[llvm] r217965 - [ARM, Fix] Fix emitLeading/TrailingFence on old ARM processors
Robin Morisset
morisset at google.com
Wed Sep 17 10:41:16 PDT 2014
Author: morisset
Date: Wed Sep 17 12:41:16 2014
New Revision: 217965
URL: http://llvm.org/viewvc/llvm-project?rev=217965&view=rev
Log:
[ARM, Fix] Fix emitLeading/TrailingFence on old ARM processors
Summary:
I had only tested this code for ARMv7 and ARMv8. This patch adds several
fallback paths if the processor does not support dmb ish:
- dmb sy if a cortex-M with support for dmb
- mcr p15, #0, r0, c7, c10, #5 for ARMv6 (special instruction equivalent to a DMB)
These fallback paths were chosen based on the code for fence seq_cst.
Thanks to luqmana for having noticed this bug.
Test Plan: Added more cases to atomic-load-store.ll + make check-all
Reviewers: jfb, t.p.northover, luqmana
Subscribers: aemerson, llvm-commits
Differential Revision: http://reviews.llvm.org/D5304
Modified:
llvm/trunk/lib/Target/ARM/ARMISelLowering.cpp
llvm/trunk/lib/Target/ARM/ARMISelLowering.h
llvm/trunk/test/CodeGen/ARM/atomic-load-store.ll
Modified: llvm/trunk/lib/Target/ARM/ARMISelLowering.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/ARM/ARMISelLowering.cpp?rev=217965&r1=217964&r2=217965&view=diff
==============================================================================
--- llvm/trunk/lib/Target/ARM/ARMISelLowering.cpp (original)
+++ llvm/trunk/lib/Target/ARM/ARMISelLowering.cpp Wed Sep 17 12:41:16 2014
@@ -10984,11 +10984,33 @@ bool ARMTargetLowering::shouldConvertCon
bool ARMTargetLowering::hasLoadLinkedStoreConditional() const { return true; }
-static void makeDMB(IRBuilder<> &Builder, ARM_MB::MemBOpt Domain) {
+Instruction* ARMTargetLowering::makeDMB(IRBuilder<> &Builder,
+ ARM_MB::MemBOpt Domain) const {
Module *M = Builder.GetInsertBlock()->getParent()->getParent();
- Function *DMB = llvm::Intrinsic::getDeclaration(M, Intrinsic::arm_dmb);
- Constant *CDomain = Builder.getInt32(Domain);
- Builder.CreateCall(DMB, CDomain);
+
+ // First, if the target has no DMB, see what fallback we can use.
+ if (!Subtarget->hasDataBarrier()) {
+ // Some ARMv6 cpus can support data barriers with an mcr instruction.
+ // Thumb1 and pre-v6 ARM mode use a libcall instead and should never get
+ // here.
+ if (Subtarget->hasV6Ops() && !Subtarget->isThumb()) {
+ Function *MCR = llvm::Intrinsic::getDeclaration(M, Intrinsic::arm_mcr);
+ ArrayRef<Value*> args = {Builder.getInt32(15), Builder.getInt32(0),
+ Builder.getInt32(0), Builder.getInt32(7),
+ Builder.getInt32(10), Builder.getInt32(5)};
+ return Builder.CreateCall(MCR, args);
+ } else {
+ // Instead of using barriers, atomic accesses on these subtargets use
+ // libcalls.
+ llvm_unreachable("makeDMB on a target so old that it has no barriers");
+ }
+ } else {
+ Function *DMB = llvm::Intrinsic::getDeclaration(M, Intrinsic::arm_dmb);
+ // Only a full system barrier exists in the M-class architectures.
