2018-03-30 08:47:31 +08:00
|
|
|
//===- IRBuilder.cpp - Builder for LLVM Instrs ----------------------------===//
|
2009-12-29 05:28:46 +08:00
|
|
|
//
|
2019-01-19 16:50:56 +08:00
|
|
|
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
|
|
|
|
// See https://llvm.org/LICENSE.txt for license information.
|
|
|
|
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
|
2009-12-29 05:28:46 +08:00
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
//
|
|
|
|
// This file implements the IRBuilder class, which is used as a convenient way
|
|
|
|
// to create LLVM instructions with a consistent and simplified interface.
|
|
|
|
//
|
|
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
|
2017-06-06 19:49:48 +08:00
|
|
|
#include "llvm/IR/IRBuilder.h"
|
2018-03-30 08:47:31 +08:00
|
|
|
#include "llvm/ADT/ArrayRef.h"
|
|
|
|
#include "llvm/ADT/None.h"
|
|
|
|
#include "llvm/IR/Constant.h"
|
|
|
|
#include "llvm/IR/Constants.h"
|
|
|
|
#include "llvm/IR/DerivedTypes.h"
|
2013-01-02 19:36:10 +08:00
|
|
|
#include "llvm/IR/Function.h"
|
2018-03-30 08:47:31 +08:00
|
|
|
#include "llvm/IR/GlobalValue.h"
|
2013-01-02 19:36:10 +08:00
|
|
|
#include "llvm/IR/GlobalVariable.h"
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
#include "llvm/IR/IntrinsicInst.h"
|
2013-01-02 19:36:10 +08:00
|
|
|
#include "llvm/IR/Intrinsics.h"
|
|
|
|
#include "llvm/IR/LLVMContext.h"
|
2018-03-30 08:47:31 +08:00
|
|
|
#include "llvm/IR/Operator.h"
|
Extend the statepoint intrinsic to allow statepoints to be marked as transitions from GC-aware code to code that is not GC-aware.
This changes the shape of the statepoint intrinsic from:
@llvm.experimental.gc.statepoint(anyptr target, i32 # call args, i32 unused, ...call args, i32 # deopt args, ...deopt args, ...gc args)
to:
@llvm.experimental.gc.statepoint(anyptr target, i32 # call args, i32 flags, ...call args, i32 # transition args, ...transition args, i32 # deopt args, ...deopt args, ...gc args)
This extension offers the backend the opportunity to insert (somewhat) arbitrary code to manage the transition from GC-aware code to code that is not GC-aware and back.
In order to support the injection of transition code, this extension wraps the STATEPOINT ISD node generated by the usual lowering lowering with two additional nodes: GC_TRANSITION_START and GC_TRANSITION_END. The transition arguments that were passed passed to the intrinsic (if any) are lowered and provided as operands to these nodes and may be used by the backend during code generation.
Eventually, the lowering of the GC_TRANSITION_{START,END} nodes should be informed by the GC strategy in use for the function containing the intrinsic call; for now, these nodes are instead replaced with no-ops.
Differential Revision: http://reviews.llvm.org/D9501
llvm-svn: 236888
2015-05-09 02:07:42 +08:00
|
|
|
#include "llvm/IR/Statepoint.h"
|
2018-03-30 08:47:31 +08:00
|
|
|
#include "llvm/IR/Type.h"
|
|
|
|
#include "llvm/IR/Value.h"
|
|
|
|
#include "llvm/Support/Casting.h"
|
|
|
|
#include "llvm/Support/MathExtras.h"
|
|
|
|
#include <cassert>
|
|
|
|
#include <cstdint>
|
|
|
|
#include <vector>
|
|
|
|
|
2009-12-29 05:28:46 +08:00
|
|
|
using namespace llvm;
|
|
|
|
|
|
|
|
/// CreateGlobalString - Make a new global variable with an initializer that
|
2010-02-11 04:04:19 +08:00
|
|
|
/// has array of i8 type filled in with the nul terminated string value
|
2009-12-29 05:28:46 +08:00
|
|
|
/// specified. If Name is specified, it is the name of the global variable
|
|
|
|
/// created.
