Commits · d572cd1b067f1177a981a4711bf2e501eaa8117b · Roger Ferrer / llvm-epi · GitLab

May 02, 2023
- Introduce MLIR Op Properties · d572cd1b
  Mehdi Amini authored Feb 26, 2023
  d572cd1b
May 01, 2023
- [mlir][linalg] Relax masked_vectorize(pad) to allow zero low Values. · b2fdb141
  Hanhan Wang authored Apr 28, 2023
  b2fdb141
- [mlir][arith] Add narrowing patterns to commute more vector ops · 7f3b0e58
  Jakub Kuderski authored May 01, 2023
  7f3b0e58
- [mlir][arith] Add narrowing patterns for other insertion ops · 3ff87088
  Jakub Kuderski authored May 01, 2023
  3ff87088
- [mlir][sparse] extend unpack operation to support unpacking a batched COO type · d4db5289
  Peiming Liu authored Apr 21, 2023
  d4db5289
Apr 29, 2023
- Simplify affine bound min or max with operands · 43735204
  Uday Bondhugula authored Apr 29, 2023
  43735204
Apr 28, 2023
- [mlir][arith] Add narrowing pattern to commute extension over insertion · 33017e5a
  Jakub Kuderski authored Apr 28, 2023
  33017e5a
- [mlir][arith] Add patterns to commute extension over vector extraction · 09043a26
  Jakub Kuderski authored Apr 28, 2023
  09043a26
- [mlir][spirv] Fix nullptr dereference in UnifyAliasedResource · 797594a0
  Jakub Kuderski authored Apr 28, 2023
  797594a0
- [mlir][tensor] Add transform to make tensor.pad loop-independent · 77124386
  Matthias Springer authored Apr 28, 2023
  77124386
Apr 27, 2023
- [MLIR][EmitC] Add empty emitc.constant check · eb22971b
  Kiung Jung authored Apr 27, 2023
  eb22971b
- Disallow to apply the op & to const · 2aa1818a
  Kiung Jung authored Apr 27, 2023
  2aa1818a
- [mlir][arith] Add missing canon pattern `trunci(ext*i(x)) -> ext*i(x)` · ab85aec1
  Jakub Kuderski authored Apr 27, 2023
  ab85aec1
- [MLIR][Linalg] Change destination logic in `bubbleUpPackOpThroughGenericOp`. · 61be9358
  Lorenzo Chelini authored Apr 26, 2023
  61be9358
- [mlir][GPUDialect] Implement memory attributions for LaunchOp · 54e96f4f
  Fabian Mora authored Apr 26, 2023
  54e96f4f
Apr 26, 2023
- [mlir][arith] Add initial integer bitwidth narrowing pass · da0730b9
  Jakub Kuderski authored Apr 25, 2023
  da0730b9
- [mlir][vector] Add pattern to break down vector.bitcast · 650f04fe
  Quinn Dawkins authored Apr 06, 2023
  650f04fe
Apr 25, 2023
- [mlir] detensorize: don't accidentally convert function entry blocks · 65eedceb
  Alex Zinenko authored Apr 22, 2023
  65eedceb
- [mlir][tosa] Fix `tosa.reshape` folder for quantized constants · 09bd5ae4
  Robert Suderman authored Apr 25, 2023
  09bd5ae4
Apr 24, 2023
- [mlir][vector] Add unroll pattern for vector.gather · 435f7d4c
  Quinn Dawkins authored Apr 05, 2023
  435f7d4c
- [mlir][spirv] Fix OOB error in TestModuleCombiner · 4f4cd963
  Jakub Kuderski authored Apr 24, 2023
  4f4cd963
- [mlir][linalg] Add support for lowering pack with outer_dims_perm. · 6f87b50b
  Hanhan Wang authored Apr 21, 2023
  6f87b50b
Apr 23, 2023
- [mlir][Vector] Add 16x16 strategy to vector.transpose lowering. · 8d163e50
  Hanhan Wang authored Apr 20, 2023
  8d163e50
- [mlir][irdl] Add IRDL registration · 8ac8c922
  Mathieu Fehr authored Mar 10, 2023
  8ac8c922
Apr 22, 2023
- Revert "Revert "Fix handling of special and large vals in expand pattern for... · 44baa655
  Ramiro Leal-Cavazos authored Apr 22, 2023
  44baa655
- [mlir] tosa.reshape - Add InferTensorType interface · 2dd396c1
  Aviad Cohen authored Apr 17, 2023
  2dd396c1
Apr 21, 2023
- [MLIR][Vector] Enable masked vectorizaton of contraction ops · c8557c7c
  Diego Caballero authored Apr 21, 2023
  c8557c7c
- [mlir][sparse][gpu] generate proper memcpy in/out host and device · 86888e42
  Aart Bik authored Apr 18, 2023
  86888e42
- [mlir][linalg] Refine `tensor.extract` vectorisation · a3ae3931
  Andrzej Warzynski authored Apr 12, 2023
  a3ae3931
- Revert "Fix handling of special and large vals in expand pattern for `round`"... · 87cef78f
  Mehdi Amini authored Apr 21, 2023
  87cef78f
Apr 20, 2023
- [mlir][openacc] Restrict types for acc.bounds operands to index or integer type · ed35b584
  Valentin Clement authored Apr 20, 2023
  ed35b584
- [mlir][sparse] fix crash when generating coiteration loop with compressed-hi DLT. · a7cfcc68
  Peiming Liu authored Apr 20, 2023
  a7cfcc68
- use heap memory for position buffer allocated for PackOp. · fd2211d8
  Peiming Liu authored Apr 20, 2023
  fd2211d8
- Add pattern that expands `math.roundeven` into `math.round` + arith · 8d2bae9a
  Ramiro Leal-Cavazos authored Apr 20, 2023
  8d2bae9a
- Fix handling of special and large vals in expand pattern for `round` · ab2fc952
  Ramiro Leal-Cavazos authored Apr 20, 2023
  ab2fc952
- Revert "[mlir][irdl] Add IRDL registration" · 5115ede6
  Arjun P authored Apr 20, 2023
  5115ede6
- [mlir][irdl] Add IRDL registration · e0d884de
  Mathieu Fehr authored Apr 20, 2023
  e0d884de
- [mlir][sparse] extend pack operation to support packing a batched COO type · 7864d736
  Peiming Liu authored Apr 18, 2023
  7864d736
- [mlir][Memref] Fold nvgpu device cp.async on src memref to dst memref · fc5c1a76
  Manish Gupta authored Apr 12, 2023
  fc5c1a76
- [mlir][sparse] support iteration over compressed-hi dimension level in loop emitter · abd66d91
  Peiming Liu authored Apr 18, 2023
  abd66d91