Help on the new 2.6 test kernel

[Pekka Pietikainen]

On Thu, Oct 09, 2003 at 10:51:45PM +0200, lars wrote:
> thats not the way it should be.
> maybe zander, who wrote the nvidia 2.6 patches can help you with
> this... had a little chat on irc with him and he tried vanilla test6-bk8 on 
> his laptop, worked ok for him and later for me too.
> you can reach him via irc @ irc.freenode.net #nvidia

I have a theory, that there are actually two nvidia + arjanv-2.6 issues

exec-shield does something that confuses the binary-only parts of the
driver, this prevents loading of the module and might not be fixable apart 
from removing exec-shield (or maybe kludging the os_mem_cmp thing that
__nvsym_00670 calls but that would be truly evil). 

The other one looks AGP 8x/KT400 related, which I explained in an earlier
mail. Basically disabling AGP or using NVAGP works, agpgart doesn't
unless you start up X with NVAGP first and then use agpgart.

Some magic pci "make it work" bit not getting flipped, perhaps?

(agpgart-X is the doesn't work scenario)

--- agpgart-X   2003-10-10 01:40:43.000000000 +0300
+++ agpgart-after-nvagp 2003-10-10 01:36:54.000000000 +0300
@@ -6,7 +6,7 @@
00:00.0 Class 0600: 1106:3189
        Subsystem: 1043:807f
        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B-
        Status: Cap+ 66Mhz+ UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort+ <MAbort+ >SERR- <PERR-
        Latency: 0
        Region 0: Memory at f8000000 (32-bit, prefetchable) [size=64M]
        Capabilities: [80] AGP version 3.5
                Status: RQ=32 Iso- ArqSz=0 Cal=2 SBA+ ITACoh- GART64-
HTrans- 64bit- FW- AGP3+ Rate=x4,x8
-               Command: RQ=32 ArqSz=0 Cal=2 SBA- AGP+ GART64- 64bit- FW-
Rate=x8
+               Command: RQ=32 ArqSz=0 Cal=2 SBA+ AGP+ GART64- 64bit- FW-
Rate=x8
        Capabilities: [c0] Power Management version 2
                Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA
PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-

Life really is too short to be debugging binary crap, so I'll leave it
at this for tonight . Interestingly enough the driver says it disables
SBA by default on x86 btw., but either I am too tired to think or the thing
actually enabled it!

(I have full lspci's in case someone actually cares)