| 2019-12-03 |
Ian Romanick | nir/algebraic: Rearrange bcsel sequences generated...
|
commit | commitdiff | tree |
| 2019-12-03 |
Ian Romanick | nir/algebraic: Simplify some Inf and NaN avoidance...
|
commit | commitdiff | tree |
| 2019-12-03 |
Ian Romanick | intel/compiler: Increase nir_opt_peephole_select threshold
|
commit | commitdiff | tree |
| 2019-12-03 |
Ian Romanick | nir/opt_peephole_select: Don't count some unary operations
|
commit | commitdiff | tree |
| 2019-11-22 |
Ian Romanick | nir/range_analysis: Make sure the table validation...
|
commit | commitdiff | tree |
| 2019-11-22 |
Ian Romanick | nir/range-analysis: Add pragmas to help loop unrolling
|
commit | commitdiff | tree |
| 2019-11-22 |
Ian Romanick | intel/fs: Disable conditional discard optimization...
|
commit | commitdiff | tree |
| 2019-11-04 |
Ian Romanick | nir/algebraic: Mark other comparison exact when removing...
|
commit | commitdiff | tree |
| 2019-11-04 |
Ian Romanick | nir/algebraic: Add the ability to mark a replacement...
|
commit | commitdiff | tree |
| 2019-10-31 |
Ian Romanick | intel/compiler: Report the number of non-spill/fill...
|
commit | commitdiff | tree |
| 2019-10-17 |
Ian Romanick | intel/vec4: Don't try both sources as immediates for DPH
|
commit | commitdiff | tree |
| 2019-10-17 |
Ian Romanick | nir/search: Fix possible NULL dereference in is_fsign
|
commit | commitdiff | tree |
| 2019-09-25 |
Ian Romanick | nir/range-analysis: Use types to provide better ranges...
|
commit | commitdiff | tree |
| 2019-09-25 |
Ian Romanick | nir/range-analysis: Use types in the hash key
|
commit | commitdiff | tree |
| 2019-09-25 |
Ian Romanick | nir/range-analysis: Bail if the types don't match
|
commit | commitdiff | tree |
| 2019-09-19 |
Ian Romanick | nir/algebraic: Additional D3D Boolean optimization
|
commit | commitdiff | tree |
| 2019-09-19 |
Ian Romanick | nir/algebraic: Do not apply late DPH optimization in...
|
commit | commitdiff | tree |
| 2019-08-29 |
Ian Romanick | nir/range-analysis: Add a lot more assertions about...
|
commit | commitdiff | tree |
| 2019-08-29 |
Ian Romanick | nir/range-analysis: Range tracking for fpow
|
commit | commitdiff | tree |
| 2019-08-29 |
Ian Romanick | nir/range-analysis: Handle constants in nir_op_mov...
|
commit | commitdiff | tree |
| 2019-08-29 |
Ian Romanick | nir/range-analysis: Fix incorrect fadd range result...
|
commit | commitdiff | tree |
| 2019-08-29 |
Ian Romanick | nir/range-analysis: Adjust result range of multiplication...
|
commit | commitdiff | tree |
| 2019-08-29 |
Ian Romanick | nir/range-analysis: Adjust result range of exp2 to...
|
commit | commitdiff | tree |
| 2019-08-29 |
Ian Romanick | nir/algebraic: Clean up value range analysis-based...
|
commit | commitdiff | tree |
| 2019-08-29 |
Ian Romanick | nir/algebraic: Mark some value range analysis-based...
|
commit | commitdiff | tree |
| 2019-08-28 |
Ian Romanick | intel/compiler: Request bitfield_reverse lowering on...
|
commit | commitdiff | tree |
| 2019-08-28 |
Ian Romanick | nir/algrbraic: Don't optimize open-coded bitfield reverse...
|
commit | commitdiff | tree |
| 2019-08-14 |
Ian Romanick | nir/algebraic: Reassociate shift-by-constant of shift...
|
commit | commitdiff | tree |
| 2019-08-14 |
Ian Romanick | nir/algebraic: Reassociate add-and-shift to be shift...
|
commit | commitdiff | tree |
| 2019-08-14 |
Ian Romanick | nir/range-analysis: Fail gracefully on non-SSA sources
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/algebraic: Use value range analysis to eliminate...
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/algebraic: Use value range analysis to convert...
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/algebraic: Use value range analysis to eliminate...
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/algebraic: Use value range analysis to eliminate...
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/range-analysis: Range tracking for ffma and flrp
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/range-analysis: Range tracking for bcsel
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/range-analysis: Tighten the range of fsat based...
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/range-analysis: Rudimentary value range analysis...
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/algebraic: Simplify some comparisons like a+constant...
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/algebraic: Recognize (a < 0 || 0 < b) as min(a...
|
commit | commitdiff | tree |
| 2019-08-06 |
Ian Romanick | nir/algebraic: Replace checks that a value is between...
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | nir/algebraic: Recognize open-coded flrp(a, b, a)
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | nir/algebraic: Rearrange 1-((1-a) * (1-b)) into flrp...
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | nir/algebraic: Reassociate fadd into fmul in DPH-like...
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | nir/algebraic: Recognize open-coded flrp(-1, 1, a)...
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | nir: intel/vec4: Add flag to disable some algebraic...
