the fast machine.
yoyo
>
> Regards,
> Thomas
>
Re: [gentoo-user] Strange boot problem - any ideas?
> I'll search for such a configuration problem, otherwise I'll create a > bug report. > > > Are you using rc_parallel="NO" ? (in /etc/rc.conf)
Re: [gentoo-user] New monitor problems. < sighs >
On 12/15/10 10:56:21, Adam Carter wrote: > > I confirmed with lsmod and even rmmod and modprobed it back in a > few > times. > > Still no joy. I really think it is either the card or the > xorg.conf file. > > Thing is, when Mandriva failed to work, that makes me wonder about > the > > card. It was using the generic "nv" driver and it still wouldn't > work. > > > > > IMO its a waste of time to worry about any of this other stuff - the > EE's in > the log file are ERRORs, directly pointing at serious issues with > missing > modules. So either the modules dont exist, or xorg cant find them. > > There's no evidence of a problem with the nvidia module or the > hardware. I don't think that's true. In my Xorg.0.log I do have (EE) Failed to load module "dri" (module does not exist, 0) (EE) Failed to load module "dri2" (module does not exist, 0) and still, X11 is running just fine. I think these modules are superfluous with a recent Xorg version. Helmut.
Re: [gentoo-user] Strange boot problem - any ideas?
On 12/15/10 11:03:43, Adam Carter wrote: > > I'll search for such a configuration problem, otherwise I'll create > a > > bug report. > > > > > > > Are you using rc_parallel="NO" ? (in /etc/rc.conf) No, why disable fast booting. If it turns out to be necessary it would be a pity! But I'll try it for a while to see if it helps, many thanks, Helmut.
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 10:54:14 Neil Bothwick wrote: > On Wed, 15 Dec 2010 10:17:01 +0100, meino.cra...@gmx.de wrote: > > 1.) Add "hal" to the USE flags of /etc/make.conf. > > This is going to be fun ;-) Yes, I remember that episode on his last machine :) -- Joost
Re: [gentoo-user] New monitor problems. < sighs >
Helmut Jarausch wrote: On 12/15/10 10:26:16, Dale wrote: Got to go take care of some medical issues. Back in a bit. I'd try to boot by SystemRescueCD and let it start X11 to rule out some hardware problems. (Version 1.6.4 is using xorg-server-1.9.2) Helmut. I tried booting it but it didn't do right. Maybe a bad burn but I had a different card and monitor in the rig then. I did boot Kanotix and it worked fine with the current monitor and card. Resolution was good and all. The Mandriva CD is GUI based and it worked fine as well. The OS however did not work. This is what is so confusing to me. I try something such as a boot CD and it works. Yet no OS works. One makes me think hardware and one makes me think the hardware is fine and that I have a bad config somewhere. Now you know why I am posting on here. Going to try some things that have been suggested now. Fingers crossed. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
> > IMO its a waste of time to worry about any of this other stuff - the > > EE's in > > the log file are ERRORs, directly pointing at serious issues with > > missing > > modules. So either the modules dont exist, or xorg cant find them. > > > > There's no evidence of a problem with the nvidia module or the > > hardware. > > I don't think that's true. > In my Xorg.0.log I do have > > (EE) Failed to load module "dri" (module does not exist, 0) > > (EE) Failed to load module "dri2" (module does not exist, 0) > > and still, X11 is running just fine. I think these modules are > superfluous with a recent Xorg version. > I found i had no X until I fixed that issue when using fglrx - so perhaps its driver dependent. In any case, Dale's setup is not working so I think the principle of fixing the obvious issue first is a sensible approach. If fixing the log file errors doesnt help, then sure, fall back to trying other things.
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 11:10:03 Helmut Jarausch wrote: > On 12/15/10 10:56:21, Adam Carter wrote: > > > I confirmed with lsmod and even rmmod and modprobed it back in a > > > > few > > times. > > > > > Still no joy. I really think it is either the card or the > > > > xorg.conf file. > > > > > Thing is, when Mandriva failed to work, that makes me wonder about > > > > the > > > > > card. It was using the generic "nv" driver and it still wouldn't > > > > work. > > > > > > IMO its a waste of time to worry about any of this other stuff - the > > EE's in > > the log file are ERRORs, directly pointing at serious issues with > > missing > > modules. So either the modules dont exist, or xorg cant find them. > > > > There's no evidence of a problem with the nvidia module or the > > hardware. > > I don't think that's true. > In my Xorg.0.log I do have > > (EE) Failed to load module "dri" (module does not exist, 0) > > (EE) Failed to load module "dri2" (module does not exist, 0) > > and still, X11 is running just fine. I think these modules are > superfluous with a recent Xorg version. > > Helmut. I agree, just checked on all the machines I am running Gentoo on with Nvidia cards and they all have those 2 EE-lines. Both of these have 2 monitors hooked up and that is working correctly. -- Joost Roeleveld
Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: Dale, Rather then rebooting, did you shut down the whole machine, eg. power off and unplug power and then start it again? I have, in the past, had issues where a card wasn't being "reset" properly during a reboot and only a cold start would lead to correct behaviour. -- Joost Roeleveld I did once but about to try that again. I recall people having that sort of issue with networks cards before too. I swapped plugs so it was without thinking that I did that. Couldn't hurt. ;-) Thanks. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
Adam Carter wrote: I confirmed with lsmod and even rmmod and modprobed it back in a few times. Still no joy. I really think it is either the card or the xorg.conf file. Thing is, when Mandriva failed to work, that makes me wonder about the card. It was using the generic "nv" driver and it still wouldn't work. IMO its a waste of time to worry about any of this other stuff - the EE's in the log file are ERRORs, directly pointing at serious issues with missing modules. So either the modules dont exist, or xorg cant find them. There's no evidence of a problem with the nvidia module or the hardware. I hope you are right. If it is missing modules, what do I need to do to fix it? I did a emerge -e nvidia-drivers and everything built fine. Dale :-) :-)
Re: [gentoo-user] Strange boot problem - any ideas?
On Wed, Dec 15, 2010 at 9:12 PM, Helmut Jarausch < jarau...@igpm.rwth-aachen.de> wrote: > On 12/15/10 11:03:43, Adam Carter wrote: > > > I'll search for such a configuration problem, otherwise I'll create > > a > > > bug report. > > > > > > > > > > > Are you using rc_parallel="NO" ? (in /etc/rc.conf) > > No, why disable fast booting. If it turns out to be necessary it > would be a pity! > But I'll try it for a while to see if it helps, > > many thanks, > Helmut. > > Its a long shot, but you were talking about bug reports and; "# WARNING: whilst we have improved parallel, it can still potentially lock # the boot process. Don't file bugs about this unless you can supply # patches that fix it without breaking other things!"
Re: [gentoo-user] Emerging package via NFS ?
Interesting approach. I'll give that a try.
Thanks,
Thomas
Am 15.12.2010 10:56, schrieb YoYo Siska:
> On Wed, Dec 15, 2010 at 10:15:12AM +0100, Thomas Drueke wrote:
>> Hi,
>>
>> is it possible to emerge packages to a $ROOT directory mounted via NFS ?
>>
>> The setup is
>> - machine A is equipped with a Quad core CPU
>> - machine B is equipped with an N330 Atom-CPU
>> - machine A is doing the system update on a local chroot-environment
>> for machine B and generates binary packages. These packages are
>> installed on machine B using the binary package feature of portage.
>>
>> I expected that the above setup would give an performance improvement
>> over letting machine B do the portage update itself. However a trial run
>> did not show significant improvement that justifies the effort. Machine
>> B still needs a reasonable amount of time to fetch unpack and install
>> the packages.
>>
>> An alternative way might be to mount machine B's / directory via NFS
>> and change make.conf's $ROOT variable to that mount point.
>>
>> Does that sound as a reasonable approach ?
>
> I had a very old machine, that was really slow. Compiles could be
> offloaded by distcc, but even the ./configure-s and portage stuff
> (checking, upacking, ...) was reaaly slow...
>
> So I just used to export / through nfs, mounted it on a fast amd64 and
> basically did (other is the slow machine)
>
> mount other:/ /mnt/other
> mount -t proc proc /mnt/other/proc
> mount --bind /dev /mnt/other/dev
> mkdir /tmp/other
> mount --bind /tmp/other /mnt/other/var/tmp/portage
> mkdir /home/gentoo-other
> mount --bind /home/gentoo-other /mnt/other/home/gentoo
>
> linux32 chroot /mnt/other /bin/bash
> emerge.
>
> For the last mkdir/mount, I have DISTDIR=/home/gentoo/distfiles and
> PKGDIR=/home/gentoo/packages in make.conf, you can do that with the
> standart /usr/portage/{distfiles,packages}
>
> This way most of the compile is done "localy" on the fast machine.
