Dave Airlie [Wed, 28 May 2008 00:03:10 +0000 (10:03 +1000)]
r500: cleanup warnings and include files
Corbin Simpson [Tue, 27 May 2008 09:12:10 +0000 (02:12 -0700)]
r5xx: Add OPCODE_XPD. In working condition, I might add.
And we're officially finished with the ARB_fragment_program
instruction set. It's worth noting that LIT is still not reliable.
SIN and COS were fixed a few commits ago. We're finished with stage 1!
Whoohoo!
Corbin Simpson [Tue, 27 May 2008 08:01:46 +0000 (01:01 -0700)]
r5xx: Just a few small LIT fixes.
Still broken; will fix tomorrow.
Corbin Simpson [Tue, 27 May 2008 05:34:32 +0000 (22:34 -0700)]
r5xx: Fix emit_mov() regression.
Specifically, fix improper swizzling.
Dave Airlie [Tue, 27 May 2008 08:15:14 +0000 (18:15 +1000)]
r500: the cs is shared per context - doh so don't use for program upload.
Also remove some unused bits of the r500 fragprog struct
Dave Airlie [Tue, 27 May 2008 07:39:35 +0000 (17:39 +1000)]
r500: need to re-setup inst offset/end for translated programs
this fixes texenv
Dave Airlie [Tue, 27 May 2008 07:11:04 +0000 (17:11 +1000)]
r500: reset fp/fp_const counts
Dave Airlie [Tue, 27 May 2008 07:08:03 +0000 (17:08 +1000)]
r300/r500: emit flush inside vap_cntl state atom
Not sure if this is a good or bad plan, it certainly doesn't make things worse here.
Dave Airlie [Tue, 27 May 2008 05:29:39 +0000 (15:29 +1000)]
r500: hopefully fix 4096 texture harder
Corbin Simpson [Tue, 27 May 2008 05:12:24 +0000 (22:12 -0700)]
r5xx: Fix FP temp counting.
One of the ref counters wasn't being added to the temp counter.
Yet another product of late-night coding...
Dave Airlie [Tue, 27 May 2008 02:46:44 +0000 (12:46 +1000)]
r500: remove debugging
Dave Airlie [Tue, 27 May 2008 02:45:15 +0000 (12:45 +1000)]
r300/r500: fix r500 fragment program texture unit references
Dave Airlie [Tue, 27 May 2008 00:59:42 +0000 (10:59 +1000)]
r500: initial support for tmu mappings
Corbin Simpson [Mon, 26 May 2008 23:00:05 +0000 (16:00 -0700)]
r5xx: Enhance emit_mov().
Now we can add arbitrary sources and swizzles. Will make many things smoother.
Corbin Simpson [Mon, 26 May 2008 22:18:41 +0000 (15:18 -0700)]
r5xx: Moar LIT.
Still not working, but getting closer.
Corbin Simpson [Mon, 26 May 2008 19:35:39 +0000 (12:35 -0700)]
r5xx: First stab at LIT.
Corbin Simpson [Mon, 26 May 2008 05:50:00 +0000 (22:50 -0700)]
Replace copyright on r500_fragprog.c
Huh, could have sworn I already did this once before...
Maybe I forgot to commit it?
Corbin Simpson [Mon, 26 May 2008 02:53:48 +0000 (19:53 -0700)]
r5xx: Unbreak texture swizzling.
Corbin Simpson [Sun, 25 May 2008 18:46:16 +0000 (11:46 -0700)]
r5xx: Massive MAD cleanup.
Common uses of MAD now use emit_mad(), the two common negation masks work,
and fixed a few off-by-one errors.
Corbin Simpson [Sun, 25 May 2008 18:35:54 +0000 (11:35 -0700)]
r5xx: Negation masks for every inst except SWZ.
Yay?
Corbin Simpson [Sun, 25 May 2008 18:35:20 +0000 (11:35 -0700)]
r5xx: More emit_alu().
Converted ADD.
Corbin Simpson [Sun, 25 May 2008 18:07:51 +0000 (11:07 -0700)]
r5xx: Add emit_mad() for FP.
If it uses MAD, emit it with emit_mad()!
(Now available at your local grocer's. Multiply and add responsibly.)
Corbin Simpson [Sat, 24 May 2008 18:30:57 +0000 (11:30 -0700)]
r5xx: Consolidate FP tex insts.
They're all the same, really.
Corbin Simpson [Sat, 24 May 2008 16:17:28 +0000 (09:17 -0700)]
r5xx: Fix SGE/SLT.