|
2015-04-04 05:33:42 +08:00
|
|
|
GlobalVariable *IRBuilderBase::CreateGlobalString(StringRef Str,
|
2015-06-19 10:12:07 +08:00
|
|
|
const Twine &Name,
|
|
|
|
unsigned AddressSpace) {
|
2012-02-05 10:29:43 +08:00
|
|
|
Constant *StrConstant = ConstantDataArray::getString(Context, Str);
|
2009-12-29 05:28:46 +08:00
|
|
|
Module &M = *BB->getParent()->getParent();
|
2018-03-30 08:47:31 +08:00
|
|
|
auto *GV = new GlobalVariable(M, StrConstant->getType(), true,
|
|
|
|
GlobalValue::PrivateLinkage, StrConstant, Name,
|
|
|
|
nullptr, GlobalVariable::NotThreadLocal,
|
|
|
|
AddressSpace);
|
2016-06-15 05:01:22 +08:00
|
|
|
GV->setUnnamedAddr(GlobalValue::UnnamedAddr::Global);
|
2019-10-15 19:24:36 +08:00
|
|
|
GV->setAlignment(Align::None());
|
2009-12-29 05:28:46 +08:00
|
|
|
return GV;
|
|
|
|
}
|
2009-12-29 05:45:40 +08:00
|
|
|
|
2011-07-12 12:14:22 +08:00
|
|
|
Type *IRBuilderBase::getCurrentFunctionReturnType() const {
|
2009-12-29 05:50:56 +08:00
|
|
|
assert(BB && BB->getParent() && "No current function!");
|
|
|
|
return BB->getParent()->getReturnType();
|
|
|
|
}
|
2010-12-27 06:49:25 +08:00
|
|
|
|
|
|
|
Value *IRBuilderBase::getCastedInt8PtrValue(Value *Ptr) {
|
2018-03-30 08:47:31 +08:00
|
|
|
auto *PT = cast<PointerType>(Ptr->getType());
|
2010-12-27 06:49:25 +08:00
|
|
|
if (PT->getElementType()->isIntegerTy(8))
|
|
|
|
return Ptr;
|
2018-07-03 20:39:52 +08:00
|
|
|
|
2010-12-27 06:49:25 +08:00
|
|
|
// Otherwise, we need to insert a bitcast.
|
|
|
|
PT = getInt8PtrTy(PT->getAddressSpace());
|
|
|
|
BitCastInst *BCI = new BitCastInst(Ptr, PT, "");
|
|
|
|
BB->getInstList().insert(InsertPt, BCI);
|
|
|
|
SetInstDebugLocation(BCI);
|
|
|
|
return BCI;
|
|
|
|
}
|
|
|
|
|
2019-02-02 04:43:25 +08:00
|
|
|
static CallInst *createCallHelper(Function *Callee, ArrayRef<Value *> Ops,
|
2014-12-30 13:55:58 +08:00
|
|
|
IRBuilderBase *Builder,
|
2018-02-24 05:16:12 +08:00
|
|
|
const Twine &Name = "",
|
|
|
|
Instruction *FMFSource = nullptr) {
|
2014-12-30 13:55:58 +08:00
|
|
|
CallInst *CI = CallInst::Create(Callee, Ops, Name);
|
2018-02-24 05:16:12 +08:00
|
|
|
if (FMFSource)
|
|
|
|
CI->copyFastMathFlags(FMFSource);
|
2010-12-27 06:49:25 +08:00
|
|
|
Builder->GetInsertBlock()->getInstList().insert(Builder->GetInsertPoint(),CI);
|
|
|
|
Builder->SetInstDebugLocation(CI);
|
2018-07-03 20:39:52 +08:00
|
|
|
return CI;
|
2010-12-27 06:49:25 +08:00
|
|
|
}
|
|
|
|
|
2019-02-02 04:43:34 +08:00
|
|
|
static InvokeInst *createInvokeHelper(Function *Invokee, BasicBlock *NormalDest,
|
2015-05-07 07:53:09 +08:00
|
|
|
BasicBlock *UnwindDest,
|
|
|
|
ArrayRef<Value *> Ops,
|
|
|
|
IRBuilderBase *Builder,
|
|
|
|
const Twine &Name = "") {
|
|
|
|
InvokeInst *II =
|
|
|
|
InvokeInst::Create(Invokee, NormalDest, UnwindDest, Ops, Name);
|
|
|
|
Builder->GetInsertBlock()->getInstList().insert(Builder->GetInsertPoint(),
