Integrate LLVM at llvm/llvm-project@0e779ad4998e (#2713)

Author: sdasgup3

BUILD.bazel

@@ -519,8 +519,6 @@ cc_library(
         "stablehlo/conversions/linalg/transforms/StablehloToLinalgRandom.cpp",
         "stablehlo/conversions/linalg/transforms/StablehloToLinalgReduce.cpp",
         "stablehlo/conversions/linalg/transforms/TypeConversion.cpp",
-        "stablehlo/transforms/conversions/TypeConversion.cpp",
-        "stablehlo/transforms/conversions/TypeConversion.h",
     ],
     hdrs = [
         "stablehlo/conversions/linalg/transforms/LegalizeToLinalgUtils.h",
@@ -531,6 +529,7 @@ cc_library(
     ],
     strip_include_prefix = ".",
     deps = [
+        "stablehlo_type_conversions",
         ":base",
         ":chlo_ops",
         ":linalg_pass_inc_gen",
@@ -995,7 +994,7 @@ cc_library(
         ":linalg_passes",
         ":reference_api",
         ":reference_configuration",
-        ":stablehlo_dialect_capi",
+        ":stablehlo_dialect_capi_objects",
         ":stablehlo_ops",
         ":stablehlo_passes",
         ":stablehlo_portable_api",

WORKSPACE.bazel

@@ -17,9 +17,9 @@ workspace(name = "stablehlo")
 
 load("@bazel_tools//tools/build_defs/repo:http.bzl", "http_archive")
 
-LLVM_COMMIT = "f8287f6c373fcf993643dd6f0e30dde304c1be73"
+LLVM_COMMIT = "0e779ad4998ef65907502101c5b82ede05ddfa4e"
 
-LLVM_SHA256 = "add2841174abc79c45aa309bdf0cf631aa8f97e7a4df57dcfca57c60df27527f"
+LLVM_SHA256 = "d5c2560b2d9ce3ced7951113f2b5d1ea428665678f4dcb1fb8780eb1219ca615"
 
 http_archive(
     name = "llvm-raw",

build_tools/llvm_version.txt

@@ -1 +1 @@
-f8287f6c373fcf993643dd6f0e30dde304c1be73
+0e779ad4998ef65907502101c5b82ede05ddfa4e

docs/generated/interpreter_passes.md

@@ -1,4 +1,5 @@
 <!-- Autogenerated by mlir-tblgen; don't manually edit -->
+
 ### `-interpreter-instrument-with-probe`
 
 _Inserts interpreter.probe ops in a StableHLO program._
@@ -26,6 +27,7 @@ sequence: `probe1`, `probe2`, ... See `interpreter.probe` for additional
 information on how data is serialized.
 
 #### Options
+
 ```
 -useDebugInfo : Whether or not to use location debug data as `probe_id` values.
 ```

docs/generated/stablehlo_linalg_passes.md

@@ -1,11 +1,11 @@
 <!-- Autogenerated by mlir-tblgen; don't manually edit -->
+
 ### `-stablehlo-legalize-to-linalg`
 
 _Legalize StableHLO to LinAlg_
 
-
-
 #### Options
+
 ```
 -enable-primitive-ops : Lower to primitive Linalg ops (map, reduce and transpose) when possible, instead of linalg.generic
 -enable-sparse-ops    : Lower to Sparse Tensor ops (sparse_tensor.concatenate)when possible, instead of linalg.generic

docs/generated/stablehlo_optimization_passes.md

@@ -1,14 +1,15 @@
 <!-- Autogenerated by mlir-tblgen; don't manually edit -->
+
 ### `-stablehlo-aggressive-folder`
 
 _Folds StableHLO operations_
 
-
-
 #### Options
+
 ```
 -fold-float : Allow for potentially lossy computations using float type.
 ```
+
 ### `-stablehlo-aggressive-simplification`
 
 _Canonicalizes StableHLO operations_
@@ -98,6 +99,7 @@ Performs graph simplifications, including:
 
 This list is pulled from code comments so is not fully exhaustive, but has
 high coverage of the pass today.
+
 ### `-stablehlo-target-independent-optimization`
 
 _Runs canonicalizers, folders, and other target-independent optimizations._

docs/generated/stablehlo_passes.md

@@ -1,9 +1,9 @@
 <!-- Autogenerated by mlir-tblgen; don't manually edit -->
+
 ### `-chlo-legalize-to-stablehlo`
 
 _Legalizes from CHLO ops flow to StableHLO and Shape ops_
 
-
 ### `-shape-legalize-to-stablehlo`
 
 _Legalize shape-related ops to StableHLO._
@@ -13,6 +13,7 @@ An experimental pass that legalizes shape-related ops to StableHLO ops.
 Bringing shape and data computations together via an optional pass will
 make it possible for the StableHLO ecosystem to potentially leverage the
 compilation pipelines that use StableHLO operations to model dynamism.
+
 ### `-stablehlo-canonicalize-dynamism`
 
 _Canonicalizes dynamic StableHLO ops into static ops._
@@ -30,6 +31,7 @@ these ops are actually constants.
 
   %0 = stablehlo.broadcast_in_dim %cst, dims = [] : (tensor<f32>) -> tensor<16xf32>
 ```
+
 ### `-stablehlo-compatibility-expander`
 
