summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* i965g: special case setup when fs has no inputsKeith Whitwell2009-11-193-25/+55
|
* i965g: add new state flag tracking fs signature changesKeith Whitwell2009-11-194-10/+20
|
* i965g: fix typo in previous commitKeith Whitwell2009-11-191-1/+1
|
* i965g: setup nr_attrs should track nr fragment shader inputsKeith Whitwell2009-11-191-2/+6
| | | | This would be equivalent to taking the nr outputs of the vp varient
* i965g: get fragment constants sort-of workingKeith Whitwell2009-11-191-8/+9
|
* i965g: make the load-before-use vs immediate path workKeith Whitwell2009-11-191-2/+3
|
* i965g: fix initialization of texture width/height/depth arraysKeith Whitwell2009-11-191-5/+6
| | | | Will remove these arrays in another branch - they're completely redundant.
* i965g: remove more references to nr_vp_outputs in wm compilationKeith Whitwell2009-11-181-3/+1
| | | | | We're really more concerned about how many inputs the fragment shader is expecting.
* i965g: handle special vs outputs speciallyKeith Whitwell2009-11-1711-55/+147
| | | | | | | | | | | Where vs output semantic tags indicate an output is signficant for fixed function processing (such as clipping, unfilled modes, etc), retain information about that output so that we can get to it easily later on. Fix up the unfilled processing, but hard-wire edgeflag to one for now. With this change, trivial/tri-unfilled works.
* i965g rename offset_edge to offset_edgeflagKeith Whitwell2009-11-173-7/+7
|
* i965g: remove droppings in xlib winsysKeith Whitwell2009-11-151-1/+0
|
* i965g: restore check on line smooth stateKeith Whitwell2009-11-141-1/+2
|
* i965g: consult fs inputs when laying out vs output regsKeith Whitwell2009-11-115-43/+113
| | | | | | | | Vertex shader now emits just the FS inputs, in the positions and order expected by the fragment shader. This means potentially regenerating the vertex shader to match different fragment shader's input layouts.
* i965g: skip over vertex position output when preallocating FS inputsKeith Whitwell2009-11-111-0/+4
|
* i965g: some more cases where file_max[] is used without obligatory +1Keith Whitwell2009-11-111-5/+5
|
* i965g: smaller upload buffer for index dataKeith Whitwell2009-11-111-1/+1
|
* i965g: fix off-by-one on curbe uploadKeith Whitwell2009-11-111-4/+10
|
* i965g: avoid use of internally generated immediatesKeith Whitwell2009-11-111-1/+4
| | | | Currently not working, so don't generate more of them.
* i965g: Fixup texture formatsJakob Bornecrantz2009-11-072-37/+74
|
* i965g: Formalize on S8Z24 as the suported depth formatJakob Bornecrantz2009-11-074-12/+4
|
* i965g: Fixup buffer creation functionJakob Bornecrantz2009-11-061-2/+12
| | | | First tri!
* i965g: Add texture transfer functionsJakob Bornecrantz2009-11-061-0/+95
| | | | They don't seem to work. Maybe we are forgetting to flush the gpu or something.
* i965g: Match pitch modification on get buffer as wellJakob Bornecrantz2009-11-061-1/+1
| | | | | | | This is a ugly hack in order to match what the intel X driver gives us. However putting this in the winsys where it fits better forces it reach more into the driver then it already does.
* i965g: First clear!Jakob Bornecrantz2009-11-061-11/+5
|
* i965g: get rid of cc key, simplify state uploadKeith Whitwell2009-11-064-129/+51
| | | | | Keep a valid reloc table active between uploads, avoid recalculating it every time.
* i965g: trivial/clear can now send stuff to hardwareJakob Bornecrantz2009-11-063-4/+21
| | | | | Added a flag if we should send commands to hardware as what we send isn't all that correct.
* i965g: Fix texture blanket functionJakob Bornecrantz2009-11-061-11/+12
|
* i965g: Add functions needed by the winsysJakob Bornecrantz2009-11-061-2/+115
|
* i965g: Winsys whitespaceJakob Bornecrantz2009-11-062-37/+37
|
* i965g: Build winsys againJakob Bornecrantz2009-11-062-57/+53
|
* i965g: Header whitespaceJakob Bornecrantz2009-11-061-42/+42
|
* i965g: get brw_state_debug.c buildingKeith Whitwell2009-11-062-3/+25
|
* i965g: propogate index size stateKeith Whitwell2009-11-061-1/+3
|
* i965g: remove references to brw_surface_bo()Keith Whitwell2009-11-062-6/+3
|
* i965g: plumb through fb_write target and eot dataKeith Whitwell2009-11-065-25/+29
|
* i965g: propogate nr_cbufs into wm prog keyKeith Whitwell2009-11-063-1/+10
|
* i965g: init saturate field in fp dst_reg helperKeith Whitwell2009-11-061-0/+1
|
* i965g: add dumping for our new pass_fp outputKeith Whitwell2009-11-066-71/+139
|
* i965g: disassemble each instruction as generatedKeith Whitwell2009-11-065-2/+11
|
* i965g: hardwire linear interpolation for nowKeith Whitwell2009-11-061-2/+5
| | | | | seems to generate saner code, need to go back and fix perspective interpolation (and remove the hard-wire) once this is working.
* i965g: fix some reloc countsKeith Whitwell2009-11-063-3/+3
|
* i965g: don't set up vs stack register for non-branching shadersKeith Whitwell2009-11-065-17/+20
|
* i965g: add DEBUG_MIN_URB flagKeith Whitwell2009-11-063-3/+13
|
* i965g: tgsi outputs cannot be used as source regsKeith Whitwell2009-11-061-2/+3
|
* i965g: use curr.vertex_element state directlyKeith Whitwell2009-11-062-11/+7
|
* i965g: don't emit line stipple packet if stipple disabledKeith Whitwell2009-11-061-1/+3
|
* i965g: populate brw_context chipset idKeith Whitwell2009-11-061-0/+1
|
* i965g: make sure blend color packet header is initializedKeith Whitwell2009-11-061-3/+9
| | | | | | | We will emit this packet at startup (dirty == ~0), even if we haven't had the state tracker call into brw_set_blend_color() yet. This way is a little more efficient also.
* i965g: remove/disable inactive state atomsKeith Whitwell2009-11-064-6/+3
|
* i965g: populate wm reloc array earlierKeith Whitwell2009-11-061-36/+46
| | | | Still have to calculate the reloc background in two places.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-22 01:25:23 +0000