summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* gallium: plumb winsys-drawable-handle through to state trackerKeith Whitwell2010-03-109-16/+42
| | | | | | | The state trackers need this value so that they can supply it as the "void *" argument to flush_frontbuffer. Fixes single-buffer rendering.
* Merge commit 'origin/master' into gallium-sw-api-2Keith Whitwell2010-03-10138-2857/+4003
|\
| * llvmpipe: Cope with null Vertex element cso.José Fonseca2010-03-101-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | CSO can often be null. For example: 1. at initialization 2. using an util module (u_blit) right after initialization (it will push state and pop the previous null state) 3. at shutdown time (state shouldn't be bound when being destroyed) Glean was hitting 2.
| * nouveau: translate PIPE_BUFFER_USAGE_UNSYNCHRONIZEDBen Skeggs2010-03-101-1/+1
| |
| * nv50: inline nv50_state_flush_notify()Ben Skeggs2010-03-102-16/+6
| | | | | | | | It's (rightly) not called from anywhere else now
| * nv50: remove nv50_context.state.instbufBen Skeggs2010-03-102-6/+0
| |
| * nv50: remove unnecessary macroBen Skeggs2010-03-101-3/+2
| |
| * nv50: move nv50_prim() into a header file rather than duplicatingBen Skeggs2010-03-103-62/+31
| |
| * nv50: add option to force immediate-mode submission, disable by defaultBen Skeggs2010-03-103-4/+14
| |
| * nv50: re-add immediate mode vertex submission pathsBen Skeggs2010-03-104-8/+399
| |
| * nv50: when doing inline indices, split elt lists so they fit into pushbufBen Skeggs2010-03-102-85/+202
| |
| * nv50: reimplement draw_elements_instance(), use for draw_elements() tooBen Skeggs2010-03-101-159/+93
| | | | | | | | | | This makes draw_elements()/draw_elements_instanced() do the right thing for the non-inline elements cases, and not require flush_notify().
| * nv50: reimplement draw_arrays_instanced(), do draw_arrays() in terms of itBen Skeggs2010-03-101-144/+134
| | | | | | | | | | | | | | | | | | This makes draw_arrays()/draw_arrays_instanced() do the right thing and not require the (probably broken anyway) flush_notify() usage. It also fixes a potential bug in the behaviour of reading InstanceID from shaders, where 0 should be read for non-instanced drawing, previously it was possible to read non-0 ids if mixing instanced/non-instanced.
| * nv50: rip out inline push draw paths temporarilyBen Skeggs2010-03-102-551/+12
| |
| * nv50: make use of scissor enable/disable methodBen Skeggs2010-03-104-25/+8
| |
| * nv50: move magic 0x142c stuff to after state emissionBen Skeggs2010-03-102-11/+8
| |
| * nv50: avoid unnecessary vp/gp/fp state changesBen Skeggs2010-03-101-0/+9
| |
| * nv50: remove pre_pipebuffer_map hack....Ben Skeggs2010-03-101-23/+0
| |
| * nv50: remove horrific abuse of nouveau_channel.flush_notifyBen Skeggs2010-03-102-1/+10
| | | | | | | | | | | | | | Really though, using this at all is just not a good idea in the 3D driver. I'm almost certain the hardware will not like a reloc appearing between begin()/end().. Anyways, this is still better than before, more fixes to come..
| * nv50: always emit dummy relocs for all still-referenced buffersBen Skeggs2010-03-101-2/+8
| |
| * nv50: ensure enough room for state changes in current pushbufBen Skeggs2010-03-104-17/+38
| | | | | | | | | | | | | | | | Also allows the nv50_state_validate() caller to request a minimum amount of space that itself requires, not all callers accurately use this yet but the simple cases are now accounted for. Rendering will also be dropped on the floor if validate fails now.
| * nv50: rework state emissionBen Skeggs2010-03-107-311/+249
| |
| * nv50: use relocs rather than re-uploading TIC all the timeBen Skeggs2010-03-103-2/+32
| |
| * r300g: Reorder regs, save one dword.Corbin Simpson2010-03-091-11/+14
| | | | | | | | Or, put another way, save 10% on rendering calls.
| * i965: Fix ENDLOOP to only patch up this loop's BREAK and CONT.Eric Anholt2010-03-091-2/+4
| | | | | | | | | | Corresponds to d225a25e21a24508aea3b877c78beb35502e942d and fixes piglit glsl-fs-loop-nested. Bug #25173.
