[llvm] [InstCombine] Handle scalable splats of constants in getMinimumFPType (PR #132960)

[Luke Lau via llvm-commits]

https://github.com/lukel97 created https://github.com/llvm/llvm-project/pull/132960

We previously handled ConstantExpr scalable splats in 5d929794a87602cfd873381e11cc99149196bb49, but only fpexts.

ConstantExpr fpexts have since been removed, and simultaneously we didn't handle splats of constants that weren't extended.

This updates it to remove the fpext check and instead see if we can shrink the result of getSplatValue.

Note that the test case doesn't get completely folded away due to #132922




>From 4b9c2437f5b666cc1962c3957a3884ce2246b6cc Mon Sep 17 00:00:00 2001
From: Luke Lau <luke at igalia.com>
Date: Wed, 26 Mar 2025 01:13:34 +0800
Subject: [PATCH 1/2] Precommit tests

---
 .../InstCombine/scalable-const-fp-splat.ll         | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/llvm/test/Transforms/InstCombine/scalable-const-fp-splat.ll b/llvm/test/Transforms/InstCombine/scalable-const-fp-splat.ll
index 731b079881f08..3648a868fe1cb 100644
--- a/llvm/test/Transforms/InstCombine/scalable-const-fp-splat.ll
+++ b/llvm/test/Transforms/InstCombine/scalable-const-fp-splat.ll
@@ -13,3 +13,17 @@ define <vscale x 2 x float> @shrink_splat_scalable_extend(<vscale x 2 x float> %
   %5 = fptrunc <vscale x 2 x double> %4 to <vscale x 2 x float>
   ret <vscale x 2 x float> %5
 }
+
+define <vscale x 2 x float> @shrink_splat_scalable_extend_rhs_constexpr(<vscale x 2 x float> %a) {
+; CHECK-LABEL: define <vscale x 2 x float> @shrink_splat_scalable_extend_rhs_constexpr(
+; CHECK-SAME: <vscale x 2 x float> [[A:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = fpext <vscale x 2 x float> [[A]] to <vscale x 2 x double>
+; CHECK-NEXT:    [[TMP2:%.*]] = fadd <vscale x 2 x double> [[TMP1]], splat (double -1.000000e+00)
+; CHECK-NEXT:    [[TMP3:%.*]] = fptrunc <vscale x 2 x double> [[TMP2]] to <vscale x 2 x float>
+; CHECK-NEXT:    ret <vscale x 2 x float> [[TMP3]]
+;
+  %2 = fpext <vscale x 2 x float> %a to <vscale x 2 x double>
+  %4 = fadd <vscale x 2 x double> %2, splat (double -1.000000e+00)
+  %5 = fptrunc <vscale x 2 x double> %4 to <vscale x 2 x float>
+  ret <vscale x 2 x float> %5
+}

>From 01be82ce028b51e54f247379459d3d994165699e Mon Sep 17 00:00:00 2001
From: Luke Lau <luke at igalia.com>
Date: Wed, 26 Mar 2025 01:16:49 +0800
Subject: [PATCH 2/2] [InstCombine] Handle scalable splats of constants in
 getMinimumFPType

We previously handled ConstantExpr scalable splats in 5d929794a87602cfd873381e11cc99149196bb49, but only fpexts.

ConstantExpr fpexts have since been removed, and simultaneously we didn't handle splats of constants that weren't extended.

This updates it to remove the fpext check and instead see if we can shrink the result of getSplatValue.

Note that the test case doesn't get completely folded away due to #132922
---
 llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp  | 11 ++++++-----
 .../Transforms/InstCombine/scalable-const-fp-splat.ll |  5 ++---
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
index 4ec1af394464b..3faaf1e52db26 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
@@ -1685,11 +1685,12 @@ static Type *getMinimumFPType(Value *V, bool PreferBFloat) {
       return T;
 
   // We can only correctly find a minimum type for a scalable vector when it is
-  // a splat. For splats of constant values the fpext is wrapped up as a
-  // ConstantExpr.
-  if (auto *FPCExt = dyn_cast<ConstantExpr>(V))
-    if (FPCExt->getOpcode() == Instruction::FPExt)
-      return FPCExt->getOperand(0)->getType();
+  // a splat.
+  if (auto *FPCE = dyn_cast<ConstantExpr>(V))
+    if (isa<ScalableVectorType>(V->getType()))
+      if (auto *Splat = dyn_cast<ConstantFP>(FPCE->getSplatValue()))
+        if (Type *T = shrinkFPConstant(Splat, PreferBFloat))
+          return T;
 
   // Try to shrink a vector of FP constants. This returns nullptr on scalable
   // vectors
diff --git a/llvm/test/Transforms/InstCombine/scalable-const-fp-splat.ll b/llvm/test/Transforms/InstCombine/scalable-const-fp-splat.ll
index 3648a868fe1cb..595486361d16e 100644
--- a/llvm/test/Transforms/InstCombine/scalable-const-fp-splat.ll
+++ b/llvm/test/Transforms/InstCombine/scalable-const-fp-splat.ll
@@ -17,9 +17,8 @@ define <vscale x 2 x float> @shrink_splat_scalable_extend(<vscale x 2 x float> %
 define <vscale x 2 x float> @shrink_splat_scalable_extend_rhs_constexpr(<vscale x 2 x float> %a) {
 ; CHECK-LABEL: define <vscale x 2 x float> @shrink_splat_scalable_extend_rhs_constexpr(
 ; CHECK-SAME: <vscale x 2 x float> [[A:%.*]]) {
-; CHECK-NEXT:    [[TMP1:%.*]] = fpext <vscale x 2 x float> [[A]] to <vscale x 2 x double>
-; CHECK-NEXT:    [[TMP2:%.*]] = fadd <vscale x 2 x double> [[TMP1]], splat (double -1.000000e+00)
-; CHECK-NEXT:    [[TMP3:%.*]] = fptrunc <vscale x 2 x double> [[TMP2]] to <vscale x 2 x float>
+; CHECK-NEXT:    [[TMP1:%.*]] = fptrunc <vscale x 2 x double> splat (double -1.000000e+00) to <vscale x 2 x float>
+; CHECK-NEXT:    [[TMP3:%.*]] = fadd <vscale x 2 x float> [[A]], [[TMP1]]
 ; CHECK-NEXT:    ret <vscale x 2 x float> [[TMP3]]
 ;
   %2 = fpext <vscale x 2 x float> %a to <vscale x 2 x double>