|
|
|
|
II);
|
|
|
|
Builder->SetInstDebugLocation(II);
|
|
|
|
return II;
|
|
|
|
}
|
|
|
|
|
2019-12-10 00:36:50 +08:00
|
|
|
CallInst *IRBuilderBase::CreateMemSet(Value *Ptr, Value *Val, Value *Size,
|
|
|
|
MaybeAlign Align, bool isVolatile,
|
|
|
|
MDNode *TBAATag, MDNode *ScopeTag,
|
|
|
|
MDNode *NoAliasTag) {
|
2010-12-27 06:49:25 +08:00
|
|
|
Ptr = getCastedInt8PtrValue(Ptr);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
Value *Ops[] = {Ptr, Val, Size, getInt1(isVolatile)};
|
2011-07-12 22:06:48 +08:00
|
|
|
Type *Tys[] = { Ptr->getType(), Size->getType() };
|
2010-12-27 06:49:25 +08:00
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn = Intrinsic::getDeclaration(M, Intrinsic::memset, Tys);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
2011-07-15 16:37:34 +08:00
|
|
|
CallInst *CI = createCallHelper(TheFn, Ops, this);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
2019-12-10 00:36:50 +08:00
|
|
|
if (Align)
|
|
|
|
cast<MemSetInst>(CI)->setDestAlignment(Align->value());
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
2010-12-27 06:49:25 +08:00
|
|
|
// Set the TBAA info if present.
|
|
|
|
if (TBAATag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa, TBAATag);
|
Add scoped-noalias metadata
This commit adds scoped noalias metadata. The primary motivations for this
feature are:
1. To preserve noalias function attribute information when inlining
2. To provide the ability to model block-scope C99 restrict pointers
Neither of these two abilities are added here, only the necessary
infrastructure. In fact, there should be no change to existing functionality,
only the addition of new features. The logic that converts noalias function
parameters into this metadata during inlining will come in a follow-up commit.
What is added here is the ability to generally specify noalias memory-access
sets. Regarding the metadata, alias-analysis scopes are defined similar to TBAA
nodes:
!scope0 = metadata !{ metadata !"scope of foo()" }
!scope1 = metadata !{ metadata !"scope 1", metadata !scope0 }
!scope2 = metadata !{ metadata !"scope 2", metadata !scope0 }
!scope3 = metadata !{ metadata !"scope 2.1", metadata !scope2 }
!scope4 = metadata !{ metadata !"scope 2.2", metadata !scope2 }
Loads and stores can be tagged with an alias-analysis scope, and also, with a
noalias tag for a specific scope:
... = load %ptr1, !alias.scope !{ !scope1 }
... = load %ptr2, !alias.scope !{ !scope1, !scope2 }, !noalias !{ !scope1 }
When evaluating an aliasing query, if one of the instructions is associated
with an alias.scope id that is identical to the noalias scope associated with
the other instruction, or is a descendant (in the scope hierarchy) of the
noalias scope associated with the other instruction, then the two memory
accesses are assumed not to alias.
