Due to some changes that must have occured since nvidia-kernel-1.0.5336 and its glx counterpart there seems to be a problem with nvidia-kernel-1.0.6106 / nvidia-glx-1.0.6106 and any video for linux applications. This breaks the ability to run video capture on my system with applications like xawtv with the rivatv backend (my system uses the saa7108e encoder in case it is any help). This issue does not occur with the stock nv driver for xorg-x11, nor does it occur if I run the nvidia-kernel-1.0.5336-r4 with nvidia-glx-1.0.5336-r2. In case it is any help, below are some outputs from /proc/driver/rivatv (which, along with devices in /dev/v4l is nonexistant if I run nvidia-kernel-1.0.6106) $ cat /proc/driver/rivatv nVidia Chip: GeForce4 Ti4x00 Model: Gainward GF4 PowerPack! Ultra/750XP Architecture: NV20 (NV25) Access: Control [0xe8000000-0xe8ffffff] FB [0xd8000000-0xdfffffff] Interrupts: 25605 out of 97023 (DMA: 0, Overlay: 7, Missing: 71411) Device: busy VideoDecoder: Philips SAA7108E Tuner: unavailable AudioDecoder: unavailable AudioProcessor: unavailable IR chip: unavailable Reproducible: Always Steps to Reproduce: 1. emerge =media-video/nvidia-kernel-1.0.6106 =media-video/nvidia-glx-1.0.6106 2. modprobe rivatv (or other v4l dependent module) 3. video capture / v4l interface breaks Actual Results: See step 3. Expected Results: Video capture should still work. Also, here is output from emerge --info: Portage 2.0.50-r8 (default-x86-1.4, gcc-3.3.3, glibc-2.3.3.20040420-r0, 2.6.7-gentoo-r8) ================================================================= System uname: 2.6.7-gentoo-r8 i686 AMD Athlon(tm) XP 2100+ Gentoo Base System version 1.4.16 distcc 2.13 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [enabled] ccache version 2.3 [enabled] Autoconf: sys-devel/autoconf-2.59-r3 Automake: sys-devel/automake-1.8.3 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-mcpu=athlon-xp -O2 -pipe" CHOST="i686-pc-linux-gnu" COMPILER="gcc3" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3.2/share/config /usr/kde/3/share/config /usr/lib/mozilla/defaults/pref /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-mcpu=athlon-xp -O2 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache distcc fixpackages sandbox userpriv usersandbox" GENTOO_MIRRORS="ftp://ftp.ucsb.edu/pub/mirrors/linux/gentoo/ ftp://gentoo.chem.wisc.edu/gentoo/ http://cudlug.cudenver.edu/gentoo/ ftp://cudlug.cudenver.edu/pub/mirrors/distributions/gentoo/ ftp://gentoo.mirrors.pair.com/" MAKEOPTS="-j8" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.us.gentoo.org/gentoo-portage" USE="X Xaw3d aalib alsa apm arts avi berkdb cdr crypt cups encode esd foomaticdb gdbm gif gnome gpm gtk gtk2 guile imlib jack jack-caps jack-tmpfs java jpeg kde ldap libg++ libwww mad mikmod motif mozilla mpeg ncurses nls oggvorbis opengl oss pam pdflib perl png python qt quicktime readline scanner sdl slang spell ssl svga tcltk tcpd tetex truetype x86 xml2 xmms xv zlib"
This is just a sidenote as I would like to have full acceleration available to my nvidia subsystem, whether I run the open source (nv) driver or the closed source (nvidia) driver. That being said, video capture under nvidia-kernel-1.0.5336 is definitely less stable than under the open source nv driver, i.e., it seems to be more prone to screen corruption and other video rendering problems (e.g., jumpy screen behavior outside the capture window) when a video capture application is active. Video capture under the nv driver works fine, and can even use overlay vs. grabdisplay on my system without any problems.
Please test this with 6111 which just went into cvs. Report back on success or failures. Thanks.
I tried 6111 - problem still seems to be there: xawtv -c /dev/v4l/video0 -global:filter "linear blend" This is xawtv-3.86, running on Linux/i686 (2.6.7-gentoo-r8) can't open /dev/v4l/video0: No such file or directory v4l-conf had some trouble, trying to continue anyway v4l2: open /dev/v4l/video0: No such file or directory v4l2: open /dev/v4l/video0: No such file or directory v4l: open /dev/v4l/video0: No such file or directory no video grabber device available Tried again since the v4l direcgtory was not there: xawtv -c /dev/video0 -global:filter "linear blend" This is xawtv-3.86, running on Linux/i686 (2.6.7-gentoo-r8) can't open /dev/video0: No such device v4l-conf had some trouble, trying to continue anyway v4l2: open /dev/video0: No such device v4l2: open /dev/video0: No such device v4l: open /dev/video0: No such device no video grabber device available
Are you sure v4l is actually loading on your system properly? Try the newest 6111 with the enable_pci patch see what happens.
As I stated previously, v4l is working on my system (i.e., it works fine with the nv driver), but does not seem to be allowed to load, whether nvidia kernel 1.0.6106 or 1.0.6111. Where is the enable pci patch for 1.0.6111? If it is included in nvidia-kernel-1.0.6111, I have already tested this configuration.
andrew@asteroid ~ $ lsmod |grep nvidia nvidia 4813748 12 andrew@asteroid ~ $ ls -l /dev/v4l total 0 crw-rw---- 1 root video 81, 64 Sep 10 2004 radio0 crw-rw---- 1 root video 81, 224 Sep 10 2004 vbi0 crw-rw---- 1 root video 81, 0 Sep 10 2004 video0 andrew@asteroid ~ $ dmesg |grep NVIDIA nvidia: module license 'NVIDIA' taints kernel. NVRM: loading NVIDIA Linux x86 NVIDIA Kernel Module 1.0-6111 Tue Jul 27 07:55:38 PDT 2004 Works for me with bttv
Bump. Does 6629 fix the issue?
Per Andrew Bevitt's suggestion, I emerged nvidia-kernel-1.0.6629, and configured xorg-x11 to use the kernel. Unfortunately it did not resolve the v4l conflict. I'm still seeing "no such device errors" even when checking to see that the device nodes are in place.
According to rivatv sourceforge page forum, nvidia blocks access some resource > 1.0.5. I was under the impression that it had something to do with ic2 and the graphics card, although it may be the same. I cannot access my rivatv card partly because it is a saa7134 type. when i load an stb bt878 im able to get a stream from a selected device but if i try to record it crashes saying it can't get buffers. there was talk in xorg cvs that dri or expended resources caused this. X modules have different ports you can check w/ xvattr or linuxvideostudio.
Is anyone working this bug with nvidia?
David, none of us on the x11 team seem to have the hardware to reproduce this issue -- can we put the onus on you to contact nvidia's bug shooters to report this (and post url to same)?
Couple of places you could report that (don't do both): 1) bugs.freedesktop.org, nvidia proprietary driver in xorg 2) linux-bugs@nvidia.com
I'll take the suggestions to post this bug upstream, either to nvidia or freedesktop.org as noted by Donnie Berkholz. In the meantime, I can positively confirm that this issue applies to the latest version of the nvidia-kernel, at time of my posting, that would be nvidia-kernel-1.0.7167 (under gentoo, this was tested with nvidia-kernel-1.0.7167-r1 / nvidia-glx-1.0.7167-r1).
FYI - this bug has been posted upstream to freedesktop.org at https://bugs.freedesktop.org/show_bug.cgi?id=2767
Thanks. =)