Corbin Simpson [Fri, 23 May 2008 09:05:24 +0000 (02:05 -0700)]
r5xx: Remove some debugging cruft.
Dave Airlie [Sat, 24 May 2008 08:18:18 +0000 (18:18 +1000)]
r500: missed a couple of inst4s.
Dave Airlie [Sat, 24 May 2008 08:12:26 +0000 (18:12 +1000)]
r500: add depth output write
Not sure how well this works yet, but we need to set the alpha to w_omask
Corbin Simpson [Fri, 23 May 2008 07:16:49 +0000 (00:16 -0700)]
r5xx: Clean up some compiler warnings.
Corbin Simpson [Fri, 23 May 2008 07:14:31 +0000 (00:14 -0700)]
r5xx: Move dumb_shader.
Was getting ticked having to scroll around it, lawl.
Corbin Simpson [Fri, 23 May 2008 07:12:37 +0000 (00:12 -0700)]
r5xx: Add OPCODE_DST.
Works completely, swizzles and everything.
Corbin Simpson [Thu, 22 May 2008 09:34:57 +0000 (02:34 -0700)]
r5xx: More trig work.
SCS now works. COS/SIN have slight issues still.
Dave Airlie [Thu, 22 May 2008 07:09:30 +0000 (17:09 +1000)]
r500: bump state atom size up for fp and fp constants
Corbin Simpson [Thu, 22 May 2008 06:35:43 +0000 (23:35 -0700)]
r5xx: Fixed LRP.
Works perfectly. It's a complex one, though, so it might fail in weird ways...
Corbin Simpson [Thu, 22 May 2008 06:33:13 +0000 (23:33 -0700)]
r5xx: Change debug info for readability.
It's weird seeing the compiled program before the assembly, that's all.
Corbin Simpson [Wed, 21 May 2008 20:51:32 +0000 (13:51 -0700)]
r5xx: Initial (broken) OPCODE_LRP.
Will compile, run, and not eat your kids, but the math is wrong.
Corbin Simpson [Wed, 21 May 2008 15:24:28 +0000 (08:24 -0700)]
r5xx: Add OPCODE_POW.
Necessary for Google Earth, among other things.
Dave Airlie [Wed, 21 May 2008 06:00:18 +0000 (16:00 +1000)]
r500: print out opcode string
Dave Airlie [Wed, 21 May 2008 02:14:42 +0000 (12:14 +1000)]
r500: set the RS unit register for R500 not R300 dangnammit..
So this appears to be my BUG. damn it to hell.
also fix sec color to be more like spec says.
Dave Airlie [Wed, 21 May 2008 00:49:26 +0000 (10:49 +1000)]
r500: finish main texture instruction decoding
Corbin Simpson [Tue, 20 May 2008 16:47:50 +0000 (09:47 -0700)]
r5xx: Count refs so we don't have to guess on temp reg allocation.
As a bonus, we can now have multiple temp temps, by slot.
Corbin Simpson [Tue, 20 May 2008 06:55:59 +0000 (23:55 -0700)]
r5xx: Fixup SOP insts.
Use the correct swizzle for alpha/SOP stuff.
Corbin Simpson [Mon, 19 May 2008 19:26:04 +0000 (12:26 -0700)]
r5xx: New fix for COS/SIN/SCS.
Not perfect yet, but getting better.
Dave Airlie [Tue, 20 May 2008 06:30:36 +0000 (16:30 +1000)]
r300/r500: fixup some of the register write sizes
Dave Airlie [Tue, 20 May 2008 06:02:19 +0000 (16:02 +1000)]
r300: some ctrl-m's wierd.
Dave Airlie [Tue, 20 May 2008 05:59:56 +0000 (15:59 +1000)]
r300/r500: fix RS col fmt bits
Corbin Simpson [Mon, 19 May 2008 18:01:00 +0000 (11:01 -0700)]
r5xx: Fixup emit_tex, add debugging info, enable temp temps.
emit_tex now chases itself with an OUT if needed.
Added airlied's dump_program, with some fixes.
Dave Airlie [Mon, 19 May 2008 11:58:28 +0000 (21:58 +1000)]
r500: add more input srcs
Dave Airlie [Mon, 19 May 2008 11:40:40 +0000 (21:40 +1000)]
r500: fix swz gets and some returns
Dave Airlie [Mon, 19 May 2008 11:11:55 +0000 (21:11 +1000)]
r500: add mask debugging
Dave Airlie [Mon, 19 May 2008 10:24:09 +0000 (20:24 +1000)]
r500: add fragment program debug dumper
Corbin Simpson [Mon, 19 May 2008 07:00:08 +0000 (00:00 -0700)]
r5xx: Fix magic offsets for output fifo write masks.