Note that is the first element of the scope metadata is a string, then it can
be combined accross functions and translation units. The string can be replaced
by a self-reference to create globally unqiue scope identifiers.
[Note: This overview is slightly stylized, since the metadata nodes really need
to just be numbers (!0 instead of !scope0), and the scope lists are also global
unnamed metadata.]
Existing noalias metadata in a callee is "cloned" for use by the inlined code.
This is necessary because the aliasing scopes are unique to each call site
(because of possible control dependencies on the aliasing properties). For
example, consider a function: foo(noalias a, noalias b) { *a = *b; } that gets
inlined into bar() { ... if (...) foo(a1, b1); ... if (...) foo(a2, b2); } --
now just because we know that a1 does not alias with b1 at the first call site,
and a2 does not alias with b2 at the second call site, we cannot let inlining
these functons have the metadata imply that a1 does not alias with b2.
llvm-svn: 213864
2014-07-24 22:25:39 +08:00
|
|
|
|
|
|
|
if (ScopeTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_alias_scope, ScopeTag);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
Add scoped-noalias metadata
This commit adds scoped noalias metadata. The primary motivations for this
feature are:
1. To preserve noalias function attribute information when inlining
2. To provide the ability to model block-scope C99 restrict pointers
Neither of these two abilities are added here, only the necessary
infrastructure. In fact, there should be no change to existing functionality,
only the addition of new features. The logic that converts noalias function
parameters into this metadata during inlining will come in a follow-up commit.
What is added here is the ability to generally specify noalias memory-access
sets. Regarding the metadata, alias-analysis scopes are defined similar to TBAA
nodes:
!scope0 = metadata !{ metadata !"scope of foo()" }
!scope1 = metadata !{ metadata !"scope 1", metadata !scope0 }
!scope2 = metadata !{ metadata !"scope 2", metadata !scope0 }
!scope3 = metadata !{ metadata !"scope 2.1", metadata !scope2 }
!scope4 = metadata !{ metadata !"scope 2.2", metadata !scope2 }
Loads and stores can be tagged with an alias-analysis scope, and also, with a
noalias tag for a specific scope:
... = load %ptr1, !alias.scope !{ !scope1 }
... = load %ptr2, !alias.scope !{ !scope1, !scope2 }, !noalias !{ !scope1 }
When evaluating an aliasing query, if one of the instructions is associated
with an alias.scope id that is identical to the noalias scope associated with
the other instruction, or is a descendant (in the scope hierarchy) of the
noalias scope associated with the other instruction, then the two memory
accesses are assumed not to alias.
Note that is the first element of the scope metadata is a string, then it can
be combined accross functions and translation units. The string can be replaced
by a self-reference to create globally unqiue scope identifiers.
[Note: This overview is slightly stylized, since the metadata nodes really need
to just be numbers (!0 instead of !scope0), and the scope lists are also global
unnamed metadata.]
Existing noalias metadata in a callee is "cloned" for use by the inlined code.
This is necessary because the aliasing scopes are unique to each call site
(because of possible control dependencies on the aliasing properties). For
example, consider a function: foo(noalias a, noalias b) { *a = *b; } that gets
inlined into bar() { ... if (...) foo(a1, b1); ... if (...) foo(a2, b2); } --
now just because we know that a1 does not alias with b1 at the first call site,
and a2 does not alias with b2 at the second call site, we cannot let inlining
these functons have the metadata imply that a1 does not alias with b2.
