diff --git a/llvm/docs/LangRef.rst b/llvm/docs/LangRef.rst
--- a/llvm/docs/LangRef.rst
+++ b/llvm/docs/LangRef.rst
@@ -14958,7 +14958,8 @@
 or the memory layout) can be expressed using the matrix intrinsics. Matrixes are
 embedded in a flat vector and the intrinsics take the dimensions as arguments.
 Currently column-major layout is assumed. The intrinsics support both integer
-and floating point matrixes.
+and floating point matrixes. Functions containing calls to matrix intrinsics
+must have the 'may-contain-matrix-intrinsics' attribute set to true.
 
 
 '``llvm.matrix.transpose.*``' Intrinsic
diff --git a/llvm/include/llvm/IR/Attributes.td b/llvm/include/llvm/IR/Attributes.td
--- a/llvm/include/llvm/IR/Attributes.td
+++ b/llvm/include/llvm/IR/Attributes.td
@@ -222,6 +222,7 @@
 def NoJumpTables : StrBoolAttr<"no-jump-tables">;
 def NoInlineLineTables : StrBoolAttr<"no-inline-line-tables">;
 def ProfileSampleAccurate : StrBoolAttr<"profile-sample-accurate">;
+def MayContainMatrixIntrinsics : StrBoolAttr<"may-contain-matrix-intrinsics">;
 
 class CompatRule<string F> {
   // The name of the function called to check the attribute of the caller and
@@ -258,6 +259,7 @@
 def : MergeRule<"setOR<NoJumpTablesAttr>">;
 def : MergeRule<"setOR<ProfileSampleAccurateAttr>">;
 def : MergeRule<"setOR<SpeculativeLoadHardeningAttr>">;
+def : MergeRule<"setOR<MayContainMatrixIntrinsicsAttr>">;
 def : MergeRule<"adjustCallerSSPLevel">;
 def : MergeRule<"adjustCallerStackProbes">;
 def : MergeRule<"adjustCallerStackProbeSize">;
diff --git a/llvm/lib/IR/Verifier.cpp b/llvm/lib/IR/Verifier.cpp
--- a/llvm/lib/IR/Verifier.cpp
+++ b/llvm/lib/IR/Verifier.cpp
@@ -4791,6 +4791,17 @@
     Assert(Size % 16 == 0, "bswap must be an even number of bytes", &Call);
     break;
   }
+  case Intrinsic::matrix_multiply:
+  case Intrinsic::matrix_transpose:
+  case Intrinsic::matrix_columnwise_load:
+  case Intrinsic::matrix_columnwise_store: {
+    Function *ParentFn = Call.getParent()->getParent();
+    Assert(ParentFn->getFnAttribute("may-contain-matrix-intrinsics")
+                   .getValueAsString() == "true",
+           "functions with matrix intrinsics must have the "
+           "may-contain-matrix-intrinsics attribute");
+    break;
+  }
   };
 }
 
diff --git a/llvm/lib/Transforms/IPO/PassManagerBuilder.cpp b/llvm/lib/Transforms/IPO/PassManagerBuilder.cpp
--- a/llvm/lib/Transforms/IPO/PassManagerBuilder.cpp
+++ b/llvm/lib/Transforms/IPO/PassManagerBuilder.cpp
@@ -149,10 +149,6 @@
     "enable-order-file-instrumentation", cl::init(false), cl::Hidden,
     cl::desc("Enable order file instrumentation (default = off)"));
 
-static cl::opt<bool>
-    EnableMatrix("enable-matrix", cl::init(false), cl::Hidden,
-                 cl::desc("Enable lowering of the matrix intrinsics"));
-
 PassManagerBuilder::PassManagerBuilder() {
     OptLevel = 2;
     SizeLevel = 0;
@@ -701,13 +697,7 @@
   MPM.add(createFloat2IntPass());
   MPM.add(createLowerConstantIntrinsicsPass());
 
-  if (EnableMatrix) {
-    MPM.add(createLowerMatrixIntrinsicsPass());
-    // CSE the pointer arithmetic of the column vectors.  This allows alias
-    // analysis to establish no-aliasing between loads and stores of different
-    // columns of the same matrix.
-    MPM.add(createEarlyCSEPass(false));
-  }
+  MPM.add(createLowerMatrixIntrinsicsPass());
 
   addExtensionsToPM(EP_VectorizerStart, MPM);
 
diff --git a/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp b/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
--- a/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
+++ b/llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
@@ -614,6 +614,8 @@
     for (Instruction *Inst : reverse(ToRemove))
       Inst->eraseFromParent();
 
+    // All matrix intrinsics have been removed.
+    Func.removeFnAttr("may-contain-matrix-intrinsics");
     return Changed;
   }
 
@@ -1479,6 +1481,11 @@
 };
 } // namespace
 
+static bool shouldRunMatrixLowering(Function &F) {
+  return F.getFnAttribute("may-contain-matrix-intrinsics").getValueAsString() ==
+         "true";
+}
+
 PreservedAnalyses LowerMatrixIntrinsicsPass::run(Function &F,
                                                  FunctionAnalysisManager &AM) {
   auto &TTI = AM.getResult<TargetIRAnalysis>(F);
@@ -1504,6 +1511,8 @@
   }
 
