brkho/mesa - mesa - Brian's Gitea

Graphique des révisions

Auteur	SHA1	Message	Date
José Fonseca	ad60552ca1	llvmpipe: Disable multithreading on windows.	il y a 15 ans
Brian Paul	7319ae0954	llvmpipe: remove tile clipping code The surface is always a multiple of the tile size now.	il y a 15 ans
Brian Paul	ec459f2aec	llvmpipe: asst. task-related clean-ups	il y a 15 ans
Brian Paul	0fccfc9cc0	llvmpipe: remove unneeded DEBUG checks, use step var	il y a 15 ans
Brian Paul	62623c4dc5	llvmpipe: added show_tiles and show_subtiles debug options These options draw lines over the tiles to show the 64x64 tile bounds and 16x16 sub-tile bounds. For debugging/visualization.	il y a 15 ans
Keith Whitwell	591401ff05	llvmpipe: use new u_ringbuffer for scene queue	il y a 15 ans
Brian Paul	2797f2bf57	llvmpipe: generate two shader varients, one omits triangle in/out testing When we know that a 4x4 pixel block is entirely inside of a triangle use the jit function which omits the in/out test code. Results in a few percent speedup in many tests.	il y a 15 ans
Brian Paul	2ba1c8189a	llvmpipe: use one loop in lp_rast_clear_zstencil() This is just a tiny bit faster.	il y a 15 ans
Brian Paul	f94a99170e	llvmpipe: optimize lp_rast_clear_color() for non-gray colors This makes a big difference in progs that clear to a non-gray color. Some demos are 30-50% faster.	il y a 15 ans
Keith Whitwell	4231006e29	llvmpipe: add bin debugger Adjust definition of empty_bin according to what's actually in empty bins. We often have a state packet before/after load commands. Still need to do something about the fence packets.	il y a 15 ans
Keith Whitwell	95ee14f147	llvmpipe: implement lp_rast_load_zstencil Load zbuffer contents for binned scenes that don't start with a clear and which have a bound zbuffer.	il y a 15 ans
Keith Whitwell	db83ad4b43	llvmpipe: improve empty-bin test further Remove unused param, add comments. Thanks to Brian for review.	il y a 15 ans
Keith Whitwell	f4b29e6ad3	llvmpipe: improve empty-bin test We emit at most two clear packets (color and z respectively).	il y a 15 ans
Keith Whitwell	da45f49cc6	llvmpipe: quick hack to short-circuit empty bins	il y a 15 ans
Keith Whitwell	c1a0441602	llvmpipe: initial mrt support Non-mrt apps work, and the code looks correct, but not many mrt test apps handy atm...	il y a 15 ans
Keith Whitwell	c9240c4c8f	llvmpipe: remove dead code	il y a 15 ans
Brian Paul	808170a0ff	llvmpipe: replace INT_MIN/2 with INT_MIN Since changing the in/out test we can just use INT_MIN to be sure the comparison against the step values always passes.	il y a 15 ans
Brian Paul	b9d33db0a4	llvmpipe: improve the in/out test a little Instead of: s = c + step m = s > 0 Do: m = step > c (with negated c)	il y a 15 ans
Brian Paul	ab94381930	llvmpipe: do final the pixel in/out triangle test in the fragment shader The test to determine which of the pixels in a 2x2 quad is now done in the fragment shader rather than in the calling C code. This is a little faster but there's a few more things to do. Note that the step[] array elements are in a different order now. Rather than being in row-major order for the 4x4 grid, they're in "quad-major" order. The setup of the step arrays is a little more complicated now. So is the course/intermediate tile test code, but some lookup tables help with that. Next steps: - early-cull 2x2 quads which are totally outside the triangle. - skip the in/out test for fully contained quads - make the in/out comparison code tighter/faster.	il y a 15 ans
Brian Paul	cdbcd96fdf	llvmpipe: tighten up an assertion	il y a 15 ans
Keith Whitwell	663750d556	llvmpipe: rename bins to scene It was pretty confusing having an entity named "bin" and another named "bins", not least because sometimes there was a need to talk about >1 of the "bins" objects, which couldn't be pluralized any further... Scene is a term used in a bunch of places to talk about what a binner operates on, so it's a decent choice here.	il y a 15 ans
Brian Paul	4b70af918d	llvmpipe: added lp_rast_fence() bin function	il y a 15 ans
