From ffd039bde1a98a6af3528216cf1a069120d8a4bf Mon Sep 17 00:00:00 2001 From: Sanjay Patel Date: Tue, 3 Feb 2015 17:13:04 +0000 Subject: [PATCH] Fix program crashes due to alignment exceptions generated for SSE memop instructions (PR22371). r224330 introduced a bug by misinterpreting the "FeatureVectorUAMem" bit. The commit log says that change did not affect anything, but that's not correct. That change allowed SSE instructions to have unaligned mem operands folded into math ops, and that's not allowed in the default specification for any SSE variant. The bug is exposed when compiling for an AVX-capable CPU that had this feature flag but without enabling AVX codegen. Another mistake in r224330 was not adding the feature flag to all AVX CPUs; the AMD chips were excluded. This is part of the fix for PR22371 ( http://llvm.org/bugs/show_bug.cgi?id=22371 ). This feature bit is SSE-specific, so I've renamed it to "FeatureSSEUnalignedMem". Changed the existing test case for the feature bit to reflect the new name and renamed the test file itself to better reflect the feature. Added runs to fold-vex.ll to check for the failing codegen. Note that the feature bit is not set by default on any CPU because it may require a configuration register setting to enable the enhanced unaligned behavior. llvm-svn: 227983 --- llvm/lib/Target/X86/X86.td | 14 +++++--------- llvm/lib/Target/X86/X86InstrFragmentsSIMD.td | 2 +- llvm/lib/Target/X86/X86Subtarget.cpp | 2 +- llvm/lib/Target/X86/X86Subtarget.h | 8 ++++---- llvm/test/CodeGen/X86/fold-vex.ll | 17 ++++++++++++++--- ...mFeature.ll => sse-unaligned-mem-feature.ll} | 6 ++++-- 6 files changed, 29 insertions(+), 20 deletions(-) rename llvm/test/CodeGen/X86/{2010-01-07-UAMemFeature.ll => sse-unaligned-mem-feature.ll} (74%) diff --git a/llvm/lib/Target/X86/X86.td b/llvm/lib/Target/X86/X86.td index ab3319afe93f..30b3b2876b8a 100644 --- a/llvm/lib/Target/X86/X86.td +++ b/llvm/lib/Target/X86/X86.td @@ -132,9 +132,9 @@ def FeatureFMA4 : SubtargetFeature<"fma4", "HasFMA4", "true", def FeatureXOP : SubtargetFeature<"xop", "HasXOP", "true", "Enable XOP instructions", [FeatureFMA4]>; -def FeatureVectorUAMem : SubtargetFeature<"vector-unaligned-mem", - "HasVectorUAMem", "true", - "Allow unaligned memory operands on vector/SIMD instructions">; +def FeatureSSEUnalignedMem : SubtargetFeature<"sse-unaligned-mem", + "HasSSEUnalignedMem", "true", + "Allow unaligned memory operands with SSE instructions">; def FeatureAES : SubtargetFeature<"aes", "HasAES", "true", "Enable AES instructions", [FeatureSSE2]>; @@ -309,7 +309,6 @@ class SandyBridgeProc : ProcessorModel : ProcessorModel : ProcessorModel : ProcessorModel : ProcessorModel; + FeatureSlowIncDec]>; def : KnightsLandingProc<"knl">; // FIXME: define SKX model @@ -399,7 +395,7 @@ class SkylakeProc : ProcessorModel; + FeatureSlowIncDec, FeatureSGX]>; def : SkylakeProc<"skylake">; def : SkylakeProc<"skx">; // Legacy alias. diff --git a/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td b/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td index 85768aeb4e47..e5de404ed218 100644 --- a/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td +++ b/llvm/lib/Target/X86/X86InstrFragmentsSIMD.td @@ -433,7 +433,7 @@ def alignedloadv8i64 : PatFrag<(ops node:$ptr), // setting a feature bit in the processor (on startup, for example). // Opteron 10h and later implement such a feature. def memop : PatFrag<(ops node:$ptr), (load node:$ptr), [{ - return Subtarget->hasVectorUAMem() + return Subtarget->hasSSEUnalignedMem() || cast(N)->getAlignment() >= 16; }]>; diff --git a/llvm/lib/Target/X86/X86Subtarget.cpp b/llvm/lib/Target/X86/X86Subtarget.cpp index 01889e887c2a..e90da0f98a65 100644 --- a/llvm/lib/Target/X86/X86Subtarget.cpp +++ b/llvm/lib/Target/X86/X86Subtarget.cpp @@ -265,7 +265,7 @@ void X86Subtarget::initializeEnvironment() { IsSHLDSlow = false; IsUAMemFast = false; IsUAMem32Slow = false; - HasVectorUAMem = false; + HasSSEUnalignedMem = false; HasCmpxchg16b = false; UseLeaForSP = false; HasSlowDivide32 = false; diff --git a/llvm/lib/Target/X86/X86Subtarget.h b/llvm/lib/Target/X86/X86Subtarget.h index 27dec6596435..417f1332c546 100644 --- a/llvm/lib/Target/X86/X86Subtarget.h +++ b/llvm/lib/Target/X86/X86Subtarget.h @@ -162,9 +162,9 @@ protected: /// True if unaligned 32-byte memory accesses are slow. bool IsUAMem32Slow; - /// HasVectorUAMem - True if SIMD operations can have unaligned memory - /// operands. This may require setting a feature bit in the processor. - bool HasVectorUAMem; + /// True if SSE operations can have unaligned memory operands. + /// This may require setting a configuration bit in the processor. + bool HasSSEUnalignedMem; /// HasCmpxchg16b - True if this processor has the CMPXCHG16B instruction; /// this is true for most x86-64 chips, but not the first AMD chips. @@ -375,7 +375,7 @@ public: bool isSHLDSlow() const { return IsSHLDSlow; } bool isUnalignedMemAccessFast() const { return IsUAMemFast; } bool isUnalignedMem32Slow() const { return IsUAMem32Slow; } - bool hasVectorUAMem() const { return HasVectorUAMem; } + bool hasSSEUnalignedMem() const { return HasSSEUnalignedMem; } bool hasCmpxchg16b() const { return HasCmpxchg16b; } bool useLeaForSP() const { return UseLeaForSP; } bool hasSlowDivide32() const { return HasSlowDivide32; } diff --git a/llvm/test/CodeGen/X86/fold-vex.ll b/llvm/test/CodeGen/X86/fold-vex.ll index a0c5e22b1c06..5a8b1d8cbfdf 100644 --- a/llvm/test/CodeGen/X86/fold-vex.ll +++ b/llvm/test/CodeGen/X86/fold-vex.ll @@ -1,12 +1,18 @@ ; Use CPU parameters to ensure that a CPU-specific attribute is not overriding the AVX definition. -; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=corei7-avx | FileCheck %s -; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=btver2 | FileCheck %s -; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=corei7-avx | FileCheck %s +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=btver2 | FileCheck %s +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=-avx | FileCheck %s --check-prefix=SSE +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=corei7-avx -mattr=-avx | FileCheck %s --check-prefix=SSE +; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=btver2 -mattr=-avx | FileCheck %s --check-prefix=SSE ; No need to load unaligned operand from memory using an explicit instruction with AVX. ; The operand should be folded into the AND instr. +; With SSE, folding memory operands into math/logic ops requires 16-byte alignment +; unless specially configured on some CPUs such as AMD Family 10H. + define <4 x i32> @test1(<4 x i32>* %p0, <4 x i32> %in1) nounwind { %in0 = load <4 x i32>* %p0, align 2 %a = and <4 x i32> %in0, %in1 @@ -16,5 +22,10 @@ define <4 x i32> @test1(<4 x i32>* %p0, <4 x i32> %in1) nounwind { ; CHECK-NOT: vmovups ; CHECK: vandps (%rdi), %xmm0, %xmm0 ; CHECK-NEXT: ret + +; SSE-LABEL: @test1 +; SSE: movups (%rdi), %xmm1 +; SSE-NEXT: andps %xmm1, %xmm0 +; SSE-NEXT: ret } diff --git a/llvm/test/CodeGen/X86/2010-01-07-UAMemFeature.ll b/llvm/test/CodeGen/X86/sse-unaligned-mem-feature.ll similarity index 74% rename from llvm/test/CodeGen/X86/2010-01-07-UAMemFeature.ll rename to llvm/test/CodeGen/X86/sse-unaligned-mem-feature.ll index bb24adb41817..15f91ee04eaf 100644 --- a/llvm/test/CodeGen/X86/2010-01-07-UAMemFeature.ll +++ b/llvm/test/CodeGen/X86/sse-unaligned-mem-feature.ll @@ -1,5 +1,4 @@ -; RUN: llc -mcpu=yonah -mattr=vector-unaligned-mem -march=x86 < %s | FileCheck %s -; CHECK: addps ( +; RUN: llc -mcpu=yonah -mattr=sse-unaligned-mem -march=x86 < %s | FileCheck %s target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" target triple = "x86_64-unknown-linux-gnu" @@ -8,4 +7,7 @@ define <4 x float> @foo(<4 x float>* %P, <4 x float> %In) nounwind { %A = load <4 x float>* %P, align 4 %B = fadd <4 x float> %A, %In ret <4 x float> %B + +; CHECK-LABEL: @foo +; CHECK: addps ( }