| Commit message (Expand) | Author | Age | Files | Lines |
* | nir: Accept nir_var_mem_global in derefs used by phis | Caio Marcelo de Oliveira Filho | 2019-05-30 | 1 | -1/+2 |
* | nir: silence three compiler warnings seen with MinGW | Brian Paul | 2019-05-29 | 2 | -4/+2 |
* | nir: Allow derefs to be used as phi sources | Caio Marcelo de Oliveira Filho | 2019-05-29 | 3 | -2/+17 |
* | nir/lower_non_uniform: safely iterate over blocks | Lionel Landwerlin | 2019-05-28 | 1 | -1/+1 |
* | st/nir: Re-vectorize shader IO | Kenneth Graunke | 2019-05-28 | 1 | -0/+6 |
* | nir: Drop imov/fmov in favor of one mov instruction | Jason Ekstrand | 2019-05-24 | 22 | -50/+38 |
* | nir/builder: Merge nir_[if]mov_alu into one nir_mov_alu helper | Jason Ekstrand | 2019-05-24 | 4 | -30/+12 |
* | nir/lower_to_source_mods: Stop turning add, sat, and neg into mov | Jason Ekstrand | 2019-05-24 | 2 | -40/+19 |
* | nir/source_mods: Add a helpers for setting source modifiers | Jason Ekstrand | 2019-05-24 | 1 | -6/+18 |
* | nir/builder: Remove the use_fmov parameter from nir_swizzle | Jason Ekstrand | 2019-05-24 | 5 | -19/+18 |
* | nir/test: add split vars tests (v2) | Dave Airlie | 2019-05-21 | 1 | -1/+419 |
* | nir: Fix clone of nir_variable state slots | Caio Marcelo de Oliveira Filho | 2019-05-20 | 1 | -3/+5 |
* | nir: Add nir_address_format_null_value() | Caio Marcelo de Oliveira Filho | 2019-05-20 | 2 | -0/+22 |
* | nir: Add nir_address_format_32bit_offset | Caio Marcelo de Oliveira Filho | 2019-05-20 | 2 | -0/+9 |
* | nir: Add nir_address_format_logical | Caio Marcelo de Oliveira Filho | 2019-05-20 | 2 | -0/+14 |
* | nir/validate: fix crash if entry is null. | Dave Airlie | 2019-05-20 | 1 | -1/+1 |
* | nir: Only convert SSA values to regs when needed | Caio Marcelo de Oliveira Filho | 2019-05-16 | 1 | -6/+22 |
* | nir: Fix nir_opt_idiv_const when negatives are involved | Caio Marcelo de Oliveira Filho | 2019-05-16 | 1 | -3/+5 |
* | nir: lower_non_uniform_access: iterate over instructions safely | Lionel Landwerlin | 2019-05-16 | 1 | -1/+1 |
* | nir/algebraic: Remove problematic "optimization" | Alyssa Rosenzweig | 2019-05-16 | 1 | -3/+0 |
* | compiler: Add lowering support for 64-bit saturate operations to software | Anuj Phogat | 2019-05-15 | 1 | -0/+3 |
* | nir: fix lower_non_uniform_access pass | Lionel Landwerlin | 2019-05-15 | 1 | -0/+1 |
* | nir/algebraic: Commute 1-fsat(a) to fsat(1-a) for all non-fmul instructions | Ian Romanick | 2019-05-14 | 2 | -5/+24 |
* | nir/algebraic: Eliminate useless fsat() on operand of comparison w/value in (... | Ian Romanick | 2019-05-14 | 2 | -0/+57 |
* | nir/algebraic: Strip double negatives from comparison sources | Ian Romanick | 2019-05-14 | 1 | -0/+28 |
* | intel/compiler: Repeat nir_opt_algebraic_late | Ian Romanick | 2019-05-14 | 1 | -0/+4 |
* | Revert "nir: add late opt to turn inot/b2f combos back to bcsel" | Ian Romanick | 2019-05-14 | 2 | -19/+0 |
* | nir/algebraic: Eliminate a tautological compare | Ian Romanick | 2019-05-14 | 1 | -0/+7 |
* | nir/algebraic: Simplify fsqrt domain guard | Ian Romanick | 2019-05-14 | 1 | -0/+1 |
* | nir/search: Don't compare 8-bit or 1-bit constants with floats | Ian Romanick | 2019-05-14 | 1 | -1/+12 |
* | nir/algebraic: Recognize open-coded fsat with modifiers | Ian Romanick | 2019-05-14 | 1 | -0/+5 |
* | nir/algebraic: Push unary operations into source operands of fsat source | Ian Romanick | 2019-05-14 | 2 | -1/+23 |
* | nir/algebraic: Recognize open-coded flrp(a, b, fsat(c)) | Ian Romanick | 2019-05-14 | 1 | -3/+12 |
* | nir/algebraic: Add missing ffma(-1, a, b) pattern | Ian Romanick | 2019-05-14 | 1 | -0/+1 |
* | nir: Mark ffma as 2src_commutative | Ian Romanick | 2019-05-14 | 2 | -17/+15 |
* | nir: Add support for 2src_commutative ops that have 3 sources | Ian Romanick | 2019-05-14 | 3 | -9/+24 |
* | nir: Rename commutative to 2src_commutative | Ian Romanick | 2019-05-14 | 4 | -48/+60 |
* | nir/validate: Use a single set for SSA def validation | Jason Ekstrand | 2019-05-13 | 1 | -78/+50 |
* | nir/validate: Use a ralloc context for our temporary data | Jason Ekstrand | 2019-05-13 | 1 | -16/+12 |
* | nir: Fix wrong sign in lower_rcp | Ruslan Kabatsayev | 2019-05-11 | 1 | -2/+2 |
* | nir: Add blend_const_color_rgba sysval | Alyssa Rosenzweig | 2019-05-10 | 1 | -1/+4 |
* | nir: allow specifying a set of opcodes in lower_alu_to_scalar | Jonathan Marek | 2019-05-10 | 2 | -6/+11 |
* | nir/flrp: Reassociate add in flrp(±1, b, c) lowering path | Ian Romanick | 2019-05-08 | 1 | -4/+4 |
* | nir/flrp: Fix typo on the flrp(±1, b, c) path | Ian Romanick | 2019-05-08 | 1 | -1/+3 |
* | nir: implement lowering for fsin and fcos | Vasily Khoruzhick | 2019-05-07 | 3 | -0/+141 |
* | nir/algebraic: Reassociate open-coded flrp(1, b, c) | Ian Romanick | 2019-05-06 | 1 | -0/+3 |
* | nir/flrp: Lower flrp(a, b, #c) differently | Ian Romanick | 2019-05-06 | 1 | -0/+17 |
* | nir/flrp: Lower flrp(a, b, c) differently if another flrp(_, b, c) exists | Ian Romanick | 2019-05-06 | 1 | -1/+56 |
* | nir/flrp: Lower flrp(a, b, c) differently if another flrp(a, _, c) exists | Ian Romanick | 2019-05-06 | 1 | -0/+89 |
* | nir/flrp: Lower flrp(±1, b, c) and flrp(a, ±1, c) differently | Ian Romanick | 2019-05-06 | 1 | -0/+134 |