2020-07-20 |
Daniel Schürmann | nir/algebraic: add distributive rules for ior/iand |
blob | commitdiff | raw |
2020-07-20 |
Daniel Schürmann | nir/algebraic: optimize (a < 0.0) ? -a : a -> fabs(a) |
blob | commitdiff | raw | diff to current |
2020-07-20 |
Daniel Schürmann | nir/algebraic: optimize fmul(x, bcsel(c, -1.0, 1.0... |
blob | commitdiff | raw | diff to current |
2020-07-20 |
Daniel Schürmann | nir/algebraic: add some more unop + bcsel optimizations |
blob | commitdiff | raw | diff to current |
2020-07-20 |
Daniel Schürmann | nir/algebraic: add optimizations for fsign/isign |
blob | commitdiff | raw | diff to current |
2020-07-20 |
Daniel Schürmann | nir/algebraic: optimize iand/ior of (n)eq zero |
blob | commitdiff | raw | diff to current |
2020-07-20 |
Daniel Schürmann | nir/algebraic: optimize bcsel(a, 0, 1) to b2i |
blob | commitdiff | raw | diff to current |
2020-07-02 |
Ian Romanick | nir/algebraic: Don't distrubte absolute-value into... |
blob | commitdiff | raw | diff to current |
2020-06-30 |
Alyssa Rosenzweig | nir: Propagate *2*16 conversions into vectors |
blob | commitdiff | raw | diff to current |
2020-06-29 |
Boris Brezillon | nir: Add new rules to optimize NOOP pack/unpack pairs |
blob | commitdiff | raw | diff to current |
2020-06-02 |
Marek Olšák | nir: add i2imp and u2ump opcodes for conversions to... |
blob | commitdiff | raw | diff to current |
2020-06-02 |
Alyssa Rosenzweig | nir: Fold f2f16(b2f32(x)) to b2f16(x) |
blob | commitdiff | raw | diff to current |
2020-05-11 |
Ian Romanick | nir/algebraic: Eliminate useless extract before unpack |
blob | commitdiff | raw | diff to current |
2020-05-11 |
Ian Romanick | nir/algebraic: Add some half packing optimizations... |
blob | commitdiff | raw | diff to current |
2020-05-11 |
Ian Romanick | nir/algebraic: Optimize ushr of pack_half, not ishr |
blob | commitdiff | raw | diff to current |
2020-05-07 |
Ian Romanick | nir/algebraic: Optimize some bfe patterns |
blob | commitdiff | raw | diff to current |
2020-05-07 |
Ian Romanick | nir/algebraic: Split ibfe and ubfe with two constant... |
blob | commitdiff | raw | diff to current |
2020-05-07 |
Ian Romanick | nir/algebraic: Recognize open-coded byte or word extrac... |
blob | commitdiff | raw | diff to current |
2020-05-07 |
Rhys Perry | nir: make fsat return 0.0 with NaN instead of passing... |
blob | commitdiff | raw | diff to current |
2020-04-28 |
Ian Romanick | nir/algebraic: Require operands to iand be 32-bit |
blob | commitdiff | raw | diff to current |
2020-04-27 |
Jonathan Marek | nir: add pack_32_2x16_split/unpack_32_2x16_split lowering |
blob | commitdiff | raw | diff to current |
2020-04-23 |
Gert Wollny | nir: Add umad24 and umul24 opcodes |
blob | commitdiff | raw | diff to current |
2020-04-23 |
Rhys Perry | nir/algebraic: don't undo lowering of 8/16-bit comparis... |
blob | commitdiff | raw | diff to current |
2020-04-20 |
Samuel Pitoiset | nir/opt_algebraic: lower 64-bit fmin3/fmax3/fmed3 |
blob | commitdiff | raw | diff to current |
2020-04-01 |
Ian Romanick | nir/algebraic: Remove a redundant fabs pattern |
blob | commitdiff | raw | diff to current |
2020-04-01 |
Ian Romanick | nir/algebraic: Use value range analysis to convert... |
blob | commitdiff | raw | diff to current |
2020-04-01 |
Ian Romanick | nir/algebraic: Distribute source modifiers into instruc... |
blob | commitdiff | raw | diff to current |
2020-03-31 |
Jason Ekstrand | nir/algebraic: Add downcast-of-pack opts |
blob | commitdiff | raw | diff to current |
2020-03-30 |
Jason Ekstrand | nir: Add b2b opcodes |
blob | commitdiff | raw | diff to current |
2020-03-30 |
Samuel Pitoiset | nir/algebraic: add fexp2(fmul(flog2(a), 0.5) -> fsqrt... |
blob | commitdiff | raw | diff to current |
2020-03-18 |
Ian Romanick | soft-fp64/flt: Perform checks in a different order |
blob | commitdiff | raw | diff to current |
2020-03-18 |
Ian Romanick | nir/algebraic: Simplify a contradiction that can occur... |
blob | commitdiff | raw | diff to current |