llvm-svn: 213864
2014-07-24 22:25:39 +08:00
|
|
|
if (NoAliasTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_noalias, NoAliasTag);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
2010-12-27 06:49:25 +08:00
|
|
|
return CI;
|
|
|
|
}
|
|
|
|
|
2018-05-31 04:02:56 +08:00
|
|
|
CallInst *IRBuilderBase::CreateElementUnorderedAtomicMemSet(
|
2019-12-19 22:41:05 +08:00
|
|
|
Value *Ptr, Value *Val, Value *Size, Align Alignment, uint32_t ElementSize,
|
2018-05-31 04:02:56 +08:00
|
|
|
MDNode *TBAATag, MDNode *ScopeTag, MDNode *NoAliasTag) {
|
|
|
|
|
|
|
|
Ptr = getCastedInt8PtrValue(Ptr);
|
|
|
|
Value *Ops[] = {Ptr, Val, Size, getInt32(ElementSize)};
|
|
|
|
Type *Tys[] = {Ptr->getType(), Size->getType()};
|
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn = Intrinsic::getDeclaration(
|
2018-05-31 04:02:56 +08:00
|
|
|
M, Intrinsic::memset_element_unordered_atomic, Tys);
|
|
|
|
|
|
|
|
CallInst *CI = createCallHelper(TheFn, Ops, this);
|
|
|
|
|
2019-12-19 22:41:05 +08:00
|
|
|
cast<AtomicMemSetInst>(CI)->setDestAlignment(Alignment);
|
2018-05-31 04:02:56 +08:00
|
|
|
|
|
|
|
// Set the TBAA info if present.
|
|
|
|
if (TBAATag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa, TBAATag);
|
|
|
|
|
|
|
|
if (ScopeTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_alias_scope, ScopeTag);
|
|
|
|
|
|
|
|
if (NoAliasTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_noalias, NoAliasTag);
|
|
|
|
|
|
|
|
return CI;
|
|
|
|
}
|
|
|
|
|
2019-12-12 22:32:19 +08:00
|
|
|
CallInst *IRBuilderBase::CreateMemCpy(Value *Dst, unsigned DstAlign, Value *Src,
|
|
|
|
unsigned SrcAlign, Value *Size,
|
|
|
|
bool isVolatile, MDNode *TBAATag,
|
|
|
|
MDNode *TBAAStructTag, MDNode *ScopeTag,
|
|
|
|
MDNode *NoAliasTag) {
|
|
|
|
return CreateMemCpy(Dst, MaybeAlign(DstAlign), Src, MaybeAlign(SrcAlign),
|
|
|
|
Size, isVolatile, TBAATag, TBAAStructTag, ScopeTag,
|
|
|
|
NoAliasTag);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateMemCpy(Value *Dst, MaybeAlign DstAlign,
|
|
|
|
Value *Src, MaybeAlign SrcAlign,
|
|
|
|
Value *Size, bool isVolatile,
|
|
|
|
MDNode *TBAATag, MDNode *TBAAStructTag,
|
|
|
|
MDNode *ScopeTag, MDNode *NoAliasTag) {
|
2010-12-27 06:49:25 +08:00
|
|
|
Dst = getCastedInt8PtrValue(Dst);
|
|
|
|
Src = getCastedInt8PtrValue(Src);
|
|
|
|
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
Value *Ops[] = {Dst, Src, Size, getInt1(isVolatile)};
|
2011-07-12 22:06:48 +08:00
|
|
|
Type *Tys[] = { Dst->getType(), Src->getType(), Size->getType() };
|
2010-12-27 06:49:25 +08:00
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn = Intrinsic::getDeclaration(M, Intrinsic::memcpy, Tys);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
2011-07-15 16:37:34 +08:00
|
|
|
CallInst *CI = createCallHelper(TheFn, Ops, this);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
2018-01-28 01:59:10 +08:00
|
|
|
auto* MCI = cast<MemCpyInst>(CI);
|
2019-12-12 22:32:19 +08:00
|
|
|
if (DstAlign)
|
|
|
|
MCI->setDestAlignment(*DstAlign);
|
|
|
|
if (SrcAlign)
|
|
|
|
MCI->setSourceAlignment(*SrcAlign);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
2010-12-27 06:49:25 +08:00
|
|
|
// Set the TBAA info if present.
