Go Back   nV News Forums > Linux Support Forums > NVIDIA Linux

Newegg Daily Deals

Reply
 
Thread Tools
Old 12-09-05, 10:48 AM   #97
CondorDes
Registered User
 
Join Date: Mar 2004
Posts: 14
Send a message via AIM to CondorDes
Default Re: 1.0-8174 for Linux x86 released

Quote:
Originally Posted by CondorDes
This is the first I've heard of such bugs, and I follow (at least) some of the X developers' weblogs through FreeDesktop. Have you (meaning nVidia) made any effort to push these bugs back upstream? Can you cite mailing list messages or bug reports, please? (Not necessarily ones you've pushed, just ones that have caused problems)

If you haven't made any effort to connect with the X.org developers about the issues you're encountering in Composite, you have no right to complain.
I take that back. I found another thread here with the Composite issues and the relevant X.org bug reports. It's good to see that nVidia is working with the X.org people to make things work.
CondorDes is offline   Reply With Quote
Old 12-09-05, 01:52 PM   #98
mriya3
Registered User
 
Join Date: Dec 2005
Posts: 1
Smile Re: 1.0-8174 for Linux x86 released

for all those who have problems with Toshiba notebooks...

I have a Toshiba Satellite 2410-304S (Geforce 4 Go 420, 32 MB)

I've successfully installed the driver on Ubuntu Breezy (kernel 2.6.12, following the guide at http://ubuntuforums.org/showthread.p...=latest+nvidia ). It is important to delete all things related to old drivers (packages, scripts in /etc/modprobe.d, etc).

Please note that drivers version 7667 and prior always crashed with composite enabled, now with 8174 everything seems stable (I only have a problem with mozilla using too much memory, 80% and more, of 512 available mb)

At first, I also had the same problem of the black screen and complete hang, but resolved it by issuing:

Code:
rmmod nvidia
mknod /dev/nvidiactl c 195 255
mknod /dev/nvidia0 c 195 0
modprobe nvidia
(as suggested in http://forums.gentoo.org/viewtopic-t...ht-nvidia.html )

I've also added some options for the nvidia module: in my case, as for other debian based distro, create a new file in /etc/modprobe.d (see attached nvidia file) containing:

Code:
options nvidia NVreg_SoftEDIDs=0 NVreg_Mobile=1
see the attached file for my xorg.conf



...finally I can enjoy some eyecandy (see screenshot)

http://attila.objectis.net/foto/tras...e.png?display=
Attached Files
File Type: zip nvidiaconf.zip (1.9 KB, 412 views)
mriya3 is offline   Reply With Quote
Old 12-09-05, 01:54 PM   #99
joffe
Registered User
 
joffe's Avatar
 
Join Date: Nov 2004
Posts: 171
Default Re: 1.0-8174 for Linux x86 released

Quote:
Originally Posted by CondorDes
I take that back. I found another thread here with the Composite issues and the relevant X.org bug reports. It's good to see that nVidia is working with the X.org people to make things work.
It feels good to release some pressure though, doesn't it? And you do have a point with your first paragraph.
__________________
GA-MA790GP-DS4H, ATI Radeon HD3300, AMD Phenom X4 9850, 4GB PC8500 [Ubuntu 8.10]
Toshiba 2450-201, GeForce4 420 Go 32MB, Pentium4 2.4GHz, 512 MB PC2700 [Ubuntu 8.10]
joffe is offline   Reply With Quote
Old 12-09-05, 05:00 PM   #100
joffe
Registered User
 
joffe's Avatar
 
Join Date: Nov 2004
Posts: 171
Thumbs up Re: 1.0-8174 for Linux x86 released

Quote:
Originally Posted by mriya3
for all those who have problems with Toshiba notebooks...
I can't believe this, but it actually works!!!

My notebook is a Toshiba Satellite 2450-201 . I don't know what I did but it helped. Now I get into X (KDE) with the new drivers for the first time since last spring... The only thing is that the console is garbled with green rectangles again, but this is a minor issue - I will probably patch the kernel with vesa-tng and the problem will most likely go away.