2020-03-18 |
Ian Romanick | nir/algebraic: Constant reassociation for bitwise opera... |
blob | commitdiff | raw | diff to current |
2020-03-18 |
Ian Romanick | nir/algebraic: Generalize some and-of-shift-right patte... |
blob | commitdiff | raw | diff to current |
2020-03-18 |
Ian Romanick | nir/algebraic: optimize ior(ine(a, 0), ine(b, 0)) to... |
blob | commitdiff | raw | diff to current |
2020-03-18 |
Ian Romanick | nir/algebraic: Simplify logic to detect sign of an... |
blob | commitdiff | raw | diff to current |
2020-03-12 |
Juan A. Suarez Romero | nir/algebraic: coalesce fmod lowering |
blob | commitdiff | raw | diff to current |
2020-03-04 |
Jason Ekstrand | nir: Flush to zero with OOB low exponents in ldexp |
blob | commitdiff | raw | diff to current |
2020-02-24 |
Hyunjun Ko | nir: Add optimization for doing removing f16/f32 conver... |
blob | commitdiff | raw | diff to current |
2020-02-24 |
Neil Roberts | nir/opcodes: Add nir_op_f2fmp |
blob | commitdiff | raw | diff to current |
2020-02-13 |
Erik Faye-Lund | Revert "nir: Add a couple trivial abs optimizations" |
blob | commitdiff | raw | diff to current |
2020-02-11 |
Samuel Pitoiset | nir: do not use De Morgan's Law rules for flt and fge |
blob | commitdiff | raw | diff to current |
2020-01-29 |
Rhys Perry | nir/algebraic: add some half packing optimizations |
blob | commitdiff | raw | diff to current |
2020-01-29 |
Rhys Perry | nir/algebraic: add patterns for a >> #b << #b |
blob | commitdiff | raw | diff to current |
2020-01-23 |
Ian Romanick | nir/algebraic: Optimize some 64-bit integer comparisons... |
blob | commitdiff | raw | diff to current |
2020-01-23 |
Ian Romanick | nir/algebraic: Add lowering for 64-bit iadd_sat and... |
blob | commitdiff | raw | diff to current |
2020-01-23 |
Ian Romanick | nir/algebraic: Add lowering for 64-bit uadd_sat |
blob | commitdiff | raw | diff to current |
2020-01-23 |
Ian Romanick | nir/algebraic: Add lowering for 64-bit usub_sat |
blob | commitdiff | raw | diff to current |
2020-01-23 |
Ian Romanick | nir/algebraic: Add lowering for 64-bit hadd and rhadd |
blob | commitdiff | raw | diff to current |
2020-01-23 |
Ian Romanick | nir/algebraic: Add lowering for uabs_usub and uabs_isub |
blob | commitdiff | raw | diff to current |
2020-01-15 |
Elie Tournier | nir/algebraic: sqrt(x)*sqrt(x) -> fabs(x) |
blob | commitdiff | raw | diff to current |
2020-01-15 |
Elie Tournier | nir/algebraic: i2f(f2i()) -> trunc() |
blob | commitdiff | raw | diff to current |
2020-01-14 |
Rhys Perry | nir/algebraic: a & ~(a >> 31) -> imax(a, 0) |
blob | commitdiff | raw | diff to current |
2019-12-17 |
Jonathan Marek | nir: add option to lower half packing opcodes |
blob | commitdiff | raw | diff to current |
2019-12-03 |
Ian Romanick | nir/algebraic: Rearrange bcsel sequences generated... |
blob | commitdiff | raw | diff to current |
2019-12-03 |
Ian Romanick | nir/algebraic: Simplify some Inf and NaN avoidance... |
blob | commitdiff | raw | diff to current |
2019-11-04 |
Ian Romanick | nir/algebraic: Mark other comparison exact when removin... |
blob | commitdiff | raw | diff to current |
2019-11-04 |
Ian Romanick | nir/algebraic: Add the ability to mark a replacement... |
blob | commitdiff | raw | diff to current |
2019-10-18 |
Rob Clark | nir: add nir_lower_amul pass |
blob | commitdiff | raw | diff to current |
2019-10-18 |
Rob Clark | nir: add address calc related opt rules |
blob | commitdiff | raw | diff to current |
2019-10-18 |
Rob Clark | nir: add amul instruction |
blob | commitdiff | raw | diff to current |
2019-09-30 |
Daniel Schürmann | nir: Remove unnecessary subtraction optimizations |
blob | commitdiff | raw | diff to current |
2019-09-30 |
Daniel Schürmann | nir: recombine nir_op_*sub when lower_sub = false |
blob | commitdiff | raw | diff to current |
2019-09-19 |
Ian Romanick | nir/algebraic: Additional D3D Boolean optimization |
blob | commitdiff | raw | diff to current |
2019-09-19 |
