I just downloaded and installed the new kernel, and rebooted. No problem. During reboot, I got the expected message that my NTFS driver failed to load because of the missing kernel-module. I was expecting a similar message from ati-fglrx but no, no message. When I go to system settings/more system settings and click on the ATI-fglrx settings, the GUI opens right up - in times past, I would get the missing kernel-module message on reboot for ATI-fglrx, and the settings GUI wouldn't open or would open with errors. What gives - did I miss something? Is an updated kernel module no longer necessary for the ATI driver from Livna?
Am Montag, den 29.08.2005, 01:14 -0400 schrieb Claude Jones:
I just downloaded and installed the new kernel, and rebooted. No problem. During reboot, I got the expected message that my NTFS driver failed to load because of the missing kernel-module. I was expecting a similar message from ati-fglrx but no, no message. When I go to system settings/more system settings and click on the ATI-fglrx settings, the GUI opens right up - in times past, I would get the missing kernel-module message on reboot for ATI-fglrx
This should still be the case. If not check if /etc/init.d/ati-fglrx is still enabled. If yes, then please post the output of
/etc/init.d/ati-fglrx start /etc/init.d/ati-fglrx stop
and the settings GUI wouldn't open or would open with errors.
Maybe ATI changed something and it works now without the kernel-module?
What gives - did I miss something? Is an updated kernel module no longer necessary for the ATI driver from Livna?
The X-Server will work fine without the kernel-module (a updated modules hopefully are pushed soon), but 3D-Support won't work.
HTH
On Monday 29 August 2005 11:46 am, Thorsten Leemhuis wrote:
Am Montag, den 29.08.2005, 01:14 -0400 schrieb Claude Jones:
I just downloaded and installed the new kernel, and rebooted. No problem. During reboot, I got the expected message that my NTFS driver failed to load because of the missing kernel-module. I was expecting a similar message from ati-fglrx but no, no message. When I go to system settings/more system settings and click on the ATI-fglrx settings, the GUI opens right up - in times past, I would get the missing kernel-module message on reboot for ATI-fglrx
This should still be the case. If not check if /etc/init.d/ati-fglrx is still enabled. If yes, then please post the output of
/etc/init.d/ati-fglrx start /etc/init.d/ati-fglrx stop
[root@viewridgeproductions cj]# chkconfig --list ati-fglrx ati-fglrx 0:off 1:off 2:on 3:on 4:on 5:on 6:off [root@viewridgeproductions cj]# /etc/init.d/ati-fglrx start Checking for ati-fglrx kernel module [FAILED]
fglrx.ko module for 2.6.12-1.1447_FC4 kernel not found DRI will not work within X with the ati-fglrx driver. Get an updated/suitable 'kernel-module-fglrx-<kernelver>' rpm from rpm.livna.org [root@viewridgeproductions cj]# /etc/init.d/ati-fglrx stop
and the settings GUI wouldn't open or would open with errors.
Maybe ATI changed something and it works now without the kernel-module?
[root@viewridgeproductions cj]# service ati-fglrx status Init-Script is not running. fglrx driver is enabled.
I opened the GUI and manipulated the color sliders, and they definitely adjusted the screen???
What gives - did I miss something? Is an updated kernel module no longer necessary for the ATI driver from Livna?
The X-Server will work fine without the kernel-module (a updated modules hopefully are pushed soon), but 3D-Support won't work.
This I knew. I was puzzled by the lack of messages on boot-up. I still am... Could the ATI GUI be adjusting the video even if the fglrx driver is not running?
