2019-07-11 |
Ian Romanick | nir/algebraic: Reassociate fadd into fmul in DPH-like... |
blob | commitdiff | raw |
2019-07-11 |
Ian Romanick | nir/algebraic: Recognize open-coded flrp(-1, 1, a)... |
blob | commitdiff | raw | diff to current |
2019-07-01 |
Sagar Ghuge | nir: Add optimization to use ROR/ROL instructions |
blob | commitdiff | raw | diff to current |
2019-06-29 |
Ian Romanick | nir/algebraic: Fail build when too many commutative... |
blob | commitdiff | raw | diff to current |
2019-06-28 |
Eric Anholt | nir: Fix lowering of bitfield_insert to shifts. |
blob | commitdiff | raw | diff to current |
2019-06-26 |
Caio Marcelo de... | nir/algebraic: Add helpers and a rule involving wrapping |
blob | commitdiff | raw | diff to current |
2019-06-26 |
Jonathan Marek | nir: remove fnot/fxor/fand/for opcodes |
blob | commitdiff | raw | diff to current |
2019-06-24 |
Daniel Schürmann | nir: introduce lowering of bitfield_insert to bfm and... |
blob | commitdiff | raw | diff to current |
2019-06-24 |
Daniel Schürmann | nir/algebraic: Use unsigned comparison when lowering... |
blob | commitdiff | raw | diff to current |
2019-06-24 |
Daniel Schürmann | nir/algebraic: Remove unnecessary iand of [iu]bfe and... |
blob | commitdiff | raw | diff to current |
2019-06-24 |
Daniel Schürmann | nir: define behavior of nir_op_bfm and nir_op_u/ibfe... |
blob | commitdiff | raw | diff to current |
2019-06-24 |
Daniel Schürmann | nir/algebraic: add optimization pattern for ('ult'... |
blob | commitdiff | raw | diff to current |
2019-06-10 |
Eduardo Lima Mitev | nir/opt_algebraic: Fix rules for imadsh_mix16 |
blob | commitdiff | raw | diff to current |
2019-06-07 |
Eduardo Lima Mitev | nir_algebraic: Add basic optimizations for umul_low... |
blob | commitdiff | raw | diff to current |
2019-06-05 |
Kenneth Graunke | nir: Combine lower_fmod16/32 back into a single lower_fmod. |
blob | commitdiff | raw | diff to current |
2019-06-05 |
Kenneth Graunke | nir: Drop lower_fmod64 option. |
blob | commitdiff | raw | diff to current |
2019-06-04 |
Alyssa Rosenzweig | nir/algebraic: Simplify max(abs(a), 0.0) -> abs(a) |
blob | commitdiff | raw | diff to current |
2019-05-31 |
Jonathan Marek | nir: add lower_bitshift option |
blob | commitdiff | raw | diff to current |
2019-05-16 |
Alyssa Rosenzweig | nir/algebraic: Remove problematic "optimization" |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Commute 1-fsat(a) to fsat(1-a) for all... |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Eliminate useless fsat() on operand... |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Strip double negatives from comparison... |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | intel/compiler: Repeat nir_opt_algebraic_late |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | Revert "nir: add late opt to turn inot/b2f combos back... |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Eliminate a tautological compare |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Simplify fsqrt domain guard |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Recognize open-coded fsat with modifiers |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Push unary operations into source operan... |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Recognize open-coded flrp(a, b, fsat(c)) |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir/algebraic: Add missing ffma(-1, a, b) pattern |
blob | commitdiff | raw | diff to current |
2019-05-14 |
Ian Romanick | nir: Mark ffma as 2src_commutative |
blob | commitdiff | raw | diff to current |
2019-05-07 |
Ian Romanick | nir/algebraic: Reassociate open-coded flrp(1, b, c) |
blob | commitdiff | raw | diff to current |
2019-05-07 |
Ian Romanick | nir: Use the flrp lowering pass instead of nir_opt_alge... |
blob | commitdiff | raw | diff to current |
2019-05-07 |
Ian Romanick | nir/algebraic: Pull common multiplication out of flrp... |
blob | commitdiff | raw | diff to current |
2019-05-07 |