> yoyo
>
>
>>
>> Regards,
>> Thomas
>>
>
>
>
>
Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: On Wednesday 15 December 2010 10:03:58 Dale wrote: Helmut Jarausch wrote: On 12/15/10 09:40:48, Dale wrote: Hi again, For those mot up to date. I built a new rig that has a Nvidia GT220 card in it. I bought a brand new monitor this morning, a LG W2253, and it worked one time. I had to reboot to move some things around and when I rebooted, the GUI doesn't come up. The BIOS screen shows up and I can see the services start up as well but when it switches to vt7, it just has a little blinking cursor at the top. Have you changed the kernel? The nvidia-drivers (as well as the ati- drivers) have to be re-emerged after (nearly) any change to the kernel. Just re-emerge it, unload and reload the corresponding kernel module and try again to startx. Helmut. Just to make sure I understand you correctly. You mean: cd /usr/src/linux make all&& make modules_install then cp the kernel to /boot ? That I didn't try. I knew if I updated the kernel I had to re-emerge the nvidia-drivers but I didn't know it was the other way around too. Dale :-) :-) Dale, Easy to test if the nvidia-driver is loaded correctly. What does "dmesg" say when loading the nvidia module? Also, is "nvidia" listed in the following file: /etc/modules.autoload.d/kernel-2.6 ? I don't have it there, for some reason it's "autoloaded", but it might help? -- Joost According to dmesg, it loads fine when booting. It also loads fine when I load it manually as well. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
> > I don't think that's true. > > In my Xorg.0.log I do have > > > > (EE) Failed to load module "dri" (module does not exist, 0) > > > > (EE) Failed to load module "dri2" (module does not exist, 0) > > > > and still, X11 is running just fine. I think these modules are > > superfluous with a recent Xorg version. > > > > Helmut. > > I agree, just checked on all the machines I am running Gentoo on with > Nvidia > cards and they all have those 2 EE-lines. > Both of these have 2 monitors hooked up and that is working correctly. > > Ok, so my argument is looking weak then - are you using the nvidia binary drivers like Dale? For interests sake, do the libdri.so and libdri2.so files exist on your system? Try find /usr -name libdri\*
Re: [gentoo-user] New monitor problems. < sighs >
On 12/15/2010 11:18 AM, Dale wrote: > Helmut Jarausch wrote: >> On 12/15/10 10:26:16, Dale wrote: >> >>> Got to go take care of some medical issues. Back in a bit. >>> >>> >> I'd try to boot by SystemRescueCD and let it start X11 >> to rule out some hardware problems. (Version 1.6.4 is using >> xorg-server-1.9.2) >> >> Helmut. how about just checking the bloody Xorg logs to see what could have gone wrong? (/var/log/Xorg.0.log) t
Re: [gentoo-user] New monitor problems. < sighs >
Helmut Jarausch wrote: On 12/15/10 10:56:21, Adam Carter wrote: I confirmed with lsmod and even rmmod and modprobed it back in a few times. Still no joy. I really think it is either the card or the xorg.conf file. Thing is, when Mandriva failed to work, that makes me wonder about the card. It was using the generic "nv" driver and it still wouldn't work. IMO its a waste of time to worry about any of this other stuff - the EE's in the log file are ERRORs, directly pointing at serious issues with missing modules. So either the modules dont exist, or xorg cant find them. There's no evidence of a problem with the nvidia module or the hardware. I don't think that's true. In my Xorg.0.log I do have (EE) Failed to load module "dri" (module does not exist, 0) (EE) Failed to load module "dri2" (module does not exist, 0) and still, X11 is running just fine. I think these modules are superfluous with a recent Xorg version. Helmut. What I found with google, those modules used to be loaded separate bit is now included into the drivers. So, it complains but they do load up but just in another way. I have had those errors on my old rig for a long time. It never caused any problem that I know of. What bothers me is the last line about glx. That I have not seen before. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
I hope you are right. If it is missing modules, what do I need to do to fix it? I did a emerge -e nvidia-drivers and everything built fine. > > Dale > > :-) :-) > Did u see my other email with this? Can you locate libdri.so, libdri2.so and libglx.so on your file system? On my ATI system they're in /usr/lib64/opengl/xorg-x11/ extensions/ so to let xorg know where to look; Section "Files" ModulePath "/usr/lib64/opengl/xorg-x11/extensions/" EndSection
Re: [gentoo-user] New monitor problems. < sighs >
Helmut Jarausch wrote: On 12/15/10 10:03:58, Dale wrote: Helmut Jarausch wrote: On 12/15/10 09:40:48, Dale wrote: Hi again, For those mot up to date. I built a new rig that has a Nvidia GT220 card in it. I bought a brand new monitor this morning, a LG W2253, and it worked one time. I had to reboot to move some things around and when I rebooted, the GUI doesn't come up. The BIOS screen shows up and I can see the services start up as well but when it switches to vt7, it just has a little blinking cursor at the top. Have you changed the kernel? The nvidia-drivers (as well as the ati- drivers) have to be re-emerged after (nearly) any change to the kernel. Just re-emerge it, unload and reload the corresponding kernel module and try again to startx. Helmut. Just to make sure I understand you correctly. You mean: cd /usr/src/linux make all&& make modules_install then cp the kernel to /boot ? That I didn't try. I knew if I updated the kernel I had to re-emerge the nvidia-drivers but I didn't know it was the other way around too. No, just emerge -1 x11-drivers/nvidia-drivers (while the /usr/src/linux symlink is OK as others have mentioned already) Helmut. (By the way, my x11-base/xorg-server-1.9.2.902 doesn't use hal anymore) I upgraded to xorg 1.9 to while trying to fix all this mess. It wasn't working before so I don't guess that hurts. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, Dec 15, 2010 at 9:32 PM, Adam Carter wrote: > I hope you are right. If it is missing modules, what do I need to do to > fix it? I did a emerge -e nvidia-drivers and everything built fine. > > fwiw xorg supplies the libdri modules, its not part of nvidia-drivers. For some reason the path it checks for these drivers is not the same as the location they're installed Not sure if that's a Gentoo specific issue or amd4 Gentoo issue...i'll see if bug has been logged
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 11:20:20 Dale wrote: > J. Roeleveld wrote: > > Dale, > > > > Rather then rebooting, did you shut down the whole machine, eg. power off > > and unplug power and then start it again? > > > > I have, in the past, had issues where a card wasn't being "reset" > > properly during a reboot and only a cold start would lead to correct > > behaviour. > > > > -- > > Joost Roeleveld > > I did once but about to try that again. I recall people having that > sort of issue with networks cards before too. I swapped plugs so it was > without thinking that I did that. Couldn't hurt. ;-) > > Thanks. > > Dale > > :-) :-) Dale, The following is not a joke. If you switched plugs, I am assuming you may have physically moved the machine. If yes, can you try moving it back to where it was before? I have solved a problem with a PC once for a friend by moving it to the other side of their house. (They were living close to a railroad with overhead lines) -- Joost
Re: [gentoo-user] Emerging package via NFS ?
Yes, I have a N330 (zotec ION) with 3G ram, no local storage and swap over nbd with portage and build area in /tmp which itself is on tmpfs. Some packages (gcc and glibc in particular) require a lot of ram and tmpfs to emerge so sometimes I have to disable tmpfs and use nfs storage. Because portage is lost on reboot when using tmpfs, an emerge sync is needed to rebuild it when an update is needed and that takes time (more than just updating portage over nfs in fact). The upside is emerges can be very fast indeed and updating portage is an unattended operation so cost (admin time) is small for me. Only disadvantage is gcc and possibly gcc need manual intervention to disable tmpfs but thats not often - if you dont mind the wasted disk space (portage) you can avoid the sync time, but having the build directories in tmpfs is a real gain - wish I had the max of 4G though :) BillK myth2 linux # df -h FilesystemSize Used Avail Use% Mounted on devtmpfs 878M 72K 878M 1% /dev shm 878M 0 878M 0% /dev/shm tmpfs 1.8G 492M 1.3G 29% /tmp tmpfs 10M 0 10M 0% /var/lock tmpfs 10M 64K 10M 1% /var/run tmpfs 10M 320K 9.7M 4% /var/cache/hald svcdir2.0M 208K 1.8M 11% /var/lib/init.d myth1:/home/MythTV/videos 1.2T 960G 251G 80% /mnt/videos myth1:/home/MythTV/posters 1.2T 960G 251G 80% /mnt/posters myth1:/home/MythTV/recordings 1.2T 960G 251G 80% /mnt/recordings myth1:/home/MythTV/music 1.2T 960G 251G 80% /mnt/music myth1:/home/MythTV/gallery 1.2T 960G 251G 80% /mnt/gallery myth2 linux # swapon -s FilenameTypeSizeUsed Priority /dev/nbd0 partition 2097148 0 -1 myth2 linux # On Wed, 2010-12-15 at 10:15 +0100, Thomas Drueke wrote: > Hi, > > is it possible to emerge packages to a $ROOT directory mounted via NFS ? > > The setup is > - machine A is equipped with a Quad core CPU > - machine B is equipped with an N330 Atom-CPU > - machine A is doing the system update on a local chroot-environment > for machine B and generates binary packages. These packages are > installed on machine B using the binary package feature of portage. > > I expected that the above setup would give an performance improvement > over letting machine B do the portage update itself. However a trial run > did not show significant improvement that justifies the effort. Machine > B still needs a reasonable amount of time to fetch unpack and install > the packages. > > An alternative way might be to mount machine B's / directory via NFS > and change make.conf's $ROOT variable to that mount point. > > Does that sound as a reasonable approach ? > > Regards, > Thomas > -- William Kenworthy Home in Perth!
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 11:30:46 Adam Carter wrote: > > > I don't think that's true. > > > In my Xorg.0.log I do have > > > > > > (EE) Failed to load module "dri" (module does not exist, 0) > > > > > > (EE) Failed to load module "dri2" (module does not exist, 0) > > > > > > and still, X11 is running just fine. I think these modules are > > > superfluous with a recent Xorg version. > > > > > > Helmut. > > > > I agree, just checked on all the machines I am running Gentoo on with > > Nvidia > > cards and they all have those 2 EE-lines. > > Both of these have 2 monitors hooked up and that is working correctly. > > > > Ok, so my argument is looking weak then - are you using the nvidia binary > drivers like Dale? Yes, I have been using the binary (nvidia-drivers) for years. Not had to fall back to open drivers yet. > For interests sake, do the libdri.so and libdri2.so files exist on your > system? Try > find /usr -name libdri\* Result: eve ~ # find /usr -name "libdri*" /usr/lib64/opengl/xorg-x11/extensions/libdri.so /usr/lib64/opengl/xorg-x11/extensions/libdri2.so eve ~ # cat /var/log/Xorg.0.log | grep EE (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (II) Loading extension MIT-SCREEN-SAVER (EE) Failed to load module "xtrap" (module does not exist, 0) (EE) Failed to load module "dri" (module does not exist, 0) (EE) Failed to load module "dri2" (module does not exist, 0) The files do exist, but are not found as I am using the nvidia-opengl, rather then the xorg-x11 opengl. -- Joost Roeleveld
