nouveau
16 years agoNV40EXA: Fix composite rotation.
Maarten Maathuis [Sat, 15 Dec 2007 12:35:45 +0000 (13:35 +0100)] 
NV40EXA: Fix composite rotation.

- Thanks to radeon (code) for inspiration and sample code.

16 years agorandr12: Add some ifdefs to make compiling better on older xserver's.
Maarten Maathuis [Sat, 15 Dec 2007 11:02:36 +0000 (12:02 +0100)] 
randr12: Add some ifdefs to make compiling better on older xserver's.

16 years agonv30: need to set default depth range
Patrice Mandin [Sat, 15 Dec 2007 10:44:41 +0000 (11:44 +0100)] 
nv30: need to set default depth range

16 years agonv30: only 4 texture units
Patrice Mandin [Sat, 15 Dec 2007 10:37:43 +0000 (11:37 +0100)] 
nv30: only 4 texture units

16 years agonv30: fix viewport clipping setup
Patrice Mandin [Sat, 15 Dec 2007 09:34:26 +0000 (10:34 +0100)] 
nv30: fix viewport clipping setup

16 years agoFix ugly bug, that creeped in.
Maarten Maathuis [Sat, 15 Dec 2007 01:00:37 +0000 (02:00 +0100)] 
Fix ugly bug, that creeped in.

16 years agorandr12: Preliminary support for rotation, please read notes.
Maarten Maathuis [Sat, 15 Dec 2007 00:44:59 +0000 (01:44 +0100)] 
randr12: Preliminary support for rotation, please read notes.

- It works, but it's slow, very slow.
- Each crtc now has it's own cursor map (fixing the bugs you never saw).
- Dualhead with one rotated screen may show minor corruption near the edges, this will disappear if a fullscreen app opens.
- Acceleration is possible, BUT:
- You need a git xserver.
- Enable NOUVEAU_EXA_PIXMAPS in nouveau_local.h (*maybe* this will become default later).
- Most of the composite hooks have problems with rotation, so left and right won't work, inverted does work for me.
- So please fix the composite hooks ;-)
- Note: No slowdowns are apparent when using proper acceleration.

16 years agoBetter restoration of TMDS regs on LVDS
Stuart Bennett [Sat, 15 Dec 2007 00:01:04 +0000 (00:01 +0000)] 
Better restoration of TMDS regs on LVDS

16 years agorandr12: It seems DVI on laptops wasn't as strange after all.
Maarten Maathuis [Fri, 14 Dec 2007 22:58:25 +0000 (23:58 +0100)] 
randr12: It seems DVI on laptops wasn't as strange after all.

16 years agoINIT_COPY_NV_REG and INIT_ZM_REG_GROUP_ADDRESS_LATCHED opcodes
Stuart Bennett [Fri, 14 Dec 2007 01:40:14 +0000 (01:40 +0000)] 
INIT_COPY_NV_REG and INIT_ZM_REG_GROUP_ADDRESS_LATCHED opcodes

Thanks to RAOF for the trace with these

Also correct a doc oopsy, and a bunch of superfluous whitespace

16 years agoInvert 18/24 bit interface flag, so that the default value (0) does not imply dithering
Stuart Bennett [Thu, 13 Dec 2007 02:30:06 +0000 (02:30 +0000)] 
Invert 18/24 bit interface flag, so that the default value (0) does not imply dithering

16 years agoName bios opcodes 4f and 50
Stuart Bennett [Thu, 13 Dec 2007 01:10:12 +0000 (01:10 +0000)] 
Name bios opcodes 4f and 50

Extrapolating from previous names, but marked as non canon

16 years agoGet VBIOS from pNv (cleanup)
Stuart Bennett [Thu, 13 Dec 2007 01:03:25 +0000 (01:03 +0000)] 
Get VBIOS from pNv (cleanup)

And a minor doc correction

16 years agorandr12: some minor cleanups, no actual changes
Stuart Bennett [Thu, 13 Dec 2007 00:15:21 +0000 (00:15 +0000)] 
randr12: some minor cleanups, no actual changes

16 years agoSetup NV40 LVDS
Stuart Bennett [Wed, 12 Dec 2007 15:54:53 +0000 (15:54 +0000)] 
Setup NV40 LVDS

16 years agoINIT_CONDITION_TIME opcode
Stuart Bennett [Wed, 12 Dec 2007 22:20:43 +0000 (22:20 +0000)] 
INIT_CONDITION_TIME opcode