What I did:
  1. removed the contents in /etc/modprobe.conf (there is no /etc/modprobe.d in slackware)
  2. added the line 'options nvidia NVreg_SoftEDIDs=0 NVreg_Mobile=1' to /etc/modprobe.conf
  3. issued '/nvidia-installer --uninstall' (this was 1.0-7676)
  4. found a libGL file in /usr/lib belonging to an old 1.0-8174 installation. this was also erased
  5. added some stuff to xorg.conf:

    Section "Module"
    Load "ddc"
    Load "vbe"

    Section "Device"
    BusID "PCI:1:0:0"
    #Option "NoDDC" "1"
    #BusID "AGP:01:00:0"
    Option "IgnoreEDID" "true"
    Option "IgnoreEdidFreqs" "true"
    Option "GenerateRTList" "0"
    Option "OverridePolarity" "1"
    Option "DigitalVibrance" "1"
    Option "CursorShadow" "1"
    Option "CursorShadowYOffset" "2"
    Option "CursorShadowXOffset" "4"
    Option "CursorShadowAlpha" "63"
    Option "NvAGP" "3"
    Option "RenderAccel" "true"
    Option "AllowGLXWithComposite" "true"

    Section "Extensions"
    Option "Composite" "Enable"
    EndSection
  6. issued '/nvidia-installer --na' (this was 1.0-8174)
  7. issued 'modprobe nvidia'
  8. startx
I don't know how much of this really matter since there are some variables in xorg.conf that are not used (see attached Xorg.0.log).
Again, thank you so much for these news! I will do more testing and let you know if I find something of interest.
Attached Files
File Type: log Xorg.0.log (31.0 KB, 462 views)
File Type: log xorg.conf.log (16.9 KB, 482 views)
__________________
GA-MA790GP-DS4H, ATI Radeon HD3300, AMD Phenom X4 9850, 4GB PC8500 [Ubuntu 8.10]
Toshiba 2450-201, GeForce4 420 Go 32MB, Pentium4 2.4GHz, 512 MB PC2700 [Ubuntu 8.10]
joffe is offline   Reply With Quote
Old 12-09-05, 05:40 PM   #101
PrakashP
gentoo ~x86_64 user
 
PrakashP's Avatar
 
Join Date: Jul 2004
Location: Germania
Posts: 213
Default Re: 1.0-8174 for Linux x86 released

Well, I have been complaining long time because of slow 2d (esp in conjunctio with aa fonts) with all drivers >6111 with stable xorg-x11. With 70rc2 everything was fast - even with newer drivers, but going to rc3 makes everything slow again. So maybe Nvidia likes to check what may cause this. I am not sure whether this is an xserver issue or Renderaccell issue, but I guess now you have more specific means to investigate.

I used these ebuilds for building xorg-x11:

http://forums.gentoo.org/viewtopic-t-396898.html

Currently I am back to rc2 due to speed and current Nvidia driver.

If you have found some specific patch, which may cause this, I could test compile xorg-x11 to make sure you found the right one.
PrakashP is offline   Reply With Quote
Old 12-09-05, 08:22 PM   #102
AaronP
NVIDIA Corporation
 
AaronP's Avatar
 
Join Date: Mar 2005
Posts: 2,487
Default Re: 1.0-8174 for Linux x86 released

PrakashP,

There is currently an issue with RenderAccel and subpixel antialiasing. Do you have subpixel antialiasing enabled? If so, do you see a performance improvement if you disable it?
AaronP is offline   Reply With Quote
Old 12-10-05, 02:05 AM   #103
PrakashP
gentoo ~x86_64 user
 
PrakashP's Avatar
 
Join Date: Jul 2004
Location: Germania
Posts: 213
Default Re: 1.0-8174 for Linux x86 released

Well, with xorg-x11 7.0-rc2. everything is fast even with aa and subpixel hinting (though hinting is mostly wrong as I explained in another thread). The last time I checked with stable xorg-x11, performance was bad with antialiased fonts, no matter if subpixel hinting was on or off. This was with 6629 drivers. Only if I disabled antialising completely performance was fast again. Only with 6111 performance was good with aa and subpixel hinting, though hinting was wrong as it is now with xorg-x11 7.0rc2 and current nvidia driver.

I cannot test 7.0rc3 with current driver, as I deinstalled it for now, but is above info sufficient?
PrakashP is offline   Reply With Quote
Old 12-10-05, 06:25 AM   #104
PrakashP
gentoo ~x86_64 user
 
PrakashP's Avatar
 
Join Date: Jul 2004
Location: Germania
Posts: 213
Default Re: 1.0-8174 for Linux x86 released

OK, I can confirm AaronP's sugegstion: With current rc3 + patch from https://bugs.freedesktop.org/show_bug.cgi?id=5258 which above ebuild seesm to include, my 2d is fast again using grey hinting instead of subpixel hinting. Using this patch gives *correct* subpixel hinting, but it is slow, as if fonts weren't accelerated.