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, Dec 15, 2010 at 11:20:33AM +0100, J. Roeleveld wrote: > On Wednesday 15 December 2010 11:10:03 Helmut Jarausch wrote: > > On 12/15/10 10:56:21, Adam Carter wrote: > > > > I confirmed with lsmod and even rmmod and modprobed it back in a > > > > > > few > > > times. > > > > > > > Still no joy. I really think it is either the card or the > > > > > > xorg.conf file. > > > > > > > Thing is, when Mandriva failed to work, that makes me wonder about > > > > > > the > > > > > > > card. It was using the generic "nv" driver and it still wouldn't > > > > > > work. > > > > > > > > > IMO its a waste of time to worry about any of this other stuff - the > > > EE's in > > > the log file are ERRORs, directly pointing at serious issues with > > > missing > > > modules. So either the modules dont exist, or xorg cant find them. > > > > > > There's no evidence of a problem with the nvidia module or the > > > hardware. > > > > I don't think that's true. > > In my Xorg.0.log I do have > > > > (EE) Failed to load module "dri" (module does not exist, 0) > > > > (EE) Failed to load module "dri2" (module does not exist, 0) > > > > and still, X11 is running just fine. I think these modules are > > superfluous with a recent Xorg version. > > > > Helmut. > > I agree, just checked on all the machines I am running Gentoo on with Nvidia > cards and they all have those 2 EE-lines. > Both of these have 2 monitors hooked up and that is working correctly. > IIRC they are used by the other (open) drivers to communicate with the dri modules in kernel... nvidia X driver communicates with the nvidia kernel driver on its own and doesn't use those two X modules eselect opengl creates those two as symlinks whnen you select xorg-x11, and when you select nvidia, they are just not created on intel machine: tabletka ~ # ls -l /usr/lib/xorg/modules/extensions/libdri.so lrwxrwxrwx 1 root root 45 Dec 13 16:02 /usr/lib/xorg/modules/extensions/libdri.so -> ../../../opengl/xorg-x11/extensions/libdri.so tabletka ~ # qfile /usr/lib/opengl/xorg-x11/extensions/libdri.so x11-base/xorg-server (/usr/lib/opengl/xorg-x11/extensions/libdri.so) that said, on all my nvidia setups (with eselect opengl nvidia) I get those errors and everything works ok, on all others (well, intel only ;) with eselect opengl xorg-x11 the modules are there... yoyo
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 11:45:34 Adam Carter wrote: > On Wed, Dec 15, 2010 at 9:32 PM, Adam Carter wrote: > > I hope you are right. If it is missing modules, what do I need to do to > > fix it? I did a emerge -e nvidia-drivers and everything built fine. > > > > fwiw xorg supplies the libdri modules, its not part of nvidia-drivers. > > For > > some reason the path it checks for these drivers is not the same as the > location they're installed Not sure if that's a Gentoo specific issue > or amd4 Gentoo issue...i'll see if bug has been logged It's not a bug, it's as designed. Those 2 files are for the "xorg-x11 " opengl implementation. When you switch to "nvidia" opengl, that location is not used. (And should not be used) -- Joost
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, 15 Dec 2010 04:22:33 -0600, Dale wrote: > I hope you are right. If it is missing modules, what do I need to do > to fix it? I did a emerge -e nvidia-drivers and everything built fine. It's not. Those "errors" about dri are a standard feature of the nvidia binary drivers. They do not indicate any sort of problem. -- Neil Bothwick "I can picture in my mind a world without war, a world without hate. And I can picture us attacking that world, because they'd never expect it." signature.asc Description: PGP signature
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 10:08:30 Dale wrote: > Neil Bothwick wrote: > > On Wed, 15 Dec 2010 02:40:48 -0600, Dale wrote: > >> [ 2081.057] (EE) Failed to load module "dri" (module does not exist, 0) > >> [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, > >> 0) [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory > >> identifier: Function not > >> [ 2082.101] (EE) NVIDIA(0): implemented > >> [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not > >> exist, 0) > >> [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA > >> X driver not found) > >> r...@smoker / # > > > > You haven't loaded the glx module, try this in xorg.conf > > > > Section "Module" > > > > Load "glx" > > Disable "dri" > > Disable "dri2" > > > > EndSection > > > > Also, what does "eselect opengl list" show? > > I reconfiged it so many times, I may have missed that the last time. I > know it was in there the other times tho. I typed it all in a couple > times. I'll give that a try again tho. > > Eselect shows nvidia. I always set it again to make sure tho. I ran > into a problem with it switching on the old rig a while back. It also > switches when I re-emerge the drivers too. I found that with google. > It seems to work for everyone else. lol > > Thanks. Will post back what happens. I got a couple suggestions now. > > Dale > > :-) :-) Dale, Did you add that section to your xorg.conf? If yes, what is the xorg-log saying when you have that added? -- Joost
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, 15 Dec 2010 11:31:54 +0100, Pintér Tibor wrote: > how about just checking the bloody Xorg logs to see what could have gone > wrong? (/var/log/Xorg.0.log) How about just reading the bloody first post in the thread where parts of the log are quoted? -- Neil Bothwick TERROR: A female Klingon with PMS. signature.asc Description: PGP signature
Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: On Wednesday 15 December 2010 11:20:20 Dale wrote: J. Roeleveld wrote: Dale, Rather then rebooting, did you shut down the whole machine, eg. power off and unplug power and then start it again? I have, in the past, had issues where a card wasn't being "reset" properly during a reboot and only a cold start would lead to correct behaviour. -- Joost Roeleveld I did once but about to try that again. I recall people having that sort of issue with networks cards before too. I swapped plugs so it was without thinking that I did that. Couldn't hurt. ;-) Thanks. Dale :-) :-) Dale, The following is not a joke. If you switched plugs, I am assuming you may have physically moved the machine. If yes, can you try moving it back to where it was before? I have solved a problem with a PC once for a friend by moving it to the other side of their house. (They were living close to a railroad with overhead lines) -- Joost Actually, it is sitting in the same place. I just wanted to sort out some cables from the old rig. I shutdown just in case I snagged soemthing, like a power cord or something. I did unhook everything a minute ago and it is still the same. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: On Wednesday 15 December 2010 11:30:46 Adam Carter wrote: I don't think that's true. In my Xorg.0.log I do have (EE) Failed to load module "dri" (module does not exist, 0) (EE) Failed to load module "dri2" (module does not exist, 0) and still, X11 is running just fine. I think these modules are superfluous with a recent Xorg version. Helmut. I agree, just checked on all the machines I am running Gentoo on with Nvidia cards and they all have those 2 EE-lines. Both of these have 2 monitors hooked up and that is working correctly. Ok, so my argument is looking weak then - are you using the nvidia binary drivers like Dale? Yes, I have been using the binary (nvidia-drivers) for years. Not had to fall back to open drivers yet. For interests sake, do the libdri.so and libdri2.so files exist on your system? Try find /usr -name libdri\* Result: eve ~ # find /usr -name "libdri*" /usr/lib64/opengl/xorg-x11/extensions/libdri.so /usr/lib64/opengl/xorg-x11/extensions/libdri2.so eve ~ # cat /var/log/Xorg.0.log | grep EE (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (II) Loading extension MIT-SCREEN-SAVER (EE) Failed to load module "xtrap" (module does not exist, 0) (EE) Failed to load module "dri" (module does not exist, 0) (EE) Failed to load module "dri2" (module does not exist, 0) The files do exist, but are not found as I am using the nvidia-opengl, rather then the xorg-x11 opengl. -- Joost Roeleveld I get the same here. Mine are in a xorg-x11 path but nothing for nvidia. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 09:03:58 Dale wrote: > I knew if I updated the kernel I had to re-emerge the nvidia-drivers > but I didn't know it was the other way around too. No, it isn't. Not here at any rate. -- Rgds Peter. Linux Counter 5290, 1994-04-23.
Re: [gentoo-user] New monitor problems. < sighs >
Pintér Tibor wrote: how about just checking the bloody Xorg logs to see what could have gone wrong? (/var/log/Xorg.0.log) That's where the errors came from that I posted in my first post. Here is the results of the latest test. I shut down the rig and unplugged the rig and monitor. Glad to see the monitor switched back to English too. It was Chinese or something. Anyway. Before the shutdown, I rebuilt the kernel, re-emerged the nvidia drivers and did a manual switch on opengl to nvidia. I let the rig sit there unplugged for about 30 minutes. I then booted it up. BIOS came up, I saw the services scroll up then it switched to vt7 and gave me a blinking cursor. It just sat there. I'm attaching both the new just tried xorg.conf and the xorg.log file. No grep or anything this time. Let me know if you see anything fishy or that needs changing. Thanks. Dale :-) :-) [ 2081.047] This is a pre-release version of the X server from The X.Org Foundation. It is not supported in any way. Bugs may be filed in the bugzilla at http://bugs.freedesktop.org/. Select the "xorg" product for bugs you find in this release. Before reporting bugs in pre-release versions please check the latest version in the X.Org Foundation git repository. See http://wiki.x.org/wiki/GitPage for git access instructions. [ 2081.047] X.Org X Server 1.9.2.902 (1.9.3 RC 2) Release Date: 2010-12-03 [ 2081.047] X Protocol Version 11, Revision 0 [ 2081.047] Build Operating System: Linux 2.6.36-gentoo-r4 x86_64 Gentoo [ 2081.047] Current Operating System: Linux fireball 2.6.36-gentoo-r4 #15 SMP PREEMPT Mon Dec 13 01:15:13 CST 2010 x86_64 [ 2081.047] Kernel command line: root=/dev/sda3 [ 2081.047] Build Date: 14 December 2010 11:26:27PM [ 2081.047] [ 2081.047] Current version of pixman: 0.18.2 [ 2081.047]Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. [ 2081.047] Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. [ 2081.047] (==) Log file: "/var/log/Xorg.0.log", Time: Tue Dec 14 23:31:59 2010 [ 2081.047] (==) Using config file: "/etc/X11/xorg.conf" [ 2081.047] (==) Using system config directory "/usr/share/X11/xorg.conf.d" [ 2081.047] (==) ServerLayout "Layout0" [ 2081.047] (**) |-->Screen "Screen0" (0) [ 2081.047] (**) | |-->Monitor "Monitor0" [ 2081.047] (**) | |-->Device "Device0" [ 2081.047] (**) |-->Input Device "Keyboard0" [ 2081.047] (**) |-->Input Device "Mouse0" [ 2081.047] (==) Automatically adding devices [ 2081.047] (==) Automatically enabling devices [ 2081.047] (WW) `fonts.dir' not found (or not valid) in "/usr/share/fonts/misc/". [ 2081.047]Entry deleted from font path. [ 2081.047](Run 'mkfontdir' on "/usr/share/fonts/misc/"). [ 2081.047] (WW) The directory "/usr/share/fonts/OTF/" does not exist. [ 2081.047]Entry deleted from font path. [ 2081.047] (WW) `fonts.dir' not found (or not valid) in "/usr/share/fonts/75dpi/". [ 2081.047]Entry deleted from font path. [ 2081.047](Run 'mkfontdir' on "/usr/share/fonts/75dpi/"). [ 2081.047] (==) FontPath set to: /usr/share/fonts/TTF/, /usr/share/fonts/Type1/, /usr/share/fonts/100dpi/ [ 2081.047] (==) ModulePath set to "/usr/lib64/xorg/modules" [ 2081.047] (WW) AllowEmptyInput is on, devices using drivers 'kbd', 'mouse' or 'vmmouse' will be disabled. [ 2081.047] (WW) Disabling Keyboard0 [ 2081.047] (WW) Disabling Mouse0 [ 2081.047] (II) Loader magic: 0x7ce120 [ 2081.047] (II) Module ABI versions: [ 2081.047]X.Org ANSI C Emulation: 0.4 [ 2081.047]X.Org Video Driver: 8.0 [ 2081.047]X.Org XInput driver : 11.0 [ 2081.047]X.Org Server Extension : 4.0 [ 2081.048] (--) PCI:*(0:1:0:0) 10de:0a20:10de:069a rev 162, Mem @ 0xfb00/16777216, 0xc000/268435456, 0xde00/33554432, I/O @ 0xef00/128, BIOS @ 0x/524288 [ 2081.048] (II) LoadModule: "extmod" [ 2081.048] (II) Loading /usr/lib64/xorg/modules/extensions/libextmod.so [ 2081.048] (II) Module extmod: vendor="X.Org Foundation" [ 2081.048]compiled for 1.9.2.902, module version = 1.0.0 [ 2081.048]Module class: X.Org Server Extension [ 2081.048]ABI class: X.Org Server Extension, version 4.0 [ 2081.048] (II) Loading extension MIT-SCREEN-SAVER [ 2081.048] (II) Loading extension XFree86-VidModeExtension [ 2081.048] (II) Loading extension XFree86-DGA [ 2081.048] (II) Loading extension DPMS [ 2081.048] (II) Loading extension XVideo [ 2081.048] (II) Loading extension XVideo-MotionCompensation [ 2081.048] (II) Loading extension X-Resource [ 2081.048] (II) LoadModule: "dbe" [ 2081.048] (II) Loading /usr/lib64/xorg/modules/extensions/libdbe.so [ 2081.049] (II) Module dbe: vendor="X.Org Foundation" [ 2081.049]compiled for 1.9.2.902, module version = 1.0.0 [ 2081.