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | intel/vec4: Try to emit immediate sources for MOV
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | intel/vec4: Try to emit a VF source in try_immediate_source
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | intel/vec4: Try to emit a single load for multiple...
|
commit | commitdiff | tree |
| 2019-07-11 |
Ian Romanick | intel/vec4: Refactor operand fixing for ffma and flrp
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | mesa: Set minimum possible GLSL version
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | intel/vec4: Delete vec4_visitor::emit_lrp
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | nir: Allow nir_ssa_alu_instr_src_components to operate...
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | nir: Handle swizzle in nir_alu_srcs_negative_equal
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | nir: nir_const_value_negative_equal compares one value...
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | nir: Port some const_value_negative_equal tests to...
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | nir: Pass fully qualified type to nir_const_value_negative_equal
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | nir: Use nir_src_bit_size instead of alu1->dest.dest...
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | intel/vec4: Reswizzle VF immediates too
|
commit | commitdiff | tree |
| 2019-07-08 |
Ian Romanick | nir: Add unit tests for nir_opt_comparison_pre
|
commit | commitdiff | tree |
| 2019-06-29 |
Ian Romanick | nir/serach: Increase maximum commutative expressions...
|
commit | commitdiff | tree |
| 2019-06-29 |
Ian Romanick | nir/algebraic: Don't mark expression with duplicate...
|
commit | commitdiff | tree |
| 2019-06-29 |
Ian Romanick | nir/search: Log Boolean constants instead of asserting
|
commit | commitdiff | tree |
| 2019-06-29 |
Ian Romanick | nir/algebraic: Fail build when too many commutative...
|
commit | commitdiff | tree |
| 2019-06-29 |
Ian Romanick | nir/algebraic: Fix whitespace error
|
commit | commitdiff | tree |
| 2019-06-29 |
Ian Romanick | intel/vec4: Try both sources as candidates for being...
|
commit | commitdiff | tree |
| 2019-06-29 |
Ian Romanick | intel/vec4: Try immediate sources for dot products too
|
commit | commitdiff | tree |
| 2019-06-29 |
Ian Romanick | intel/vec4: Try emitting non-scalar immediates
|
commit | commitdiff | tree |
| 2019-06-24 |
Ian Romanick | glsl: Don't increase the iteration count when there...
|
commit | commitdiff | tree |
| 2019-06-11 |
Ian Romanick | intel/fs: Mark source 0 of bcsel as needing Boolean...
|
commit | commitdiff | tree |
| 2019-06-06 |
Ian Romanick | intel/compiler: Treat b32csel as potentially producing...
|
commit | commitdiff | tree |
| 2019-06-06 |
Ian Romanick | intel/fs: Improve discard_if code generation
|
commit | commitdiff | tree |
| 2019-06-06 |
Ian Romanick | intel/fs: Add need_dest parameter to fs_visitor::nir_emit_alu
|
commit | commitdiff | tree |
| 2019-06-06 |
Ian Romanick | intel/fs: Allow cmod propagation across reads and writes...
|
commit | commitdiff | tree |
| 2019-06-06 |
Ian Romanick | intel/fs: Fix flag_subreg handling in cmod propagation
|
commit | commitdiff | tree |
| 2019-06-06 |
Ian Romanick | intel/fs: Add missing tests for cmod_propagate_not
|
commit | commitdiff | tree |
| 2019-05-31 |
Ian Romanick | intel/compiler: Use compare rematerialization pass
|
commit | commitdiff | tree |
| 2019-05-31 |
Ian Romanick | nir: Rematerialize compare instructions
|
commit | commitdiff | tree |
| 2019-05-31 |
Ian Romanick | nir: Add a shallow clone function for nir_alu_instr
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Commute 1-fsat(a) to fsat(1-a) for all...
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Eliminate useless fsat() on operand...
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Strip double negatives from comparison...
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | intel/compiler: Repeat nir_opt_algebraic_late
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | Revert "nir: add late opt to turn inot/b2f combos back...
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Eliminate a tautological compare
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Simplify fsqrt domain guard
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/search: Don't compare 8-bit or 1-bit constants...
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Recognize open-coded fsat with modifiers
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Push unary operations into source operands...
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Recognize open-coded flrp(a, b, fsat(c))
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | intel/fs: Allow cmod propagation to instructions with...
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir/algebraic: Add missing ffma(-1, a, b) pattern
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir: Mark ffma as 2src_commutative
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir: Add support for 2src_commutative ops that have...
|
commit | commitdiff | tree |
| 2019-05-14 |
Ian Romanick | nir: Rename commutative to 2src_commutative
|
commit | commitdiff | tree |
| 2019-05-09 |
Ian Romanick | nir: Initialize lower_flrp_progress everywhere
|
commit | commitdiff | tree |
| 2019-05-08 |
Ian Romanick | nir/flrp: Reassociate add in flrp(±1, b, c) lowering...
|
commit | commitdiff | tree |
| 2019-05-08 |
Ian Romanick | nir/flrp: Fix typo on the flrp(±1, b, c) path
|
commit | commitdiff | tree |
| 2019-05-07 |
Ian Romanick | intel/compiler: Don't always require precise lowering...
|
commit | commitdiff | tree |
| 2019-05-07 |
Ian Romanick | nir/algebraic: Reassociate open-coded flrp(1, b, c)
|
commit | commitdiff | tree |
| next |
projects / mesa.git / search