| * i965: Unalias all GLSL source regs from the destination regs used.Eric Anholt2010-03-091-113/+25
| | | | | | | | | | | | | | | | | | | | | | We were doing it ad-hoc before, as instructions with potential aliasing problems were identified. But thanks to swizzling basically anything can have aliasing, so just do it generally at source reg setup time. This is somewhat inefficient, because sometimes an operation doesn't need unaliasing protection if the swizzling is safe, but the unaliasing before didn't cover those cases either. Fixes piglit glsl-fs-loop.
| * i965: Print the offsets for WHILE and BREAK in disasm.Eric Anholt2010-03-091-2/+2
| |
| * i965: Fix nested loops in the VS.Eric Anholt2010-03-091-2/+4
| | | | | | | | | | | | | | We were patching up all the break and continues between the start of our loop and the end of our loop, even if they were breaks/continues for an inner loop. Avoiding patching already patched breaks/continues fixes piglit glsl-vs-loop-nested.
| * i965: Fix up VP constbuf leak on program delete.Eric Anholt2010-03-091-3/+11
| |
| * r300: don't crash if there's no colorbufferMaciej Cencora2010-03-091-1/+6
| |
| * radeon: mark framebuffer as incomplete if renderbuffer format isn't ↵Maciej Cencora2010-03-092-67/+62
| | | | | | | | supported by hw
| * r300: add support for more rendering formatsMaciej Cencora2010-03-092-17/+79
| |
| * radeon/r200/r300/r600: add is_format_renderable functionMaciej Cencora2010-03-099-0/+70
| |
| * progs/tests: use first color attachementMaciej Cencora2010-03-093-9/+9
| | | | | | | | Allows the tests to be run on drivers that supports only one color attachement.
| * r300: enable HW accelerated gl(Read/Copy/Draw)PixelsMaciej Cencora2010-03-091-2/+7
| |
| * radeon: add hw accelerated glReadPixel support (not enabled yet)Maciej Cencora2010-03-099-2/+203
| |
| * glapi: Fix scons build -- list new file.José Fonseca2010-03-091-0/+1
| |
| * glapi: allow for any mangle prefixGeorge Sapountzis2010-03-091-2/+3
| |
| * glapi: split out arch-specific code for entrypointsGeorge Sapountzis2010-03-094-295/+357
| |
| * glapi: mv table functions to glapi_getproc.c and add local headerGeorge Sapountzis2010-03-094-66/+98
| |
| * drop stray XFree86Server, XGLServerGeorge Sapountzis2010-03-092-9/+4
| |
| * svga: Don't do culling while the software pipeline is active.Keith Whitwell2010-03-091-2/+11
| | | | | | | | | | It does it for us, and additionally introduces potentially back-facing triangles.
| * mesa/st: initialize velements state properlyRoland Scheidegger2010-03-091-0/+1
| | | | | | | | | | | | one variable is a bitfield where the rest is never written to, which caused valgrind to complain. Might have caused cso to not recognize an already stored state. Reported by Christoph Bumiller.
| * cso: don't forget to release vertex elements stateRoland Scheidegger2010-03-091-0/+1
| |
| * r300g: kill pf_component_countMarek Olšák2010-03-091-28/+1
| |
| * gallivm: checkpoint: nearest mipmap filteringBrian Paul2010-03-095-292/+601
| | | | | | | | | | | | | | | | | | The LOD is computed from texcoord partial derivatives and used to select a mipmap level. Still some bugs in texel fetching. Lots of rough edges and unfinished parts but the basics are in place. Lots of changes to the lp_bld_arit.c code to support non-vector/scalar datatypes.
| * util: simplify util_format_get_nr_components helperRoland Scheidegger2010-03-091-8/+1
| | | | | | | | | | since the number of components is now stored in the pipe format description, we obviously no longer need to figure this out ourselves.
| * util: Use bitshift arithmetic for pack too.José Fonseca2010-03-091-19/+58
| |
| * cso: Fix typo in assert.Corbin Simpson2010-03-091-1/+1
| |
| * r300g: Precalculate and CSO-bind PSC state.Corbin Simpson2010-03-094-205/+190
| |
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-04 17:25:57 +0000