Re: [gentoo-user] New monitor problems. < sighs >
Peter Humphrey wrote: On Wednesday 15 December 2010 09:03:58 Dale wrote: I knew if I updated the kernel I had to re-emerge the nvidia-drivers but I didn't know it was the other way around too. No, it isn't. Not here at any rate. I tried it anyway. No change. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: On Wednesday 15 December 2010 10:08:30 Dale wrote: Neil Bothwick wrote: On Wed, 15 Dec 2010 02:40:48 -0600, Dale wrote: [ 2081.057] (EE) Failed to load module "dri" (module does not exist, 0) [ 2081.057] (EE) Failed to load module "dri2" (module does not exist, 0) [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: Function not [ 2082.101] (EE) NVIDIA(0): implemented [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does not exist, 0) [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA X driver not found) r...@smoker / # You haven't loaded the glx module, try this in xorg.conf Section "Module" Load "glx" Disable "dri" Disable "dri2" EndSection Also, what does "eselect opengl list" show? I reconfiged it so many times, I may have missed that the last time. I know it was in there the other times tho. I typed it all in a couple times. I'll give that a try again tho. Eselect shows nvidia. I always set it again to make sure tho. I ran into a problem with it switching on the old rig a while back. It also switches when I re-emerge the drivers too. I found that with google. It seems to work for everyone else. lol Thanks. Will post back what happens. I got a couple suggestions now. Dale :-) :-) Dale, Did you add that section to your xorg.conf? If yes, what is the xorg-log saying when you have that added? -- Joost I posted the new xorg.conf file and log file in another post. You should have it or get it any minute. Glad I got a memory stick to move those from one rig to another. ;-) Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 12:39:27 Dale wrote: > Pintér Tibor wrote: > > how about just checking the bloody Xorg logs to see what could have gone > > wrong? (/var/log/Xorg.0.log) > > That's where the errors came from that I posted in my first post. Here > is the results of the latest test. I shut down the rig and unplugged > the rig and monitor. Glad to see the monitor switched back to English > too. It was Chinese or something. Anyway. Before the shutdown, I > rebuilt the kernel, re-emerged the nvidia drivers and did a manual > switch on opengl to nvidia. I let the rig sit there unplugged for about > 30 minutes. I then booted it up. BIOS came up, I saw the services > scroll up then it switched to vt7 and gave me a blinking cursor. It > just sat there. > > I'm attaching both the new just tried xorg.conf and the xorg.log file. > No grep or anything this time. > > Let me know if you see anything fishy or that needs changing. > > Thanks. > > Dale > > :-) :-) Ok, check your kernel config. The following error into google gave me a hint, [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: Function not [ 2082.101] (EE) NVIDIA(0): implemented http://www.nvnews.net/vbulletin/showthread.php?t=118109 This one says: Try enabling CONFIG_SYSVIPC. -- Joost
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 12:39:27 Dale wrote: > Pintér Tibor wrote: > > how about just checking the bloody Xorg logs to see what could have gone > > wrong? (/var/log/Xorg.0.log) > > That's where the errors came from that I posted in my first post. Here > is the results of the latest test. I shut down the rig and unplugged > the rig and monitor. Glad to see the monitor switched back to English > too. It was Chinese or something. Anyway. Before the shutdown, I > rebuilt the kernel, re-emerged the nvidia drivers and did a manual > switch on opengl to nvidia. I let the rig sit there unplugged for about > 30 minutes. I then booted it up. BIOS came up, I saw the services > scroll up then it switched to vt7 and gave me a blinking cursor. It > just sat there. > > I'm attaching both the new just tried xorg.conf and the xorg.log file. > No grep or anything this time. > > Let me know if you see anything fishy or that needs changing. > > Thanks. > > Dale > > :-) :-) this page might also give a hint: http://www.nvnews.net/vbulletin/showthread.php?t=13734 Do you have "/dev/shm" on your system? -- Joost Ps. the third hit on google is this thread
Re: [gentoo-user] Strange boot problem - any ideas?
On Wednesday 15 December 2010 10:12:23 Helmut Jarausch wrote: > On 12/15/10 11:03:43, Adam Carter wrote: > > Are you using rc_parallel="NO" ? (in /etc/rc.conf) > > No, why disable fast booting. If it turns out to be necessary it > would be a pity! > But I'll try it for a while to see if it helps, On the other hand, why gamble on a clean boot for the sake of a couple of seconds? -- Rgds Peter. Linux Counter 5290, 1994-04-23.
Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 12:44:59 Dale wrote: > J. Roeleveld wrote: > > On Wednesday 15 December 2010 10:08:30 Dale wrote: > >> Neil Bothwick wrote: > >>> On Wed, 15 Dec 2010 02:40:48 -0600, Dale wrote: > [ 2081.057] (EE) Failed to load module "dri" (module does not exist, > 0) [ 2081.057] (EE) Failed to load module "dri2" (module does not > exist, 0) [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared > memory identifier: Function not > [ 2082.101] (EE) NVIDIA(0): implemented > [ 2082.118] (EE) NVIDIA: Failed to load module "dri2" (module does > not exist, 0) > [ 2082.120] (EE) Failed to initialize GLX extension (Compatible > NVIDIA X driver not found) > r...@smoker / # > >>> > >>> You haven't loaded the glx module, try this in xorg.conf > >>> > >>> Section "Module" > >>> > >>> Load "glx" > >>> Disable "dri" > >>> Disable "dri2" > >>> > >>> EndSection > >>> > >>> Also, what does "eselect opengl list" show? > >> > >> I reconfiged it so many times, I may have missed that the last time. I > >> know it was in there the other times tho. I typed it all in a couple > >> times. I'll give that a try again tho. > >> > >> Eselect shows nvidia. I always set it again to make sure tho. I ran > >> into a problem with it switching on the old rig a while back. It also > >> switches when I re-emerge the drivers too. I found that with google. > >> It seems to work for everyone else. lol > >> > >> Thanks. Will post back what happens. I got a couple suggestions now. > >> > >> Dale > >> > >> :-) :-) > > > > Dale, > > > > Did you add that section to your xorg.conf? > > If yes, what is the xorg-log saying when you have that added? > > > > -- > > Joost > > I posted the new xorg.conf file and log file in another post. You > should have it or get it any minute. > > Glad I got a memory stick to move those from one rig to another. ;-) > > Dale > > :-) :-) Saw it and replied. the "glx" appears to be loaded, as per that log, but found something else. See my other reply for that. -- Joost
Re: [gentoo-user] Questions about SATA and hot plugging.
Volker Armin Hemmann wrote: you seem to have misunderstood Mark. The power is delivered by the harddisks case. AKA external power supply. esata has 0 power distribution capabilities. Always remember: first unplug the sata cable, wait, then power. This allows the device to flush the cache. That was what I understood I just didn't type it in very well. Basically, it is not like USB. The connector only carries data and no power for devices. Since I don't have any external P/Ss then eSATA isn't doing me any good right now. Maybe down the road some day. Dale :-) :-)
Re: [gentoo-user] New monitor problems. < sighs >
> >>> >> Result: >> eve ~ # find /usr -name "libdri*" >> /usr/lib64/opengl/xorg-x11/extensions/libdri.so >> /usr/lib64/opengl/xorg-x11/extensions/libdri2.so >> >> eve ~ # cat /var/log/Xorg.0.log | grep EE >> (WW) warning, (EE) error, (NI) not implemented, (??) unknown. >> (II) Loading extension MIT-SCREEN-SAVER >> (EE) Failed to load module "xtrap" (module does not exist, 0) >> (EE) Failed to load module "dri" (module does not exist, 0) >> (EE) Failed to load module "dri2" (module does not exist, 0) >> >> The files do exist, but are not found as I am using the nvidia-opengl, >> rather >> then the xorg-x11 opengl. >> > DRI != opengl > >> > > I get the same here. Mine are in a xorg-x11 path but nothing for nvidia. > > Sorry - Total fail on my behalf, the dri libraries arent required for fglrx either - I just tried it.
Re: [gentoo-user] Questions about SATA and hot plugging.
Stroller wrote: On 14/12/2010, at 4:57pm, Mark Knecht wrote: The SATA spec allows for hot plugging, so technically yes ... My recollection of my understanding (multiple disclaimers) was that SATA *allowed* for SATA hot-plugging but didn't *mandate* it. If I could have found a SATA motherboard&/or controller with multiple ports that supported SATA hot-plugging then, when I built my storage server, I would have bought that and saved myself quite a bit of money against the hardware RAID card I bought instead. The impression I got when reading, however, was that many SATA ports can be hot-plugged, but their manufacturers don't explicitly state the fact, so one can't be 100% sure it's safe to do so. Also: learn to snip, guys. Stroller. According to the mobo manual, if I enable AHCI, it is hot swappable. I just ain't to comfy doing it. I'd like to see it done with no smoke getting out first. Dale :-) :_)
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, Dec 15, 2010 at 05:39:27AM -0600, Dale wrote: ... > > I'm attaching both the new just tried xorg.conf and the xorg.log > file. No grep or anything this time. > > Let me know if you see anything fishy or that needs changing. > > Thanks. > > Dale ... > > [ 2082.083] (II) Loading extension NV-GLX > [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: > Function not > [ 2082.101] (EE) NVIDIA(0): implemented ... > [ 2082.120] (EE) Failed to initialize GLX extension (Compatible NVIDIA X > driver not found) > [ 2082.138] > Fatal server error: > [ 2082.138] Failed to initialize the OpenGL server this looks like the problem, but I don't know whats wrong (it certainly should not be a mismatch between nvidia X driver and nvidia kernel module, that produces a different error ;) google seems to find a lot of stuff for the "Failed to obtain", but I didn't look into it.. yoyo
!! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: On Wednesday 15 December 2010 12:39:27 Dale wrote: Pintér Tibor wrote: how about just checking the bloody Xorg logs to see what could have gone wrong? (/var/log/Xorg.0.log) That's where the errors came from that I posted in my first post. Here is the results of the latest test. I shut down the rig and unplugged the rig and monitor. Glad to see the monitor switched back to English too. It was Chinese or something. Anyway. Before the shutdown, I rebuilt the kernel, re-emerged the nvidia drivers and did a manual switch on opengl to nvidia. I let the rig sit there unplugged for about 30 minutes. I then booted it up. BIOS came up, I saw the services scroll up then it switched to vt7 and gave me a blinking cursor. It just sat there. I'm attaching both the new just tried xorg.conf and the xorg.log file. No grep or anything this time. Let me know if you see anything fishy or that needs changing. Thanks. Dale :-) :-) Ok, check your kernel config. The following error into google gave me a hint, [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: Function not [ 2082.101] (EE) NVIDIA(0): implemented http://www.nvnews.net/vbulletin/showthread.php?t=118109 This one says: Try enabling CONFIG_SYSVIPC. -- Joost You da man !!! That was it. Yeppie !!! I got my nice new monitor back. I bet this was what was wrong with the other monitor too. I'm happy so I'm not going to try it. Whew. I was about ready to try the ATI card too. Yuck !! lol THANKS MUCH !! Dale :-) :-)
Re: !! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs >
On Wednesday 15 December 2010 13:37:33 Dale wrote: > J. Roeleveld wrote: > > On Wednesday 15 December 2010 12:39:27 Dale wrote: > >> Pintér Tibor wrote: > >>> how about just checking the bloody Xorg logs to see what could have > >>> gone wrong? (/var/log/Xorg.0.log) > >> > >> That's where the errors came from that I posted in my first post. Here > >> is the results of the latest test. I shut down the rig and unplugged > >> the rig and monitor. Glad to see the monitor switched back to English > >> too. It was Chinese or something. Anyway. Before the shutdown, I > >> rebuilt the kernel, re-emerged the nvidia drivers and did a manual > >> switch on opengl to nvidia. I let the rig sit there unplugged for about > >> 30 minutes. I then booted it up. BIOS came up, I saw the services > >> scroll up then it switched to vt7 and gave me a blinking cursor. It > >> just sat there. > >> > >> I'm attaching both the new just tried xorg.conf and the xorg.log file. > >> No grep or anything this time. > >> > >> Let me know if you see anything fishy or that needs changing. > >> > >> Thanks. > >> > >> Dale > >> > >> :-) :-) > > > > Ok, > > > > check your kernel config. > > > > The following error into google gave me a hint, > > [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: > > Function not > > [ 2082.101] (EE) NVIDIA(0): implemented > > > > > > http://www.nvnews.net/vbulletin/showthread.php?t=118109 > > > > This one says: > > Try enabling CONFIG_SYSVIPC. > > > > -- > > Joost > > You da man !!! That was it. Yeppie !!! I got my nice new monitor > back. I bet this was what was wrong with the other monitor too. I'm > happy so I'm not going to try it. > > Whew. I was about ready to try the ATI card too. Yuck !! lol > > THANKS MUCH !! > > Dale > > :-) :-) Am glad you got it working. Enjoy the new machine :) -- Joost
Re: !! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs >
J. Roeleveld wrote: On Wednesday 15 December 2010 13:37:33 Dale wrote: J. Roeleveld wrote: Ok, check your kernel config. The following error into google gave me a hint, [ 2082.101] (EE) NVIDIA(0): Failed to obtain a shared memory identifier: Function not [ 2082.101] (EE) NVIDIA(0): implemented http://www.nvnews.net/vbulletin/showthread.php?t=118109 This one says: Try enabling CONFIG_SYSVIPC. -- Joost You da man !!! That was it. Yeppie !!! I got my nice new monitor back. I bet this was what was wrong with the other monitor too. I'm happy so I'm not going to try it. Whew. I was about ready to try the ATI card too. Yuck !! lol THANKS MUCH !! Dale :-) :-) Am glad you got it working. Enjoy the new machine :) -- Joost I went back and looked at the nvidia driver guide on gentoo.org. There is no mention of nvidia needing that. Should I tell the doc team so they can mention that for others or is my machine a little unique? Since someone else ran into it on the link you posted, it may be something that at least needs a mention even tho it is not nvidia specific. Thoughts? I like this thing. The rig is super fast and the monitor is really nice. These bad eyes can read this better. Thanks. Dale :-) :-)
Re: !! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs>
On Wednesday 15 December 2010 08:26:11 Dale wrote: > J. Roeleveld wrote: > > On Wednesday 15 December 2010 13:37:33 Dale wrote: > >> J. Roeleveld wrote: > > > > Am glad you got it working. Enjoy the new machine :) > > > > -- > > Joost > > I went back and looked at the nvidia driver guide on gentoo.org. There > is no mention of nvidia needing that. Should I tell the doc team so > they can mention that for others or is my machine a little unique? > Since someone else ran into it on the link you posted, it may be > something that at least needs a mention even tho it is not nvidia specific. > > Thoughts? Might be usefull to have it in the docs, but I think, as it is something other software requires as well, it might be an idea to stick it in the install- guide and in the ebuild-notes as well? Am wondering, isn't it in the X11 documents on the site yet? > I like this thing. The rig is super fast and the monitor is really > nice. These bad eyes can read this better. I've got the same CPU as you have, actually, just on an older mainboard. When I bought mine, I did stock it full with the max memory the board can take (8 gig). Not sure how much memory you have, but a 6Gig ramdisk mounted at /var/tmp/portage is sufficient to compile openoffice. That speeds things up even more ;) -- Joost
[gentoo-user] Weird (?) permission problem...
