[clang] [llvm] Fix/172104 clang cl simd intrinsics (PR #172116)

[Priyanshu Singh via llvm-commits]

https://github.com/dev-priyanshu15 updated https://github.com/llvm/llvm-project/pull/172116

>From 4d15af1f63bf72dcedd4ff18b9c9a63cf3e7c7e6 Mon Sep 17 00:00:00 2001
From: spriyanshucoder <s.priyanshu.coder at gmail.com>
Date: Sat, 13 Dec 2025 03:50:18 +0530
Subject: [PATCH 1/2] [LLVM][LoopVectorize] Fix SIGFPE crash in
 getPredBlockCostDivisor

When computing block frequency division in getPredBlockCostDivisor,
the function could perform division by zero if BBFreq equals 0.
This caused a floating point exception crash when processing blocks
with zero block frequency information.

Add a guard check to return 1 when BBFreq is 0, avoiding the undefined
behavior while maintaining correct cost model behavior. A divisor of 1
indicates the block is executed at the same frequency as the header,
which is a safe default when frequency information is unavailable.

Fixes issue #172049
---
 .../Transforms/Vectorize/LoopVectorize.cpp    |  4 ++
 .../LoopVectorize/crash-sigfpe-zero-freq.ll   | 39 +++++++++++++++++++
 2 files changed, 43 insertions(+)
 create mode 100644 llvm/test/Transforms/LoopVectorize/crash-sigfpe-zero-freq.ll

diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 0108351f821f4..ed83aa3447bf9 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -2907,6 +2907,10 @@ uint64_t LoopVectorizationCostModel::getPredBlockCostDivisor(
   uint64_t BBFreq = getBFI().getBlockFreq(BB).getFrequency();
   assert(HeaderFreq >= BBFreq &&
          "Header has smaller block freq than dominated BB?");
+  // Guard against division by zero when BBFreq is 0.
+  // In such cases, return 1 to avoid undefined behavior.
+  if (BBFreq == 0)
+    return 1;
   return std::round((double)HeaderFreq / BBFreq);
 }
 
diff --git a/llvm/test/Transforms/LoopVectorize/crash-sigfpe-zero-freq.ll b/llvm/test/Transforms/LoopVectorize/crash-sigfpe-zero-freq.ll
new file mode 100644
index 0000000000000..3b6f3b67d23c5
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/crash-sigfpe-zero-freq.ll
@@ -0,0 +1,39 @@
+; Test case for crash with Floating point Exception in loop-vectorize pass
+; This test verifies that the loop vectorizer does not crash with SIGFPE
+; when processing blocks with zero block frequency.
+; See issue #172049
+
+; RUN: opt -passes=loop-vectorize -S %s
+
+; ModuleID = 'reduced.ll'
+source_filename = "reduced.ll"
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128-ni:1-p2:32:8:8:32-ni:2"
+target triple = "x86_64-unknown-linux-gnu"
+
+define ptr addrspace(1) @wombat() gc "statepoint-example" {
+bb:
+  br label %bb2
+
+bb1:
+  ret ptr addrspace(1) null
+
+bb2:
+  %phi = phi i64 [ %add, %bb6 ], [ 0, %bb ]
+  br i1 false, label %bb3, label %bb6
+
+bb3:
+  br i1 false, label %bb4, label %bb5, !prof !0
+
+bb4:
+  br label %bb6
+
+bb5:
+  br label %bb6
+
+bb6:
+  %add = add i64 %phi, 1
+  %icmp = icmp eq i64 %phi, 0
+  br i1 %icmp, label %bb2, label %bb1
+}
+
+!0 = !{!"branch_weights", i32 1, i32 0}

>From b42e14eb03dcdd440037a0b03efd5b72652954fd Mon Sep 17 00:00:00 2001
From: spriyanshucoder <s.priyanshu.coder at gmail.com>
Date: Sat, 13 Dec 2025 06:09:51 +0530
Subject: [PATCH 2/2] [clang] Enable SSE4.1 intrinsics for Windows MSVC targets
 - Issue #172104

---
 clang/lib/Basic/Targets/X86.cpp                |  8 ++++++++
 clang/test/Sema/simd-intrinsic-sse41-default.c | 16 ++++++++++++++++
 2 files changed, 24 insertions(+)
 create mode 100644 clang/test/Sema/simd-intrinsic-sse41-default.c

diff --git a/clang/lib/Basic/Targets/X86.cpp b/clang/lib/Basic/Targets/X86.cpp
index f00d435937b92..157ddd8aca8ad 100644
--- a/clang/lib/Basic/Targets/X86.cpp
+++ b/clang/lib/Basic/Targets/X86.cpp
@@ -157,6 +157,14 @@ bool X86TargetInfo::initFeatureMap(
   if (getTriple().getArch() == llvm::Triple::x86_64)
     setFeatureEnabled(Features, "sse2", true);
 
+  // Enable SSE4.1 for Windows MSVC targets to support SIMD intrinsics like
+  // _mm_mullo_epi32 without requiring explicit /arch: flags.
+  if ((getTriple().getArch() == llvm::Triple::x86_64 ||
+       getTriple().getArch() == llvm::Triple::x86) &&
+      getTriple().isWindowsMSVCEnvironment()) {
+    setFeatureEnabled(Features, "sse4.1", true);
+  }
+
   using namespace llvm::X86;
 
   SmallVector<StringRef, 16> CPUFeatures;
diff --git a/clang/test/Sema/simd-intrinsic-sse41-default.c b/clang/test/Sema/simd-intrinsic-sse41-default.c
new file mode 100644
index 0000000000000..aa5f4729da8d0
--- /dev/null
+++ b/clang/test/Sema/simd-intrinsic-sse41-default.c
@@ -0,0 +1,16 @@
+// RUN: %clang_cc1 -x c -triple i386-pc-windows-msvc -fsyntax-only %s
+// RUN: %clang_cc1 -x c -triple x86_64-pc-windows-msvc -fsyntax-only %s
+// RUN: %clang_cc1 -x c -triple i386-unknown-linux-gnu -fsyntax-only %s
+// RUN: %clang_cc1 -x c -triple x86_64-unknown-linux-gnu -fsyntax-only %s
+
+// This test verifies that SSE4.1 intrinsics are available by default
+// without requiring explicit /arch: or -msse4.1 flags.
+
+#include <immintrin.h>
+
+void test_sse41_intrinsics(void) {
+  __m128i a = _mm_set1_epi32(1);
+  __m128i b = _mm_set1_epi32(2);
+  __m128i result = _mm_mullo_epi32(a, b);
+  (void)result;
+}