   bool runOnFunction(Function &F) override {
+    if (!shouldRunMatrixLowering(F))
+      return false;
     auto &TTI = getAnalysis<TargetTransformInfoWrapperPass>().getTTI(F);
     auto &ORE = getAnalysis<OptimizationRemarkEmitterWrapperPass>().getORE();
     LowerMatrixIntrinsics LMT(F, TTI, &ORE);
diff --git a/llvm/test/Other/opt-O2-pipeline.ll b/llvm/test/Other/opt-O2-pipeline.ll
--- a/llvm/test/Other/opt-O2-pipeline.ll
+++ b/llvm/test/Other/opt-O2-pipeline.ll
@@ -200,6 +200,10 @@
 ; CHECK-NEXT:       Lower constant intrinsics
 ; CHECK-NEXT:       Dominator Tree Construction
 ; CHECK-NEXT:       Natural Loop Information
+; CHECK-NEXT:       Lazy Branch Probability Analysis
+; CHECK-NEXT:       Lazy Block Frequency Analysis
+; CHECK-NEXT:       Optimization Remark Emitter
+; CHECK-NEXT:       Lower the matrix intrinsics
 ; CHECK-NEXT:       Canonicalize natural loops
 ; CHECK-NEXT:       LCSSA Verifier
 ; CHECK-NEXT:       Loop-Closed SSA Form Pass
diff --git a/llvm/test/Other/opt-O3-pipeline.ll b/llvm/test/Other/opt-O3-pipeline.ll
--- a/llvm/test/Other/opt-O3-pipeline.ll
+++ b/llvm/test/Other/opt-O3-pipeline.ll
@@ -205,6 +205,10 @@
 ; CHECK-NEXT:       Lower constant intrinsics
 ; CHECK-NEXT:       Dominator Tree Construction
 ; CHECK-NEXT:       Natural Loop Information
+; CHECK-NEXT:       Lazy Branch Probability Analysis
+; CHECK-NEXT:       Lazy Block Frequency Analysis
+; CHECK-NEXT:       Optimization Remark Emitter
+; CHECK-NEXT:       Lower the matrix intrinsics
 ; CHECK-NEXT:       Canonicalize natural loops
 ; CHECK-NEXT:       LCSSA Verifier
 ; CHECK-NEXT:       Loop-Closed SSA Form Pass
diff --git a/llvm/test/Other/opt-Os-pipeline.ll b/llvm/test/Other/opt-Os-pipeline.ll
--- a/llvm/test/Other/opt-Os-pipeline.ll
+++ b/llvm/test/Other/opt-Os-pipeline.ll
@@ -187,6 +187,10 @@
 ; CHECK-NEXT:       Lower constant intrinsics
 ; CHECK-NEXT:       Dominator Tree Construction
 ; CHECK-NEXT:       Natural Loop Information
+; CHECK-NEXT:       Lazy Branch Probability Analysis
+; CHECK-NEXT:       Lazy Block Frequency Analysis
+; CHECK-NEXT:       Optimization Remark Emitter
+; CHECK-NEXT:       Lower the matrix intrinsics
 ; CHECK-NEXT:       Canonicalize natural loops
 ; CHECK-NEXT:       LCSSA Verifier
 ; CHECK-NEXT:       Loop-Closed SSA Form Pass
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/bigger-expressions-double.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/bigger-expressions-double.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/bigger-expressions-double.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/bigger-expressions-double.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck  %s
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
-define void @transpose_multiply(<9 x double>* %A.Ptr, <9 x double>* %B.Ptr, <9 x double>* %C.Ptr) {
+define void @transpose_multiply(<9 x double>* %A.Ptr, <9 x double>* %B.Ptr, <9 x double>* %C.Ptr) #0 {
 ; CHECK-LABEL: @transpose_multiply(
 ; CHECK-NEXT:  entry:
 
@@ -248,7 +248,7 @@
 declare <9 x double> @llvm.matrix.transpose(<9 x double>, i32, i32)
 declare <9 x double> @llvm.matrix.multiply.v9f64.v9f64.v9f64(<9 x double>, <9 x double>, i32, i32, i32)
 
-define void @transpose_multiply_add(<9 x double>* %A.Ptr, <9 x double>* %B.Ptr, <9 x double>* %C.Ptr) {
+define void @transpose_multiply_add(<9 x double>* %A.Ptr, <9 x double>* %B.Ptr, <9 x double>* %C.Ptr) #0 {
 ; CHECK-LABEL: @transpose_multiply_add(
 ; CHECK-NEXT:  entry:
 
@@ -511,3 +511,5 @@
   store <9 x double> %res, <9 x double>* %C.Ptr
   ret void
 }
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/lowering-drops-attr.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/lowering-drops-attr.ll
new file mode 100644
--- /dev/null
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/lowering-drops-attr.ll
@@ -0,0 +1,42 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck  %s
+; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
+
+; Check that the running -lower-matrix-intrinsics removes the
+; "may-contain-matrix-intrinsics". Because it's the only attribute for
+; @transpose, the attribute reference will be dropped from the definition.
+
+define <8 x i32> @transpose(<8 x i32> %a) #0 {
+; CHECK-LABEL: define <8 x i32> @transpose(<8 x i32> %a) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <2 x i32> <i32 2, i32 3>
+; CHECK-NEXT:    [[SPLIT2:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <2 x i32> <i32 4, i32 5>
+; CHECK-NEXT:    [[SPLIT3:%.*]] = shufflevector <8 x i32> [[A]], <8 x i32> undef, <2 x i32> <i32 6, i32 7>
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <2 x i32> [[SPLIT]], i64 0
+; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <4 x i32> undef, i32 [[TMP0]], i64 0
+; CHECK-NEXT:    [[TMP2:%.*]] = extractelement <2 x i32> [[SPLIT1]], i64 0
+; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i32> [[TMP1]], i32 [[TMP2]], i64 1
+; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <2 x i32> [[SPLIT2]], i64 0
+; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i32> [[TMP3]], i32 [[TMP4]], i64 2
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <2 x i32> [[SPLIT3]], i64 0
+; CHECK-NEXT:    [[TMP7:%.*]] = insertelement <4 x i32> [[TMP5]], i32 [[TMP6]], i64 3
+; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <2 x i32> [[SPLIT]], i64 1
+; CHECK-NEXT:    [[TMP9:%.*]] = insertelement <4 x i32> undef, i32 [[TMP8]], i64 0
+; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <2 x i32> [[SPLIT1]], i64 1
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <4 x i32> [[TMP9]], i32 [[TMP10]], i64 1
+; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <2 x i32> [[SPLIT2]], i64 1
+; CHECK-NEXT:    [[TMP13:%.*]] = insertelement <4 x i32> [[TMP11]], i32 [[TMP12]], i64 2
+; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <2 x i32> [[SPLIT3]], i64 1
+; CHECK-NEXT:    [[TMP15:%.*]] = insertelement <4 x i32> [[TMP13]], i32 [[TMP14]], i64 3
+; CHECK-NEXT:    [[TMP16:%.*]] = shufflevector <4 x i32> [[TMP7]], <4 x i32> [[TMP15]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+; CHECK-NEXT:    ret <8 x i32> [[TMP16]]
+;
+entry:
+  %c  = call <8 x i32> @llvm.matrix.transpose(<8 x i32> %a, i32 2, i32 4)
+  ret <8 x i32> %c
+}
+
+declare <8 x i32> @llvm.matrix.transpose(<8 x i32>, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-contraction-fmf.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-contraction-fmf.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-contraction-fmf.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-contraction-fmf.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 
-define <4 x double> @multiply_2x2(<4 x double> %a, <4 x double> %b) {
+define <4 x double> @multiply_2x2(<4 x double> %a, <4 x double> %b) #0 {
 ; CHECK-LABEL: @multiply_2x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x double> [[A:%.*]], <4 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -67,3 +67,5 @@
 }
 