Hi, my /tmp has the permissions set to 1777 . And it is the mountpoint for an extra partitions holding the stuff of '/tmp/' When booting into single-user mode and unmounting /tmp and doing a ls -ld /tmp it shows drwxrwxrwt 45 root root 61440 2010-12-15 15:07 /tmp BUT as soon as I mount the device on /tmp and doing the same ls -ld again it shows drwsrwsrwt 45 root root 61440 2010-12-15 15:07 /tm That looks not ok to me. /etc/fstab has an entry which "options" field is set to "default" And this happens to any mountpoint I mount that device on regardless of its perm settings before the mount What is the reason for this? Best regards, mcc
Re: !! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs>
J. Roeleveld wrote: On Wednesday 15 December 2010 08:26:11 Dale wrote: J. Roeleveld wrote: On Wednesday 15 December 2010 13:37:33 Dale wrote: J. Roeleveld wrote: Am glad you got it working. Enjoy the new machine :) -- Joost I went back and looked at the nvidia driver guide on gentoo.org. There is no mention of nvidia needing that. Should I tell the doc team so they can mention that for others or is my machine a little unique? Since someone else ran into it on the link you posted, it may be something that at least needs a mention even tho it is not nvidia specific. Thoughts? Might be usefull to have it in the docs, but I think, as it is something other software requires as well, it might be an idea to stick it in the install- guide and in the ebuild-notes as well? Am wondering, isn't it in the X11 documents on the site yet? I'll post it on gentoo-doc and see what they say. I like this thing. The rig is super fast and the monitor is really nice. These bad eyes can read this better. I've got the same CPU as you have, actually, just on an older mainboard. When I bought mine, I did stock it full with the max memory the board can take (8 gig). Not sure how much memory you have, but a 6Gig ramdisk mounted at /var/tmp/portage is sufficient to compile openoffice. That speeds things up even more ;) -- Joost Mine maxes out at 16Gbs. I have a single 4Gb stick in it right now. I plan to add a stick every few months until I get it full. Then I will have plenty of room to put portage on the ramdisk like you. It should be pretty fast then. I rarely use more than 1Gb tho. Even then, I have a lot of images open in Gimp or something to use that much. Now that my rig is fixed, I'm going to take a nap. I'm beat. Dale :-) :-)
Re: [gentoo-user] Weird (?) permission problem...
Apparently, though unproven, at 16:13 on Wednesday 15 December 2010, meino.cra...@gmx.de did opine thusly: > Hi, > > my /tmp has the permissions set to 1777 . And it is the mountpoint for > an extra partitions holding the stuff of '/tmp/' > > > When booting into single-user mode and unmounting /tmp and doing > a > > ls -ld /tmp > > it shows > > drwxrwxrwt 45 root root 61440 2010-12-15 15:07 /tmp > > BUT > > as soon as I mount the device on /tmp and doing the same ls -ld > again it shows > > > drwsrwsrwt 45 root root 61440 2010-12-15 15:07 /tm > > > That looks not ok to me. > > /etc/fstab has an entry which "options" field is set to "default" > > And this happens to any mountpoint I mount that device on regardless > of its perm settings before the mount > > > What is the reason for this? What filesystem type? You probably have some defaults set that make it suid/sgid, but these things are filesystem-dependant and for that we need to know the type -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Weird (?) permission problem...
On Wed, 15 Dec 2010 15:13:31 +0100, meino.cra...@gmx.de wrote: > And this happens to any mountpoint I mount that device on regardless > of its perm settings before the mount With nothing mounted on it, the mount point's permission are those of the directory. As soon as you mount something on it, the mount point has the ownership and permissions of the root of the filesystem that you just mounted there. In the same way that the contents of the filesystem appear at the mount point, so does the metadata, so change the permissions after mounting. -- Neil Bothwick A real programmer never documents his code. It was hard to make, it should be hard to read signature.asc Description: PGP signature
Re: [gentoo-user] Weird (?) permission problem...
Neil Bothwick [10-12-15 15:40]: > On Wed, 15 Dec 2010 15:13:31 +0100, meino.cra...@gmx.de wrote: > > > And this happens to any mountpoint I mount that device on regardless > > of its perm settings before the mount > > With nothing mounted on it, the mount point's permission are those of the > directory. As soon as you mount something on it, the mount point has the > ownership and permissions of the root of the filesystem that you just > mounted there. In the same way that the contents of the filesystem appear > at the mount point, so does the metadata, so change the permissions > after mounting. > > > -- > Neil Bothwick > > A real programmer never documents his code. > It was hard to make, it should be hard to read ...unfortunately (as root) cd /tmp chmod 1777 . does not help... mcc
Re: [gentoo-user] Weird (?) permission problem...
On Wednesday 15 December 2010 15:41:25 meino.cra...@gmx.de wrote: > Neil Bothwick [10-12-15 15:40]: > > On Wed, 15 Dec 2010 15:13:31 +0100, meino.cra...@gmx.de wrote: > > > And this happens to any mountpoint I mount that device on regardless > > > of its perm settings before the mount > > > > With nothing mounted on it, the mount point's permission are those of the > > directory. As soon as you mount something on it, the mount point has the > > ownership and permissions of the root of the filesystem that you just > > mounted there. In the same way that the contents of the filesystem > > appear at the mount point, so does the metadata, so change the > > permissions after mounting. > > ...unfortunately (as root) > > cd /tmp > chmod 1777 . > > does not help... I don't think you can change the permissions like that. Try: cd / chmod 1777 /tmp To remove the "s"-bits, try the following: cd / chmod u-s /tmp chmod g-s /tmp This, however, needs to be done while the "/tmp" filesystem is mounted. Otherwise you are only changing the mount-point (directory) not the actual filesystem. -- Joost
Re: !! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs>
On Wednesday 15 December 2010 13:33:03 J. Roeleveld wrote: > Not sure how much memory you have, but a 6Gig ramdisk mounted at > /var/tmp/portage is sufficient to compile openoffice. That speeds > things up even more ;) The same size of tmpfs on /tmp also works even if you have only 4GB RAM, as I have. When it gets too full it just starts swapping. Actually, now that I check again, I see I've raised the size to 16G: $ grep 16G /etc/fstab tmpfs /tmp tmpfs nodev,nosuid,size=16G 0 0 $ grep swap /etc/fstab /dev/sda3 none swap sw,pri=10 0 0 /dev/sdb3 none swap sw,pri=10 0 0 /dev/sda7 none swap sw,pri=10 0 /dev/sdb7 none swap sw,pri=10 0 /dev/sdX3 are 2GB and /dev/sdX7 are 20GB (probably far too much swap, but disks are cheap). -- Rgds Peter. Linux Counter 5290, 1994-04-23.
Re: [gentoo-user] Weird (?) permission problem...
J. Roeleveld [10-12-15 16:00]: > On Wednesday 15 December 2010 15:41:25 meino.cra...@gmx.de wrote: > > Neil Bothwick [10-12-15 15:40]: > > > On Wed, 15 Dec 2010 15:13:31 +0100, meino.cra...@gmx.de wrote: > > > > And this happens to any mountpoint I mount that device on regardless > > > > of its perm settings before the mount > > > > > > With nothing mounted on it, the mount point's permission are those of the > > > directory. As soon as you mount something on it, the mount point has the > > > ownership and permissions of the root of the filesystem that you just > > > mounted there. In the same way that the contents of the filesystem > > > appear at the mount point, so does the metadata, so change the > > > permissions after mounting. > > > > ...unfortunately (as root) > > > > cd /tmp > > chmod 1777 . > > > > does not help... > > I don't think you can change the permissions like that. > Try: > cd / > chmod 1777 /tmp > > To remove the "s"-bits, try the following: > cd / > chmod u-s /tmp > chmod g-s /tmp > > This, however, needs to be done while the "/tmp" filesystem is mounted. > Otherwise you are only changing the mount-point (directory) not the actual > filesystem. > > -- > Joost > interesting... Until now, I thought '.' is equal to the directory I am in. Ok, times is changing, me too, but as it seems not fast enough ;) Thanks a lot... thats fix it! Best regards, mcc
Re: [gentoo-user] Weird (?) permission problem...