|
|
|
|
if (TBAATag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa, TBAATag);
|
2012-09-27 06:17:14 +08:00
|
|
|
|
|
|
|
// Set the TBAA Struct info if present.
|
|
|
|
if (TBAAStructTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa_struct, TBAAStructTag);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
Add scoped-noalias metadata
This commit adds scoped noalias metadata. The primary motivations for this
feature are:
1. To preserve noalias function attribute information when inlining
2. To provide the ability to model block-scope C99 restrict pointers
Neither of these two abilities are added here, only the necessary
infrastructure. In fact, there should be no change to existing functionality,
only the addition of new features. The logic that converts noalias function
parameters into this metadata during inlining will come in a follow-up commit.
What is added here is the ability to generally specify noalias memory-access
sets. Regarding the metadata, alias-analysis scopes are defined similar to TBAA
nodes:
!scope0 = metadata !{ metadata !"scope of foo()" }
!scope1 = metadata !{ metadata !"scope 1", metadata !scope0 }
!scope2 = metadata !{ metadata !"scope 2", metadata !scope0 }
!scope3 = metadata !{ metadata !"scope 2.1", metadata !scope2 }
!scope4 = metadata !{ metadata !"scope 2.2", metadata !scope2 }
Loads and stores can be tagged with an alias-analysis scope, and also, with a
noalias tag for a specific scope:
... = load %ptr1, !alias.scope !{ !scope1 }
... = load %ptr2, !alias.scope !{ !scope1, !scope2 }, !noalias !{ !scope1 }
When evaluating an aliasing query, if one of the instructions is associated
with an alias.scope id that is identical to the noalias scope associated with
the other instruction, or is a descendant (in the scope hierarchy) of the
noalias scope associated with the other instruction, then the two memory
accesses are assumed not to alias.
Note that is the first element of the scope metadata is a string, then it can
be combined accross functions and translation units. The string can be replaced
by a self-reference to create globally unqiue scope identifiers.
[Note: This overview is slightly stylized, since the metadata nodes really need
to just be numbers (!0 instead of !scope0), and the scope lists are also global
unnamed metadata.]
Existing noalias metadata in a callee is "cloned" for use by the inlined code.
This is necessary because the aliasing scopes are unique to each call site
(because of possible control dependencies on the aliasing properties). For
example, consider a function: foo(noalias a, noalias b) { *a = *b; } that gets
inlined into bar() { ... if (...) foo(a1, b1); ... if (...) foo(a2, b2); } --
now just because we know that a1 does not alias with b1 at the first call site,
and a2 does not alias with b2 at the second call site, we cannot let inlining
these functons have the metadata imply that a1 does not alias with b2.
llvm-svn: 213864
2014-07-24 22:25:39 +08:00
|
|
|
if (ScopeTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_alias_scope, ScopeTag);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
Add scoped-noalias metadata
This commit adds scoped noalias metadata. The primary motivations for this
feature are:
1. To preserve noalias function attribute information when inlining
2. To provide the ability to model block-scope C99 restrict pointers
Neither of these two abilities are added here, only the necessary
infrastructure. In fact, there should be no change to existing functionality,
only the addition of new features. The logic that converts noalias function
parameters into this metadata during inlining will come in a follow-up commit.
What is added here is the ability to generally specify noalias memory-access
sets. Regarding the metadata, alias-analysis scopes are defined similar to TBAA
nodes:
!scope0 = metadata !{ metadata !"scope of foo()" }
!scope1 = metadata !{ metadata !"scope 1", metadata !scope0 }
!scope2 = metadata !{ metadata !"scope 2", metadata !scope0 }
!scope3 = metadata !{ metadata !"scope 2.1", metadata !scope2 }
!scope4 = metadata !{ metadata !"scope 2.2", metadata !scope2 }
Loads and stores can be tagged with an alias-analysis scope, and also, with a
noalias tag for a specific scope:
... = load %ptr1, !alias.scope !{ !scope1 }
... = load %ptr2, !alias.scope !{ !scope1, !scope2 }, !noalias !{ !scope1 }
When evaluating an aliasing query, if one of the instructions is associated
with an alias.scope id that is identical to the noalias scope associated with
the other instruction, or is a descendant (in the scope hierarchy) of the
noalias scope associated with the other instruction, then the two memory
accesses are assumed not to alias.