 declare <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double>, <4 x double>, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-contraction.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-contraction.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-contraction.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double-contraction.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -matrix-allow-contract -S < %s | FileCheck %s
 
 
-define <4 x double> @multiply_2x2(<4 x double> %a, <4 x double> %b) {
+define <4 x double> @multiply_2x2(<4 x double> %a, <4 x double> %b) #0 {
 ; CHECK-LABEL: @multiply_2x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x double> [[A:%.*]], <4 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -67,3 +67,5 @@
 }
 
 declare <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double>, <4 x double>, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-double.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 
-define <4 x double> @multiply_2x2(<4 x double> %a, <4 x double> %b) {
+define <4 x double> @multiply_2x2(<4 x double> %a, <4 x double> %b) #0 {
 ; CHECK-LABEL: @multiply_2x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x double> [[A:%.*]], <4 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -72,7 +72,7 @@
 
 declare <4 x double> @llvm.matrix.multiply.v4f64.v4f64.v4f64(<4 x double>, <4 x double>, i32, i32, i32)
 
-define <4 x double> @multiply_1x2(<2 x double> %a, <2 x double> %b) {
+define <4 x double> @multiply_1x2(<2 x double> %a, <2 x double> %b) #0 {
 ; CHECK-LABEL: @multiply_1x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <2 x double> [[A:%.*]], <2 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -116,7 +116,7 @@
 
 declare <4 x double> @llvm.matrix.multiply.v4f64.v2f64.v2f64(<2 x double>, <2 x double>, i32, i32, i32)
 
-define <9 x double> @multiply_2x3(<6 x double> %a, <6 x double> %b) {
+define <9 x double> @multiply_2x3(<6 x double> %a, <6 x double> %b) #0 {
 ; CHECK-LABEL: @multiply_2x3(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x double> [[A:%.*]], <6 x double> undef, <3 x i32> <i32 0, i32 1, i32 2>
@@ -252,3 +252,5 @@
 }
 
 declare <9 x double> @llvm.matrix.multiply.v6f64.v6f64.v6f64(<6 x double>, <6 x double>, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float-contraction-fmf.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float-contraction-fmf.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float-contraction-fmf.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float-contraction-fmf.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 
-define <4 x float> @multiply_2x2(<4 x float> %a, <4 x float> %b) {
+define <4 x float> @multiply_2x2(<4 x float> %a, <4 x float> %b) #0 {
 ; CHECK-LABEL: @multiply_2x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x float> [[A:%.*]], <4 x float> undef, <2 x i32> <i32 0, i32 1>
@@ -67,3 +67,5 @@
 }
 
 declare <4 x float> @llvm.matrix.multiply.v4f32.v4f32.v4f32(<4 x float>, <4 x float>, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float-contraction.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float-contraction.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float-contraction.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float-contraction.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -matrix-allow-contract -S < %s | FileCheck %s
 
 
-define <4 x float> @multiply_2x2(<4 x float> %a, <4 x float> %b) {
+define <4 x float> @multiply_2x2(<4 x float> %a, <4 x float> %b) #0 {
 ; CHECK-LABEL: @multiply_2x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x float> [[A:%.*]], <4 x float> undef, <2 x i32> <i32 0, i32 1>
@@ -67,3 +67,5 @@
 }
 
 declare <4 x float> @llvm.matrix.multiply.v4f32.v4f32.v4f32(<4 x float>, <4 x float>, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-float.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 
-define <4 x float> @multiply_2x2(<4 x float> %a, <4 x float> %b) {
+define <4 x float> @multiply_2x2(<4 x float> %a, <4 x float> %b) #0 {
 ; CHECK-LABEL: @multiply_2x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x float> [[A:%.*]], <4 x float> undef, <2 x i32> <i32 0, i32 1>
@@ -72,7 +72,7 @@
 
 declare <4 x float> @llvm.matrix.multiply.v4f32.v4f32.v4f32(<4 x float>, <4 x float>, i32, i32, i32)
 
-define <4 x float> @multiply_1x2(<2 x float> %a, <2 x float> %b) {
+define <4 x float> @multiply_1x2(<2 x float> %a, <2 x float> %b) #0 {
 ; CHECK-LABEL: @multiply_1x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <2 x float> [[A:%.*]], <2 x float> undef, <2 x i32> <i32 0, i32 1>
@@ -116,7 +116,7 @@
 
 declare <4 x float> @llvm.matrix.multiply.v4f32.v2f32.v2f32(<2 x float>, <2 x float>, i32, i32, i32)
 
-define <9 x float> @multiply_2x3(<6 x float> %a, <6 x float> %b) {
+define <9 x float> @multiply_2x3(<6 x float> %a, <6 x float> %b) #0 {
 ; CHECK-LABEL: @multiply_2x3(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x float> [[A:%.*]], <6 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
@@ -252,3 +252,5 @@
 }
 
 declare <9 x float> @llvm.matrix.multiply.v6f32.v6f32.v6f32(<6 x float>, <6 x float>, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-i32.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-i32.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-i32.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/multiply-i32.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 
-define <4 x i32> @multiply_2x2(<4 x i32> %a, <4 x i32> %b) {
+define <4 x i32> @multiply_2x2(<4 x i32> %a, <4 x i32> %b) #0 {
 ; CHECK-LABEL: @multiply_2x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <4 x i32> [[A:%.*]], <4 x i32> undef, <2 x i32> <i32 0, i32 1>
@@ -72,7 +72,7 @@
 
 declare <4 x i32> @llvm.matrix.multiply.v4i32.v4i32.v4i32(<4 x i32>, <4 x i32>, i32, i32, i32)
 