Brian Paul	9323740738	llvmpipe: added lp_rast_get_num_threads()	il y a 15 ans
Brian Paul	92dc0f92b0	llvmpipe: implement lp_rast_load_color()	il y a 15 ans
Brian Paul	2bce5c195f	llvmpipe: checkpoint: more thread/queuing changes Now mapping/unmapping the framebuffer is done by a rasteizer thread rather than the main calling thread.	il y a 15 ans
Brian Paul	de31b0e60c	llvmpipe: remove unused lp_rasterizer::fb field	il y a 15 ans
Brian Paul	156eabbaf9	llvmpipe: improve framebuffer/surface code	il y a 15 ans
Brian Paul	205da96fc6	llvmpipe: remove unused fb parameter	il y a 15 ans
Brian Paul	4e67f10331	llvmpipe: minor comment fix	il y a 15 ans
Brian Paul	9509f73c21	llvmpipe: checkpoint: use empty/full bin queues	il y a 15 ans
Brian Paul	96689d451a	llvmpipe: added some debug/info code	il y a 15 ans
Brian Paul	3bee8c2e7c	llvmpipe: use the empty_bins queue now	il y a 15 ans
Brian Paul	d7dbc66636	llvmpipe: checkpoint: begin plugging in bin queue code	il y a 15 ans
José Fonseca	7f457acabc	llvmpipe: Use number of CPUs as default number of threads. Also bump MAX_THREADS to 8.	il y a 15 ans
Brian Paul	73e13c33fd	llvmpipe: remove some left-over debug code	il y a 15 ans
Brian Paul	aab1ceceec	llvmpipe: implement threaded rasterization The LP_NUM_THREADS env var controls how many threads are created. The default (and max) is 4, for now. If LP_NUM_THREADS = 0, threading is not used.	il y a 15 ans
Brian Paul	3a06c113c7	llvmpipe: repartition lp_rasterizer state for threading Some of the state is per-thread. Put that state in new lp_rasterizer_task struct.	il y a 15 ans
Brian Paul	cdaea049c9	llvmpipe: use bin iteration functions when rasterizing bins	il y a 15 ans
Brian Paul	8a23105fa0	llvmpipe: still more bin code reorganization Move tiles_x,y fields from setup state into bin state. Move more bin-adding commands into lp_bin.[ch].	il y a 15 ans
Brian Paul	01b1900084	llvmpipe: reorganization of binning data structions and funtions New lp_bins struct contains all bin information. More move bin-related code into lp_bin.[ch] Use new/updated bin-access functions to hide implementation details. The result is more/cleaner separation between the setup and rast components. This will make double-buffering of the bins easier, etc.	il y a 15 ans
Brian Paul	b533b56750	llvmpipe: move lp_rasterize_bin() into lp_rast.c First step of moving bin rasterization/execution code out of lp_setup.c	il y a 15 ans
Brian Paul	a08d630216	llvmpipe: use LP_DBG() macro everywhere	il y a 15 ans
Brian Paul	b1659b9213	llvmpipe: bin state-change commands Previously, each triangle had a pointer to the state to use for shading. Now we insert state-change commands into the bins. When we execute one of those commands we just update a 'current state' pointer and use that pointer when calling the jit shader. When inserting state-change commands into a bin we check if the previous command was also a state-change command and simply replace it. This avoids accumulating useless/redundant state-change commands.	il y a 15 ans
Brian Paul	e2f4634456	llvmpipe: checkpoint some initial state binning code	il y a 15 ans
Brian Paul	ffd0759973	llvmpipe: comments and a stub for lp_rast_set_state()	il y a 15 ans
Brian Paul	51663f0506	llvmpipe: quick & dirty implementation of lp_rast_store_zstencil() This allows us to do a glReadPixels(GL_DEPTH_COMPONENT) to see what's in the depth buffer to help debugging.	il y a 15 ans
Brian Paul	866e6856d3	llvmpipe: execute shaders on 4x4 blocks instead of 8x2 This matches the convention used by the recursive rasterizer. Also fixed assorted typos, comments, etc. Now tri-z.c, gears.c, etc look basically right but there's still some cracks in triangle rasterization.	il y a 15 ans
Brian Paul	7505510c7b	llvmpipe: add a bunch of comments	il y a 15 ans
Keith Whitwell	280c8acc43	llvmpipe: fix the worst of the depth regressions since switch to 4x4	il y a 16 ans
Keith Whitwell	694f05ac18	llvmpipe: remove one of two definitions of TILESIZE	il y a 16 ans

1 2

79 Révisions (ad60552ca19bce657eadaffd31db752e34e9cda3)