On Mon August 29 2005 1:57 pm, Claude Jones wrote:
On Monday 29 August 2005 11:46 am, Thorsten Leemhuis wrote:
Am Montag, den 29.08.2005, 01:14 -0400 schrieb Claude Jones:
I just downloaded and installed the new kernel, and rebooted. No problem. During reboot, I got the expected message that my NTFS driver failed to load because of the missing kernel-module. I was expecting a similar message from ati-fglrx but no, no message. When I go to system settings/more system settings and click on the ATI-fglrx settings, the GUI opens right up - in times past, I would get the missing kernel-module message on reboot for ATI-fglrx
This should still be the case. If not check if /etc/init.d/ati-fglrx is still enabled. If yes, then please post the output of
/etc/init.d/ati-fglrx start /etc/init.d/ati-fglrx stop
[root@viewridgeproductions cj]# chkconfig --list ati-fglrx ati-fglrx 0:off 1:off 2:on 3:on 4:on 5:on 6:off [root@viewridgeproductions cj]# /etc/init.d/ati-fglrx start Checking for ati-fglrx kernel module [FAILED]
fglrx.ko module for 2.6.12-1.1447_FC4 kernel not found DRI will not work within X with the ati-fglrx driver. Get an updated/suitable 'kernel-module-fglrx-<kernelver>' rpm from rpm.livna.org [root@viewridgeproductions cj]# /etc/init.d/ati-fglrx stop
and the settings GUI wouldn't open or would open with errors.
Maybe ATI changed something and it works now without the kernel-module?
[root@viewridgeproductions cj]# service ati-fglrx status Init-Script is not running. fglrx driver is enabled.
I opened the GUI and manipulated the color sliders, and they definitely adjusted the screen???
What gives - did I miss something? Is an updated kernel module no longer necessary for the ATI driver from Livna?
The X-Server will work fine without the kernel-module (a updated modules hopefully are pushed soon), but 3D-Support won't work.
This I knew. I was puzzled by the lack of messages on boot-up. I still am... Could the ATI GUI be adjusting the video even if the fglrx driver is not running?
Just to follow this up: I just did a confirm boot or whatever it's called, pressing 'i' at the point where the kernel is being unpacked. The ntfs file system mount was rejected due to the lack of a current module. When it got to the ati-fglrx driver, it loaded with no error message. It used to return a message after it checked for the current module - it's not doing that, even though I have the latest kernel, and the previous kernel module. This is going on on two separate machines. The driver seems to be loaded, and the ATI GUI control window comes up and the various functions work. I'm running the Radeon 9200SE card on both machines - both machines seem to be running 3D. Strange... The messages above were for my machine at the office. Below are the same commands run on this machine, in my home: [root@viewridgeproductions2 cj]# service ati-fglrx status Init-Script is running. [root@viewridgeproductions2 cj]# chkconfig --list ati-fglrx ati-fglrx 0:off 1:off 2:on 3:on 4:on 5:on 6:off [root@viewridgeproductions2 cj]# /etc/init.d/ati-fglrx start [root@viewridgeproductions2 cj]# /etc/init.d/ati-fglrx stop [root@viewridgeproductions2 cj]#
As you can see, there appears to be no response to the start/stop commands on this machine. The one at the office returned a [Failed] message when I tried to start the driver after checking for the module, so the behavior isn't identical.
Claude Jones wrote:
On Mon August 29 2005 1:57 pm, Claude Jones wrote: I'm running the Radeon 9200SE card on both machines - both machines seem to be running 3D.
The 9200SE uses an RV280 series chip, which is well-supported by the Free drivers which are a part of the kernel and X.org's X11 server implementation. Just for the sake of my curiosity, is there something special that ATi's proprietary driver gives you that you cannot get with the Free drivers? I think, considering that GNU/Linux is based on the principles of Free/open-source software, it would make sense to use Free drivers where supported, right? :-O
Thanks.
On 8/30/05, Peter Gordon admin@ramshacklestudios.com wrote:
The 9200SE uses an RV280 series chip, which is well-supported by the Free drivers which are a part of the kernel and X.org http://X.org's X11 server implementation. Just for the sake of my curiosity, is there something special that ATi's proprietary driver gives you that you cannot get with the Free drivers? I think, considering that GNU/Linux is based on the principles of Free/open-source software, it would make sense to use Free drivers where supported, right? :-O
2 words: Doom 3 ;)
(last time I looked the Free drivers did cause lots of graphic distortions in that game).
Klaasjan
On Tue, 30 Aug 2005, Klaasjan Brand wrote:
server implementation. Just for the sake of my curiosity, is there something special that ATi's proprietary driver gives you that you cannot get with the Free drivers? I think, considering that GNU/Linux is based on the principles of Free/open-source software, it would make sense to use Free drivers where supported, right? :-O
2 words: Doom 3 ;)
(last time I looked the Free drivers did cause lots of graphic distortions in that game).