-define <4 x i32> @multiply_1x2(<2 x i32> %a, <2 x i32> %b) {
+define <4 x i32> @multiply_1x2(<2 x i32> %a, <2 x i32> %b) #0 {
 ; CHECK-LABEL: @multiply_1x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <2 x i32> [[A:%.*]], <2 x i32> undef, <2 x i32> <i32 0, i32 1>
@@ -116,7 +116,7 @@
 
 declare <4 x i32> @llvm.matrix.multiply.v4i32.v2i32.v2i32(<2 x i32>, <2 x i32>, i32, i32, i32)
 
-define <9 x i32> @multiply_2x3(<6 x i32> %a, <6 x i32> %b) {
+define <9 x i32> @multiply_2x3(<6 x i32> %a, <6 x i32> %b) #0 {
 ; CHECK-LABEL: @multiply_2x3(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x i32> [[A:%.*]], <6 x i32> undef, <3 x i32> <i32 0, i32 1, i32 2>
@@ -252,3 +252,5 @@
 }
 
 declare <9 x i32> @llvm.matrix.multiply.v6i32.v6i32.v6i32(<6 x i32>, <6 x i32>, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backward.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backward.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backward.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backward.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck  %s
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
-define <8 x double> @fadd_transpose(<8 x double> %a, <8 x double> %b) {
+define <8 x double> @fadd_transpose(<8 x double> %a, <8 x double> %b) #0 {
 ; CHECK-LABEL: @fadd_transpose(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x double> [[A:%.*]], <8 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -42,7 +42,7 @@
   ret <8 x double> %c
 }
 
-define <8 x double> @load_fadd_transpose(<8 x double>* %A.Ptr, <8 x double> %b) {
+define <8 x double> @load_fadd_transpose(<8 x double>* %A.Ptr, <8 x double> %b) #0 {
 ; CHECK-LABEL: @load_fadd_transpose(
 ; CHECK-NEXT:  entry:
 
@@ -94,3 +94,5 @@
 }
 
 declare <8 x double> @llvm.matrix.transpose(<8 x double>, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backwards-unsupported.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backwards-unsupported.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backwards-unsupported.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-backwards-unsupported.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 ; Check that we we use flattened vectors for PHI operands and extract the columns afterwards.
-define <9 x double> @unsupported_phi(i1 %cond, <9 x double> %A, <9 x double> %B, <9 x double> %C) {
+define <9 x double> @unsupported_phi(i1 %cond, <9 x double> %A, <9 x double> %B, <9 x double> %C) #0 {
 ; CHECK-LABEL: @unsupported_phi(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    br i1 [[COND:%.*]], label [[IF_THEN:%.*]], label [[IF_ELSE:%.*]]
@@ -90,7 +90,7 @@
 }
 
 ; Make sure we use a flattened vector when calling @foo and the use its flat vector result properly.
-define <9 x double> @unsupported_call(i1 %cond, <9 x double> %A, <9 x double> %B) {
+define <9 x double> @unsupported_call(i1 %cond, <9 x double> %A, <9 x double> %B) #0 {
 ; CHECK-LABEL: @unsupported_call(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <9 x double> [[A:%.*]], <9 x double> undef, <3 x i32> <i32 0, i32 1, i32 2>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <9 x double> [[A]], <9 x double> undef, <3 x i32> <i32 3, i32 4, i32 5>
@@ -133,3 +133,5 @@
 declare <9 x double> @llvm.matrix.multiply.v9f64.v9f64.v9f64(<9 x double>, <9 x double>, i32 immarg, i32 immarg, i32 immarg)
 declare <9 x double> @llvm.matrix.transpose.v9f64(<9 x double>, i32 immarg, i32 immarg)
 declare <9 x double> @foo(<9 x double>)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-forward.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-forward.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-forward.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-forward.ll
@@ -4,7 +4,7 @@
 
 ; Check that we do not emit shufflevectors to flatten the result of the
 ; transpose and store the columns directly.
-define void @transpose_store(<8 x double> %a, <8 x double>* %Ptr) {
+define void @transpose_store(<8 x double> %a, <8 x double>* %Ptr) #0 {
 ; CHECK-LABEL: @transpose_store(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x double> [[A:%.*]], <8 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -43,7 +43,7 @@
 
 declare <8 x double> @llvm.matrix.transpose(<8 x double>, i32, i32)
 
-define <8 x double> @transpose_fadd(<8 x double> %a) {
+define <8 x double> @transpose_fadd(<8 x double> %a) #0 {
 ; CHECK-LABEL: @transpose_fadd(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x double> [[A:%.*]], <8 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -79,7 +79,7 @@
   ret <8 x double> %res
 }
 
-define <8 x double> @transpose_fmul(<8 x double> %a) {
+define <8 x double> @transpose_fmul(<8 x double> %a) #0 {
 ; CHECK-LABEL: @transpose_fmul(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x double> [[A:%.*]], <8 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -114,3 +114,5 @@
   %res = fmul <8 x double> %c, %a
   ret <8 x double> %res
 }
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-mixed-users.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-mixed-users.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-mixed-users.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-mixed-users.ll
@@ -4,7 +4,7 @@
 