16 years agoOpcode 4F
Stuart Bennett [Wed, 12 Dec 2007 18:08:07 +0000 (18:08 +0000)] 
Opcode 4F

16 years agoFP BIOS mode finding for nv40
Stuart Bennett [Mon, 10 Dec 2007 22:50:40 +0000 (22:50 +0000)] 
FP BIOS mode finding for nv40

16 years agoCR57 doc addition
Stuart Bennett [Tue, 11 Dec 2007 23:23:27 +0000 (23:23 +0000)] 
CR57 doc addition

16 years agorandr12: A good guess at the lower bits of SEL_CLK (i hope ;-)).
Maarten Maathuis [Wed, 12 Dec 2007 21:12:33 +0000 (22:12 +0100)] 
randr12: A good guess at the lower bits of SEL_CLK (i hope ;-)).

16 years agorandr12: Add some extra safeties (to mode checking).
Maarten Maathuis [Wed, 12 Dec 2007 20:03:34 +0000 (21:03 +0100)] 
randr12: Add some extra safeties (to mode checking).

16 years agorandr12: Enable dithering on lvds when needed.
Maarten Maathuis [Tue, 11 Dec 2007 20:43:02 +0000 (21:43 +0100)] 
randr12: Enable dithering on lvds when needed.

16 years agorandr12: SEL_CLK attempt 10, return to simplicity, which may actually work ;-)
Maarten Maathuis [Tue, 11 Dec 2007 19:22:23 +0000 (20:22 +0100)] 
randr12: SEL_CLK attempt 10, return to simplicity, which may actually work ;-)

16 years agorandr12: (mostly) Some fixes for NV40.
Maarten Maathuis [Tue, 11 Dec 2007 16:52:39 +0000 (17:52 +0100)] 
randr12: (mostly) Some fixes for NV40.

- The NV40 is a strange card, as it partially behaves like an NV3x.

16 years agorandr12: Fix SEL_CLK assignment again (for non-mobile cards) + fix crosswiring.
Maarten Maathuis [Tue, 11 Dec 2007 10:50:28 +0000 (11:50 +0100)] 
randr12: Fix SEL_CLK assignment again (for non-mobile cards) + fix crosswiring.

- I made a good guess for mobile cards, but i need more mmio-traces.
- I accidentally broke crosswiring for my card (and other's like it).

16 years agorandr12: Fix an earlier mistake, make a decent guess about dvi on laptops.
Maarten Maathuis [Mon, 10 Dec 2007 23:04:38 +0000 (00:04 +0100)] 
randr12: Fix an earlier mistake, make a decent guess about dvi on laptops.

- I could use some more mmiotraces of LVDS+DVI on NV4x.

16 years agorandr12: Improve crosswiring for digital outputs + minor changes and cleanup.
Maarten Maathuis [Mon, 10 Dec 2007 19:43:26 +0000 (20:43 +0100)] 
randr12: Improve crosswiring for digital outputs + minor changes and cleanup.

16 years agorandr12: Properly set SEL_CLK based on bus, not based on output.
Maarten Maathuis [Mon, 10 Dec 2007 15:31:47 +0000 (16:31 +0100)] 
randr12: Properly set SEL_CLK based on bus, not based on output.

- This hopefully fixes non-working dvi on some desktop cards and on mobile cards.

16 years agorandr12: Add config option for scaling mode.
Maarten Maathuis [Sun, 9 Dec 2007 23:03:49 +0000 (00:03 +0100)] 
randr12: Add config option for scaling mode.

16 years agorandr12: Some cleanup.
Maarten Maathuis [Sun, 9 Dec 2007 21:46:39 +0000 (22:46 +0100)] 
randr12: Some cleanup.

16 years agoBios register write function special case and some INIT_RESET improvements
Stuart Bennett [Sun, 9 Dec 2007 19:58:58 +0000 (19:58 +0000)] 
Bios register write function special case and some INIT_RESET improvements

nv32_wr weirdness only seen on NV40+, but it shouldn't hurt for earlier cards

16 years agoSome renaming for bios indexed port io
Stuart Bennett [Sun, 9 Dec 2007 17:09:03 +0000 (17:09 +0000)] 
Some renaming for bios indexed port io

16 years agorandr12: Revert dither value to that used by nv.
Maarten Maathuis [Sun, 9 Dec 2007 17:36:26 +0000 (18:36 +0100)] 
randr12: Revert dither value to that used by nv.

