[mythtv-users] Re: Strange X behavior with nvidia drivers

Paul Hamm paulhamm at verizon.net
Wed Dec 15 15:01:01 UTC 2004


Axel

I took another look and found this patch for 6111.  Which should be
better.  Originally I was searching on the error and just grabbed the
first fix I came across.  I have not yet tried it.  I have 6629 running
now.  From this site http://kerneltrap.org/node/4022?
PHPSESSID=e1de6827263236c1ea9dd7cd5161f18d



--- linux-orig/arch/i386/mm/init.c      2004-09-26 03:43:57.944613000
+1000
+++ linux-new/arch/i386/mm/init.c       2004-09-28 02:37:21.787922000
+1000
@@ -41,6 +41,7 @@
 #include <asm/sections.h>
 unsigned int __VMALLOC_RESERVE = 128 << 20;
+EXPORT_SYMBOL(__VMALLOC_RESERVE);

 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 unsigned long highstart_pfn, highend_pfn;

On Wed, 2004-12-15 at 01:23 +0100, Axel Thimm wrote:
> On Tue, Dec 14, 2004 at 12:41:29PM -0500, Paul Hamm wrote:
> > No joy, I seem to have the same issue with 6111.  BTW I had to download
> > and install the nvidia drivers manually as 6111 and 6106 both give this
> > error if you use atrpms.
> > 
> > nvidia: Unknown symbol __VMALLOC_RESERVE
> > 
> > the fix is to unpack the NVIDIA-Linux-x86-1.0-6111-pkg1.run and edit
> > nv.c by adding the following line below the Global state comment.
> > 
> > unsigned int __VMALLOC_RESERVE;
> 
> But that's the wrong "fix", as it introduces a different
> __VMALLOC_RESERVE symbol than the one that is not anymore exported
> (but still used). So while there are no errors at module load time,
> any access to __VMALLOC_RESERVE will be to the above symbol and not
> the kernel's.
> 
> The fix is either in nvidia's sources (e.g. 6629) or in rexprting the
> symbol in the kernel (requires patching the kernel).
> 
> > I also tried Jarods tip of adding, which did nothing.
> > 
> > Option      "ConnectedMonitor" "crt"
> > 
> > I will drop the FC2 hard drive back in an see if I get the same
> > behavior.
> > 
> > On Tue, 2004-12-14 at 09:31, Doug Larrick wrote:
> > > Paul Hamm wrote:
> > > > I am currently going through an OS upgrade cycle to FC3.  One of my
> > > > machines has a Chaintech 7nif2 igp system board 512mb XP 3200+.  This
> > > > machine has been my test box for mythtv.  I was planning to switch it
> > > > over to my primary work station and change my work station over to a
> > > > mythtv server.  After installing FC3 (everything) and running apt-get
> > > > dist-upgrade, install KDE 3.3, latest FC3 kernel, and nvidia drivers.  I
> > > > noticed something odd.  I usually run my workstation at 1280x1024.  The
> > > > machine with the 7nif2 has always been run at 1024x768.  Now I want it
> > > > to run at 1280x1024, it does so using the nv driver.  The problem is
> > > > that if I use the nvidia driver 6629 I only get 1024x768 on the display
> > > > with a virtual 1280x1024.  Is this a known issue or am I missing
> > > > something blindingly obvious?
> > > 
> > > IIRC 6629 fixes some bugs in the code that reads resolution information 
> > > from the monitor.  Try adding Option IgnoreEDID in the appropriate 
> > > location (sorry, not in front of me).
> > > 
> > > You're lucky, though...  6629 is completely hosed for people with cards 
> > > earlier than about an FX5200.
> > > 
> > > -Doug
> > > 
> > > ______________________________________________________________________
> > > _______________________________________________
> > > mythtv-users mailing list
> > > mythtv-users at mythtv.org
> > > http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
> > 
> 
> > _______________________________________________
> > mythtv-users mailing list
> > mythtv-users at mythtv.org
> > http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
> 
> 
> _______________________________________________
> mythtv-users mailing list
> mythtv-users at mythtv.org
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users



More information about the mythtv-users mailing list