 ; Currently we only lower stores with shape information, but need to embed the
 ; matrix in a flat vector for function calls and returns.
-define <8 x double> @strided_load_4x4(<8 x double> %in, <8 x double>* %Ptr) {
+define <8 x double> @strided_load_4x4(<8 x double> %in, <8 x double>* %Ptr) #0 {
 ; CHECK-LABEL: @strided_load_4x4(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x double> [[IN:%.*]], <8 x double> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <8 x double> [[IN]], <8 x double> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
@@ -51,3 +51,5 @@
 declare <8 x double> @llvm.matrix.transpose(<8 x double>, i32, i32)
 
 declare void @foo(<8 x double>)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-multiple-iterations.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-multiple-iterations.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-multiple-iterations.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/propagate-multiple-iterations.ll
@@ -6,7 +6,7 @@
 ; Make sure we propagate in multiple iterations. First, we back-propagate the
 ; shape information from the transpose to %A, in the next iteration we
 ; forward-propagate it to %Mul, and then back to %B.
-define <16 x double> @backpropagation_iterations(<16 x double>* %A.Ptr, <16 x double>* %B.Ptr) {
+define <16 x double> @backpropagation_iterations(<16 x double>* %A.Ptr, <16 x double>* %B.Ptr) #0 {
 ; CHECK-LABEL: @backpropagation_iterations(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <16 x double>* [[A_PTR:%.*]] to double*
 ; CHECK-NEXT:    [[TMP2:%.*]] = bitcast double* [[TMP1]] to <4 x double>*
@@ -82,3 +82,5 @@
 
 declare <16 x double> @llvm.matrix.multiply.v16f64.v16f64.v16f64(<16 x double>, <16 x double>, i32 immarg, i32 immarg, i32 immarg)
 declare <16 x double> @llvm.matrix.transpose.v16f64(<16 x double>, i32 immarg, i32 immarg)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/remarks-inlining.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/remarks-inlining.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/remarks-inlining.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/remarks-inlining.ll
@@ -92,7 +92,7 @@
 ; CHECK-LABEL: remark: transpose.h:13:11: Lowered with 0 stores, 0 loads, 8 compute ops
 ; CHECK-NEXT:  transpose.1x2.float(transpose.2x1.float(addr %D))
 
-define void @toplevel(<15 x double>* %A, <15 x double>* %B, <15 x double>* %C, <2 x float>* %D) !dbg !16 {
+define void @toplevel(<15 x double>* %A, <15 x double>* %B, <15 x double>* %C, <2 x float>* %D) #0 !dbg !16 {
 entry:
   %a = load <15 x double>, <15 x double> *%A, align 16, !dbg !3791
   %b = call <15 x double> @llvm.matrix.columnwise.load(<15 x double>* %B, i32 5, i32 3, i32 5), !dbg !3793
@@ -164,3 +164,5 @@
 
 !108 = !DILocation(line: 66, column: 11, scope: !23, inlinedAt: !109)
 !109 = !DILocation(line: 510, column: 0, scope: !16)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/remarks-shared-subtrees.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/remarks-shared-subtrees.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/remarks-shared-subtrees.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/remarks-shared-subtrees.ll
@@ -87,7 +87,7 @@
 ; STDERR-NEXT:     columnwise.load.2x15.double(addr %arg3, scalar))),
 ; STDERR-NEXT:   addr %arg2,
 ; STDERR-NEXT:   10)
-define void @test_2leafs(double* %arg1, double* %arg2, double* %arg3, i32 %stride, i32 %offset) !dbg !8 {
+define void @test_2leafs(double* %arg1, double* %arg2, double* %arg3, i32 %stride, i32 %offset) #0 !dbg !8 {
 bb:
   %shared.load = tail call <8 x double> @llvm.matrix.columnwise.load.v8f64.p0f64(double* %arg1, i32 %stride, i32 2, i32 4), !dbg !10, !noalias !10
   %shared.load.2 = tail call <30 x double> @llvm.matrix.columnwise.load.v30f64.p0f64(double* %arg3, i32 %stride, i32 2, i32 15), !dbg !10, !noalias !10
@@ -108,6 +108,8 @@
 declare void @llvm.matrix.columnwise.store.v8f64.p0f64(<8 x double>, double* writeonly, i32, i32 immarg, i32 immarg)
 declare <60 x double> @llvm.matrix.multiply.v60f64.v8f64.v30f64(<8 x double>, <30 x double>, i32 immarg, i32 immarg, i32 immarg)
 
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
+
 !llvm.module.flags = !{!0, !1, !2, !3}
 !llvm.dbg.cu = !{!4}
 !llvm.ident = !{!7}
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/remarks.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/remarks.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/remarks.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/remarks.ll
@@ -8,7 +8,7 @@
 ; CHECK-NEXT: store(
 ; CHECK-NEXT:  transpose.2x6.double(load(addr %A)),
 ; CHECK-NEXT:  addr %B)
-define void @transpose(<12 x double>* %A, <12 x double>* %B) !dbg !23 {
+define void @transpose(<12 x double>* %A, <12 x double>* %B) #0 !dbg !23 {
   %load = load <12 x double>, <12 x double>* %A, !dbg !24
   %t = call <12 x double> @llvm.matrix.transpose.v12f64.v12f64(<12 x double> %load, i32 2, i32 6), !dbg !24
   store <12 x double> %t, <12 x double>* %B, !dbg !24
@@ -24,7 +24,7 @@
 ; CHECK-NEXT:    load(addr %A),
 ; CHECK-NEXT:    load(addr %B)),
 ; CHECK-NEXT:   addr %C)
-define void @multiply(<12 x double>* %A, <12 x double>* %B, <4 x double>* %C) !dbg !25 {
+define void @multiply(<12 x double>* %A, <12 x double>* %B, <4 x double>* %C) #0 !dbg !25 {
   %A.matrix = load <12 x double>, <12 x double>* %A, !dbg !26
   %B.matrix = load <12 x double>, <12 x double>* %B, !dbg !26
   %t = call <4 x double> @llvm.matrix.multiply(<12 x double> %A.matrix, <12 x double> %B.matrix, i32 2, i32 6, i32 2), !dbg !26
@@ -38,7 +38,7 @@
 ; CHECK-NEXT:  store(
 ; CHECK-NEXT:   columnwise.load.3x3.double(addr %A, 5),
 ; CHECK-NEXT:   addr %B)
-define void @columnwise.load(<9 x double>* %A, <9 x double>* %B) !dbg !27 {
+define void @columnwise.load(<9 x double>* %A, <9 x double>* %B) #0 !dbg !27 {
   %A.matrix = call <9 x double> @llvm.matrix.columnwise.load(<9 x double>* %A, i32 5, i32 3, i32 3), !dbg !28
   store <9 x double> %A.matrix, <9 x double>* %B, !dbg !28
   ret void
@@ -51,7 +51,7 @@
 ; CHECK-NEXT:   columnwise.load.3x3.double(addr %A, 5),
 ; CHECK-NEXT:   addr %B,
 ; CHECK-NEXT:   10)
-define void @columnwise.store(<9 x double>* %A, <9 x double>* %B) !dbg !29 {
+define void @columnwise.store(<9 x double>* %A, <9 x double>* %B) #0 !dbg !29 {
   %A.matrix = call <9 x double> @llvm.matrix.columnwise.load(<9 x double>* %A, i32 5, i32 3, i32 3), !dbg !30
   call void @llvm.matrix.columnwise.store(<9 x double> %A.matrix, <9 x double>* %B, i32 10, i32 3, i32 3), !dbg !30
   ret void
@@ -69,7 +69,7 @@
 ; CHECK-NEXT:   addr %B,
 ; CHECK-NEXT:   10)
 