I hope Nvidia will accelerate the subpixel case, at it will become more and more important as users will probably working more on LCDs than CRTs in short future.

Ok, I created a poll, to find out how important perfect LCD support, would be for the users:


http://www.nvnews.net/vbulletin/showthread.php?t=61385
PrakashP is offline   Reply With Quote

Old 12-10-05, 01:09 PM   #105
GGtheMaD
Registered User
 
Join Date: Jun 2004
Location: North Italy (near Maggiore Lake)
Posts: 12
Default Re: 1.0-8174 for Linux x86 released

Now I use driver 1.0-7174 because after that release I had the same problem with every release, also with this!

When I resume from suspend to RAM or suspend to disk, if I switch in a console, with Ctrl+F1, Ctrl+F2, etc, the result is the corruption of all consoles!

This is the image:



An other problem is with DPMS, when the screen switch off for the screen standby, the screen doesn't switch on when I touch any keys or the touchpad... only if I press Ctrl+F1 and after Ctrl+F7 the screen switch on!

Laptop Toshiba M30-801
Intel Centrino - CPU 1.5GHz - RAM 512MB
FX Go 5200 64MB
Slackware 10.2
Kernel 2.6.14.3 (but with previous kernels is the same thing!)

Some suggestions?
GGtheMaD is offline   Reply With Quote
Old 12-10-05, 07:52 PM   #106
wschutzer
Registered User
 
wschutzer's Avatar
 
Join Date: Dec 2005
Location: Sao Carlos, Brazil
Posts: 55
Default Re: 1.0-8174 for Linux x86 released

Quote:
Originally Posted by GGtheMaD
When I resume from suspend to RAM or suspend to disk, if I switch in a console, with Ctrl+F1, Ctrl+F2, etc, the result is the corruption of all consoles!
Some suggestions?
Same problem here ever since. I can never use the text-mode consoles after suspending to ram. Only a reboot will fix that.

I have the very same problem with DPMS. It seems that the backlight won't come back unless I did CTRL-F1 followed by CTRL-F7.

Anybody else out there with the same problems?

My computer is a notebook with the following configuration:
Toshiba Satellite 1415-S173
Intel Celeron 1.8Ghz
512Mb/30Gb
nVidia Corporation NV17 [GeForce4 420 Go] (rev a3), driver 1.0-7676 (I get a black screen with 1.0-8174)
Fedora Core 4, kernel 2.6.14-1.1644_FC4, KDE 3.5

Waldeck
wschutzer is offline   Reply With Quote
Old 12-11-05, 07:09 AM   #107
olivn
Registered User
 
Join Date: Dec 2005
Posts: 11
Default Re: 1.0-8174 for Linux x86 released

OpenSuse 10 with latest Suse kernel (2.6.13-15.7-smp) on a dual CPU (AMD Athlon MP 1800) with a GeForce 6800 LE


1) X process consumes 100% each time kdm starts
Code:
Dec 10 18:43:45 bia kernel: ACPI: PCI Interrupt 0000:01:05.0[A] -> GSI 17 (level, low) -> IRQ 185
Dec 10 18:43:45 bia kernel: NVRM: loading NVIDIA Linux x86 NVIDIA Kernel Module  1.0-8174  Tue Nov 22 17:48:37 PST 2005
Dec 10 18:43:45 bia kernel: agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
Dec 10 18:43:45 bia kernel: agpgart: Putting AGP V2 device at 0000:00:00.0 into 4x mode
Dec 10 18:43:45 bia kernel: agpgart: Putting AGP V2 device at 0000:01:05.0 into 4x mode
Dec 10 18:43:45 bia kernel: agpgart: Found an AGP 2.0 compliant device at 0000:00:00.0.
Dec 10 18:43:45 bia kernel: agpgart: Putting AGP V2 device at 0000:00:00.0 into 4x mode
Dec 10 18:43:45 bia kernel: agpgart: Putting AGP V2 device at 0000:01:05.0 into 4x mode
Dec 10 18:43:47 bia kernel: NVRM: Xid: 6, PE0000 1500 ff1b55ad 0000e124 00000000 ff4c4c4c


