summaryrefslogtreecommitdiff
path: root/src/gallium/drivers
AgeCommit message (Collapse)Author
2010-09-12llvmpipe: fix wierd performance regression in isosurfKeith Whitwell
I really don't understand the mechanism behind this, but it seems like the way data blocks for a scene are malloced, and in particular whether we treat them as stack or a queue, and whether we retain the most recently allocated or least recently allocated has a real affect (~5%) on isosurf framerates... This is probably specific to my distro or even just my machine, but none the less, it's nicer not to see the framerates go in the wrong direction.
2010-09-12llvmpipe: Only generate the whole shader specialization for opaque shaders.José Fonseca
If not opaque, then the color buffer will have to be read any way, therefore the specialization is pointless.
2010-09-12pb: add void * for flush ctx to mapping functionsDave Airlie
If the buffer we are attempting to map is referenced by the unsubmitted command stream for this context, we need to flush the command stream, however to do that we need to be able to access the context at the lowest level map function, currently we set the buffer in the toplevel map, but this racy between context. (we probably have a lot more issues than that.) I'll look into a proper solution as suggested by jrfonseca when I get some time.
2010-09-11nv30: fix breakage due to 10 texcoord support on nv40Luca Barbieri
2010-09-11r600g: Undo bo placement change.Tilman Sauerbeck
This reverts a part of e795ca8f3175fa6fd97b6b2ef2775e3f8803012a that causes artefacts and a performance drop. Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-09-11llvmpipe: Silence some warnings.José Fonseca
2010-09-11llvmpipe: Don't store display the alpha ref value in the key.José Fonseca
It's never used.
2010-09-11gallivm: s/lp_build_broadcast_aos/lp_build_swizzle_scalar_aos/José Fonseca
More accurate description of this function purpose.
2010-09-11llvmpipe: restore larger command blocksKeith Whitwell
2010-09-11llvmpipe: move some debug to DEBUG_SCENEKeith Whitwell
2010-09-11llvmpipe: add DEBUG_MEM optionKeith Whitwell
2010-09-11llvmpipe: allow bigger scenesKeith Whitwell
2010-09-10r300/compiler: Enable presubtract sourcesTom Stellard
The r300 compiler can now emit instructions that select from the presubtract source. A peephole optimization has been added to convert instructions like: ADD Temp[0].x, none.1, -Temp[1].x into the INV (1 - src0) presubtract operation.
2010-09-10r600g: Fixed a bo reference leak in the draw module.Tilman Sauerbeck
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-09-10r600g: Fixed a bo leak in r600_texture_from_handle().Tilman Sauerbeck
We would leak bo if the argument check failed. Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-09-10r600g: Don't leave stale references in query_list when we cannot create bo.Tilman Sauerbeck
Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-09-10r600g: Implemented the y component write for the LOG opcode.Tilman Sauerbeck
This makes the 'vp1-LOG test' piglit test work. Signed-off-by: Tilman Sauerbeck <tilman@code-monkey.de>
2010-09-10r600g: fixup state calculations for picking states.Dave Airlie
for evergreen I ended up using a non-contig array of states, but this code needs a bit of fixing up to deal with that.
2010-09-10r600g: fixup CB state numbering headerDave Airlie
2010-09-10r600g: fix warning in r600 pipe driverDave Airlie
2010-09-10r600g: evergreen CBs are more sane to support with a single stateDave Airlie
2010-09-10r600g: add initial evergreen supportDave Airlie
adds shader opcodes + assembler support (except ARL) uses constant buffers add interp instructions in fragment shader adds all evergreen hw states adds evergreen pm4 support. this runs gears for me on my evergreen
2010-09-10r600g: align flushing of cb/db with DDX/r600c.Dave Airlie
the DDX and r600c both flush cb/db after the draw is emitted, as long as they do that, r600g can't be different, as it races. We end up with r600g flush, set CB, DDX set CB, flush. This was causing misrendering on my evergreen, where sometimes the drawing would go to an old CB.
2010-09-10r600g: don't need 3 bos here.Dave Airlie
the code should reloc correctly a single BO 3 times.
2010-09-09llvmpipe: remove redundant tgsi_dup_tokens() callBrian Paul
We were calling this twice so the first allocation was orphaned/leaked.
2010-09-08r600g: add initial bank swizzle support.Dave Airlie
this is ported from r600c mostly, bank swizzling is real messy and I don't think I got enough sleep last night to fully understand it.
2010-09-08r600g: Added missing r600_bc_add_literal() calls to XPD implementation.Tilman Sauerbeck
Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-09-08r600g: Added preliminary support for the LOG opcode.Tilman Sauerbeck
Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-09-08r600g: Added missing abs-ification to RSQ.Tilman Sauerbeck
This makes the 'fp1-RSQ test 2' piglit test work. Signed-off-by: Dave Airlie <airlied@redhat.com>
2010-09-08r600g: split opcodes out and add wrapper around usage.Dave Airlie
This splits the r600 opcodes out of the sq file and adds a wrapper so we can convert to evergreen opcodes later without touching these functions too much.
2010-09-08r600g: add support for constants in memory buffers.Dave Airlie
DX9 constants were in the constant file, and evergreen no longer support cfile. r600/700 can also use constants in memory buffers, so add the code (disabled for now) to enable that as precursor for evergreen.
2010-09-08r600g: abstract the hw states out behind a vtbl.Dave Airlie
this is step one towards evergreen support, it lets us plug in whole new hw level states.
2010-09-07llvmpipe: Add lp_rast_debug.c to Makefile.Chia-I Wu
2010-09-07llvmpipe: Refactor lp_scene_add_resource_referenceJosé Fonseca
Less goto spaghetti.
2010-09-07llvmpipe: Fix negated logic in lp_scene_add_resource_reference().José Fonseca
Fixes performance regression.
2010-09-07llvmpipe: move more coef setup into lp_setup_coef.cKeith Whitwell
2010-09-07llvmpipe: declare fence handle structKeith Whitwell
2010-09-07llvmpipe: rasterization debugging helpersKeith Whitwell
2010-09-07llvmpipe: use opcodes instead of function pointers in binsKeith Whitwell
Also, move some state from rasterizer struct to the scene.
2010-09-07llvmpipe: rearrange queriesKeith Whitwell
2010-09-07llvmpipe: enforce fixed memory limit on scenesKeith Whitwell
2010-09-07llvmpipe: clean up deferred zstencil clearsKeith Whitwell
2010-09-07llvmpipe: rework fences and queriesKeith Whitwell
2010-09-07llvmpipe: move whole-tile emit into a functionKeith Whitwell
2010-09-07llvmpipe: put fs variant dumping in a functionKeith Whitwell
2010-09-06nvfx: fix return in vp mainLuca Barbieri
2010-09-06r600g: add error print for no literals for r700s as wellDave Airlie
2010-09-06r600g: fixup r700 assembler for clamp/relative addressingDave Airlie
2010-09-06r600g: add script to generate header file with offsets into state objects.Dave Airlie
This was inherently fragile as any changes to r600_states.h would also need manual updating of all of the bits in radeon.h. Just add a simple python script to do the conversion, its not hooked up to make at all. This also will make adding evergreen a bit easier.
2010-09-06r600g: fix up surface referencesDave Airlie
We end up referencing the new surf and derefing the old surface which is a copy of the pointer to the new surf. So just bump the ref count directly.
generated by cgit v1.2.3 (git 2.39.1) at 2025-03-04 07:07:41 +0000