-define void @binaryops(<9 x double>* %A, <9 x double>* %B) !dbg !31 {
+define void @binaryops(<9 x double>* %A, <9 x double>* %B) #0 !dbg !31 {
   %A.matrix = call <9 x double> @llvm.matrix.columnwise.load(<9 x double>* %A, i32 5, i32 3, i32 3), !dbg !32
   %R1.matrix = fadd <9 x double> %A.matrix, %A.matrix, !dbg !32
   %R2.matrix = fmul <9 x double> %R1.matrix, %A.matrix, !dbg !32
@@ -93,7 +93,7 @@
 ; CHECK-NEXT:    load(addr %D)),
 ; CHECK-NEXT:   addr %E)
 
-define void @multiple_expressions(<9 x double>* %A, <9 x double>* %B, <12 x double>* %C, <12 x double>* %D, <4 x double>* %E) !dbg !33 {
+define void @multiple_expressions(<9 x double>* %A, <9 x double>* %B, <12 x double>* %C, <12 x double>* %D, <4 x double>* %E) #0 !dbg !33 {
   %A.matrix = call <9 x double> @llvm.matrix.columnwise.load(<9 x double>* %A, i32 5, i32 3, i32 3), !dbg !34
   %R1.matrix = fadd <9 x double> %A.matrix, %A.matrix, !dbg !34
   %R2.matrix = fmul <9 x double> %R1.matrix, %A.matrix, !dbg !34
@@ -116,7 +116,7 @@
 ; CHECK-NEXT:    (reused) columnwise.load.3x3.double(addr %A, 5)),
 ; CHECK-NEXT:   stack addr %B,
 ; CHECK-NEXT:   10)
-define void @stackaddresses(<9 x double>* %A) !dbg !35 {
+define void @stackaddresses(<9 x double>* %A) #0 !dbg !35 {
   %B = alloca <9 x double>
   %A.matrix = call <9 x double> @llvm.matrix.columnwise.load(<9 x double>* %A, i32 5, i32 3, i32 3), !dbg !36
   %R1.matrix = fadd <9 x double> %A.matrix, %A.matrix, !dbg !36
@@ -130,7 +130,7 @@
 ; CHECK-NEXT:   transpose.5x3.double(load(addr %A)),
 ; CHECK-NEXT:   stack addr %s1)
 %S1 = type {<15 x double>*}
-define void @get_underlying_object(%S1* %A) !dbg !21 {
+define void @get_underlying_object(%S1* %A) #0 !dbg !21 {
 entry:
   %s1 = alloca <15 x double>, !dbg !22
   %a1 = getelementptr %S1, %S1* %A, i32 0, i32 0, !dbg !22
@@ -148,6 +148,8 @@
 
 declare <15 x double> @llvm.matrix.transpose.v15f64.v15f64(<15 x double>, i32, i32)
 
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
+
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!3, !4}
 
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-double.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-double.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-double.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-double.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck %s
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
-define <9 x double> @strided_load_3x3(<9 x double>* %in, i32 %stride) {
+define <9 x double> @strided_load_3x3(<9 x double>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_3x3(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <9 x double>* [[IN:%.*]] to double*
@@ -30,7 +30,7 @@
 
 declare <9 x double> @llvm.matrix.columnwise.load(<9 x double>*, i32, i32, i32)
 
-define <9 x double> @strided_load_9x1(<9 x double>* %in, i32 %stride) {
+define <9 x double> @strided_load_9x1(<9 x double>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_9x1(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <9 x double>* [[IN:%.*]] to double*
@@ -47,7 +47,7 @@
 
 declare <8 x double> @llvm.matrix.columnwise.load.v8f64(<8 x double>*, i32, i32, i32)
 
-define <8 x double> @strided_load_4x2(<8 x double>* %in, i32 %stride) {
+define <8 x double> @strided_load_4x2(<8 x double>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_4x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x double>* [[IN:%.*]] to double*
@@ -72,3 +72,5 @@
 ; CHECK: declare <8 x double> @llvm.matrix.columnwise.load.v8f64.p0v8f64(<8 x double>*, i32, i32 immarg, i32 immarg) [[READONLY]]
 
 ; CHECK: attributes [[READONLY]] = { argmemonly nounwind readonly willreturn }
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-float.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-float.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-float.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-float.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck %s
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
-define <9 x float> @strided_load_3x3(<9 x float>* %in, i32 %stride) {
+define <9 x float> @strided_load_3x3(<9 x float>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_3x3(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <9 x float>* [[IN:%.*]] to float*
@@ -30,7 +30,7 @@
 
 declare <9 x float> @llvm.matrix.columnwise.load(<9 x float>*, i32, i32, i32)
 
-define <9 x float> @strided_load_9x1(<9 x float>* %in, i32 %stride) {
+define <9 x float> @strided_load_9x1(<9 x float>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_9x1(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <9 x float>* [[IN:%.*]] to float*
@@ -47,7 +47,7 @@
 
 declare <8 x float> @llvm.matrix.columnwise.load.v8f32(<8 x float>*, i32, i32, i32)
 