16 years agorandr12: Set FP_CONTROL correctly for NV2x cards.
Maarten Maathuis [Sun, 9 Dec 2007 17:34:10 +0000 (18:34 +0100)] 
randr12: Set FP_CONTROL correctly for NV2x cards.

16 years agorandr12: Improve output selection.
Maarten Maathuis [Sun, 9 Dec 2007 17:10:29 +0000 (18:10 +0100)] 
randr12: Improve output selection.

- Now detect if there is a dual link capable output at all.
- NV2x cards can't switch crtc on analog outputs.

16 years agoAdditional LVDS bios script sanity check
Stuart Bennett [Sun, 9 Dec 2007 15:51:02 +0000 (15:51 +0000)] 
Additional LVDS bios script sanity check

Avoids current incorrect situation when LVDS attempted on NV40

16 years agorandr12: Fix minor (potentional) bug.
Maarten Maathuis [Sun, 9 Dec 2007 15:29:14 +0000 (16:29 +0100)] 
randr12: Fix minor (potentional) bug.

16 years agorandr12: Add noscale mode and some misc fixes.
Maarten Maathuis [Sun, 9 Dec 2007 11:47:55 +0000 (12:47 +0100)] 
randr12: Add noscale mode and some misc fixes.

16 years agorandr12: Fix typo, it's "our" not "or".
Maarten Maathuis [Sat, 8 Dec 2007 17:34:04 +0000 (18:34 +0100)] 
randr12: Fix typo, it's "our" not "or".

16 years agorandr12: Assuming an output always has a crtc is a bad idea.
Maarten Maathuis [Sat, 8 Dec 2007 17:25:41 +0000 (18:25 +0100)] 
randr12: Assuming an output always has a crtc is a bad idea.

- Don't rely on current crtc to restore mode.

16 years agorandr12: Avoid that two outputs have the same clock selected.
Maarten Maathuis [Sat, 8 Dec 2007 15:49:21 +0000 (16:49 +0100)] 
randr12: Avoid that two outputs have the same clock selected.

16 years agorandr12: Fix bug in sel_clk assignment.
Maarten Maathuis [Sat, 8 Dec 2007 14:50:39 +0000 (15:50 +0100)] 
randr12: Fix bug in sel_clk assignment.

16 years agoBase error :)
Stuart Bennett [Sat, 8 Dec 2007 02:18:41 +0000 (02:18 +0000)] 
Base error :)

16 years agoFix failure to cope with varying table lengths
Stuart Bennett [Fri, 7 Dec 2007 16:56:26 +0000 (16:56 +0000)] 
Fix failure to cope with varying table lengths

16 years agoAdd a new valid register range, which appears in some gf7 biosen.
Stuart Bennett [Fri, 7 Dec 2007 02:54:36 +0000 (02:54 +0000)] 
Add a new valid register range, which appears in some gf7 biosen.

NVClock has it down as appearing in gf8 as a new PBUS range,
but the use on gf7 looks PBUS-like too

16 years agoSome more bios script opcodes
Stuart Bennett [Fri, 7 Dec 2007 03:13:52 +0000 (03:13 +0000)] 
Some more bios script opcodes

Thanks to swany for the trace for these

16 years agoSome M table parsing, and defer I table execution
Stuart Bennett [Fri, 7 Dec 2007 02:17:37 +0000 (02:17 +0000)] 
Some M table parsing, and defer I table execution

16 years agoImprove TMDS04 setting for LVDS
Stuart Bennett [Fri, 7 Dec 2007 00:18:25 +0000 (00:18 +0000)] 
Improve TMDS04 setting for LVDS

16 years agoTMDS regs setting for pre NV40
Stuart Bennett [Thu, 6 Dec 2007 23:37:01 +0000 (23:37 +0000)] 
TMDS regs setting for pre NV40

16 years agoBIOS version parsing
Stuart Bennett [Fri, 7 Dec 2007 00:30:14 +0000 (00:30 +0000)] 
BIOS version parsing

16 years agoRefactor BIT specific details out of TMDS setting
Stuart Bennett [Thu, 6 Dec 2007 17:01:49 +0000 (17:01 +0000)] 
Refactor BIT specific details out of TMDS setting