Well, this sure explains a lot.
Corbin Simpson [Mon, 19 May 2008 06:52:54 +0000 (23:52 -0700)]
r5xx: Swap sources for CMP.
Follows the same pattern as the op on r3xx/r4xx. Thanks airlied.
Corbin Simpson [Mon, 19 May 2008 06:35:07 +0000 (23:35 -0700)]
r5xx: Fix typo of epic proportions.
Corbin Simpson [Mon, 19 May 2008 05:38:28 +0000 (22:38 -0700)]
r5xx: ALU/OUT fixups.
Lots of small changes. Intentionally breaks some tex stuffs.
Dave Airlie [Sun, 18 May 2008 07:58:29 +0000 (17:58 +1000)]
r300: fixup US_OUT_FMT bits
Dave Airlie [Sun, 18 May 2008 05:24:38 +0000 (15:24 +1000)]
r500: you can have a single texcoord
Corbin Simpson [Sat, 17 May 2008 20:38:35 +0000 (13:38 -0700)]
r5xx: Add OPCODE_KIL.
Corbin Simpson [Sat, 17 May 2008 19:45:46 +0000 (12:45 -0700)]
r5xx: Added OPCODE_DPH.
Like DP4, but with one swizzle change.
Corbin Simpson [Sat, 17 May 2008 16:27:35 +0000 (09:27 -0700)]
r5xx: Fix FRC.
This makes tri-frc work.
(Remind me again why I'm allowed near a compiler, lawl.)
Corbin Simpson [Fri, 16 May 2008 18:46:26 +0000 (11:46 -0700)]
r5xx: Fix SCS.
Output instructions need to be marked OUT so they can write to the fifo.
Also, negation doesn't work with SWZ yet.
Corbin Simpson [Thu, 15 May 2008 17:51:52 +0000 (10:51 -0700)]
r5xx: Add OPCODE_SWZ.
It's so easy!
Corbin Simpson [Thu, 15 May 2008 17:29:38 +0000 (10:29 -0700)]
r5xx: Add OPCODE_SCS.
It's disabled, though, because it doesn't work. I'll figure it out later...
Corbin Simpson [Thu, 15 May 2008 07:49:32 +0000 (00:49 -0700)]
r5xx: Adding more opcodes.
EX2, FRC, LG2, SIN, RCP, and RSQ, if you care.
All of these except FRC are like COS. This pretty much rounds out the set of
opcodes which can be done in one ALU inst.
Corbin Simpson [Thu, 15 May 2008 07:18:08 +0000 (00:18 -0700)]
r5xx: First swing at OPCODE_COS.
Corbin Simpson [Thu, 15 May 2008 07:11:10 +0000 (00:11 -0700)]
r5xx: Unbreak MAX and MIN.
Both of them had faulty copypasta.
Dave Airlie [Sat, 17 May 2008 03:31:14 +0000 (13:31 +1000)]
r500: set fragprog end to correct place
Alex Deucher [Sat, 17 May 2008 00:40:47 +0000 (10:40 +1000)]
r300: SC register naming cleanup
Alex Deucher [Sat, 17 May 2008 00:27:11 +0000 (10:27 +1000)]
r500: write out the correct FP registers
Dave Airlie [Thu, 15 May 2008 10:38:41 +0000 (20:38 +1000)]
r500: default rsunit swizzle like fglrx
Dave Airlie [Thu, 15 May 2008 06:24:54 +0000 (16:24 +1000)]
r500: shift tex src properly
Dave Airlie [Thu, 15 May 2008 05:05:40 +0000 (15:05 +1000)]
r500: fixup r500 rs unit texture coordinate counting
Dave Airlie [Thu, 15 May 2008 04:54:51 +0000 (14:54 +1000)]
r500: remove some debugging
Dave Airlie [Thu, 15 May 2008 04:52:39 +0000 (14:52 +1000)]
r500: split output/pixel masks and emit in the correct places
Dave Airlie [Thu, 15 May 2008 04:51:50 +0000 (14:51 +1000)]
r3/500: emit RS state before VAP
Dave Airlie [Thu, 15 May 2008 03:48:31 +0000 (13:48 +1000)]
r500: fixup the program allocations to be the correct sizes
Dave Airlie [Thu, 15 May 2008 03:47:47 +0000 (13:47 +1000)]
r300: set screen so that context init can find out chip ids
Dave Airlie [Thu, 15 May 2008 03:24:01 +0000 (13:24 +1000)]
r500: add cmp support in theory
Dave Airlie [Thu, 15 May 2008 03:21:50 +0000 (13:21 +1000)]
r500: some trivial fixups to get tri working.