-define <8 x float> @strided_load_4x2(<8 x float>* %in, i32 %stride) {
+define <8 x float> @strided_load_4x2(<8 x float>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_4x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x float>* [[IN:%.*]] to float*
@@ -66,3 +66,5 @@
   %load = call <8 x float> @llvm.matrix.columnwise.load.v8f32(<8 x float>* %in, i32 %stride, i32 4, i32 2)
   ret <8 x float> %load
 }
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-i32.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-i32.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-i32.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-i32.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck %s
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
-define <9 x i32> @strided_load_3x3(<9 x i32>* %in, i32 %stride) {
+define <9 x i32> @strided_load_3x3(<9 x i32>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_3x3(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <9 x i32>* [[IN:%.*]] to i32*
@@ -30,7 +30,7 @@
 
 declare <9 x i32> @llvm.matrix.columnwise.load(<9 x i32>*, i32, i32, i32)
 
-define <9 x i32> @strided_load_9x1(<9 x i32>* %in, i32 %stride) {
+define <9 x i32> @strided_load_9x1(<9 x i32>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_9x1(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <9 x i32>* [[IN:%.*]] to i32*
@@ -47,7 +47,7 @@
 
 declare <8 x i32> @llvm.matrix.columnwise.load.v8i32(<8 x i32>*, i32, i32, i32)
 
-define <8 x i32> @strided_load_4x2(<8 x i32>* %in, i32 %stride) {
+define <8 x i32> @strided_load_4x2(<8 x i32>* %in, i32 %stride) #0 {
 ; CHECK-LABEL: @strided_load_4x2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <8 x i32>* [[IN:%.*]] to i32*
@@ -66,3 +66,5 @@
   %load = call <8 x i32> @llvm.matrix.columnwise.load.v8i32(<8 x i32>* %in, i32 %stride, i32 4, i32 2)
   ret <8 x i32> %load
 }
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-double.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-double.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-double.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-double.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck %s
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
-define void @strided_store_3x2(<6 x double> %in, double* %out) {
+define void @strided_store_3x2(<6 x double> %in, double* %out) #0 {
 ; CHECK-LABEL: @strided_store_3x2(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x double> [[IN:%.*]], <6 x double> undef, <3 x i32> <i32 0, i32 1, i32 2>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <6 x double> [[IN]], <6 x double> undef, <3 x i32> <i32 3, i32 4, i32 5>
@@ -17,7 +17,7 @@
   ret void
 }
 
-define void @strided_store_3x2_nonconst_stride(<6 x double> %in, i32 %stride, double* %out) {
+define void @strided_store_3x2_nonconst_stride(<6 x double> %in, i32 %stride, double* %out) #0 {
 ; CHECK-LABEL: @strided_store_3x2_nonconst_stride(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x double> [[IN:%.*]], <6 x double> undef, <3 x i32> <i32 0, i32 1, i32 2>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <6 x double> [[IN]], <6 x double> undef, <3 x i32> <i32 3, i32 4, i32 5>
@@ -38,7 +38,7 @@
 
 declare void @llvm.matrix.columnwise.store(<6 x double>, double*, i32, i32, i32)
 
-define void @strided_store_2x3(<10 x double> %in, double* %out) {
+define void @strided_store_2x3(<10 x double> %in, double* %out) #0 {
 ; CHECK-LABEL: @strided_store_2x3(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <10 x double> [[IN:%.*]], <10 x double> undef, <2 x i32> <i32 0, i32 1>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <10 x double> [[IN]], <10 x double> undef, <2 x i32> <i32 2, i32 3>
@@ -72,3 +72,6 @@
 ; CHECK: declare void @llvm.matrix.columnwise.store.v10f64.p0f64(<10 x double>, double* writeonly, i32, i32 immarg, i32 immarg) [[WRITEONLY]]
 
 ; CHECK: attributes [[WRITEONLY]] = { argmemonly nounwind willreturn writeonly }
+
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-float.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-float.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-float.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-float.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck %s
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
-define void @strided_store_3x2(<6 x float> %in, float* %out) {
+define void @strided_store_3x2(<6 x float> %in, float* %out) #0 {
 ; CHECK-LABEL: @strided_store_3x2(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x float> [[IN:%.*]], <6 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <6 x float> [[IN]], <6 x float> undef, <3 x i32> <i32 3, i32 4, i32 5>
@@ -17,7 +17,7 @@
   ret void
 }
 
-define void @strided_store_3x2_nonconst_stride(<6 x float> %in, i32 %stride, float* %out) {
+define void @strided_store_3x2_nonconst_stride(<6 x float> %in, i32 %stride, float* %out) #0 {
 ; CHECK-LABEL: @strided_store_3x2_nonconst_stride(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x float> [[IN:%.*]], <6 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <6 x float> [[IN]], <6 x float> undef, <3 x i32> <i32 3, i32 4, i32 5>
@@ -38,7 +38,7 @@
 
 declare void @llvm.matrix.columnwise.store(<6 x float>, float*, i32, i32, i32)
 
-define void @strided_store_2x3(<10 x float> %in, float* %out) {
+define void @strided_store_2x3(<10 x float> %in, float* %out) #0 {
 ; CHECK-LABEL: @strided_store_2x3(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <10 x float> [[IN:%.*]], <10 x float> undef, <2 x i32> <i32 0, i32 1>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <10 x float> [[IN]], <10 x float> undef, <2 x i32> <i32 2, i32 3>
@@ -66,3 +66,5 @@
 }
 
 declare void @llvm.matrix.columnwise.store.v10f32(<10 x float>, float*, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-i32.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-i32.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-i32.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-i32.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -lower-matrix-intrinsics -S < %s | FileCheck %s
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
-define void @strided_store_3x2(<6 x i32> %in, i32* %out) {
+define void @strided_store_3x2(<6 x i32> %in, i32* %out) #0 {
 ; CHECK-LABEL: @strided_store_3x2(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x i32> [[IN:%.*]], <6 x i32> undef, <3 x i32> <i32 0, i32 1, i32 2>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <6 x i32> [[IN]], <6 x i32> undef, <3 x i32> <i32 3, i32 4, i32 5>
@@ -17,7 +17,7 @@
   ret void
 }
 
