[mlir] Use llvm::copy (NFC) (#168213)

Identified with llvm-use-ranges.

mlir/lib/Analysis/Presburger/IntegerRelation.cpp

@@ -2265,11 +2265,11 @@ IntegerRelation::unionBoundingBox(const IntegerRelation &otherCst) {
     newLb[d] = lbFloorDivisor;
     newUb[d] = -lbFloorDivisor;
     // Copy over the symbolic part + constant term.
-    std::copy(minLb.begin(), minLb.end(), newLb.begin() + getNumDimVars());
+    llvm::copy(minLb, newLb.begin() + getNumDimVars());
     std::transform(newLb.begin() + getNumDimVars(), newLb.end(),
                    newLb.begin() + getNumDimVars(),
                    std::negate<DynamicAPInt>());
-    std::copy(maxUb.begin(), maxUb.end(), newUb.begin() + getNumDimVars());
+    llvm::copy(maxUb, newUb.begin() + getNumDimVars());
 
     boundingLbs.emplace_back(newLb);
     boundingUbs.emplace_back(newUb);

mlir/lib/CAPI/IR/BuiltinTypes.cpp

@@ -536,7 +536,7 @@ MlirLogicalResult mlirMemRefTypeGetStridesAndOffset(MlirType type,
   if (failed(memrefType.getStridesAndOffset(strides_, *offset)))
     return mlirLogicalResultFailure();
 
-  (void)std::copy(strides_.begin(), strides_.end(), strides);
+  (void)llvm::copy(strides_, strides);
   return mlirLogicalResultSuccess();
 }
 

mlir/lib/Dialect/Affine/Transforms/PipelineDataTransfer.cpp

@@ -82,7 +82,7 @@ static bool doubleBuffer(Value oldMemRef, AffineForOp forOp) {
     ArrayRef<int64_t> oldShape = oldMemRefType.getShape();
     SmallVector<int64_t, 4> newShape(1 + oldMemRefType.getRank());
     newShape[0] = 2;
-    std::copy(oldShape.begin(), oldShape.end(), newShape.begin() + 1);
+    llvm::copy(oldShape, newShape.begin() + 1);
     return MemRefType::Builder(oldMemRefType).setShape(newShape).setLayout({});
   };
 

mlir/lib/Dialect/Linalg/Transforms/HoistPadding.cpp

@@ -913,8 +913,7 @@ static Value replaceByPackingResult(RewriterBase &rewriter,
       llvm_unreachable("loop independence prerequisite not met");
 
     // offsets = [maybe_leading_ivs = originalLoopIvs, 0 .. 0].
-    std::copy(loopIterationCounts.begin(), loopIterationCounts.end(),
-              offsets.begin());
+    llvm::copy(loopIterationCounts, offsets.begin());
     hoistedPackedTensor =
         scf::getForInductionVarOwner(packingResult.clonedLoopIvs.front())
             ->getResult(0);

mlir/lib/Dialect/SparseTensor/Transforms/Utils/SparseTensorIterator.h

@@ -197,7 +197,7 @@ public:
   // Sets the iterate to the specified position.
   void seek(ValueRange vals) {
     assert(vals.size() == cursorValsCnt);
-    std::copy(vals.begin(), vals.end(), cursorValsStorageRef.begin());
+    llvm::copy(vals, cursorValsStorageRef.begin());
     // Now that the iterator is re-positioned, the coordinate becomes invalid.
     crd = nullptr;
   }

mlir/lib/Dialect/XeGPU/Utils/XeGPUUtils.cpp

@@ -308,7 +308,7 @@ xegpu::extractVectorsWithShapeFromValue(OpBuilder &builder, Location loc,
   int64_t rankDiff = srcShapeRank - targetShapeRank;
   std::fill(adjustedTargetShape.begin(), adjustedTargetShape.begin() + rankDiff,
             1);
-  std::copy(shape.begin(), shape.end(), adjustedTargetShape.begin() + rankDiff);
+  llvm::copy(shape, adjustedTargetShape.begin() + rankDiff);
 
   SmallVector<Value> result;
   for (SmallVector<int64_t> offsets :

mlir/lib/Tools/PDLL/AST/Nodes.cpp

@@ -21,7 +21,7 @@ static StringRef copyStringWithNull(Context &ctx, StringRef str) {
     return str;
 
   char *data = ctx.getAllocator().Allocate<char>(str.size() + 1);
-  std::copy(str.begin(), str.end(), data);
+  llvm::copy(str, data);
   data[str.size()] = 0;
   return StringRef(data, str.size());
 }