Lines Matching full:ir3
168 - ir3: Infinite loop in scheduler when splitting
1227 - ir3: fix mem_ctx for ir3_disasm_info::nir
1243 - ir3: update ir3_const_state comment
1244 - ir3: copy req_local_mem for MESA_SHADER_KERNEL
1245 - ir3: fix output_loc size
1246 - ir3: remove unused patch_vertices_in
1247 - ir3: increment constoff right after it is assigned
1248 - ir3: fix tess param allocation
1249 - ir3: set UL flag before ir3_lower_subgroups
1279 - ir3: fix predicate splitting in scheduler
1287 - ir3: Create at least one thread for async shader compile
1307 - ir3, freedreno, tu: Make ir3_shader_variant store stream_output
1308 - ir3, fd, tu: Copy misc. info from ir3_shader to ir3_shader_variant
1309 - ir3: Use ir3_shader_variant::type more often
1310 - ir3: Allocate disasm_info under variant
1311 - ir3: Add ir3_shader_variant::compiler
1313 - ir3: Remove ir3_shader_variant::shader
1314 - ir3: Add functions to serialize variants
1315 - ir3: Support disabling the pipeline cache
1316 - ir3: Add ir3_shader_create_variant()
1337 - ir3/sched: Fix could_sched() determination
1339 - ir3: Fix vectorizer condition for SSBOs
1371 - ir3/spill: Fix extracting from a vector at the end of a block
1461 - ir3: handle gl_Layer and gl_ViewportIndex when there is TES + GS
1462 - ir3: Force late-z if FS has global store/atomic
1480 - ir3: Use NIR's info.writes_memory to detect when when to force late-z
1497 - ir3/ra: Always insert interval for precolored inputs
1498 - ir3: Never remove GS_HEADER_IR3 sysval input
1501 - ir3: Prevent reordering movmsk with kill
1764 - freedreno/ir3: Make sched nodes before adding deps.
1766 - freedreno/ir3: Call nir_opt_find_array_copies().
1840 - freedreno/ir3: Fix 16-bit bit_count.
1844 - freedreno/ir3: Add support for 16-bit nir_texop_lod.
1845 - freedreno/ir3: Lower texture instructions used only for f2f16 to 16-bit.
1851 - freedreno/ir3: Enable load/store vectorization for SSBO access, too.
1852 - freedreno/ir3: Make the types of tex coord padding match.
1853 - freedreno/ir3: Fix validation of half-precision image store values.
1854 - freedreno/ir3: Fold 16-bit conversions into image load/store src/dsts.
1904 - freedreno/ir3: Disable image/ssbo 16-bit conversion folding pre-a6xx.
1925 - ir3: Retire the cp postsched pass now that we do RA in SSA.
1942 - ir3: Make sure to pass the interp_mode through in our load_bary lowering.
1943 - nir+ir3: Rename load_size_ir3 to load_center_rhw_ir3.
1945 - ir3: Clarify what's happening in the interpolateAtOffset() math.
1946 - ir3: Use non-persp interpolation when appropriate for interpolateAtOffset.
1958 - ir3: Fix the no-emitted-vertex condition emission in geom lowering.
1965 - freedreno/ir3: Enable core NIR's 16-bit ALU optimizations.
2291 - ir3: Implement [iu]sub_sat.
2297 - ir3: Only run 16bit tex NIR passes on a5xx+.
2298 - ir3: Lower alu to scalar if nir_legalize_16bit_sampler_srcs made progress.
2301 - ir3: Stop using nir_legalize_16bit_sampler_srcs.
2485 - ir3: handle intrinsic_load_draw_id when scanning driver constants
2488 - freedreno,ir3: rename Z_CLAMP_ENABLE to Z_CLIP_DISABLE
2493 - ir3: change maximum size of const files.
2494 - ir3: handle shared consts.
4196 - freedreno/ir3: Move the texture array coord fixup to nir
4197 - freedreno/ir3: Pass 16-bit sampler coordinates when possible.
5304 - freedreno/ir3: Don't lower_gs multiple times
5305 - freedreno/ir3: Add copy_vars() helper
5306 - freedreno/ir3: Copy vars if needed on EndPrimitive()
5307 - freedreno/ir3: Add more tess varying slots
5310 - freedreno/ir3: Fix GS clip-plane lowering
5320 - freedreno/ir3: Remove unneeded forward declaration
5772 - freedreno/ir3: tidy up duplication of common nir options