16 years agoThe 'T' table is clearly a TMDS table, call it as such
Stuart Bennett [Thu, 6 Dec 2007 18:48:04 +0000 (18:48 +0000)] 
The 'T' table is clearly a TMDS table, call it as such

16 years agoThe TMDS scripts being non-stubs does not seem to matter - they're still not executed
Stuart Bennett [Thu, 6 Dec 2007 05:03:49 +0000 (05:03 +0000)] 
The TMDS scripts being non-stubs does not seem to matter - they're still not executed

Message left as log clutter in case there's some future case of broken TMDS when the scripts might be useful

16 years agoTest for LVDS_INIT script
Stuart Bennett [Thu, 6 Dec 2007 04:14:56 +0000 (04:14 +0000)] 
Test for LVDS_INIT script

16 years agorandr12: Small fix for dual link dvi.
Maarten Maathuis [Thu, 6 Dec 2007 20:48:13 +0000 (21:48 +0100)] 
randr12: Small fix for dual link dvi.

16 years agorandr12: Improve compatibility with DCB 2.0
Maarten Maathuis [Thu, 6 Dec 2007 18:54:08 +0000 (19:54 +0100)] 
randr12: Improve compatibility with DCB 2.0

16 years agorandr12: oops, lost the restore case
Stuart Bennett [Thu, 6 Dec 2007 02:28:18 +0000 (02:28 +0000)] 
randr12: oops, lost the restore case

16 years agorandr12: a common function to fix TMDS04, and new use of LVDS_RESET
Stuart Bennett [Thu, 6 Dec 2007 01:54:45 +0000 (01:54 +0000)] 
randr12: a common function to fix TMDS04, and new use of LVDS_RESET

Plus some corrections/additions to call_lvds_script, and a minor CR5758 doc update

16 years agorandr12: Fix comment.
Maarten Maathuis [Wed, 5 Dec 2007 23:28:49 +0000 (00:28 +0100)] 
randr12: Fix comment.

16 years agorandr12: Some assorted fixes, which hopefully improve things for dual link dvi.
Maarten Maathuis [Wed, 5 Dec 2007 23:25:48 +0000 (00:25 +0100)] 
randr12: Some assorted fixes, which hopefully improve things for dual link dvi.

16 years agorandr12: replace preferred_ramdac with preferred_output.
Maarten Maathuis [Wed, 5 Dec 2007 19:41:27 +0000 (20:41 +0100)] 
randr12: replace preferred_ramdac with preferred_output.

16 years agorandr12: Clean up the unneeded ramdac variable.
Maarten Maathuis [Wed, 5 Dec 2007 19:27:57 +0000 (20:27 +0100)] 
randr12: Clean up the unneeded ramdac variable.

16 years agorandr12: Pick right register set, to correctly "guess" the bios/VT clock.
Maarten Maathuis [Wed, 5 Dec 2007 18:03:52 +0000 (19:03 +0100)] 
randr12: Pick right register set, to correctly "guess" the bios/VT clock.

- A wrong register set leads to improper detection of crosswiring, thus selecting the wrong clock.
- I noticed a 0.675 MHz difference in the clock (which is vpll2), when instead i should have had vpll1.

16 years agorandr12: Small fix.
Maarten Maathuis [Wed, 5 Dec 2007 15:51:04 +0000 (16:51 +0100)] 
randr12: Small fix.

16 years agorandr12: Final commit, restore TMDS regs before everything else.
Maarten Maathuis [Wed, 5 Dec 2007 15:47:07 +0000 (16:47 +0100)] 
randr12: Final commit, restore TMDS regs before everything else.

- Use the bios to program the tmds registers.
- Restore TMDS registers in nv_crtc_restore, to make sure it's early enough (see code note).
- It works for me, let me know if it breaks stuff ;-)

16 years agorandr12: Intermediate bios based flat panel init commit, still a few minor issues.
Maarten Maathuis [Wed, 5 Dec 2007 06:57:30 +0000 (07:57 +0100)] 
randr12: Intermediate bios based flat panel init commit, still a few minor issues.