On Wednesday 15 December 2010 16:20:32 meino.cra...@gmx.de wrote: > J. Roeleveld [10-12-15 16:00]: > > On Wednesday 15 December 2010 15:41:25 meino.cra...@gmx.de wrote: > > > Neil Bothwick [10-12-15 15:40]: > > > > On Wed, 15 Dec 2010 15:13:31 +0100, meino.cra...@gmx.de wrote: > > > > > And this happens to any mountpoint I mount that device on > > > > > regardless of its perm settings before the mount > > > > > > > > With nothing mounted on it, the mount point's permission are those of > > > > the directory. As soon as you mount something on it, the mount point > > > > has the ownership and permissions of the root of the filesystem that > > > > you just mounted there. In the same way that the contents of the > > > > filesystem appear at the mount point, so does the metadata, so > > > > change the permissions after mounting. > > > > > > ...unfortunately (as root) > > > > > > cd /tmp > > > chmod 1777 . > > > > > > does not help... > > > > I don't think you can change the permissions like that. > > Try: > > cd / > > chmod 1777 /tmp > > > > To remove the "s"-bits, try the following: > > cd / > > chmod u-s /tmp > > chmod g-s /tmp > > > > This, however, needs to be done while the "/tmp" filesystem is mounted. > > Otherwise you are only changing the mount-point (directory) not the > > actual filesystem. > > > > -- > > Joost > > interesting... > Until now, I thought '.' is equal to the directory I am in. That's true, but not entirely :) I don't think "chmod" is supposed to work that way :) > Ok, times is changing, me too, but as it seems not fast enough ;) Times are changing, so are people, but there are too many changes occuring for people to pick the "right" changes :) > Thanks a lot... thats fix it! You're welcome :) -- Joost
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, Dec 15, 2010 at 2:40 AM, Dale wrote: > Hi again, > > For those mot up to date. I built a new rig that has a Nvidia GT220 card in > it. I bought a brand new monitor this morning, a LG W2253, and it worked > one time. I had to reboot to move some things around and when I rebooted, > the GUI doesn't come up. The BIOS screen shows up and I can see the > services start up as well but when it switches to vt7, it just has a little > blinking cursor at the top. At first glance my guess is that your kernel is configured incorrectly somehow. I have a similar card (Nvidia GT 240). I am using fully ~amd64 system and it all works. FWIW, below are my settings & a couple thoughts too. Right away I think you need to enable the glx module in your xorg.conf. I don't think the monitor and screen sections are necessary at all (it should autodetect everything if you have a modern monitor with EDID support (anything from past 10 years probably has it), and forcing its hsync/vsync could be a cause of problems if they are incorrect). So I'd try removing those. The README that comes with the nvidia-drivers has answers to many of the common error messages, including one of yours: from /usr/share/doc/nvidia-drivers-260.19.21/README.bz2: Q. X crashes during 'startx', and my X log file contains this error message: (EE) NVIDIA(0): Failed to obtain a shared memory identifier. A. The NVIDIA OpenGL driver and the NVIDIA X driver require shared memory to communicate; you must have 'CONFIG_SYSVIPC' enabled in your kernel. One thing that I don't know if anyone suggested is to check your framebuffer settings in kernel. I seem to remember that enabling the nvidia framebuffer support in kernel could prevent the drivers from loading. I'm personally using uvesafb from sys-apps/v86d to get a high-res console and it works nicely with the closed nvidia-drivers. here's what I get on my working system: /etc/X11/xorg.conf in its entirety: Section "Module" Load "glx" EndSection Section "Device" Identifier "nVidia GT 240" Driver "nvidia" Option "NoLogo" "1" EndSection "dmesg | grep nvidia" shows me this: [9.485269] nvidia: module license 'NVIDIA' taints kernel. [ 10.118846] nvidia :03:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 [ 10.118857] nvidia :03:00.0: setting latency timer to 64 "grep -i nvidia /var/log/Xorg.0.log" shows this: [49.453] (**) | |-->Device "nVidia GT 240" [50.891] (II) Module glx: vendor="NVIDIA Corporation" [50.906] (II) NVIDIA GLX Module 260.19.21 Thu Nov 4 21:42:11 PDT 2010 [51.008] (II) LoadModule: "nvidia" [51.008] (II) Loading /usr/lib64/xorg/modules/drivers/nvidia_drv.so [51.126] (II) Module nvidia: vendor="NVIDIA Corporation" [51.205] (II) NVIDIA dlloader X Driver 260.19.21 Thu Nov 4 21:18:43 PDT 2010 [51.205] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs [51.400] (II) NVIDIA(0): Creating default Display subsection in Screen section [51.400] (==) NVIDIA(0): Depth 24, (==) framebuffer bpp 32 [51.400] (==) NVIDIA(0): RGB weight 888 [51.400] (==) NVIDIA(0): Default visual is TrueColor [51.400] (==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0) [51.400] (**) NVIDIA(0): Option "NoLogo" "1" [51.401] (**) NVIDIA(0): Enabling RENDER acceleration [51.401] (II) NVIDIA(0): Support for GLX with the Damage and Composite X extensions is [51.401] (II) NVIDIA(0): enabled. [52.061] (II) NVIDIA(0): NVIDIA GPU GeForce GT 240 (GT215) at PCI:3:0:0 (GPU-0) [52.061] (--) NVIDIA(0): Memory: 1048576 kBytes [52.061] (--) NVIDIA(0): VideoBIOS: 70.15.24.00.00 [52.061] (II) NVIDIA(0): Detected PCI Express Link width: 16X [52.061] (--) NVIDIA(0): Interlaced video modes are supported on this GPU [52.061] (--) NVIDIA(0): Connected display device(s) on GeForce GT 240 at PCI:3:0:0 [52.061] (--) NVIDIA(0): DELL SP2309W (DFP-0) [52.061] (--) NVIDIA(0): DELL SP2309W (DFP-0): 330.0 MHz maximum pixel clock [52.061] (--) NVIDIA(0): DELL SP2309W (DFP-0): Internal Dual Link TMDS [52.107] (II) NVIDIA(0): Assigned Display Device: DFP-0 [52.107] (==) NVIDIA(0): [52.107] (==) NVIDIA(0): No modes were requested; the default mode "nvidia-auto-select" [52.107] (==) NVIDIA(0): will be used as the requested mode. [52.107] (==) NVIDIA(0): [52.107] (II) NVIDIA(0): Validated modes: [52.107] (II) NVIDIA(0): "nvidia-auto-select" [52.107] (II) NVIDIA(0): Virtual screen size determined to be 2048 x 1152 [52.138] (--) NVIDIA(0): DPI set to (101, 100); computed from "UseEdidDpi" X config [52.139] (--) NVIDIA(0): option [52.139] (==) NVIDIA(0): Enabling 32-bit ARGB GLX visuals. [52.139] (II) NVIDIA: Using 768.00 MB of virtual memory for indirect memory access. [52.140] (II) NVIDIA(0): Initialized GPU GART. [52.147] (II) NVIDIA(0): Setting mode "nvidia-auto-select" [52.219] (II) NVIDIA(0): Initialized OpenGL Acceler
Re: [gentoo-user] Weird (?) permission problem...
Apparently, though unproven, at 17:20 on Wednesday 15 December 2010, meino.cra...@gmx.de did opine thusly: > J. Roeleveld [10-12-15 16:00]: > > On Wednesday 15 December 2010 15:41:25 meino.cra...@gmx.de wrote: > > > Neil Bothwick [10-12-15 15:40]: > > > > On Wed, 15 Dec 2010 15:13:31 +0100, meino.cra...@gmx.de wrote: > > > > > And this happens to any mountpoint I mount that device on > > > > > regardless of its perm settings before the mount > > > > > > > > With nothing mounted on it, the mount point's permission are those of > > > > the directory. As soon as you mount something on it, the mount point > > > > has the ownership and permissions of the root of the filesystem that > > > > you just mounted there. In the same way that the contents of the > > > > filesystem appear at the mount point, so does the metadata, so > > > > change the permissions after mounting. > > > > > > ...unfortunately (as root) > > > > > > cd /tmp > > > chmod 1777 . > > > > > > does not help... > > > > I don't think you can change the permissions like that. > > Try: > > cd / > > chmod 1777 /tmp > > > > To remove the "s"-bits, try the following: > > cd / > > chmod u-s /tmp > > chmod g-s /tmp > > > > This, however, needs to be done while the "/tmp" filesystem is mounted. > > Otherwise you are only changing the mount-point (directory) not the > > actual filesystem. > > > > -- > > Joost > > interesting... > Until now, I thought '.' is equal to the directory I am in. Usually it is, this is a special case Every other action you could do with it resolves to the same thing no matter what point of view you take. chmod/chown changes the filesystem or mount point, which are different things. So there's two command interpretations. It's all quite logical once you've figured it out but even then most of us still never remember which is which... > > Ok, times is changing, me too, but as it seems not fast enough ;) > > Thanks a lot... thats fix it! > > Best regards, > mcc -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Questions about SATA and hot plugging.
E-SATA != SATA Nah. They are *exactly* the same. Evidently someone realized that the original SATA connector is way too fragile to be regularly used to plug/unplug a cable by hand, so they engineered in some features which make it a bit more resilient. But apart from the shape of the connector there is really no difference. (Well, in the old days of SATAI not many chipsets supported hotplug; often boards came with a couple of eSATA ports wired to a separate chip with hotplug support. But on virtually all new boards all ports support hotplug). andrea
Re: [gentoo-user] OT HD video camera & Gentoo
On Wed, Dec 1, 2010 at 3:20 PM, James wrote: > Howdy, > > I'm going to get myself an HD video camera, that > works well with open source software on Gentoo. > Problem is (head scratching), I recall when I went > through this some years ago I had lots of problems > making a decision that worked with linux. I ended up with a > Sony DCR-SR42. It works and the disk is mounted > via the usb port. I can cd into the /dev/disk/ > mount point and used scp to remotely copy > the mpg files and jpg photos to any system > on the network. That to me is minimal functionality. > > > Surely somebody has a HD video camera that works > extra cool with Gentoo and gives a variety of video output > formats, like h.264; or the video output can be > easily converted to h.264 and other formats? > > > Surely, there is an easy method (software app) to just > burn a dvd from the contents of the video camera directly > with (gentoo) linux; just like I can do, from the old > sony(XP) software that came with the camera? > > > Any recommendations are most welcome. I mostly record > fast moving sports, like basketball, football and > water skiing, so the 'debouncing technologies' or > motion compensation if you like, are also of keen > interest to me. Hi James, sorry nobody replied to your message yet. I was on vacation but I have an HD video camera and have used it in Gentoo to edit the videos etc. :) I have a Canon Vixia HG10 camera, which is a bit old by now and I don't think it is made anymore (the newer models in the same line are similar but better in basically every way). It does 720p and "1080i" (I put the latter in quotes because the actual resolution is anamorphic 16:9 in 1440x1080, not the 1920x1080 that one might expect when the packaging proclaims its 1080 HD support...). It has a hard drive and stores the movies in avchd format. It gets mounted as a normal USB mass storage device, no monkey business is necessary in getting to the files. In fact, this format is H.264 video and AC3 audio already, and from what I understand it is compatible with blu-ray players. You can simply burn the files onto a disc (even a DVD if the video is short enough) and play it in HD on your blu-ray player, or play them on your PC with mplayer. I don't have a BR player so I've never tried (you can also hook the camera directly to the HDMI port of your TV). I think that any camera that stores its files in avchd format would be just as easy to use from that standpoint. >From an editing-in-linux standpoint, by far the best software I've found is kdenlive, which is free and open-source and uses mlt as its video processing backend. It supports the modern HD formats and is very actively developed, and lets you do everything you'd expect from a video editing software. It can export files in almost any format imaginable, and has presets for most of them, including mobile devices and sharing sites like youtube and vimeo in addition to the standard files like avi, mkv, etc. It also has a transcoder if you simply want to convert files as-is. There is a pretty active mailing list as well. One thing to keep in mind is that dealing with HD video is hugely processor and disk space intensive. If you don't have at least a Core 2, or equivalent, or newer, it might be prohibitively slow. Be prepared to use up hundreds of gigs of disk space, too. Processing time is likely to be several times longer than the length of the video itself. Good luck!