the counter was being used one instruction over the end
Dave Airlie [Thu, 15 May 2008 01:46:23 +0000 (11:46 +1000)]
r500: we just need to emit a colour for clear drop tex instruction
Alex Deucher [Tue, 13 May 2008 20:12:57 +0000 (16:12 -0400)]
R300: clean up GA registers
Alex Deucher [Tue, 13 May 2008 19:46:23 +0000 (15:46 -0400)]
R3xx: clean up ZB registers
Alex Deucher [Tue, 13 May 2008 18:32:30 +0000 (14:32 -0400)]
R300: clean up CB registers
Alex Deucher [Tue, 13 May 2008 18:02:29 +0000 (14:02 -0400)]
R300: clean up Fog registers
Alex Deucher [Tue, 13 May 2008 17:38:30 +0000 (13:38 -0400)]
R500: fixup r300EmitClearState() FP for r5xx
Alex Deucher [Tue, 13 May 2008 12:37:58 +0000 (08:37 -0400)]
R300: cleanup FS code and fill in missing details
Alex Deucher [Tue, 13 May 2008 10:09:55 +0000 (06:09 -0400)]
R3xx: more PVS cleanup
Alex Deucher [Mon, 12 May 2008 15:58:35 +0000 (11:58 -0400)]
Merge branch 'r500-support' of git+ssh://agd5f@git.freedesktop.org/git/mesa/mesa into r500-support
Alex Deucher [Mon, 12 May 2008 15:57:08 +0000 (11:57 -0400)]
R500: add support for 4k textures
Dave Airlie [Wed, 7 May 2008 05:59:21 +0000 (15:59 +1000)]
r500: cleanup r500 RS setup
Dave Airlie [Wed, 7 May 2008 05:16:27 +0000 (15:16 +1000)]
r500: for rectangular textures set to unscaled coordinates.
Corbin Simpson [Wed, 7 May 2008 07:06:26 +0000 (00:06 -0700)]
r5xx: Fix FP inputs. (For good?)
FP inputs are now counted and mapped correctly, and temps
are allocated tightly and correctly.
Corbin Simpson [Wed, 7 May 2008 06:36:50 +0000 (23:36 -0700)]
r5xx: Fix false error with DP3/DP4.
DP3/DP4 only takes two arguments, but tried to load three, causing
a false fallback to the dumb shader.
Corbin Simpson [Wed, 7 May 2008 05:18:28 +0000 (22:18 -0700)]
r5xx: Index inputs and temps.
This is not the same as r3xx indexing. It only tries to protect inputs on
the pixel stack from getting clobbered by temps or texs.
Texs don't need special treatment since they read from special input regs
and write to the same temp regs as ALU/FC instructions.
Corbin Simpson [Wed, 7 May 2008 01:14:21 +0000 (18:14 -0700)]
r5xx: FP: Add OPCODE_TXB.
Tex lookup with biased LOD. Should magically work.
Corbin Simpson [Wed, 7 May 2008 00:21:30 +0000 (17:21 -0700)]
r5xx: FP: Make MOV/ABS look pretty.
We can't really do anything like emit_alu, so we're doing emit_mov instead.
Corbin Simpson [Tue, 6 May 2008 19:44:53 +0000 (12:44 -0700)]
r5xx: Emit an OUT instruction at the end of execution.
This should make TEX/TXP work right. (Note: "Should" is not "does.")
Corbin Simpson [Tue, 6 May 2008 19:42:40 +0000 (12:42 -0700)]
r5xx: We update max_temp_idx now, so no need to hard-code it.
This roughly doubles the speed of glxgears (GINAB) by allowing
more pixels to run concurrently.
Corbin Simpson [Tue, 6 May 2008 19:18:07 +0000 (12:18 -0700)]
r5xx: Fix typo.
Gotta be more careful with my cut'n'paste, lawl.
Corbin Simpson [Tue, 6 May 2008 18:57:24 +0000 (11:57 -0700)]
r5xx: Use max_temp_idx.
Dave Airlie [Mon, 5 May 2008 08:42:27 +0000 (18:42 +1000)]
r500: consolidate tex instructions
you cannot change a tex into an output so this means we have to actually
do another instruction after this one to mov if its an output