16 years agoAdd LVDS script function, and use it for DPMS
Stuart Bennett [Wed, 5 Dec 2007 00:35:35 +0000 (00:35 +0000)] 
Add LVDS script function, and use it for DPMS

16 years agoParse the LVDS Manufacturer table
Stuart Bennett [Wed, 5 Dec 2007 00:22:22 +0000 (00:22 +0000)] 
Parse the LVDS Manufacturer table

Use dual link flag to set bit 31 of FP_CONTROL properly

16 years agoModify DCB dual link flag, and DCB config parsing for LVDS
Stuart Bennett [Tue, 4 Dec 2007 20:31:06 +0000 (20:31 +0000)] 
Modify DCB dual link flag, and DCB config parsing for LVDS

The DCB duallink flag indicates that this setup may be used for duallink;
it obviously can't know whether a duallink display is plugged or not

Use LVDS DCB conf flags for mode detection / selection

16 years agorandr12: Add fullscreen gpu scaling (only as randr property for the moment).
Maarten Maathuis [Tue, 4 Dec 2007 08:06:10 +0000 (09:06 +0100)] 
randr12: Add fullscreen gpu scaling (only as randr property for the moment).

16 years agoBetter T table stuff.
Stuart Bennett [Tue, 4 Dec 2007 01:31:24 +0000 (01:31 +0000)] 
Better T table stuff.

Hopefully now usable by calling parse_t_table(...) at the appropriate place

Also collapse some messages, and add an array bound check

16 years agoAllow BIOS execution at runtime, if wanted
Stuart Bennett [Tue, 4 Dec 2007 01:40:37 +0000 (01:40 +0000)] 
Allow BIOS execution at runtime, if wanted

And comment the opcodes I haven't double checked yet

16 years agoMake the bios a bit more accessible
Stuart Bennett [Tue, 4 Dec 2007 01:31:24 +0000 (01:31 +0000)] 
Make the bios a bit more accessible

And make a few things use the new structure

16 years agorandr12: set CR58 appropriately for 0 and 2 cases
Stuart Bennett [Tue, 4 Dec 2007 00:00:40 +0000 (00:00 +0000)] 
randr12: set CR58 appropriately for 0 and 2 cases

16 years agorandr12: refactor dcb usage
Stuart Bennett [Mon, 3 Dec 2007 23:35:01 +0000 (23:35 +0000)] 
randr12: refactor dcb usage

heads is also a better name than head

16 years agorandr12: some code economy
Stuart Bennett [Mon, 3 Dec 2007 22:26:33 +0000 (22:26 +0000)] 
randr12: some code economy

Only addition is adding call to output2->funcs->commit when stealing a ramdac,
no logical changes otherwise AFAICT

16 years agorandr12: Improve CR57 and CR58 infrastructure and saving/restoration
Stuart Bennett [Tue, 4 Dec 2007 00:06:05 +0000 (00:06 +0000)] 
randr12: Improve CR57 and CR58 infrastructure and saving/restoration

16 years agorandr12: A few minor comment changes.
Maarten Maathuis [Mon, 3 Dec 2007 19:55:26 +0000 (20:55 +0100)] 
randr12: A few minor comment changes.

16 years agorandr12: Select CRTC 1 on pre-nv40 (stupid bug) + initial support choosing scaling...
Maarten Maathuis [Mon, 3 Dec 2007 19:51:16 +0000 (20:51 +0100)] 
randr12: Select CRTC 1 on pre-nv40 (stupid bug) + initial support choosing scaling method on tmds.

- The output property SCALING_MODE accepts "gpu" and "panel".
- The first obeys aspect ratio, the second always goes fullscreen.
- This is just a first attemp at exposing an output property.

16 years agoShow the id's of unknown bit structures in the bios.
Maarten Maathuis [Sun, 2 Dec 2007 21:10:56 +0000 (22:10 +0100)] 
Show the id's of unknown bit structures in the bios.

16 years agorandr12: dual-dvi cards need extra consideration.
Maarten Maathuis [Sun, 2 Dec 2007 11:02:08 +0000 (12:02 +0100)] 
randr12: dual-dvi cards need extra consideration.