9200se is way underpowered for doom3, having half the memory bandwidth of the regular 9200 doesn't help it either.
Klaasjan
On 8/30/05, Joel Jaeggli joelja@darkwing.uoregon.edu wrote:
9200se is way underpowered for doom3, having half the memory bandwidth of the regular 9200 doesn't help it either.
It's not underpowered; just the minimum required. I know casual gamers who finished doom3 on lesser cards...
Klaasjan
On Tue August 30 2005 1:34 am, Peter Gordon wrote:
Claude Jones wrote:
On Mon August 29 2005 1:57 pm, Claude Jones wrote: I'm running the Radeon 9200SE card on both machines - both machines
seem to be running 3D.
The 9200SE uses an RV280 series chip, which is well-supported by the Free drivers which are a part of the kernel and X.org's X11 server implementation. Just for the sake of my curiosity, is there something special that ATi's proprietary driver gives you that you cannot get with the Free drivers? I think, considering that GNU/Linux is based on the principles of Free/open-source software, it would make sense to use Free drivers where supported, right? :-O
This is the first time I've seen any one say that the ATI drivers included in the kernel were well-supported. Everything I've otherwise read has said that the built in drivers were lacking in 3D/Open GL performance. I am no expert, and don't know anything about the RV280 chip. But, I am in video production, and I was concerned about maximizing my video configuration for that reason. The only reference I have are the screen savers, and my anecdotal experience is that the video performance is visibly better on some of the more complex screen savers with the ATI-fglrx driver then with the one built in to the kernel.
On Tue, 2005-08-30 at 09:17 -0400, Claude Jones wrote:
This is the first time I've seen any one say that the ATI drivers included in the kernel were well-supported. Everything I've otherwise read has said that the built in drivers were lacking in 3D/Open GL performance.
I have a 9200 (not SE), and I get much better performance with the built in drivers than the fglrx drivers from livna. Perhaps I'm doing something wrong when installing the livna packages...
On Tue August 30 2005 11:36 am, Brian Mury wrote:
On Tue, 2005-08-30 at 09:17 -0400, Claude Jones wrote:
This is the first time I've seen any one say that the ATI drivers included in the kernel were well-supported. Everything I've otherwise read has said that the built in drivers were lacking in 3D/Open GL performance.
I have a 9200 (not SE), and I get much better performance with the built in drivers than the fglrx drivers from livna. Perhaps I'm doing something wrong when installing the livna packages...
Interesting - I shall monitor this thread, to see if anyone else comments on this. The new kernel-modules were put up this morning on Livna, and I've installed them. I still can't tell if they're active because of the anomalies I reported in my previous posts. I continue to get no messages on boot, with the ATI-fglrx driver seemingly being loaded (but it was behaving the same way before I installed the new module - the script is supposed to check for the presence of the correct module, and return an error message if it's not there - it didn't seem to be doing this before or after the new kernel-module installation). If I step through the boot, when it gets to the ATI driver, I say yes when it asks whether to load it, and it returns no message, but moves on to the next service.
The new kernel-modules were put up this morning on Livna,
News as in 8.16.20? I don't see them, at least not through yum. The driver is there but not the kernel module.
The new kernel-modules were put up this morning on Livna,
News as in 8.16.20? I don't see them, at least not through yum. The driver is there but not the kernel module.
I should mention I am running x86_64
On Tue August 30 2005 12:21 pm, Robert Bell wrote:
The new kernel-modules were put up this morning on Livna,
News as in 8.16.20? I don't see them, at least not through yum. The driver is there but not the kernel module.
I should mention I am running x86_64
Don't know about those. Non-64 version is up.
Am Dienstag, den 30.08.2005, 09:21 -0700 schrieb Robert Bell:
The new kernel-modules were put up this morning on Livna,
News as in 8.16.20? I don't see them, at least not through yum. The driver is there but not the kernel module.
I should mention I am running x86_64
From: http://rpm.livna.org/fedora/4/x86_64/RPMS.lvn/
[...] kernel-module-fglrx-2.6.12-1.1447_FC4-8.16.20.1-0.lvn.1.4.x86_64.rpm 29-Aug-2005 22:50 359K [...]
