X problems with nvidia drivers under linux

Jeff Vaughan wrote:
psychobudgie wrote:

On Tue, 30 Dec 2003 12:55:44 -0500, Jeff Vaughan wrote:


I'm having a problem where X freezes when using nvidia's 5328 or 4363
drivers under linux Mandake. The symptoms are that my keyboard is
not responsive but my mouse cursor still moves. Even though I can't
switch to a virtual console, I am able to ssh in and find that X is
using 90% of cpu time. If I restart X everything works fine until
the next crash. I haven't found anything suspicious in
/var/log/XFree86.0.log, and I don't know of other logs I should check.

I've seen several posts claiming this is bug in the nvidia drivers
(and I don't experience it with the open source "nv" drivers). Does
anyone know of any workarounds? What about downgrading to older
drivers or changing kernel versions? I'm running 2.4.22-10mdk.

Thank you.

Jeff



http://www.sh.nu/download/nvidia/linux-2.4/

try this patched driver. The latest driver has a rather large AGP problem
on newer VIA chipsets. Also set Force no ACPI in LILO. Should fix any
probs you are having.


Thanks to everyone for the help. I'm trying the patched drivers right
now and will give 4496 and 4363 a shot if problems persist. Hopefully X
will last the night

Jeff


I've cycled through a bunch of driver versions. So far my best results
have been with the patched 5328 drivers, which crash about twice a day.
I haven't tried 4496 yet, but I need stability over 3d acceleration
for a few days. So I'm back with the nv drivers for now.

Is there a good way to verify that ACPI abd APIC are disabled? I'm
wondering if I may not have completely turned those services off.

Jeff

Jeff Vaughan wrote:


I've cycled through a bunch of driver versions. So far my best results
have been with the patched 5328 drivers, which crash about twice a day.
I haven't tried 4496 yet, but I need stability over 3d acceleration
for a few days. So I'm back with the nv drivers for now.

Is there a good way to verify that ACPI abd APIC are disabled? I'm
wondering if I may not have completely turned those services off.

Jeff

Disable ACPI in Bios may help also make sure that "fast writes" is disabled,
this can cause probs with some cards.

boot with the parameter of acpi=off at the grub/lilo screen, just type it in
it should be echoed on the bottom line, you can then check the boot.msg which
should say :-


---snip---

Inspecting /boot/System.map-2.4.21-144-athlon
Loaded 21343 symbols from /boot/System.map-2.4.21-144-athlon.
Symbols match kernel version 2.4.21.
Loaded 165 symbols from 7 modules.
klogd 1.4.1, log source = ksyslog started.
4Linux version 2.4.21-144-athlon ) (gcc version 3.3.1
(SuSE Linux)) #1 Fri Nov 14 00:01:10 UTC 2003
6BIOS-provided physical RAM map:
4 BIOS-e820: 0000000000000000 - 00000000000a0000 (usable)
4 BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
4 BIOS-e820: 0000000000100000 - 0000000040000000 (usable)
4 BIOS-e820: 00000000fec00000 - 00000000fec01000 (reserved)
4 BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)
4 BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)
5128MB vmalloc/ioremap area available.
5128MB HIGHMEM available.
5896MB LOWMEM available.
7ACPI: have wakeup address 0xc0001000
4On node 0 totalpages: 262144
4zone(0): 4096 pages.
4zone(1): 225280 pages.
4zone(2): 32768 pages.
4Building zonelist for node : 0
4Kernel command line: root=/dev/hdb3 vga=0x31a acpi=off desktop hdd=ide-scsi
hddlun=0 splash=silent
6ide_setup: hdd=ide-scsi
6ide_setup: hddlun=0
6bootsplash: silent mode.
6Initializing CPU#0
4Detected 2254.893 MHz processor.
4Console: colour dummy device 80x25
4Calibrating delay loop... 4456.44 BogoMIPS
6Memory: 1032460k/1048576k available (1727k kernel code, 15732k reserved,
605k data, 176k init, 131072k highmem)
6Dentry cache hash table entries: 131072 (order: 8, 1048576 bytes)
6Inode cache hash table entries: 65536 (order: 7, 524288 bytes)
6Mount cache hash table entries: 512 (order: 0, 4096 bytes)
4Buffer-cache hash table entries: 65536 (order: 6, 262144 bytes)
4Page-cache hash table entries: 262144 (order: 8, 1048576 bytes)
6CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
6CPU: L2 Cache: 512K (64 bytes/line)
6Intel machine check architecture supported.
6Intel machine check reporting enabled on CPU#0.
7CPU: After generic, caps: 0383fbff c1c3fbff 00000000 00000000
7CPU: Common caps: 0383fbff c1c3fbff 00000000 00000000
4CPU: AMD Athlon(tm) stepping 00
6Enabling fast FPU save and restore... done.
6Enabling unmasked SIMD FPU exception support... done.
6Checking 'hlt' instruction... OK.
4POSIX conformance testing by UNIFIX
4mtrr: v1.40 (20010327) Richard Gooch )
4mtrr: detected mtrr type: Intel

vvvvvvvv--here--vvvvvvvvvvvv


6ACPI: Subsystem revision 20030619
6ACPI: Disabled via command line (acpi=off)
6PCI: PCI BIOS revision 2.10 entry at 0xfb540, last bus=2
6PCI: Using configuration type 1
6PCI: Probing PCI hardware
4PCI: Probing PCI hardware (bus 00)
6PCI: Discovered primary peer bus ff [IRQ]
6PCI: Using IRQ router default [10de/01e0] at 00:00.0


---pins---


HTH

--
Mark
Twixt hill and high water.
N.Wales, UK.
Email is spam trap try baskitcaise at gmx dot co dot uk

On Thu, 01 Jan 2004 11:10:21 -0500, Jeff Vaughan wrote:

I've cycled through a bunch of driver versions. So far my best results
have been with the patched 5328 drivers, which crash about twice a day.
I haven't tried 4496 yet, but I need stability over 3d acceleration
for a few days. So I'm back with the nv drivers for now.

If you would do as suggested (use the 4496), you would have both stability
and acceleration. Shrug.

--
If you're not on the edge, you're taking up too much space.
Linux Registered User #327951