summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* i915tex: Set framebuffer size to match window before calling _mesa_make_current.Michel Dänzer2007-03-071-10/+10
| | | | Fixes issues with apps that don't call glViewport by default.
* i915tex: Sync pages differently when crossing pipe borders.Michel Dänzer2007-02-281-12/+18
| | | | | Don't flip (up to twice) immediately but just arrange things such that the pages will be in sync on both pipes on the next flip.
* i915tex: Check that intel_rb is valid before trying to add it to an fbo.Michel Dänzer2007-02-281-2/+4
|
* i915tex: Also update intel_rb->vbl_pending when scheduled swap is not a flip.Michel Dänzer2007-02-281-3/+3
|
* i915tex: Schedule flips when possible.Michel Dänzer2007-02-226-179/+228
| | | | | Also move vsync related state from context to window, so it's possible to schedule several flips ahead of time with triple buffering.
* i915tex: Triple buffering support, only effective with page flipping so far.Michel Dänzer2007-02-207-63/+166
| | | | | | Pending flips are tracked per renderbuffer and the colour renderbuffer attachments of window framebuffer objects are rotated on flips to avoid stalling the pipeline for pending flips unnecessarily.
* intelWindowMoved: Some debugging output was accidentally left always enabled.Michel Dänzer2007-02-201-1/+1
|
* intelRotateWindow: Only use back buffer and cliprects in very specific cases.Michel Dänzer2007-02-201-7/+6
|
* intelCopyBuffer: Remove incorrect intel_fb->pf_current_page tests.Michel Dänzer2007-02-201-17/+4
| | | | | | | intel_get_rb_region already takes the current page into account. This would result in broken rendering when multiple 3D windows are visible and the pages are reversed.
* i915tex: Support page flipping on both CRTCs independently.Michel Dänzer2007-02-2010-117/+203
| | | | | No longer track page flipping state per context but per window, via struct intel_framebuffer which wraps struct gl_framebuffer for windows.
* i915tex: Make page flipping work again.Michel Dänzer2007-02-207-46/+94
|
* i915tex: Fix performance regression with new vbo code and latest drm.Thomas Hellstrom2007-02-191-1/+4
| | | | | | | With current memory preferences, vbos ended up in AGP space where reading from them got a bit slow. Make sure buffer objects are initially created in system memory.
* Merge branch 'master' of ↵Sean D'Epagnier2007-02-1616-69/+332
|\ | | | | | | git+ssh://[email protected]/git/mesa/mesa
| * doxygen: update doxygen project number to 6.5Jerome Glisse2007-02-161-1/+1
| | | | | | | | doxygen number was laggin behind a bit.
| * Update doxygen doc to reflet vbo changes.Jerome Glisse2007-02-1612-16/+16
| | | | | | | | | | | | Update doxygen doc, array_cache no longuer exist, new shiny vbo modules is there. Tested on unix, but i think i didn't broke that bat :).
| * nouveau: nv50: updatePatrice Mandin2007-02-161-1/+75
| |
| * nouveau: nv10,20: wrong viewport clip setupPatrice Mandin2007-02-162-2/+6
| |
| * nouveau: nv20: texture matrixPatrice Mandin2007-02-161-1/+4
| |
| * nouveau: nv20: updatePatrice Mandin2007-02-161-32/+176
| |
| * nouveau: nv10: set clear value for 16 and 24 bits depthPatrice Mandin2007-02-161-4/+15
| |
| * nouveau: nv10: rescale depth near and farPatrice Mandin2007-02-161-2/+8
| |
| * nouveau: nv10: fix viewport scale and originPatrice Mandin2007-02-161-12/+33
| |
* | Commiting a few minor fixes to glut fbdev (segfault on a separate stack)Sean D'Epagnier2007-02-1621-2239/+55
|/ | | | | | Updated all drivers that used the old _ac_ functions to use the new _vbo_ functions. I know this fixed the fbdev driver, it looks like it might fix some other drivers too.
* r300: Fix cos & add scs to fragment program.Rune Peterson2007-02-142-88/+184
| | | | | | | | | | So this do : - Fixes COS. - Does range reductions for SIN & COS. - Adds SCS. - removes the optimized version of SIN & COS. - tweaked weight (should help on precision). - fixed a copy paste typo in emit_arith().
* test for POT before using sample_2d_linear_repeat()Etienne Clement2007-02-141-1/+3
|
* fix typo preventing generic arrays from working with ARB_vp (bug #9952)Roland Scheidegger2007-02-141-1/+1
|
* nouveau: nv10: there are 8 clipping regionsPatrice Mandin2007-02-141-1/+1
|
* fix stupid bug in the optimized per-vertex fog generation codeRoland Scheidegger2007-02-141-1/+2
|
* i915tex: Take into account various mapping states when dropping theThomas Hellstrom2007-02-141-2/+13
| | | | batch buffer after a resolution / rotation switch.