?
HTH
CU thl
Am Dienstag, den 30.08.2005, 11:48 -0400 schrieb Claude Jones:
[...] I still can't tell if they're active because of the anomalies I reported in my previous posts.
Look at your /var/log/Xorg.0.log -- you should see there if the driver and DRI are active.
I'll look into your "anomalies", but I did not find the time yet.
HTH
CU thl
On Wed August 31 2005 1:36 am, Thorsten Leemhuis wrote:
Am Dienstag, den 30.08.2005, 11:48 -0400 schrieb Claude Jones:
[...] I still can't tell if they're active because of the anomalies I reported in my previous posts.
Look at your /var/log/Xorg.0.log -- you should see there if the driver and DRI are active.
I looked, but I can't say I know what I was looking at. There's definitely fglrx activity, but there appear numerous error messages. It's a huge section of that log, so I am not sure whether I should post it.
A sample:
(II) Loading /usr/X11R6/lib/modules/drivers/fglrx_drv.o (II) Module fglrx: vendor="FireGL - ATI Technologies Inc." compiled for 6.8.0, module version = 8.16.20 Module class: X.Org Video Driver ABI class: X.Org Video Driver, version 0.7
It would seem that the driver is being loaded, though all sorts of error messages follow later on:
II) ATI Proprietary Linux Driver Version Identifier:8.16.20 (II) ATI Proprietary Linux Driver Release Identifier: @@UNRELEASED_AND_UNSUPPORTED_DRIVER@@ (II) ATI Proprietary Linux Driver Build Date: Aug 16 2005 00:15:14 (II) ATI Proprietary Linux Driver Build Information: autobuild-rel-r6-8.16.1-driver-lnx-206829 (WW) fglrx: No matching Device section for instance (BusID PCI:1:0:1) found
Then later, regarding DRI:
(II) fglrx(0): doing DRIScreenInit drmOpenDevice: node name is /dev/dri/card0 drmOpenDevice: open result is -1, (Unknown error 999) drmOpenDevice: open result is -1, (Unknown error 999) drmOpenDevice: Open failed drmOpenDevice: node name is /dev/dri/card0 drmOpenDevice: open result is -1, (Unknown error 999) drmOpenDevice: open result is -1, (Unknown error 999) drmOpenDevice: Open failed drmOpenByBusid: Searching for BusID PCI:1:0:0 drmOpenDevice: node name is /dev/dri/card0 drmOpenDevice: open result is 7, (OK) drmOpenByBusid: drmOpenMinor returns 7 drmOpenByBusid: drmGetBusid reports drmOpenDevice: node name is /dev/dri/card1 drmOpenDevice: open result is -1, (No such device) drmOpenDevice: open result is -1, (No such device)
But then, later:
(II) fglrx(0): [drm] register handle = 0xfe9f0000 (II) fglrx(0): [agp] Mode=0x1f004a1b bridge: 0x8086/0x2570 (II) fglrx(0): [agp] AGP v1/2 disable mask 0x00000000 (II) fglrx(0): [agp] AGP v3 disable mask 0x00000000 (II) fglrx(0): [agp] enabling AGP with mode=0x1f004b1a (II) fglrx(0): [agp] AGP protocol is enabled for graphics board. (cmd=0x1f004312) (II) fglrx(0): [agp] graphics chipset has AGP v3.0 (native mode) (II) fglrx(0): [drm] ringbuffer size = 0x00100000 bytes (II) fglrx(0): [drm] DRM buffer queue setup: nbufs = 100 bufsize = 28672 (II) fglrx(0): [drm] texture shared area handle = 0xf8f81000 (II) fglrx(0): shared FSAAScale=1 (II) fglrx(0): DRI initialization successfull!
So, it would appear that DRI is up. These samples probably represent less than 10% of the lines containing ATI or fgrlx - I can post the whole thing if necessary.
I'll look into your "anomalies", but I did not find the time yet.
Ok- thanks
Claude Jones kirjoitti viestissään (lähetysaika keskiviikko, 31. elokuuta 2005 09:15):
It would seem that the driver is being loaded, though all sorts of error messages follow later on:
What error messages? Your posting didn't show _any_ errors, just informational log messages (II) and one warning (WW). Errors would have (EE) in the beginning of the line, just as it says in the start of the log file: "Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown."