Ian Romanick | nir/algebraic: Pull common addition out of flrp arguments |
blob | commitdiff | raw | diff to current |
2019-04-26 |
Jason Ekstrand | nir/algebraic: Optimize integer cast-of-cast |
blob | commitdiff | raw | diff to current |
2019-04-19 |
Erico Nunes | nir/algebraic: add lowering for fsign |
blob | commitdiff | raw | diff to current |
2019-04-18 |
Ian Romanick | nir/algebraic: Strength reduce some compares of x and -x |
blob | commitdiff | raw | diff to current |
2019-04-18 |
Ian Romanick | nir/algebraic: Fix some 1-bit Boolean weirdness |
blob | commitdiff | raw | diff to current |
2019-04-18 |
Ian Romanick | nir/algebraic: Replace a pattern where iand with a... |
blob | commitdiff | raw | diff to current |
2019-04-18 |
Ian Romanick | nir/algebraic: Recognize open-coded copysign(1.0, a) |
blob | commitdiff | raw | diff to current |
2019-04-13 |
Christian Gmeiner | nir: add lower_ftrunc |
blob | commitdiff | raw | diff to current |
2019-04-09 |
Caio Marcelo de... | nir/algebraic: Lower CS derivatives to zero when no... |
blob | commitdiff | raw | diff to current |
2019-04-05 |
Jason Ekstrand | nir/algebraic: Add some logical OR and AND patterns |
blob | commitdiff | raw | diff to current |
2019-04-05 |
Jason Ekstrand | nir/algebraic: Drop some @bool specifiers |
blob | commitdiff | raw | diff to current |
2019-03-28 |
Ian Romanick | nir/algebraic: Add missing 16-bit extract_[iu]8 patterns |
blob | commitdiff | raw | diff to current |
2019-03-28 |
Ian Romanick | nir/algebraic: Add missing 64-bit extract_[iu]8 patterns |
blob | commitdiff | raw | diff to current |
2019-03-28 |
Ian Romanick | nir/algebraic: Remove redundant extract_[iu]8 patterns |
blob | commitdiff | raw | diff to current |
2019-03-28 |
Ian Romanick | nir/algebraic: Fix up extract_[iu]8 after loop unrolling |
blob | commitdiff | raw | diff to current |
2019-03-25 |
Iago Toral Quiroga | compiler/nir: add lowering for 16-bit ldexp |
blob | commitdiff | raw | diff to current |
2019-03-25 |
Iago Toral Quiroga | compiler/nir: add lowering for 16-bit flrp |
blob | commitdiff | raw | diff to current |
2019-03-25 |
Iago Toral Quiroga | compiler/nir: add lowering option for 16-bit fmod |
blob | commitdiff | raw | diff to current |
2019-03-15 |
Jason Ekstrand | nir/algebraic: Add a couple optimizations for iabs... |
blob | commitdiff | raw | diff to current |
2019-03-09 |
Kenneth Graunke | Revert MR 369 (Fix extract_i8 and extract_u8 for 64... |
blob | commitdiff | raw | diff to current |
2019-03-09 |
Ian Romanick | nir/algebraic: Add missing 16-bit extract_[iu]8 patterns |
blob | commitdiff | raw | diff to current |
2019-03-09 |
Ian Romanick | nir/algebraic: Add missing 64-bit extract_[iu]8 patterns |
blob | commitdiff | raw | diff to current |
2019-03-09 |
Ian Romanick | nir/algebraic: Remove redundant extract_[iu]8 patterns |
blob | commitdiff | raw | diff to current |
2019-03-09 |
Ian Romanick | nir/algebraic: Fix up extract_[iu]8 after loop unrolling |
blob | commitdiff | raw | diff to current |
2019-03-05 |
Karol Herbst | nir/spirv: initial handling of OpenCL.std extension... |
blob | commitdiff | raw | diff to current |
2019-03-04 |
Sagar Ghuge | nir/algebraic: Optimize low 32 bit extraction |
blob | commitdiff | raw | diff to current |
2019-03-04 |
Sagar Ghuge | nir/glsl: Add another way of doing lower_imul64 for... |
blob | commitdiff | raw | diff to current |
2019-03-02 |
Ian Romanick | nir/algebraic: Optimize away an fsat of a b2f |
blob | commitdiff | raw | diff to current |
2019-03-01 |
Ian Romanick | nir/algebraic: Replace a-fract(a) with floor(a) |
blob | commitdiff | raw | diff to current |
2019-03-01 |
Ian Romanick | nir/algebraic: Replace a bcsel of a b2f sources with... |
blob | commitdiff | raw | diff to current |
2019-03-01 |
Ian Romanick | nir/algebraic: Replace i2b used by bcsel or if-statemen... |
blob | commitdiff | raw | diff to current |
2019-02-25 |
Daniel Schürmann | nir: Use SM5 properties to optimize shift(a@32, iand... |
blob | commitdiff | raw | diff to current |