+ Domain = Subtarget->isMClass() ? ARM_MB::SY : Domain;
+ Constant *CDomain = Builder.getInt32(Domain);
+ return Builder.CreateCall(DMB, CDomain);
+ }
}
// Based on http://www.cl.cam.ac.uk/~pes20/cpp/cpp0xmappings.html
Modified: llvm/trunk/lib/Target/ARM/ARMISelLowering.h
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/ARM/ARMISelLowering.h?rev=217965&r1=217964&r2=217965&view=diff
==============================================================================
--- llvm/trunk/lib/Target/ARM/ARMISelLowering.h (original)
+++ llvm/trunk/lib/Target/ARM/ARMISelLowering.h Wed Sep 17 12:41:16 2014
@@ -393,6 +393,7 @@ namespace llvm {
Type *Ty, CallingConv::ID CallConv, bool isVarArg) const override;
bool hasLoadLinkedStoreConditional() const override;
+ Instruction *makeDMB(IRBuilder<> &Builder, ARM_MB::MemBOpt Domain) const;
Value *emitLoadLinked(IRBuilder<> &Builder, Value *Addr,
AtomicOrdering Ord) const override;
Value *emitStoreConditional(IRBuilder<> &Builder, Value *Val,
Modified: llvm/trunk/test/CodeGen/ARM/atomic-load-store.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/ARM/atomic-load-store.ll?rev=217965&r1=217964&r2=217965&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/ARM/atomic-load-store.ll (original)
+++ llvm/trunk/test/CodeGen/ARM/atomic-load-store.ll Wed Sep 17 12:41:16 2014
@@ -3,6 +3,8 @@
; RUN: llc < %s -mtriple=thumbv7-apple-ios -verify-machineinstrs | FileCheck %s -check-prefix=THUMBTWO
; RUN: llc < %s -mtriple=thumbv6-apple-ios | FileCheck %s -check-prefix=THUMBONE
; RUN: llc < %s -mtriple=armv4-apple-ios | FileCheck %s -check-prefix=ARMV4
+; RUN: llc < %s -mtriple=armv6-apple-ios | FileCheck %s -check-prefix=ARMV6
+; RUN: llc < %s -mtriple=thumbv7m-apple-ios | FileCheck %s -check-prefix=THUMBM
define void @test1(i32* %ptr, i32 %val1) {
; ARM-LABEL: test1
@@ -15,6 +17,14 @@ define void @test1(i32* %ptr, i32 %val1)
; THUMBTWO: dmb {{ish$}}
; THUMBTWO-NEXT: str
; THUMBTWO-NEXT: dmb {{ish$}}
+; ARMV6-LABEL: test1
+; ARMV6: mcr p15, #0, {{r[0-9]*}}, c7, c10, #5
+; ARMV6: str
+; ARMV6: mcr p15, #0, {{r[0-9]*}}, c7, c10, #5
+; THUMBM-LABEL: test1
+; THUMBM: dmb sy
+; THUMBM: str
+; THUMBM: dmb sy
store atomic i32 %val1, i32* %ptr seq_cst, align 4
ret void
}
@@ -28,6 +38,12 @@ define i32 @test2(i32* %ptr) {
; THUMBTWO-LABEL: test2
; THUMBTWO: ldr
; THUMBTWO-NEXT: dmb {{ish$}}
+; ARMV6-LABEL: test2
+; ARMV6: ldr
+; ARMV6: mcr p15, #0, {{r[0-9]*}}, c7, c10, #5
+; THUMBM-LABEL: test2
+; THUMBM: ldr
+; THUMBM: dmb sy
%val = load atomic i32* %ptr seq_cst, align 4
ret i32 %val
}
@@ -55,6 +71,11 @@ define void @test3(i8* %ptr1, i8* %ptr2)
; THUMBONE-NOT: dmb
; THUMBONE: strb
; THUMBONE-NOT: dmb
+
+; ARMV6-LABEL: test3
+; ARMV6-NOT: mcr
+; THUMBM-LABEL: test3
+; THUMBM-NOT: dmb sy
%val = load atomic i8* %ptr1 unordered, align 1
store atomic i8 %val, i8* %ptr2 unordered, align 1
ret void
@@ -64,6 +85,8 @@ define void @test4(i8* %ptr1, i8* %ptr2)
; THUMBONE-LABEL: test4
; THUMBONE: ___sync_val_compare_and_swap_1
; THUMBONE: ___sync_lock_test_and_set_1
+; ARMV6-LABEL: test4
+; THUMBM-LABEL: test4
%val = load atomic i8* %ptr1 seq_cst, align 1
store atomic i8 %val, i8* %ptr2 seq_cst, align 1
ret void
More information about the llvm-commits
mailing list