Re: [gentoo-user] Questions about SATA and hot plugging.
The SATA spec allows for hot plugging, so technically yes ...
My recollection of my understanding (multiple disclaimers) was that SATA
*allowed* for SATA hot-plugging but didn't *mandate* it.
a good summary of the hardware/driver situation wrt hotplugging can be
found here:
https://ata.wiki.kernel.org/index.php/SATA_hardware_features
Short version: most controllers nowadays support hotplug, provided they
are not operated in compatibility ("IDE") mode.
We have quite a number of software RAID setups with SATA disks in
hot-swap backplanes; so far we found that hotplug works quite reliably
on Intel (ICH9R/ICH10R), AMD (SB700/SB800) and Silicon Image (sil3132)
controllers.
andrea
[gentoo-user] OT: cheap "make yourself" server: i7-950 or phenom 1100T?
Hi, a friend of mine asked me to prepare a small server for him. Unfortunatelly he can not afford to buy brand-name server so he asked me to build one for him, from "consumer" components (yes, I already warned him about "zero-support" consequences). It should be some kind of "multi-purpose" server (web, ftp, mail, dns, virtualisation, etc). His budget is ~600-700€ (for cpu, mobo, ram), and he wants the best value for the money... Now, the crucial decision is what cpu (&mobo) I should use: A: Intel Core i7-950, 4x 3.06GHz (4 cores, + hyper-threading) B: AMD Phenom II X6 1100T Black Edition, 6x 3.30GHz (6 cores) Is it better to use phenom with 6 true cores, or i7 with 4 real and 4 "fake" cores (hyper-threading)? Concerning price, there is no difference, both of the above mentioned cpus cost ~250€ here in Europe. btw, mobos for phenom have up to 4x dimm, while mobos for core-i7 can have up to 6x dimm (that might be a valid point, he is going to need a lot of memory). So what should I pick for him? i7-950, or phenom-1100t? Or yet some cheap 4/6-core opteron 4xxx/6xxx? Jarry -- ___ This mailbox accepts e-mails only from selected mailing-lists! Everything else is considered to be spam and therefore deleted.
Re: [gentoo-user] Questions about SATA and hot plugging.
On Wednesday 15 December 2010 05:54:40 Dale wrote: > > According to the mobo manual, if I enable AHCI, it is hot swappable. and enabling AHCI is the only sane option. So do it. > I > just ain't to comfy doing it. I'd like to see it done with no smoke > getting out first. no smoke. Worst case: controller hangs, you have to reboot.
Re: [gentoo-user] Questions about SATA and hot plugging.
On Wed, Dec 15, 2010 at 9:49 AM, Volker Armin Hemmann wrote: > On Wednesday 15 December 2010 05:54:40 Dale wrote: > >> >> According to the mobo manual, if I enable AHCI, it is hot swappable. > > and enabling AHCI is the only sane option. So do it. > >> I >> just ain't to comfy doing it. I'd like to see it done with no smoke >> getting out first. > > no smoke. Worst case: controller hangs, you have to reboot. > > The worst case is _slightly_ worse than that, but certainly no smoke. Some systems (my 6 drive RAID compute server for instance) changes drive mapping between AHCI and compatibility modes so I had to adjust /etc/fstab. If Dale is using labels of some type he will likely be better off than I was. There certainly won't be any harm caused by changing the BIOS setting to AHCI and trying it out. Reboot and change BIOS back is all he would have to do in my experience. - Mark
Re: [gentoo-user] Remove redundant entries in "world" - howto
The following lines creates an auditWorldFile.log log file which will
show packages requires by other packages, so theones you can safely
remove.
#!/bin/bash
n=`wc -l /var/lib/portage/world|awk '{ print $1 }'`;
for i in `seq 1 $n`;do
pkg=`cat /var/lib/portage/world|head -n$i|tail -n1`;
echo -e "Packages depending on $pkg." >> /tmp/auditWorldFile.log
equery d $pkg >> /tmp/auditWorldFile.log
echo -e "" >> /tmp/auditWorldFile.log
done;
2010/12/8 Johannes Kimmel :
> On 12/08/2010 12:23 PM, Helmut Jarausch wrote:
>>
>> Hi,
>>
>> does anybody know about an easy method to remove all entries from
>> /var/lib/portage/world
>> which would have been pulled in anyway
>> even if they were not contained in world.
>>
>> My current attempt would be to write a script
>> which executes emerge -vpc on each entry in world.
>> If it wouldn't be removed it's obsolete in world.
>>
>> Unfortunately this has to be done in several rounds.
>>
>> Many thanks for a hint,
>> Helmut.
>>
>>
>
> Hi,
>
> I wanted to add, that a minimal world in my opinion isn't always what you
> want. For example in the time I searched for a suitable window manager for
> me I did a lot of depclean these days and accidently removed the xserver.
> There was no harm done, but I figured, that my world file should contain all
> packages, that should never removed automatically unless I want to. This way
> there is a little less danger involved using depclean. So this type of work
> might only be done by hand.
>
> Johannes Kimmel
>
>
Re: [gentoo-user] Questions about SATA and hot plugging.
Mark Knecht wrote: On Wed, Dec 15, 2010 at 9:49 AM, Volker Armin Hemmann wrote: On Wednesday 15 December 2010 05:54:40 Dale wrote: According to the mobo manual, if I enable AHCI, it is hot swappable. and enabling AHCI is the only sane option. So do it. I just ain't to comfy doing it. I'd like to see it done with no smoke getting out first. no smoke. Worst case: controller hangs, you have to reboot. The worst case is _slightly_ worse than that, but certainly no smoke. Some systems (my 6 drive RAID compute server for instance) changes drive mapping between AHCI and compatibility modes so I had to adjust /etc/fstab. If Dale is using labels of some type he will likely be better off than I was. There certainly won't be any harm caused by changing the BIOS setting to AHCI and trying it out. Reboot and change BIOS back is all he would have to do in my experience. - Mark Are the drives any faster when using AHCI tho? If the speed is the same then I may try it next time I reboot but not real sure why it would matter. I was hoping for something even faster. Dale :-) :-)
Re: [gentoo-user] OT: cheap "make yourself" server: i7-950 or phenom 1100T?
Jarry [10-12-15 19:08]: > Hi, > > a friend of mine asked me to prepare a small server for him. > Unfortunatelly he can not afford to buy brand-name server so > he asked me to build one for him, from "consumer" components > (yes, I already warned him about "zero-support" consequences). > It should be some kind of "multi-purpose" server (web, ftp, > mail, dns, virtualisation, etc). His budget is ~600-700€ (for > cpu, mobo, ram), and he wants the best value for the money... > > Now, the crucial decision is what cpu (&mobo) I should use: > > A: Intel Core i7-950, 4x 3.06GHz (4 cores, + hyper-threading) > B: AMD Phenom II X6 1100T Black Edition, 6x 3.30GHz (6 cores) > > Is it better to use phenom with 6 true cores, or i7 with > 4 real and 4 "fake" cores (hyper-threading)? Concerning price, > there is no difference, both of the above mentioned cpus cost > ~250€ here in Europe. > > btw, mobos for phenom have up to 4x dimm, while mobos for > core-i7 can have up to 6x dimm (that might be a valid point, > he is going to need a lot of memory). > > So what should I pick for him? i7-950, or phenom-1100t? > Or yet some cheap 4/6-core opteron 4xxx/6xxx? > > Jarry > > -- > ___ > This mailbox accepts e-mails only from selected mailing-lists! > Everything else is considered to be spam and therefore deleted. > I think the formula "performance/money" is fitted better by AMD than by Intel. If you take "performance" and "forget the money" Intel will be your friend. I myself choose an AMD Phenom X6 1090T (which can easily by pushed to be a 1100T by the way) on a ASUS Crosshair IV Formula. But a few days agao I heard Gigabyte would be more AMD friendly... I uses this mainly for rendering -- all cores can be used by Blender in parallel. Only my two cent ... you currency may vary. I DONT WANT to start a flamewar here! Its only my opinion I wanted to express :) Best regards, mcc PS: If the siftware you will use is not capable to put full load on the machine you will pay for more hardware than it is used. PPS: Hyperthreading uses unused parts of a core to run stuff which does not need the used parts of the same core. When you have 8 identical jobs running, there have to be 8 identical parts in the cores available otherwise there is nothing to hypethread. It depends heavily on the kind and mixture of jobs running on the machine whether hyperthreading is a win or a marketing joke... I myself (own opinion) feel better to have six physical cores capable of running six identical threads doing six things real parallel, than to guess, whether 4 of the eight threads Blender is showing me is /possibly/ waiting for getting access to an unused part of one of the four cores. Yes, I am an AMD friend since Intel way of "handling" the P90 floating point bug for their customers. And since this 15 (?) years this decision was ok -- at least for me. ;)
Re: [gentoo-user] New monitor problems. < sighs >
Paul Hartman wrote: On Wed, Dec 15, 2010 at 2:40 AM, Dale wrote: Hi again, For those mot up to date. I built a new rig that has a Nvidia GT220 card in it. I bought a brand new monitor this morning, a LG W2253, and it worked one time. I had to reboot to move some things around and when I rebooted, the GUI doesn't come up. The BIOS screen shows up and I can see the services start up as well but when it switches to vt7, it just has a little blinking cursor at the top. At first glance my guess is that your kernel is configured incorrectly somehow. I have a similar card (Nvidia GT 240). I am using fully ~amd64 system and it all works. FWIW, below are my settings& a couple thoughts too. << SNIP>> I can send you my kernel .config if you want. Let me know! Good luck :) You are correct, it was a kernel config issue. Nvidia needs CONFIG_SYSVIPC enabled in the kernel to work correctly. Dale :-) :-)
Re: [gentoo-user] Questions about SATA and hot plugging.
On Wed, Dec 15, 2010 at 10:24 AM, Dale wrote: > Mark Knecht wrote: >> >> On Wed, Dec 15, 2010 at 9:49 AM, Volker Armin Hemmann >> wrote: >> >>> >>> On Wednesday 15 December 2010 05:54:40 Dale wrote: >>> >>> According to the mobo manual, if I enable AHCI, it is hot swappable. >>> >>> and enabling AHCI is the only sane option. So do it. >>> >>> I just ain't to comfy doing it. I'd like to see it done with no smoke getting out first. >>> >>> no smoke. Worst case: controller hangs, you have to reboot. >>> >>> >>> >> >> The worst case is _slightly_ worse than that, but certainly no smoke. >> Some systems (my 6 drive RAID compute server for instance) changes >> drive mapping between AHCI and compatibility modes so I had to adjust >> /etc/fstab. If Dale is using labels of some type he will likely be >> better off than I was. >> >> There certainly won't be any harm caused by changing the BIOS setting >> to AHCI and trying it out. Reboot and change BIOS back is all he would >> have to do in my experience. >> >> - Mark >> >> > > Are the drives any faster when using AHCI tho? If the speed is the same > then I may try it next time I reboot but not real sure why it would matter. > I was hoping for something even faster. > > Dale I suspect they might be faster but you'd have to benchmark them yourself to find out. I never did that or don't remember the results if I did. - Mark
Re: !! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs>
Peter Humphrey wrote: On Wednesday 15 December 2010 13:33:03 J. Roeleveld wrote: Not sure how much memory you have, but a 6Gig ramdisk mounted at /var/tmp/portage is sufficient to compile openoffice. That speeds things up even more ;) The same size of tmpfs on /tmp also works even if you have only 4GB RAM, as I have. When it gets too full it just starts swapping. Actually, now that I check again, I see I've raised the size to 16G: $ grep 16G /etc/fstab tmpfs /tmp tmpfs nodev,nosuid,size=16G 0 0 $ grep swap /etc/fstab /dev/sda3 none swap sw,pri=10 0 0 /dev/sdb3 none swap sw,pri=10 0 0 /dev/sda7 none swap sw,pri=10 0 /dev/sdb7 none swap sw,pri=10 0 /dev/sdX3 are 2GB and /dev/sdX7 are 20GB (probably far too much swap, but disks are cheap). Interesting. I didn't know it would go to swap when it started getting full. Considering a emerge -e world takes about 3 days on my old rig and only takes about 10 hours on my new one, I already got a pretty good increase. We are always looking for more tho ain't we? lol Dale :-) :-)
Re: [gentoo-user] Questions about SATA and hot plugging.