2019-02-17 |
Kenneth Graunke | nir: Don't reassociate add/mul chains containing only... |
blob | commitdiff | raw | diff to current |
2019-02-15 |
Ian Romanick | nir/algebraic: Simplify comparison with sequential... |
blob | commitdiff | raw | diff to current |
2019-02-15 |
Ian Romanick | nir/algebraic: Convert some f2u to f2i |
blob | commitdiff | raw | diff to current |
2019-02-14 |
Eric Anholt | nir: Move panfrost's isign lowering to nir_opt_algebraic. |
blob | commitdiff | raw | diff to current |
2018-12-19 |
Ian Romanick | nir/algebraic: Don't put quotes around floating point... |
blob | commitdiff | raw | diff to current |
2018-12-16 |
Jason Ekstrand | nir/algebraic: Add some optimizations for D3D-style... |
blob | commitdiff | raw | diff to current |
2018-12-16 |
Jason Ekstrand | nir/algebraic: Optimize 1-bit Booleans |
blob | commitdiff | raw | diff to current |
2018-12-16 |
Jason Ekstrand | nir/algebraic: Generalize an optimization |
blob | commitdiff | raw | diff to current |
2018-12-16 |
Jason Ekstrand | nir/algebraic: Make an optimization more specific |
blob | commitdiff | raw | diff to current |
2018-12-16 |
Jason Ekstrand | nir: Drop support for lower_b2f |
blob | commitdiff | raw | diff to current |
2018-12-16 |
Jason Ekstrand | nir/algebraic: Optimize x2b(xneg(a)) -> a |
blob | commitdiff | raw | diff to current |
2018-12-05 |
Jason Ekstrand | nir: Make boolean conversions sized just like the others |
blob | commitdiff | raw | diff to current |
2018-12-05 |
Jason Ekstrand | nir/opt_algebraic: Add 32-bit specifiers to a bunch... |
blob | commitdiff | raw | diff to current |
2018-12-05 |
Jason Ekstrand | nir/opt_algebraic: Drop bit-size suffixes from conversions |
blob | commitdiff | raw | diff to current |
2018-12-05 |
Jason Ekstrand | nir/opt_algebraic: Simplify an optimization using the... |
blob | commitdiff | raw | diff to current |
2018-11-27 |
Jonathan Marek | nir: add fceil lowering |
blob | commitdiff | raw | diff to current |
2018-11-12 |
Christian Gmeiner | nir: add lowering for ffloor |
blob | commitdiff | raw | diff to current |
2018-10-22 |
Jason Ekstrand | nir/algebraic: Generalize an optimization |
blob | commitdiff | raw | diff to current |
2018-10-11 |
Jason Ekstrand | nir: Add a bunch of b2[if] optimizations |
blob | commitdiff | raw | diff to current |
2018-10-09 |
Ian Romanick | nir/algebraic: Simplify fsat of fsign |
blob | commitdiff | raw | diff to current |
2018-10-09 |
Ian Romanick | nir/algebraic: sign(x)*x*x is abs(x)*x |
blob | commitdiff | raw | diff to current |
2018-08-29 |
Jason Ekstrand | nir/algebraic: Add some max/min optimizations |
blob | commitdiff | raw | diff to current |
2018-08-29 |
Jason Ekstrand | nir/algebraic: Add more extract_[iu](8|16) optimizations |
blob | commitdiff | raw | diff to current |
2018-08-29 |
Jason Ekstrand | nir/algebraic: Be more careful converting ushr to extra... |
blob | commitdiff | raw | diff to current |
2018-08-09 |
Mathieu Bridon | python: Use the right function for the job |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Transform expressions of b2f(a) and b2f(b) to... |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Transform expressions of b2f(a) and b2f(b) to... |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Transform expressions of b2f(a) and b2f(b) to... |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Transform expressions of b2f(a) and b2f(b) to... |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Transform expressions of b2f(a) and b2f(b) to... |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Transform -fabs(a) >= 0 to a == 0 |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Transform expressions of b2f(a) and b2f(b) to... |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Transform -fabs(a) < 0 to a != 0 |
blob | commitdiff | raw | diff to current |
2018-08-04 |
Ian Romanick | nir: Rearrange bcsel with two bcsel sources |
blob | commitdiff | raw | diff to current |
next |