Commits · f20ace6f333fa56af1879f7480a0e7979201c374 · Roger Ferrer / llvm-epi · GitLab

May 27, 2020
- [mlir][shape] Add `shape.get_extent`. · cf42b704
  Sean Silva authored May 21, 2020
  cf42b704
May 26, 2020
- [mlir] Hotfix - Add inline to avoid multiple symbols on trivial functions · c990bdf7
  Nicolas Vasilache authored May 26, 2020
  
  c990bdf7
- [mlir] Hotfix - Drop spurious constexpr that breaks build · e9003207
  Nicolas Vasilache authored May 26, 2020
  
  e9003207
- [mlir][Vector] Add more vector.contract -> outerproduct lowerings and fix... · ba10daa8
  Nicolas Vasilache authored May 26, 2020
  ba10daa8
- [MLIR] Helper class referencing MemRefType to unify runner implementations. · 222e0e58
  Christian Sigg authored May 25, 2020
  222e0e58
- [mlir][Vector] Add vector contraction to outerproduct lowering · 9578a54f
  Nicolas Vasilache authored May 26, 2020
  9578a54f
- Update DialectConversion.md · a3b5ccdd
  Tharindu Rusira authored May 26, 2020
  a3b5ccdd
May 25, 2020
- Make mlir::Value's bool conversion operator explicit · a9b5edc5
  Benjamin Kramer authored May 24, 2020
  a9b5edc5
- Added pow intrinsic to LLVMIR dialect · 7293dd5b
  George Mitenkov authored May 25, 2020
  7293dd5b
- [mlir] Expand operand adapter to take attributes · 4b8632e1
  Jacques Pienaar authored May 24, 2020
  4b8632e1
May 21, 2020
- [NFC] Make assertion more informative. · be88ba09
  Sean Silva authored May 21, 2020
  be88ba09
- [mlir][spirv] Enable composite instructions for cooperative matrix type. · 0712eac7
  Thomas Raoux authored May 21, 2020
  0712eac7
- [mlir][spirv] Add remaining cooperative matrix instructions · 15389cdc
  Thomas Raoux authored May 21, 2020
  15389cdc
- [mlir][rocdl] Exposing buffer load/store intrinsic · 9c53ac08
  jerryyin authored May 19, 2020
  9c53ac08
- [mlir][gpu] Refactor ConvertGpuLaunchFuncToCudaCalls pass. · 2cbbc266
  Wen-Heng (Jack) Chung authored May 18, 2020
  2cbbc266
- [mlir] NFC - Add a builder to vector.transpose · 941005f5
  Nicolas Vasilache authored May 21, 2020
  941005f5
- Revert "[mlir][gpu] Refactor ConvertGpuLaunchFuncToCudaCalls pass." · 5c3ebd77
  Mehdi Amini authored May 21, 2020
  5c3ebd77
May 20, 2020
- [mlir] NFC - Appease GCC 5 again.. · 3393cc4c
  Nicolas Vasilache authored May 20, 2020
  
  3393cc4c
- [mlir][gpu] Refactor functions for workgroup and private buffer attributions. · ad398164
  Wen-Heng (Jack) Chung authored May 06, 2020
  ad398164
- [mlir][gpu] Refactor ConvertGpuLaunchFuncToCudaCalls pass. · cdb6f05e
  Wen-Heng (Jack) Chung authored May 18, 2020
  cdb6f05e
- [mlir] NFC - Appease GCC 5 again.. · ebf14d9b
  Nicolas Vasilache authored May 20, 2020
  
  ebf14d9b
- [mlir][spirv] Adapt subview legalization to the updated op semantics. · 0e88eb5c
  MaheshRavishankar authored May 20, 2020
  0e88eb5c
- [mlir][Linalg] Add producer-consumer fusion when producer is a ConstantOp · 071358e0
  MaheshRavishankar authored May 20, 2020
  071358e0
- [mlir][Vector] Add option to fully unroll for VectorTransfer to SCF lowering · 7c3c5b11
  Nicolas Vasilache authored May 20, 2020
  7c3c5b11
- [mlir] ensureRegionTerminator: take OpBuilder · 3ccf4a5b
  Alex Zinenko authored May 20, 2020
  3ccf4a5b
- [mlir] SCFToStandard: support any ops in and around the control flow ops · a7d88a90
  Alex Zinenko authored May 20, 2020
  a7d88a90
- [mlir] Erase or clear blocks through ConversionPatternRewriter when applicable · 57cbeaa8
  Alex Zinenko authored May 20, 2020
  57cbeaa8
- [mlir] DialectConversion: support erasing blocks · df48026b
  Alex Zinenko authored May 20, 2020
  df48026b
- [mlir] DialectConversion: avoid double-free when rolling back op creation · 5d5df06a
  Alex Zinenko authored May 20, 2020
  5d5df06a
- [mlir] Toy tutorial: insert terminators at the end of the loop during rewrite · a655144f
  Alex Zinenko authored May 20, 2020
  a655144f
- [mlir][Linalg] NFC - Simplify GenericNestLoop builder · 19e5b2bc
  Nicolas Vasilache authored May 20, 2020
  19e5b2bc
- [mlir][Linalg] Refactor linalg tiling · 004a3d4f
  Nicolas Vasilache authored May 20, 2020
  004a3d4f
- [mlir] NFC: rename tests related to SCF dialect from Loops to SCF · eab4a199
  Alex Zinenko authored May 19, 2020
  eab4a199
- [mlir] Add custom assembly formats to shape.witness ops. · 02035580
  Tres Popp authored May 19, 2020
  02035580
- [mlir] Custom printing/parsing for Shape::AssumingOp · fb6986ef
  Tres Popp authored May 19, 2020
  fb6986ef
- [mlir] Mark witness related Shape dialect ops as NoSideEffect. · 44226c1f
  Tres Popp authored May 19, 2020
  44226c1f
- [mlir][SystemZ] Fix incompatible datalayout in SystemZ · 9f2ce5b9
  Haruki Imai authored May 20, 2020
  9f2ce5b9
- Mark AffineMap::replaceDimsAndSymbols as const (NFC) · 78453e37
  Chintan Kaur authored May 20, 2020
  78453e37
- [mlir] NFC - Appease gcc 5 · da95a0d8
  Nicolas Vasilache authored May 19, 2020
  da95a0d8
- [mlir][spirv] First step to support spirv cooperative matrix extension. · b359bbaa
  Thomas Raoux authored May 19, 2020
  b359bbaa