|
|
#1 | |
|
Registered User
Join Date: Nov 2002
Posts: 4
|
Dear All,
I am using RH7.3 with kernel 2.4.18-19.7.x. (i686, non-smp) on an HP Vectra VL420 with a low-profile Quadro 2MX card. The NVIDIA_kernel was self-compiled (since NVidia provides a binary module for 2.4.18-18.7.x but not for 2.4.18-19.7.x). The NVidia_GLX rpm is from NVidia's site. I have the following problem: Local and remote X11 apps work fine. Local GLX apps work fine but remote GLX apps crash the X server -immediately- with a sig11 (remote X11-only apps -do- work fine). I tried this with apps exported from an Sgi Octane running IRIX 6.5.15f and also with a SunBlade 2000 running Solaris 8 with SUN's OpenGL 1.2.3. I think to remember that it worked fine on my machine with 2.4.18-18.7.x but I am unsure. I am reporting this here in the hope that someone will know why it is happening or will know how to obtain debug output from the NVidia driver (have not found a 'debug' mode in the documentation). I have included info from my XFfree86-4 file, info from the kernel log (I'm using agpgart on my intel AGP chipset) and info from the /proc/drive/nvidia filesystem.. Thanks for reading. Vincent S. Cojot, coyote@step.polymtl.ca root@myhostname:[~][1006]# dmesg [...snip...] 0: nvidia: loading NVIDIA Linux x86 nvidia.o Kernel Module 1.0-4191 Mon Dec 9 11:49:01 PST 2002 Linux agpgart interface v0.99 (c) Jeff Hartmann agpgart: Maximum main memory to use for agp memory: 94M agpgart: Detected Intel i845 chipset agpgart: AGP aperture is 64M @ 0xe0000000 [...snip...] root@myhostname:[~][1008]# lsmod|grep -i agp agpgart 40256 3 (autoclean) root@myhostname:[~][1023]# modinfo nvidia filename: /lib/modules/2.4.18-19.7.x/kernel/drivers/video/nvidia.o description: <none> author: <none> license: "NVIDIA" parm: NVreg_VideoMemoryTypeOverride int parm: NVreg_EnableVia4x int parm: NVreg_EnableALiAGP int parm: NVreg_ReqAGPRate int parm: NVreg_UpdateKernelAGP int parm: NVreg_EnableAGPSBA int parm: NVreg_EnableAGPFW int parm: NVreg_SoftEDIDs int parm: NVreg_Mobile int parm: NVreg_ResmanDebugLevel int parm: NVreg_FlatPanelMode int root@myhostname:[~][1024]# modinfo agpgart filename: /lib/modules/2.4.18-19.7.x/kernel/drivers/char/agp/agpgart.o description: <none> author: "Jeff Hartmann <jhartmann@precisioninsight.com>" license: "GPL and additional rights" parm: agp_try_unsupported int Config file: section "files" fontpath "/usr/X11R6/lib/X11/fonts/local" fontpath "/usr/X11R6/lib/X11/fonts/misc" fontpath "/usr/X11R6/lib/X11/fonts/75dpi:unscaled" fontpath "/usr/X11R6/lib/X11/fonts/100dpi:unscaled" fontpath "/usr/X11R6/lib/X11/fonts/Type1" fontpath "/usr/X11R6/lib/X11/fonts/CID" fontpath "/usr/X11R6/lib/X11/fonts/75dpi" fontpath "/usr/X11R6/lib/X11/fonts/100dpi" fontpath "/opt/SoftWindows/fonts" endsection section "module" subsection "extmod" option "omit xfree86-dga" endsubsection load "dbe" load "glx" load "type1" load "freetype" endsection section "inputdevice" identifier "keyboard0" driver "keyboard" option "autorepeat" "100 30" endsection section "inputdevice" identifier "mouse0" driver "mouse" option "device" "/dev/psaux" option "protocol" "ps/2" endsection /etc/X11/XF86Config-4: section "modes" identifier "modes0" modeline "512x384" 19.392 512 528 592 640 384 385 388 404 modeline "512x384" 21.978 512 528 592 640 384 385 388 404 modeline "1280x1024@60" 108 1280 1328 1440 1688 1024 1025 1028 1066 +hsync +vsync modeline "1600x1024" 108 1600 1616 1656 1704 1024 1027 1030 1056 Modeline "1920x1200" 245.500 1920 1984 2240 2584 1200 1203 1206 1250 endsection section "monitor" identifier "SGI_21inch_GDM-5411" vendorname "SGI" modelname "GDM-5411" horizsync 30-121 vertrefresh 48-75 # monitor-upper-limit=160; specify desired max freq usemodes "modes0" option "dpms" endsection section "device" identifier "device0" driver "nvidia" endsection section "screen" identifier "screen0" device "device0" monitor "SGI_21inch_GDM-5411" # specify appropriate monitor defaultdepth 24 Option "nvagp" "2" #uncomment this to disable AGP Option "CursorShadow" "1" subsection "display" depth 24 modes "1152x864" "1024x768" "800x600" "640x480" "512x384" endsubsection subsection "display" depth 16 modes "1152x864" "1024x768" "800x600" "640x480" "512x384" endsubsection endsection Modules.conf: alias char-major-195 nvidia options nvidia NVreg_EnableAGPFW=0 Linux myhostname 2.4.18-19.7.x #1 Thu Dec 12 