-define void @strided_store_3x2_nonconst_stride(<6 x i32> %in, i32 %stride, i32* %out) {
+define void @strided_store_3x2_nonconst_stride(<6 x i32> %in, i32 %stride, i32* %out) #0 {
 ; CHECK-LABEL: @strided_store_3x2_nonconst_stride(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <6 x i32> [[IN:%.*]], <6 x i32> undef, <3 x i32> <i32 0, i32 1, i32 2>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <6 x i32> [[IN]], <6 x i32> undef, <3 x i32> <i32 3, i32 4, i32 5>
@@ -38,7 +38,7 @@
 
 declare void @llvm.matrix.columnwise.store(<6 x i32>, i32*, i32, i32, i32)
 
-define void @strided_store_2x3(<10 x i32> %in, i32* %out) {
+define void @strided_store_2x3(<10 x i32> %in, i32* %out) #0 {
 ; CHECK-LABEL: @strided_store_2x3(
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <10 x i32> [[IN:%.*]], <10 x i32> undef, <2 x i32> <i32 0, i32 1>
 ; CHECK-NEXT:    [[SPLIT1:%.*]] = shufflevector <10 x i32> [[IN]], <10 x i32> undef, <2 x i32> <i32 2, i32 3>
@@ -66,3 +66,5 @@
 }
 
 declare void @llvm.matrix.columnwise.store.v10i32(<10 x i32>, i32*, i32, i32, i32)
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-double.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-double.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-double.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-double.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 
-define <8 x double> @transpose(<8 x double> %a) {
+define <8 x double> @transpose(<8 x double> %a) #0 {
 ; CHECK-LABEL: @transpose(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x double> [[A:%.*]], <8 x double> undef, <2 x i32> <i32 0, i32 1>
@@ -36,7 +36,7 @@
 
 declare <8 x double> @llvm.matrix.transpose(<8 x double>, i32, i32)
 
-define <8 x double> @transpose_single_column(<8 x double> %a) {
+define <8 x double> @transpose_single_column(<8 x double> %a) #0 {
 ; CHECK-LABEL: @transpose_single_column(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x double> [[A:%.*]], <8 x double> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
@@ -72,7 +72,7 @@
 
 declare <12 x double> @llvm.matrix.transpose.v12f64(<12 x double>, i32, i32)
 
-define <12 x double> @transpose_double_3x4(<12 x double> %a) {
+define <12 x double> @transpose_double_3x4(<12 x double> %a) #0 {
 ; CHECK-LABEL: @transpose_double_3x4(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <12 x double> [[A:%.*]], <12 x double> undef, <3 x i32> <i32 0, i32 1, i32 2>
@@ -112,3 +112,5 @@
   %c  = call <12 x double> @llvm.matrix.transpose.v12f64(<12 x double> %a, i32 3, i32 4)
   ret <12 x double> %c
 }
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-float.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-float.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-float.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-float.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 
-define <8 x float> @transpose(<8 x float> %a) {
+define <8 x float> @transpose(<8 x float> %a) #0 {
 ; CHECK-LABEL: @transpose(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x float> [[A:%.*]], <8 x float> undef, <2 x i32> <i32 0, i32 1>
@@ -36,7 +36,7 @@
 
 declare <8 x float> @llvm.matrix.transpose(<8 x float>, i32, i32)
 
-define <8 x float> @transpose_single_column(<8 x float> %a) {
+define <8 x float> @transpose_single_column(<8 x float> %a) #0 {
 ; CHECK-LABEL: @transpose_single_column(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x float> [[A:%.*]], <8 x float> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
@@ -72,7 +72,7 @@
 
 declare <12 x float> @llvm.matrix.transpose.v12f32(<12 x float>, i32, i32)
 
-define <12 x float> @transpose_float_3x4(<12 x float> %a) {
+define <12 x float> @transpose_float_3x4(<12 x float> %a) #0 {
 ; CHECK-LABEL: @transpose_float_3x4(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <12 x float> [[A:%.*]], <12 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
@@ -112,3 +112,5 @@
   %c  = call <12 x float> @llvm.matrix.transpose.v12f32(<12 x float> %a, i32 3, i32 4)
   ret <12 x float> %c
 }
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }
diff --git a/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-i32.ll b/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-i32.ll
--- a/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-i32.ll
+++ b/llvm/test/Transforms/LowerMatrixIntrinsics/transpose-i32.ll
@@ -3,7 +3,7 @@
 ; RUN: opt -passes='lower-matrix-intrinsics' -S < %s | FileCheck %s
 
 
-define <8 x i32> @transpose(<8 x i32> %a) {
+define <8 x i32> @transpose(<8 x i32> %a) #0 {
 ; CHECK-LABEL: @transpose(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <2 x i32> <i32 0, i32 1>
@@ -36,7 +36,7 @@
 
 declare <8 x i32> @llvm.matrix.transpose(<8 x i32>, i32, i32)
 
-define <8 x i32> @transpose_single_column(<8 x i32> %a) {
+define <8 x i32> @transpose_single_column(<8 x i32> %a) #0 {
 ; CHECK-LABEL: @transpose_single_column(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <8 x i32> [[A:%.*]], <8 x i32> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
@@ -72,7 +72,7 @@
 
 declare <12 x i32> @llvm.matrix.transpose.v12i32(<12 x i32>, i32, i32)
 
-define <12 x i32> @transpose_i32_3x4(<12 x i32> %a) {
+define <12 x i32> @transpose_i32_3x4(<12 x i32> %a) #0 {
 ; CHECK-LABEL: @transpose_i32_3x4(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SPLIT:%.*]] = shufflevector <12 x i32> [[A:%.*]], <12 x i32> undef, <3 x i32> <i32 0, i32 1, i32 2>
@@ -112,3 +112,5 @@
   %c  = call <12 x i32> @llvm.matrix.transpose.v12i32(<12 x i32> %a, i32 3, i32 4)
   ret <12 x i32> %c
 }
+
+attributes #0 = { "may-contain-matrix-intrinsics"="true" }