Ian Romanick | nir/algebraic: Do not apply late DPH optimization in... |
blob | commitdiff | raw | diff to current |
2019-09-17 |
Andres Gomez | nir/algebraic: mark float optimizations returning one... |
blob | commitdiff | raw | diff to current |
2019-08-29 |
Ian Romanick | nir/algebraic: Clean up value range analysis-based... |
blob | commitdiff | raw | diff to current |
2019-08-29 |
Ian Romanick | nir/algebraic: Mark some value range analysis-based... |
blob | commitdiff | raw | diff to current |
2019-08-28 |
Ian Romanick | nir/algrbraic: Don't optimize open-coded bitfield rever... |
blob | commitdiff | raw | diff to current |
2019-08-21 |
Daniel Schürmann | nir/algebraic: some subtraction optimizations |
blob | commitdiff | raw | diff to current |
2019-08-16 |
Rhys Perry | nir/algebraic: add a few masking-before-unpack optimiza... |
blob | commitdiff | raw | diff to current |
2019-08-14 |
Ian Romanick | nir/algebraic: Reassociate shift-by-constant of shift... |
blob | commitdiff | raw | diff to current |
2019-08-14 |
Ian Romanick | nir/algebraic: Reassociate add-and-shift to be shift... |
blob | commitdiff | raw | diff to current |
2019-08-06 |
Ian Romanick | nir/algebraic: Use value range analysis to eliminate... |
blob | commitdiff | raw | diff to current |
2019-08-06 |
Ian Romanick | nir/algebraic: Use value range analysis to convert... |
blob | commitdiff | raw | diff to current |
2019-08-06 |
Ian Romanick | nir/algebraic: Use value range analysis to eliminate... |
blob | commitdiff | raw | diff to current |
2019-08-06 |
Ian Romanick | nir/algebraic: Use value range analysis to eliminate... |
blob | commitdiff | raw | diff to current |
2019-08-06 |
Ian Romanick | nir/algebraic: Simplify some comparisons like a+constan... |
blob | commitdiff | raw | diff to current |
2019-08-06 |
Ian Romanick | nir/algebraic: Recognize (a < 0 || 0 < b) as min(a... |
blob | commitdiff | raw | diff to current |
2019-08-06 |
Ian Romanick | nir/algebraic: Replace checks that a value is between... |
blob | commitdiff | raw | diff to current |
2019-07-31 |
Erico Nunes | nir/algebraic: rename lower_bitshift to lower_bitops |
blob | commitdiff | raw | diff to current |
2019-07-31 |
Erico Nunes | nir/algebraic: add new fsum ops and fdot lowering |
blob | commitdiff | raw | diff to current |
2019-07-24 |
Jonathan Marek | nir/algebraic: add scmp algebraic optimizations |
blob | commitdiff | raw | diff to current |
2019-07-24 |
Jonathan Marek | nir/algebraic: add option to lower fall_equalN/fany_nequalN |
blob | commitdiff | raw | diff to current |
2019-07-24 |
Jonathan Marek | nir/algebraic: add fdot2 optimizations |
blob | commitdiff | raw | diff to current |
2019-07-24 |
Jonathan Marek | nir/algebraic: add option to lower fdph |
blob | commitdiff | raw | diff to current |
2019-07-24 |
Jonathan Marek | nir: replace lower_sincos with algebraic opt |
blob | commitdiff | raw | diff to current |
2019-07-24 |
Jonathan Marek | nir/algebraic: allow swizzle in nir_algebraic replace... |
blob | commitdiff | raw | diff to current |
2019-07-19 |
Rhys Perry | nir/algebraic: mark a few comparison simplifications... |
blob | commitdiff | raw | diff to current |
2019-07-19 |
Rhys Perry | nir/algebraic: optimize contradictory iand operands |
blob | commitdiff | raw | diff to current |
2019-07-17 |
Jason Ekstrand | nir/algebraic: Optimize comparisons and up-casts |
blob | commitdiff | raw | diff to current |
2019-07-17 |
Jason Ekstrand | nir/algebraic: Optimize comparing unpacked values |
blob | commitdiff | raw | diff to current |
2019-07-11 |
Ian Romanick | nir/algebraic: Recognize open-coded flrp(a, b, a) |
blob | commitdiff | raw | diff to current |
2019-07-11 |
Ian Romanick | nir/algebraic: Rearrange 1-((1-a) * (1-b)) into flrp... |
blob | commitdiff | raw | diff to current |
2019-07-11 |
Ian Romanick | nir/algebraic: Reassociate fadd into fmul in DPH-like... |
blob | commitdiff | raw | diff to current |
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 |
next |