 _Compatibility expander for StableHLO operations._
@@ -68,9 +70,11 @@ func.func @tan_op_non_complex(%arg0: tensor<4xf64>) -> tensor<4xf64> {
 ```
 
 #### Options
+
 ```
 -target : The target version. Must be a version of the form #.#.#.
 ```
+
 ### `-stablehlo-complex-math-expander`
 
 _Expander for StableHLO complex math operations._
@@ -98,11 +102,11 @@ func.func @sqrt_op_complex(%arg0: tensor<4xcomplex<f64>>) -> tensor<4xcomplex<f6
   return %2 : tensor<4xcomplex<f64>>
 }
 ```
+
 ### `-stablehlo-convert-to-signless`
 
 _Pass to transform the IR to be on signless integers._
 
-
 ### `-stablehlo-legalize-composite-to-call`
 
 _Replaces composite ops with a call to their decomposition._
@@ -133,9 +137,11 @@ stablehlo-opt --stablehlo-legalize-composite-to-call=except='foo.baz,foo.qux'
 ```
 
 #### Options
+
 ```
 -except : Names of composites that should not be replaced with calls.
 ```
+
 ### `-stablehlo-legalize-deprecated-ops`
 
 _Legalize deprecated ops to well-supported ops._
@@ -146,9 +152,11 @@ removals in various compilation pipelines by legalizing them to their
 long-term supported counterparts.
 
 #### Options
+
 ```
 -fail-on-unused : Fail on (mostly) unused ops that are deprecated without any fallback.
 ```
+
 ### `-stablehlo-legalize-qdq-to-quantized-op`
 
 _Fuse (de-quantize, floating-point operation and quantize) pattern into StableHLO quantized operation_
@@ -177,6 +185,7 @@ func.func @add(%arg0: tensor<16x16x!quant.uniform<u8:f32, 3.400000e+01:16>>) ->
   return %2 : tensor<16x16x!quant.uniform<u8:f32, 3.400000e+01:16>>
 }
 ```
+
 ### `-stablehlo-legalize-quant-to-math`
 
 _Convert from StableHLO quantized ops to StableHLO primitive math ops._
@@ -219,6 +228,7 @@ func.func @add(%arg0: tensor<i8>, %arg1: tensor<i8>) -> tensor<i8> {
   return %13 : tensor<i8>
 }
 ```
+
 ### `-stablehlo-legalize-quantized-op-to-qdq`
 
 _Decompose quantized StableHLO operation to (de-quantize, floating-point operation and quantize) pattern._
@@ -244,11 +254,11 @@ func.func @add(%arg0: tensor<!quant.uniform<i8:f32, 1.000000e+00>>, %arg1: tenso
   return %3 : tensor<!quant.uniform<i8:f32, 3.000000e+00:2>>
 }
 ```
+
 ### `-stablehlo-legalize-to-vhlo`
 
 _Legalize StableHLO to VHLO._
 
-
 ### `-stablehlo-refine-arguments`
 
 _Refines the argument shapes of the main function._
@@ -277,9 +287,11 @@ passed to the pass create method. The refinement type list must specify the
 type of every argument to the `main` method being refined.
 
 #### Options
+
 ```
 -types : The new types to be used for the main function's arguments, specified as an MLIR TypeRange 'tensor<1x2xf32>, ...'
 ```
+
 ### `-stablehlo-refine-shapes`
 
 _Refines shapes across a StableHLO program._
@@ -325,18 +337,17 @@ Modules valid for shape refinement must have the following properties:
     returning constant values after refinement. These functions are inlined.
   * All calls to a single function resolve to the same argument shapes, and no
     recursive / co-recursive function calls are made.
+
 ### `-vhlo-legalize-to-stablehlo`
 
 _Legalize VHLO to StableHLO._
 
-
 ### `-vhlo-to-version`
 
 _Convert between versions of VHLO._
 
-
-
 #### Options
+
 ```
 -target : The target version. Must be a version of the form #.#.# .
 ```

docs/generated/stablehlo_tosa_passes.md

@@ -1,22 +1,24 @@
 <!-- Autogenerated by mlir-tblgen; don't manually edit -->
+
 ### `-stablehlo-legalize-to-tosa`
 
 _Legalize StableHLO to TOSA_
 
-
 ### `-stablehlo-prepare-for-tosa`
 
 _Prepare StableHLO for legalization to TOSA_
 
 This pass adds rewriters to make StableHLO ops more compatible with TOSA ops.
 Currently simplifies stablehlo.dot_general into stablehlo.dot for easier lowering.
+
 ### `-stablehlo-quant-legalize-to-tosa-rescale`
 
 _Legalize StableHLO Quantized operations to TOSA rescale operations_
 
 This pass rewrites StableHLO quantized operations to integer operations
 by inserting TOSA rescale operations at the inputs and outputs of the
 integer operations.
+
 ### `-tosa-rescale-legalize-to-stablehlo`
 
 _Legalize TOSA rescales to StableHlo primitive math operations_