index
:
~cwabbott0/mesa
cwabbott-lima
cwabbott-lima-2
glsl-to-nir-builder
i965-fp64
i965-fp64-v2
i965-fp64-v3
i965-payload-interference
i965-sched
i965-sched-conservative
i965-sched-conservative-v2
i965-sched-test
i965-sched-v2
i965-sched-v3
i965-use-ssa
i965-use-ssa-v2
ir3-sched
jenkins
master
nir-cf-insert-instr
nir-control-flow-mod
nir-cse-hash
nir-cse-hash-v2
nir-dead-cf-v2
nir-dead-cf-v3
nir-dead-cf-v4
nir-dead-cf-v5
nir-deref-instr
nir-divergence
nir-divergence-v2
nir-divergence-v3
nir-divergence-v4
nir-docs
nir-equality-saturation
nir-factor-phis
nir-foreach-block-rewrite
nir-foreach-block-rewrite-v2
nir-foreach-ssa-src
nir-gvn
nir-gvn-v2
nir-opcodes-cleanup
nir-opcodes-cleanup-v2
nir-opt-remove-phis
nir-reassociate-consts
nir-review-v1
nir-serialize
nir-v1.0.1
nir-value-range
nir-vec4-out-of-ssa
nir-vectorize
nir-worklist
radv-amd-shader-ballot
radv-anv-64bit-fixes
radv-doom-exts
radv-rewrite-vars
radv-shader-ballot
radv-shader-ballot-v3
radv-shader-ballot-v4
random-fp64-fixes
shader-time-hacks
test-compressed-wa
test-fp64-compressed-wa
ue4
wip/nir-vtn
Connor's silly Mesa stuff.
UNKNOWN
summary
refs
log
tree
commit
diff
log msg
author
committer
range
Age
Commit message (
Expand
)
Author
Files
Lines
2015-08-13
i965/vec4/nir: translate d2f/f2d
i965-fp64
Connor Abbott
1
-0
/
+32
2015-08-13
i965/vec4: generate d2f/f2d pseudo-instructions
Connor Abbott
1
-0
/
+25
2015-08-13
i965/vec4: handle doubles in type_size()
Connor Abbott
1
-3
/
+6
2015-08-13
i965/vec4: divide width and exec_size by 2 for 64-bit types
Connor Abbott
1
-7
/
+31
2015-08-13
i965/vec4: teach DCE about 64-bit types
Connor Abbott
1
-0
/
+10
2015-08-13
i965/vec4: teach DCE about d2f/f2d
Connor Abbott
1
-0
/
+2
2015-08-13
i965/vec4: don't constant propagate 64-bit immediates
Connor Abbott
1
-0
/
+7
2015-08-13
i965/vec4: teach opt_reduce_swizzle about d2f/f2d
Connor Abbott
1
-0
/
+6
2015-08-13
i965/vec4: add double/float conversion pseudo-opcodes
Connor Abbott
2
-0
/
+6
2015-08-13
i965/vec4/nir: implement double packing/unpacking
Connor Abbott
1
-0
/
+56
2015-08-13
i965/vec4/nir: implement double comparisons
Connor Abbott
1
-2
/
+22
2015-08-13
i965/vec4/nir: fix emitting 64-bit immediates
Connor Abbott
1
-4
/
+15
2015-08-13
i965/vec4/nir: duplicate swizzles and writemasks for doubles
Connor Abbott
1
-7
/
+25
2015-08-13
i965/vec4: set the right type for 64-bit registers
Connor Abbott
1
-0
/
+3
2015-08-13
i965/vec4/nir: allocate two registers for dvec3/dvec4
Connor Abbott
1
-3
/
+22
2015-08-13
i965: add brw_vecn_grf()
Connor Abbott
1
-0
/
+6
2015-08-13
i965: add new BRW_SWIZZLE defines
Connor Abbott
1
-0
/
+4
2015-08-13
i965/vec4: add support for printing DF immediates
Connor Abbott
1
-0
/
+3
2015-08-13
i965/vec4: add a constructor for DF immediates
Connor Abbott
2
-0
/
+10
2015-08-13
XXX split_doubles fixes
Connor Abbott
1
-6
/
+50
2015-08-11
i965/fs: print writemask_all when it's enabled
Connor Abbott
1
-0
/
+3
2015-08-11
XXX undo some of lower SIMD fix
Connor Abbott
1
-1
/
+1
2015-08-11
i965/fs: don't propagate cmod when the exec sizes differ
Connor Abbott
1
-0
/
+3
2015-08-11
i965: add pass for splitting dvec3/dvec4 ALU operations
Connor Abbott
4
-0
/
+322
2015-08-11
i965/fs: respect force_sechalf/force_writemask_all in CSE
Connor Abbott
1
-0
/
+2
2015-08-11
i965/fs: fix dst width calculation in CSE
Connor Abbott
1
-1
/
+1
2015-08-11
i965/fs: fix regs_written in LOAD_PAYLOAD for doubles
Connor Abbott
1
-2
/
+5
2015-08-11
i965/fs: make SIMD-splitting respect the original stride/offset
Connor Abbott
1
-5
/
+11
2015-08-11
i965/fs: fix is_copy_payload() for doubles
Connor Abbott
1
-2
/
+2
2015-08-10
i965/vec4/nir: simplify glsl_type_for_nir_alu_type()
Connor Abbott
1
-14
/
+2
2015-08-10
i965: fix brw_glsl_base_type_for_nir_type() for doubles
Connor Abbott
1
-3
/
+6
2015-08-10
i965: fix brw_type_for_nir_type() for doubles
Connor Abbott
1
-4
/
+12
2015-08-10
XXX rebase fixes
Connor Abbott
2
-30
/
+0
2015-08-10
i965/fs: force nir_type_int64/uint64 to double
Connor Abbott
1
-0
/
+6
2015-08-10
XXX fixup rcp
Connor Abbott
1
-3
/
+3
2015-08-10
XXX actually fixup nir_algebraic
Connor Abbott
3
-48
/
+228
2015-08-10
i965: use pack/unpackDouble lowering
Connor Abbott
1
-0
/
+2
2015-08-10
XXX expand on can_take_stride
Connor Abbott
1
-3
/
+32
2015-08-10
XXX i965/fs: fix uniform emission again
Connor Abbott
1
-5
/
+4
2015-08-10
XXX use split double pack
Connor Abbott
1
-11
/
+13
2015-08-10
XXX fix constants (need to rework this)
Connor Abbott
1
-0
/
+5
2015-08-10
XXX use split opcodes in lower_double_ops
Connor Abbott
1
-61
/
+25
2015-08-10
nir: add a pass for lowering pack/unpackDouble2x32
Connor Abbott
3
-0
/
+111
2015-08-10
nir: add split versions of pack/unpackDouble2x32
Connor Abbott
1
-0
/
+36
2015-08-10
XXX re-enable copy prop
Connor Abbott
1
-1
/
+1
2015-08-10
XXX stash
Connor Abbott
6
-8
/
+52
2015-08-10
i965/fs: print non-1 strides when dumping instructions
Connor Abbott
1
-0
/
+14
2015-08-10
i965/fs: fix load_uniform intrinsic for doubles
Connor Abbott
1
-1
/
+1
2015-08-10
i965/fs: fix assign_constant_locations() when regs_read() > 1
Connor Abbott
1
-2
/
+2
2015-08-10
i965/fs: fix regs_read() for uniforms
Connor Abbott
1
-1
/
+1
[next]