09:00:42 EST 2002 i686 unknown root@myhostname:[/var/log][1010]# rpm -q NVIDIA_GLX NVIDIA_GLX-1.0-4191 root@myhostname:[/var/log][1011]# rpm -q NVIDIA_kernel NVIDIA_kernel-1.0-4191 root@myhostname:[/var/log][1013]# cat /proc/driver/nvidia/cards/0 Model: Quadro2 MXR/EX/Go IRQ: 11 Video BIOS: 03.11.01.26.14 Card Type: AGP root@myhostname:[/var/log][1014]# cat /proc/driver/nvidia/agp/card Fast Writes: Supported SBA: Not Supported AGP Rates: 4x 2x 1x Registers: 0x1f000017:0x1f000104 root@myhostname:[/var/log][1015]# cat /proc/driver/nvidia/agp/host-bridge Host Bridge: Intel i845 Fast Writes: Supported SBA: Supported AGP Rates: 4x 2x 1x Registers: 0x1f000217:0x00000104 root@myhostname:[/var/log][1016]# cat /proc/driver/nvidia/agp/status Status: Enabled Driver: AGPGART AGP Rate: 4x Fast Writes: Disabled SBA: Disabled root@myhostname:[/var/log][1017]# cat /proc/driver/nvidia/version NVRM version: NVIDIA Linux x86 nvidia.o Kernel Module 1.0-4191 Mon Dec 9 11:49:01 PST 2002 GCC version: gcc version 2.96 20000731 (Red Hat Linux 7.3 2.96-113) root@myhostname:[~][1015]# grep WW /var/log/XFree86.0.log (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (WW) No Layout section. Using the first Screen section. (WW) `fonts.dir' not found (or not valid) in "/usr/X11R6/lib/X11/fonts/local". (WW) `fonts.dir' not found (or not valid) in "/usr/X11R6/lib/X11/fonts/CID". (WW) The directory "/opt/SoftWindows/fonts" does not exist. (WW) NVIDIA(0): Failure reading EDID parameters for display device 0. |
|
|
|
|
|
|
#2 | |
|
Registered User
Join Date: Sep 2002
Posts: 52
|
This is something I have complained about several times. In my experience 4191 will not handle GLX connections from anything other than another linux/nvidia box, even though earlier drivers work just fine. Please send your experiences along to linux-bugs@nvidia.com ... remotely displaying GL apps to a linux box is IMNSHO an extremely important feature that (at least in the past) has a been a very strong point in favor of the nvidia drivers since certain other drivers <cough>DRI<cough> don't do this properly.
In the past linux-bugs@nvidia.com has been pretty repsonsive ... at least as far as the bugs I report appear to be fixed in the next release. So lets make sure that nvidia is aware of how their customers are using their products so they will be sure to properly test them before future releases. |
|
|
|
|
|
|
#3 |
|
Registered User
Join Date: Nov 2002
Posts: 4
|
Merlin42,
Thanks for the reply.. now at least I know that I'm not dreaming. ![]() I have another machine that also has 4191 but with kernel-2.4.18-18.7.xsmp and a Quadro4 750XGL. I will check to see if I can reproduce this on that other machine.. If the problem is in 4191, then it's really really bad.. I thought the problem was in my configuration or perhaps in the -19.7.x kernel from Redhat but I did not think that this could come from the driver.. *Sigh*. Thanks. BTW: I e-mailed linux-bugs@nvidia just after I posted here.. So perhaps I'll get a reply.. Vincent S. Cojot, coyote@step.polymtl.ca |
|
|
|
|
|
#4 | |
|
Registered User
Join Date: Nov 2002
Posts: 4
|
Hello All,
I tested that this morning on my Sgi 330 with 1.5GB Ram, a Quadro4 750XGL, Redhat 7.2 (Kernel 2.4.18-18.7.x + updates). Both rpm's (NVIDIA_Kernel and NVIDIA_GLX) are taken from NVidia's site. Remote GLX apps from an Sgi Octane running IRIX 6.5 crash my local display. Can NVidia fix that problem? (pretty please). 1.0-4071 is supposed to be -required- for Maya 4.5 under Linux but it was never released. 41.91 was relased with Overlays (THANK YOU NVidia) but it came with the remote GLX crash bug.. ![]() Thank you. Vincent S. Cojot, coyote@step.polymtl.ca |
|
|
|
|
|
|
#5 |
|
International God of Sex
Join Date: Jul 2002
Location: en.gb.uk
Posts: 655
|
Yes, I've seen this issue re-appear.
This isn't the first time that fixed problems have become "un-fixed" in a subsequent official driver release. Does make one wonder about the level of regression testing at NV...
__________________
Got nuts? |
|
|
|
|
|
#6 | |
|
Registered User
Join Date: Nov 2002
Posts: 4
|
Merlin42, Just so you know. I got e-mail back from an NVidia engineer and also from Andy Mecham. NVidia is aware of the problem and is researching a fix so I guess this issue will eventually get fixed and make us happy again.
![]() Vincent Quote:
|
|
|
|
|
![]() |
| Thread Tools | |
|
|