16 years agoFix endian bug on BIT biosen
Stuart Bennett [Sun, 2 Dec 2007 02:33:57 +0000 (02:33 +0000)] 
Fix endian bug on BIT biosen

16 years agoDon't use PROM if the checksum is bad
Stuart Bennett [Sun, 2 Dec 2007 02:14:32 +0000 (02:14 +0000)] 
Don't use PROM if the checksum is bad

16 years agorandr12: print "or" _pre_ ffs-ing
Stuart Bennett [Sat, 1 Dec 2007 09:55:31 +0000 (09:55 +0000)] 
randr12: print "or" _pre_ ffs-ing

16 years agoDCB parsed flag for duallink outputs
Stuart Bennett [Sat, 1 Dec 2007 09:53:05 +0000 (09:53 +0000)] 
DCB parsed flag for duallink outputs

Light on proof, strong on suspicion with this one

16 years agorandr12: Always use vpll2_true on nv4x cards.
Maarten Maathuis [Sun, 2 Dec 2007 01:08:20 +0000 (02:08 +0100)] 
randr12: Always use vpll2_true on nv4x cards.

16 years agorandr12: A safer bet for TMDS register 0x2e.
Maarten Maathuis [Sun, 2 Dec 2007 00:48:01 +0000 (01:48 +0100)] 
randr12: A safer bet for TMDS register 0x2e.

16 years agorandr12: Remove preferred_crtc, since it's probably incorrect.
Maarten Maathuis [Sun, 2 Dec 2007 00:34:49 +0000 (01:34 +0100)] 
randr12: Remove preferred_crtc, since it's probably incorrect.

16 years agorandr12: Improve output register setting.
Maarten Maathuis [Sun, 2 Dec 2007 00:08:41 +0000 (01:08 +0100)] 
randr12: Improve output register setting.

16 years agorandr12: Fix up some dvi related issues, thanks to MighMoS.
Maarten Maathuis [Sat, 1 Dec 2007 17:40:46 +0000 (18:40 +0100)] 
randr12: Fix up some dvi related issues, thanks to MighMoS.

16 years agorandr12: Fix NV4x dvi clock selection, mostly affects G70 cards.
Maarten Maathuis [Sat, 1 Dec 2007 10:23:50 +0000 (11:23 +0100)] 
randr12: Fix NV4x dvi clock selection, mostly affects G70 cards.

16 years agoMake the debug variable global to all DMA calls.
Stephane Marchesin [Sat, 1 Dec 2007 09:33:53 +0000 (10:33 +0100)] 
Make the debug variable global to all DMA calls.

16 years agoThe BIT bios T table.
Stuart Bennett [Sat, 1 Dec 2007 06:29:09 +0000 (06:29 +0000)] 
The BIT bios T table.

T might stand for TMDS.
Parsing once on start may not be the correct answer, could be a runtime thing

16 years agoBump DCB table init so that INIT_50 can use it
Stuart Bennett [Sat, 1 Dec 2007 06:37:17 +0000 (06:37 +0000)] 
Bump DCB table init so that INIT_50 can use it

16 years ago"or" can be 0xc
Stuart Bennett [Sat, 1 Dec 2007 05:35:40 +0000 (05:35 +0000)] 
"or" can be 0xc

Quite what this means, or why Nvidia couldn't give it a more descriptive
variable name in the g80 driver is anyone's guess

16 years agoUse Nvidia's naming for CRTC_INDEX_COLOR
Stuart Bennett [Sat, 1 Dec 2007 05:17:47 +0000 (05:17 +0000)] 
Use Nvidia's naming for CRTC_INDEX_COLOR

16 years agoBios script code 0x50
Stuart Bennett [Sat, 1 Dec 2007 05:16:20 +0000 (05:16 +0000)] 
Bios script code 0x50

16 years agoAdd dcb "location" field. 0 for on-chip, !0 for external
Stuart Bennett [Sat, 1 Dec 2007 00:45:32 +0000 (00:45 +0000)] 
Add dcb "location" field. 0 for on-chip, !0 for external

16 years agorandr12: Don't unset bit 8-15 in sel_clk.
Maarten Maathuis [Fri, 30 Nov 2007 15:47:26 +0000 (16:47 +0100)] 
randr12: Don't unset bit 8-15 in sel_clk.

16 years agorandr12: Big commit, read message.
Maarten Maathuis [Fri, 30 Nov 2007 15:43:33 +0000 (16:43 +0100)] 
randr12: Big commit, read message.

- Outputs (on nv4x) should now select vpll2 instead of vpll1 when neccesary (this took a long time to figure out).
- Register 0x00680594 is now set correctly to allow analog secondary outputs to work (for me at least).
- I found the bit that triggers non-VGA modesetting in nv10reg, so in the future we could copy the blob in that area.
- Misc fixes to a lot of things.