On Wednesday 31 August 2005 9:57 am, Markku Kolkka wrote:
............................................................................Errors would have (EE) in the beginning of the line, just as it says in the start of the log file:
OK - missed that section at the beginning - I saw words like "error" and "failed" in the messages and made intuitive assumptions
On Tue, 2005-08-30 at 09:17 -0400, Claude Jones wrote:
the video performance is visibly better on some of the more complex screen savers with the ATI-fglrx driver then with the one built in to the kernel.
Does anybody know, Why nvidia 3D-graphics support is excluded from the default kernel, as opposed to ATI-ones?
Strong wrote:
Does anybody know, Why nvidia 3D-graphics support is excluded from the default kernel, as opposed to ATI-ones?
Fedora won't, as a matter of policy, include non-open source software. Both Nvidia and ATi's proprietary drivers require kernel support. They will not release their kernel modifications or their user-space drivers under an Open Source license.
I understand that source is only available (under a non-Open Source license) for a kernel-mode "glue layer", not for the entire kernel module, and not at all for the user-space.
(Besides, Fedora want to keep the Fedora kernel relatively close to upstream. There's practically no chance that either Fedora or Linus Torvalds would include functionality that could only realistically be used by closed source programs. So even if Nvidia or ATi GPLed their kernel support, this wouldn't help get it in the kernel).
Older ATi cards have Open Source 3D acceleration, so this *is* included. I understand that Intel graphics have Open Source drivers, and most adapters have working 2D acceleration.
http://fedoraproject.org/wiki/ForbiddenItems http://fedora.redhat.com/about/objectives.html
Hope this helps,
James.
Peter Gordon wrote:
The 9200SE uses an RV280 series chip, which is well-supported by [...] sense to use Free drivers where supported, right? :-O
Ok. Thanks for the input all. I was just curious. :)
Am Montag, den 29.08.2005, 13:57 -0400 schrieb Claude Jones:
On Monday 29 August 2005 11:46 am, Thorsten Leemhuis wrote:
Am Montag, den 29.08.2005, 01:14 -0400 schrieb Claude Jones:
This should still be the case. If not check if /etc/init.d/ati-fglrx is still enabled. If yes, then please post the output of
/etc/init.d/ati-fglrx start /etc/init.d/ati-fglrx stop
[root@viewridgeproductions cj]# chkconfig --list ati-fglrx ati-fglrx 0:off 1:off 2:on 3:on 4:on 5:on 6:off
Seems to be on. And calling it manually...
[root@viewridgeproductions cj]# /etc/init.d/ati-fglrx start Checking for ati-fglrx kernel module [FAILED]
fglrx.ko module for 2.6.12-1.1447_FC4 kernel not found DRI will not work within X with the ati-fglrx driver. Get an updated/suitable 'kernel-module-fglrx-<kernelver>' rpm from rpm.livna.org
...works fine. I rechecked here and everything works as expected. Probably something is wrong on your machine, but I don't have an idea what. Sorry.
On Wednesday 31 August 2005 1:56 pm, Thorsten Leemhuis wrote:
...works fine. I rechecked here and everything works as expected. Probably something is wrong on your machine, but I don't have an idea what. Sorry.
Yes, it's kind of hard to diagnose at a distance. In any event, to just add to the record: I have updated my machine at the office just now, and if I step through the boot, I get the message "Checking for kernel-module..." or whatever those words are, and it returns an "Ok" - in all other respects, the behavior of the two machines is identical. The drivers seem to be running, the ATI GUI's controls function, and my 3D seems to good. The only difference now is that the machine at home doesn't return the "Checking for kernel-module..." message - is it possible I've inadvertently changed something in your script? I guess I should compare the two scripts, the one from my home machine, and the one on the office box, and see if they're the same.
-
Brian Mury -
Claude Jones -
James Wilkinson -
Joel Jaeggli -
Klaasjan Brand -
Markku Kolkka -
Peter Gordon -
Robert Bell -
Strong -
Thorsten Leemhuis