* I965: fix a failure on waiting irq.Xiang, Haihao2007-02-121-1/+2
| | | | | Wait until getting the right fence if drm/i915 resets the counter.
* fd.o #BZ 9684Zou Nan hai2007-02-121-1/+2
| | | | | gcc may cast an overflowed float value to an negative int value on i386, fix it to make sure at least some apps will not crash if it contains wrong normal data.
* r300: Add proper support for sin/cos instruction in fragment programRune Peterson2007-02-126-121/+175
| | | | | | | | | | | Getting proper SIN and COS wasn't as easy as it appeared. I had to make make some changes to the fragment program code. general FP changes: - support HHH swizzle for vector instructions. - don't copy a source to a temp when it is not XYZW swizzled, but combine the two and have the swizzle resolve any issues. (saves temps/instructions with more elaborate shader code) - fix overflow in cnstv[].
* ctx->Point._Size fixRune Peterson2007-02-103-26/+7
|
* optimize per-pixel fog programRoland Scheidegger2007-02-092-50/+31
| | | | | | | use the same internal state as for vertex fog. From the old implemenentation, this changes a SUB+MUL into a MAD for linear fog, and saves a MUL for EXP/EXP2, plus saves a (constant) parameter. While here, fix the broken (forgotten) swizzling. (untested)
* array overflow fixZou Nan hai2007-02-091-2/+2
|
* r200: simplify / unify input map handling for vp and fftnlRoland Scheidegger2007-02-095-334/+141
| | | | | | Use the same input map handling for fftnl and vertex programs. It doesn't enable any new functionality (should make it easy to support per-vertex materials though), but the code is much cleaner.
* optimize generated vertex programs a bitRoland Scheidegger2007-02-095-32/+71
| | | | | | | | Use new internal state to avoid per-vertex normalization of static spot direction vector. Use internal state for simpler per-vertex fog computations (MAD instead of SUB/MUL for linear fog, EX2 instead of POW for EXP/EXP2 fog). Simplify point size calc (2 MADs instead of MOV, MUL, MUL, DP3), and while there fix it up (RSQ instead of RCP). All untested...
* fix issues in vp when using FRAG_BIT_WPOS in a fragment program (bug #9910)Roland Scheidegger2007-02-091-20/+15
| | | | | | Redirect all VERT_RESULT_HPOS writes to a temp and use that for fixup. The viewport transformation still seems to take some shortcuts, and it still does not seem to work at all...
* nouveau: a couple of NV3x fixesBen Skeggs2007-02-065-5/+45
|
* Add support for CN700 chipset in miniglxJakob Bornecrantz2007-02-062-0/+2
|
* fix issues when using mixed-mode pos-invariant vp and ff tnl (bug #9856)Roland Scheidegger2007-02-051-19/+32
| | | | | | | | | ARB_vp requires vertex transformation to be invariant to fixed function tnl if the position_invariant option is used. So the same function needs to be used, otherwise z-fighting artifacts may happen with applications which rely on the results being really the same due to precision issues when dealing with floating point values (may not be a problem when using a non-optimizing compiler strictly following IEEE rules).
* Merge branch 'master' of git+ssh://[email protected]/git/mesa/mesaBrian2007-02-051-2/+2
|\
| * nouveau: OUT_RINGp expects the size in dwords, not bytes.Ben Skeggs2007-02-061-2/+2
| | | | | | | | This fixes the *actual* bug that the previous commit was supposed to fix..
* | fix mem leakPanagiotis Papadakos2007-02-051-0/+3
|/
* fix mem leaksPanagiotis Papadakos2007-02-052-0/+5
|
* nouveau: fix a small bug in nv10 swtclBen Skeggs2007-02-061-1/+1
|
* nouveau: new bufferobj code.Ben Skeggs2007-02-063-169/+562
| | | | | | | | | | | The old code suffered from a number of issues, the most severe being that with the Mesa VBO merge even swtcl used the driver's bufferobj interface. On most VBO types (or non-AGP cards) the buffer ended up in vram, and killed swtcl performance greatly. All bufferobj's start in system memory now, until they get referenced as a "real" VBO. The other big change is that only potentially "damaged" areas are uploaded/downloaded to/from the hardware.
* break if we removed driver from Drivers list in driDestroyDisplayPanagiotis Papadakos2007-02-051-0/+1
| | | | Else we read memory we just released, in for statement.
* nouveau: nv10: add init for unknown 0x120Patrice Mandin2007-02-041-0/+8
|
* nouveau: fix nv04 and nv10 swtcl, more work on nv04 state.Stephane Marchesin2007-02-044-14/+118
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-18 14:54:30 +0000