OpenHarmony/third_party_mesa3d
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

agx: Lower VBOs in NIR

Browse Source 
Now we support all the vertex formats! This means we don't hit u_vbuf for format
translation, which helps performance in lots of applications. By doing the
lowering in NIR, the vertex fetch code itself can be optimized by NIR (e.g.
nir_opt_algebraic) which can improve generated code quality.

In my first implementation of this, I had a big switch statement mapping format
enums to interchange formats and post-processing code. This ends up being really
unwieldly, the combinatorics of bit packing + conversion + swizzles is
enormous and for performance we want to support everything (no u_vbuf
fallbacks). To keep the combinatorics in check, we rely on parsing the
util_format_description to separate out the issues of bit packing, conversion,
and swizzling, allowing us to handle bizarro formats like B10G10R10A2_SNORM with
no special casing.

In an effort to support everything in one shot, this handles all the formats
needed for the extensions EXT_vertex_array_bgra, ARB_vertex_type_2_10_10_10_rev,
and ARB_vertex_type_10f_11f_11f_rev.

Passes dEQP-GLES3.functional.vertex_arrays.*

Signed-off-by: Alyssa Rosenzweig <alyssa@rosenzweig.io>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19996>
This commit is contained in:
Alyssa Rosenzweig
2022-11-24 20:40:50 -05:00
committed by Marge Bot
parent fb49715a2c
commit 8dcf7648f1
11 changed files with 318 additions and 222 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docs/features.txt
View File

@@ -108,7 +108,7 @@ GL 3.3, GLSL 3.30 --- all DONE: freedreno, i965, nv50, nvc0, r600, radeonsi, llv
GL_ARB_texture_swizzle DONE (v3d, vc4, panfrost, lima, asahi)
GL_ARB_timer_query DONE ()
GL_ARB_instanced_arrays DONE (etnaviv/HALTI2, v3d, panfrost)
GL_ARB_vertex_type_2_10_10_10_rev DONE (v3d, panfrost)
GL_ARB_vertex_type_2_10_10_10_rev DONE (v3d, panfrost, asahi)
GL 4.0, GLSL 4.00 --- all DONE: freedreno/a6xx, i965/gen7+, nvc0, r600, radeonsi, llvmpipe, virgl, zink, d3d12
@@ -208,7 +208,7 @@ GL 4.4, GLSL 4.40 -- all DONE: freedreno/a6xx, i965/gen8+, nvc0, r600, radeonsi,
GL_ARB_query_buffer_object DONE (freedreno/a6xx, i965/hsw+, virgl)
GL_ARB_texture_mirror_clamp_to_edge DONE (freedreno, i965, nv50, softpipe, virgl, v3d, panfrost)
GL_ARB_texture_stencil8 DONE (freedreno, i965/hsw+, nv50, softpipe, virgl, v3d, panfrost, d3d12, asahi)
GL_ARB_vertex_type_10f_11f_11f_rev DONE (freedreno, i965, nv50, softpipe, virgl, panfrost, d3d12)
GL_ARB_vertex_type_10f_11f_11f_rev DONE (freedreno, i965, nv50, softpipe, virgl, panfrost, d3d12, asahi)
GL 4.5, GLSL 4.50 -- all DONE: freedreno/a6xx, nvc0, r600, radeonsi, llvmpipe, zink