| Commit message (Expand) | Author | Age | Files | Lines |
* | nir/algebraic: support for power-of-two optimizations | Rob Clark | 2016-06-03 | 1 | -3/+12 |
* | nir/algebraic: Separate ffma lowering from fusing | Jason Ekstrand | 2016-05-11 | 1 | -1/+1 |
* | nir: Separate 32 and 64-bit fmod lowering | Samuel Iglesias Gonsálvez | 2016-05-04 | 1 | -2/+3 |
* | nir/algebraic: Support lowering for both 64 and 32-bit ldexp | Jason Ekstrand | 2016-04-28 | 1 | -9/+22 |
* | nir/opcodes: Make ldexp take an explicitly 32-bit int | Jason Ekstrand | 2016-04-28 | 1 | -1/+1 |
* | nir: Add lrp lowering for doubles in opt_algebraic | Samuel Iglesias Gonsálvez | 2016-04-28 | 1 | -3/+6 |
* | nir: rename lower_flrp to lower_flrp32 | Samuel Iglesias Gonsálvez | 2016-04-28 | 1 | -6/+6 |
* | nir/opt_algebraic: Fix some expressions with ambiguous bit sizes | Jason Ekstrand | 2016-04-27 | 1 | -3/+3 |
* | nir/search: Respect the bit_size parameter on nir_search_value | Jason Ekstrand | 2016-04-27 | 1 | -1/+4 |
* | nir/algebraic: Add a mechanism for specifying the bit size of a value | Jason Ekstrand | 2016-04-27 | 1 | -0/+4 |
* | nir/algebraic: Add lowering for ldexp | Jason Ekstrand | 2016-04-13 | 1 | -0/+31 |
* | nir: Add more modulus opcodes | Jason Ekstrand | 2016-04-13 | 1 | -0/+1 |
* | nir: Merge redudant integer clamping. | Markus Wick | 2016-04-11 | 1 | -1/+4 |
* | nir: Do basic constant reassociation. | Kenneth Graunke | 2016-04-11 | 1 | -0/+8 |
* | nir: Simplify a bcsel to logical-or | Ian Romanick | 2016-03-31 | 1 | -0/+1 |
* | nir: Fix typo from commit 6702f1acde9. | Matt Turner | 2016-03-30 | 1 | -1/+1 |
* | nir: Propagate negates up multiplication chains. | Matt Turner | 2016-03-30 | 1 | -0/+4 |
* | nir/algebraic: Flag inexact optimizations | Jason Ekstrand | 2016-03-23 | 1 | -59/+62 |
* | nir/algebraic: Fix fmin detection to match the spec | Jason Ekstrand | 2016-03-23 | 1 | -1/+1 |
* | nir/algebraic: Get rid of an invlid fxor optimization | Jason Ekstrand | 2016-03-23 | 1 | -1/+0 |
* | nir/algebraic: Allow for flagging operations as being inexact | Jason Ekstrand | 2016-03-23 | 1 | -1/+8 |
* | nir: Don't abs slt and friends | Ian Romanick | 2016-03-22 | 1 | -0/+4 |
* | nir: Don't abs the result of b2f or b2i | Ian Romanick | 2016-03-22 | 1 | -0/+2 |
* | nir: Simplify 0 < fabs(a) | Ian Romanick | 2016-03-22 | 1 | -0/+6 |
* | nir: Simplify 0 >= b2f(a) | Ian Romanick | 2016-03-22 | 1 | -0/+7 |
* | nir: Simplify i2b with negated or abs operand | Ian Romanick | 2016-03-22 | 1 | -0/+2 |
* | nir: Lower flrp with Boolean interpolator to bcsel | Ian Romanick | 2016-03-22 | 1 | -2/+5 |
* | nir: Recognize open-coded extract_u16. | Matt Turner | 2016-03-04 | 1 | -0/+5 |
* | nir: Recognize open-coded extract_u8. | Matt Turner | 2016-03-04 | 1 | -0/+7 |
* | nir: Recognize open-coded bitfield_reverse. | Matt Turner | 2016-02-08 | 1 | -0/+13 |
* | nir: Recognize product of open-coded pow()s. | Matt Turner | 2016-02-08 | 1 | -0/+2 |
* | nir: Add opt_algebraic rules for xor with zero. | Matt Turner | 2016-02-08 | 1 | -0/+2 |
* | nir: Add lowering support for unpacking opcodes. | Matt Turner | 2016-02-01 | 1 | -0/+28 |
* | nir: Add lowering support for packing opcodes. | Matt Turner | 2016-02-01 | 1 | -0/+20 |
* | nir: Add opcodes to extract bytes or words. | Matt Turner | 2016-02-01 | 1 | -0/+16 |
* | nir: move to compiler/ | Emil Velikov | 2016-01-26 | 1 | -0/+285 |