On Wednesday 15 December 2010 12:24:58 Dale wrote: > Mark Knecht wrote: > > On Wed, Dec 15, 2010 at 9:49 AM, Volker Armin Hemmann > > > > wrote: > >> On Wednesday 15 December 2010 05:54:40 Dale wrote: > >>> According to the mobo manual, if I enable AHCI, it is hot swappable. > >> > >> and enabling AHCI is the only sane option. So do it. > >> > >>> I > >>> just ain't to comfy doing it. I'd like to see it done with no smoke > >>> getting out first. > >> > >> no smoke. Worst case: controller hangs, you have to reboot. > > > > The worst case is _slightly_ worse than that, but certainly no smoke. > > Some systems (my 6 drive RAID compute server for instance) changes > > drive mapping between AHCI and compatibility modes so I had to adjust > > /etc/fstab. If Dale is using labels of some type he will likely be > > better off than I was. > > > > There certainly won't be any harm caused by changing the BIOS setting > > to AHCI and trying it out. Reboot and change BIOS back is all he would > > have to do in my experience. > > > > - Mark > > Are the drives any faster when using AHCI tho? If the speed is the same > then I may try it next time I reboot but not real sure why it would > matter. I was hoping for something even faster. ahci is robust, in case of an error you don't have to wait for the 30-ide- timeout. NCQ can speed up some stuff. AHCI is just the right thing to do.
Re: [gentoo-user] OT: cheap "make yourself" server: i7-950 or phenom 1100T?
meino.cra...@gmx.de wrote: Jarry [10-12-15 19:08]: Hi, a friend of mine asked me to prepare a small server for him. Unfortunatelly he can not afford to buy brand-name server so he asked me to build one for him, from "consumer" components (yes, I already warned him about "zero-support" consequences). It should be some kind of "multi-purpose" server (web, ftp, mail, dns, virtualisation, etc). His budget is ~600-700€ (for cpu, mobo, ram), and he wants the best value for the money... Now, the crucial decision is what cpu (&mobo) I should use: A: Intel Core i7-950, 4x 3.06GHz (4 cores, + hyper-threading) B: AMD Phenom II X6 1100T Black Edition, 6x 3.30GHz (6 cores) Is it better to use phenom with 6 true cores, or i7 with 4 real and 4 "fake" cores (hyper-threading)? Concerning price, there is no difference, both of the above mentioned cpus cost ~250€ here in Europe. btw, mobos for phenom have up to 4x dimm, while mobos for core-i7 can have up to 6x dimm (that might be a valid point, he is going to need a lot of memory). So what should I pick for him? i7-950, or phenom-1100t? Or yet some cheap 4/6-core opteron 4xxx/6xxx? Jarry -- ___ This mailbox accepts e-mails only from selected mailing-lists! Everything else is considered to be spam and therefore deleted. I think the formula "performance/money" is fitted better by AMD than by Intel. If you take "performance" and "forget the money" Intel will be your friend. I myself choose an AMD Phenom X6 1090T (which can easily by pushed to be a 1100T by the way) on a ASUS Crosshair IV Formula. But a few days agao I heard Gigabyte would be more AMD friendly... I uses this mainly for rendering -- all cores can be used by Blender in parallel. Only my two cent ... you currency may vary. I DONT WANT to start a flamewar here! Its only my opinion I wanted to express :) Best regards, mcc I'll give my $0.02 worth as well. I just built a new rig. It has a Gigabyte GA-770T-USB3 mobo and a AMD Phenom II X4 and it is really fast and efficient. Price was very good too. From what I have read, Gigabyte is highly rated. This may be something you want to use as a starting point if available in you area. I'd be glad to send you links to the core parts so that you know what works well together. Dale :-) :-)
Re: [gentoo-user] OT: cheap "make yourself" server: i7-950 or phenom 1100T?
On 12/15/2010 09:35 AM, Jarry wrote: > So what should I pick for him? i7-950, or phenom-1100t? > Or yet some cheap 4/6-core opteron 4xxx/6xxx? Uh oh..don your flameproof underwear and open the floodgates I've not used the new six-core AMDs but I love my XII 940. And with Intel's new plumbing, my experience with the i7 seems to point to only slightly to the i7 side, as long as you have speedy RAM. I think in the long run, these two are so close, it just comes down to how much memory bandwidth you can wring from the machine. I/O is still I/O so, what else is left, really?
Re: [gentoo-user] OT: cheap "make yourself" server: i7-950 or phenom 1100T?
Dale [10-12-15 19:52]: > meino.cra...@gmx.de wrote: > >Jarry [10-12-15 19:08]: > > > >>Hi, > >> > >>a friend of mine asked me to prepare a small server for him. > >>Unfortunatelly he can not afford to buy brand-name server so > >>he asked me to build one for him, from "consumer" components > >>(yes, I already warned him about "zero-support" consequences). > >>It should be some kind of "multi-purpose" server (web, ftp, > >>mail, dns, virtualisation, etc). His budget is ~600-700€ (for > >>cpu, mobo, ram), and he wants the best value for the money... > >> > >>Now, the crucial decision is what cpu (&mobo) I should use: > >> > >>A: Intel Core i7-950, 4x 3.06GHz (4 cores, + hyper-threading) > >>B: AMD Phenom II X6 1100T Black Edition, 6x 3.30GHz (6 cores) > >> > >>Is it better to use phenom with 6 true cores, or i7 with > >>4 real and 4 "fake" cores (hyper-threading)? Concerning price, > >>there is no difference, both of the above mentioned cpus cost > >>~250€ here in Europe. > >> > >>btw, mobos for phenom have up to 4x dimm, while mobos for > >>core-i7 can have up to 6x dimm (that might be a valid point, > >>he is going to need a lot of memory). > >> > >>So what should I pick for him? i7-950, or phenom-1100t? > >>Or yet some cheap 4/6-core opteron 4xxx/6xxx? > >> > >>Jarry > >> > >>-- > >>___ > >>This mailbox accepts e-mails only from selected mailing-lists! > >>Everything else is considered to be spam and therefore deleted. > >> > >> > > > >I think the formula "performance/money" is fitted better by AMD > >than by Intel. > >If you take "performance" and "forget the money" Intel will be your > >friend. > > > >I myself choose an AMD Phenom X6 1090T (which can easily by pushed > >to be a 1100T by the way) on a ASUS Crosshair IV Formula. But a few > >days agao I heard Gigabyte would be more AMD friendly... > > > >I uses this mainly for rendering -- all cores can be used by Blender > >in parallel. > > > >Only my two cent ... you currency may vary. > > > >I DONT WANT to start a flamewar here! > >Its only my opinion I wanted to express :) > > > >Best regards, > >mcc > > > > > > I'll give my $0.02 worth as well. I just built a new rig. It has a > Gigabyte GA-770T-USB3 mobo and a AMD Phenom II X4 and it is really fast > and efficient. Price was very good too. From what I have read, > Gigabyte is highly rated. > > This may be something you want to use as a starting point if available > in you area. I'd be glad to send you links to the core parts so that > you know what works well together. > > Dale > > :-) :-) > Silent cooling ? Scythe Mulgen 2 Rev B! ;) mcc
Re: !! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs>
On Wed, Dec 15, 2010 at 9:09 AM, Peter Humphrey wrote: > On Wednesday 15 December 2010 13:33:03 J. Roeleveld wrote: > >> Not sure how much memory you have, but a 6Gig ramdisk mounted at >> /var/tmp/portage is sufficient to compile openoffice. That speeds >> things up even more ;) > > The same size of tmpfs on /tmp also works even if you have only 4GB RAM, > as I have. When it gets too full it just starts swapping. In my experience if it starts swapping, that is much slower than just using disk for /tmp in the first place. YMMV :)
Re: [gentoo-user] New monitor problems. < sighs >
On Wed, Dec 15, 2010 at 12:35 PM, Dale wrote: > Paul Hartman wrote: >> >> On Wed, Dec 15, 2010 at 2:40 AM, Dale wrote: >> >>> >>> Hi again, >>> >>> For those mot up to date. I built a new rig that has a Nvidia GT220 card >>> in >>> it. I bought a brand new monitor this morning, a LG W2253, and it worked >>> one time. I had to reboot to move some things around and when I >>> rebooted, >>> the GUI doesn't come up. The BIOS screen shows up and I can see the >>> services start up as well but when it switches to vt7, it just has a >>> little >>> blinking cursor at the top. >>> >> >> At first glance my guess is that your kernel is configured incorrectly >> somehow. >> >> I have a similar card (Nvidia GT 240). I am using fully ~amd64 system >> and it all works. FWIW, below are my settings& a couple thoughts too. >> << SNIP>> >> >> I can send you my kernel .config if you want. Let me know! Good luck :) >> >> > > You are correct, it was a kernel config issue. Nvidia needs CONFIG_SYSVIPC > enabled in the kernel to work correctly. Yep, sorry I only noticed the "Solved" thread after I wrote my reply. I was on vacation and have about 100 threads here to catch up on. The README file that comes with the nvidia drivers explains that problem (as I wrote) and much more, so it might be a good read in case there's any other info in there that might be helpful in optimizing your set-up now that it's working. Also emerge nvidia-settings if you haven't already, it has a nice GUI for tweaking some of the options rather than going through xorg.conf. Have fun!
Re: !! SOLVED !! Re: [gentoo-user] New monitor problems. < sighs>
Paul Hartman wrote: On Wed, Dec 15, 2010 at 9:09 AM, Peter Humphrey wrote: On Wednesday 15 December 2010 13:33:03 J. Roeleveld wrote: Not sure how much memory you have, but a 6Gig ramdisk mounted at /var/tmp/portage is sufficient to compile openoffice. That speeds things up even more ;) The same size of tmpfs on /tmp also works even if you have only 4GB RAM, as I have. When it gets too full it just starts swapping. In my experience if it starts swapping, that is much slower than just using disk for /tmp in the first place. YMMV :) My thought was that at least portage wouldn't stop the compile when it ran out of room. I usually put /var on a separate partition and always forget that OOo needs some space. I usually don't give it enough the first time around. Doesn't OOo need about 4 or 5Gbs now? Dale :-) :-)
[gentoo-user] Re: Questions about SATA and hot plugging.
On 2010-12-15, Andrea Conti wrote: >> E-SATA != SATA > > Nah. They are *exactly* the same. Not according to Wikipedia -- it says the electrical specs for eSATA are different than the specs for "normal" SATA. I've seen that stated in other places as well. I don't have copies of the two specs, so I can't say I'm 100%, but I believe the Wikipedia page. -- Grant Edwards grant.b.edwardsYow! I'm a fuschia bowling at ball somewhere in Brittany gmail.com