2) If I start X manually, the kernel has crashed once ( I can't reproduce it)
Code:
Dec 10 22:02:00 bia kernel: kernel BUG at arch/i386/mm/pageattr.c:136!
Dec 10 22:02:00 bia kernel: invalid operand: 0000 [#1]
Dec 10 22:02:00 bia kernel: SMP 
Dec 10 22:02:00 bia kernel: Modules linked in: w83781d i2c_sensor i2c_isa snd_pcm_oss snd_mixer_oss snd_seq_dummy bluetooth nvidia h
Dec 10 22:02:00 bia kernel: CPU:    0
Dec 10 22:02:00 bia kernel: EIP:    0060:[<c011da75>]    Tainted: P     U VLI
Dec 10 22:02:00 bia kernel: EFLAGS: 00210082   (2.6.13-15.7-smp) 
Dec 10 22:02:00 bia kernel: EIP is at __change_page_attr+0xf5/0x1e0
Dec 10 22:02:00 bia kernel: eax: 00000163   ebx: c10ed860   ecx: c1008860   edx: c1000000
Dec 10 22:02:00 bia kernel: esi: c0443c74   edi: 00000163   ebp: c76c3000   esp: c2a43ec8
Dec 10 22:02:00 bia kernel: ds: 007b   es: 007b   ss: 0068
Dec 10 22:02:00 bia kernel: Process amarokapp (pid: 11080, threadinfo=c2a42000 task=c2260580)
Dec 10 22:02:00 bia kernel: Stack: cd5290a0 c10ed860 00000000 00000001 c011dbca 00000163 00200292 cd5290a0 
Dec 10 22:02:00 bia kernel:        00000000 d39749a0 caa42b00 e1f2b841 00000000 e1f69204 00000102 cd5290a0 
Dec 10 22:02:00 bia kernel:        c60e6bcc c60e6bcc e1f28416 00000000 e1f682c0 cd5290a0 00000000 d48fd000 
Dec 10 22:02:00 bia kernel: Call Trace:
Dec 10 22:02:00 bia kernel:  [<c011dbca>] change_page_attr+0x6a/0x80
Dec 10 22:02:00 bia kernel:  [<e1f2b841>] nv_vm_free_pages+0x7f/0xe7 [nvidia]
Dec 10 22:02:00 bia kernel:  [<e1f28416>] nv_kern_vma_release+0x7d/0x98 [nvidia]
Dec 10 22:02:00 bia kernel:  [<c016183f>] remove_vm_struct+0x4f/0x80
Dec 10 22:02:00 bia kernel:  [<c0163c7a>] exit_mmap+0x11a/0x170
Dec 10 22:02:01 bia kernel:  [<c01237d3>] mmput+0x33/0xa0
Dec 10 22:02:01 bia kernel:  [<c0128b27>] do_exit+0xd7/0x3d0
Dec 10 22:02:01 bia kernel:  [<c0128e97>] do_group_exit+0x37/0xa0
Dec 10 22:02:01 bia kernel:  [<c01042ab>] sysenter_past_esp+0x54/0x79
Dec 10 22:02:01 bia kernel: Code: 42 04 85 c0 74 30 31 c0 8b 1c 24 8b 74 24 04 8b 7c 24 08 8b 6c 24 0c 83 c4 10 c3 8b 51 0c eb c0 8d
Dec 10 22:02:01 bia kernel:  <1>Fixing recursive fault but reboot is needed!
Attached Files
File Type: txt message-oops.txt (3.1 KB, 212 views)
File Type: zip nvidia-bug-report.zip (30.0 KB, 159 views)
olivn is offline   Reply With Quote
Old 12-11-05, 10:57 AM   #108
FreeCatalonia
Registered User
 
Join Date: Jul 2004
Posts: 19
Default Re: 1.0-8174 for Linux x86 released

Quote:
Originally Posted by wschutzer
Same problem here ever since. I can never use the text-mode consoles after suspending to ram. Only a reboot will fix that.

I have the very same problem with DPMS. It seems that the backlight won't come back unless I did CTRL-F1 followed by CTRL-F7.

Anybody else out there with the same problems?
I have the very same problem with my Dell Inspiron with an NVIDIA Geforce 4, after suspending to disk (software suspend 2 with the nv.c hack, why hasn't nvidia included that simple patch yet?), my consoles are completely scrambled and unusable, EVEN after stripping the FB console support from the kernel. Setting vga=normal in lilo.conf doesn't help either. Only rebooting will restore the consoles, and another supend to disk will disable them again.

The 7174 driver didn't have that problem, could suspend to disk just fine WITH AGP ENABLED! All the later drivers suffer from this issue.

P.S. Is it me or the LCD screen is not as bright after suspending? It has a dull look, as if it had worn out. A complete reboot seems to restore the full brightness of the screen. Has anyone noticed?
FreeCatalonia is offline   Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 07:32 PM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 1998 - 2014, nV News.