| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | nir: Add lowering for find_lsb. | Eric Anholt | 2018-06-06 | 1 | -0/+4 |
| * | nir: Add lowering for ifind_msb to ufind_msb. | Eric Anholt | 2018-06-06 | 1 | -0/+4 |
| * | nir: Add lowering from ibitfield_extract/ubitfield_extract to shifts. | Eric Anholt | 2018-06-06 | 1 | -0/+16 |
| * | nir: Add lowering for bitfieldInsert without using bfi. | Eric Anholt | 2018-06-06 | 1 | -0/+14 |
| * | nir: Lower !f2b(x) to x == 0.0 | Ian Romanick | 2018-06-01 | 1 | -0/+1 |
| * | nir: Add some missing "optimization undo" patterns | Ian Romanick | 2018-06-01 | 1 | -1/+19 |
| * | nir: optimize iand(ieq(a, 0), ieq(b, 0)) to ieq(ior(a, b), 0) | Samuel Pitoiset | 2018-05-31 | 1 | -0/+2 |
| * | nir: add unsigned comparison simplifications | Timothy Arceri | 2018-05-30 | 1 | -0/+2 |
| * | nir: Implement optional b2f->iand lowering | Alyssa Rosenzweig | 2018-05-18 | 1 | -1/+4 |
| * | nir: Don't condition 'a-b < 0' -> 'a < b' on is_not_used_by_conditional | Ian Romanick | 2018-03-26 | 1 | -3/+1 |
| * | nir: Don't compare b2f or b2i with zero | Ian Romanick | 2018-03-19 | 1 | -0/+5 |
| * | nir: Don't i2b a value that is already Boolean | Ian Romanick | 2018-03-08 | 1 | -0/+1 |
| * | nir: Narrow some dot product operations | Ian Romanick | 2018-03-08 | 1 | -0/+8 |
| * | nir: Simplify some comparisons like a+b < a | Ian Romanick | 2018-03-06 | 1 | -0/+9 |
| * | nir: Use De Morgan's Law on logic compounded comparisons | Ian Romanick | 2018-03-06 | 1 | -0/+9 |
| * | nir: Replace fmin(b2f(a), b) with a bcsel | Ian Romanick | 2018-03-06 | 1 | -0/+9 |
| * | nir: Pull b2f out of bcsel | Ian Romanick | 2018-03-06 | 1 | -0/+1 |
| * | nir: Replace an odd comparison involving fmin of -b2f | Ian Romanick | 2018-03-06 | 1 | -0/+13 |
| * | nir: Mark bcsel-to-fmin (or fmax) transformations as inexact | Ian Romanick | 2018-03-06 | 1 | -2/+2 |
| * | nir: Recognize some more open-coded fmin / fmax | Ian Romanick | 2018-03-06 | 1 | -0/+2 |
| * | nir: add lower_ldexp to nir compiler options | Timothy Arceri | 2018-02-28 | 1 | -2/+2 |
| * | nir: lower fexp2(fmul(flog2(a), 2)) to fmul(a, a) | Samuel Pitoiset | 2018-02-22 | 1 | -0/+2 |
| * | nir: add is_used_once for fmul(fexp2(a), fexp2(b)) to fexp2(fadd(a, b)) | Samuel Pitoiset | 2018-02-22 | 1 | -1/+1 |
| * | nir: Distribute binary operations with constants into bcsel | Ian Romanick | 2018-01-30 | 1 | -0/+15 |
| * | nir: Rearrange logic op-compounded integer compares | Ian Romanick | 2018-01-30 | 1 | -0/+35 |
| * | nir: Rearrange and-compounded float compares | Ian Romanick | 2018-01-30 | 1 | -0/+8 |
| * | nir: Separate a weird compare with zero to two compares with zero | Ian Romanick | 2018-01-30 | 1 | -0/+2 |
| * | nir: Simplify min and max of b2f | Ian Romanick | 2018-01-30 | 1 | -0/+5 |
| * | nir: Undo possible damage caused by rearranging or-compounded float compares | Ian Romanick | 2018-01-30 | 1 | -0/+9 |
| * | nir: Be more conservative about rearranging or-compounded compares | Ian Romanick | 2018-01-30 | 1 | -4/+9 |
| * | nir: See through an fneg to apply existing optimizations | Ian Romanick | 2018-01-30 | 1 | -3/+6 |
| * | nir: fix algebraic optimizations | Connor Abbott | 2017-08-01 | 1 | -2/+2 |
| * | nir: Optimize find_lsb/imsb/umsb error checks | Matt Turner | 2017-07-20 | 1 | -0/+11 |
| * | nir/i965: add before ffma algebraic opts | Timothy Arceri | 2017-04-24 | 1 | -0/+23 |
| * | nir: shuffle constants to the top | Timothy Arceri | 2017-04-24 | 1 | -1/+7 |
| * | nir: add flt comparision simplification | Timothy Arceri | 2017-04-24 | 1 | -0/+4 |
| * | nir: Rework conversion opcodes | Jason Ekstrand | 2017-03-14 | 1 | -20/+20 |
| * | nir: remove shebang from python scripts | Emil Velikov | 2017-03-10 | 1 | -1/+0 |
| * | nir/algebraic: Optimize 64bit pack/unpack | Jason Ekstrand | 2017-02-16 | 1 | -0/+6 |
| * | nir: Combine the int and double [un]pack opcodes | Jason Ekstrand | 2017-02-16 | 1 | -1/+1 |
| * | nir: Shift count for shift opcodes is always 32-bits | Ian Romanick | 2017-01-20 | 1 | -4/+4 |
| * | nir: add min/max optimisation | Elie TOURNIER | 2017-01-19 | 1 | -0/+12 |
| * | nir: optimise min/max fadd combos | Timothy Arceri | 2017-01-14 | 1 | -0/+4 |
| * | nir: don't turn ieq/ine into inot if used by an if | Timothy Arceri | 2017-01-12 | 1 | -2/+2 |
| * | nir: add late opt to turn inot/b2f combos back to bcsel | Timothy Arceri | 2017-01-12 | 1 | -0/+3 |
| * | nir: add imprecise flrp optimisation | Timothy Arceri | 2017-01-12 | 1 | -0/+1 |
| * | nir: add another comparison simplification | Timothy Arceri | 2017-01-09 | 1 | -0/+2 |
| * | nir: Turn bcsel of +/- 1.0 and 0.0 into b2f sequences. | Kenneth Graunke | 2017-01-09 | 1 | -0/+4 |
| * | nir: Convert ineg(b2i(a)) to a if it's a boolean. | Kenneth Graunke | 2017-01-09 | 1 | -0/+2 |
| * | nir/algebraic: Add optimizations for "a == a && a CMP b" | Jason Ekstrand | 2016-12-22 | 1 | -0/+8 |