Note that is the first element of the scope metadata is a string, then it can
be combined accross functions and translation units. The string can be replaced
by a self-reference to create globally unqiue scope identifiers.
[Note: This overview is slightly stylized, since the metadata nodes really need
to just be numbers (!0 instead of !scope0), and the scope lists are also global
unnamed metadata.]
Existing noalias metadata in a callee is "cloned" for use by the inlined code.
This is necessary because the aliasing scopes are unique to each call site
(because of possible control dependencies on the aliasing properties). For
example, consider a function: foo(noalias a, noalias b) { *a = *b; } that gets
inlined into bar() { ... if (...) foo(a1, b1); ... if (...) foo(a2, b2); } --
now just because we know that a1 does not alias with b1 at the first call site,
and a2 does not alias with b2 at the second call site, we cannot let inlining
these functons have the metadata imply that a1 does not alias with b2.
llvm-svn: 213864
2014-07-24 22:25:39 +08:00
|
|
|
if (NoAliasTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_noalias, NoAliasTag);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
2018-07-03 20:39:52 +08:00
|
|
|
return CI;
|
2010-12-27 06:49:25 +08:00
|
|
|
}
|
|
|
|
|
2017-06-16 22:43:59 +08:00
|
|
|
CallInst *IRBuilderBase::CreateElementUnorderedAtomicMemCpy(
|
[Alignment][NFC] Use Align with CreateElementUnorderedAtomicMemCpy
Summary:
This is patch is part of a series to introduce an Alignment type.
See this thread for context: http://lists.llvm.org/pipermail/llvm-dev/2019-July/133851.html
See this patch for the introduction of the type: https://reviews.llvm.org/D64790
Reviewers: courbet, nicolasvasilache
Subscribers: hiraditya, jfb, mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, csigg, arpith-jacob, mgester, lucyrfox, herhut, liufengdb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D73041
2020-01-20 21:47:28 +08:00
|
|
|
Value *Dst, Align DstAlign, Value *Src, Align SrcAlign, Value *Size,
|
2017-11-11 03:38:12 +08:00
|
|
|
uint32_t ElementSize, MDNode *TBAATag, MDNode *TBAAStructTag,
|
|
|
|
MDNode *ScopeTag, MDNode *NoAliasTag) {
|
|
|
|
assert(DstAlign >= ElementSize &&
|
|
|
|
"Pointer alignment must be at least element size");
|
|
|
|
assert(SrcAlign >= ElementSize &&
|
|
|
|
"Pointer alignment must be at least element size");
|
2017-06-07 00:45:25 +08:00
|
|
|
Dst = getCastedInt8PtrValue(Dst);
|
|
|
|
Src = getCastedInt8PtrValue(Src);
|
|
|
|
|
2017-06-16 22:43:59 +08:00
|
|
|
Value *Ops[] = {Dst, Src, Size, getInt32(ElementSize)};
|
|
|
|
Type *Tys[] = {Dst->getType(), Src->getType(), Size->getType()};
|
2017-06-07 00:45:25 +08:00
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn = Intrinsic::getDeclaration(
|
2017-06-16 22:43:59 +08:00
|
|
|
M, Intrinsic::memcpy_element_unordered_atomic, Tys);
|
2017-06-07 00:45:25 +08:00
|
|
|
|
|
|
|
CallInst *CI = createCallHelper(TheFn, Ops, this);
|
|
|
|
|
2017-11-11 03:38:12 +08:00
|
|
|
// Set the alignment of the pointer args.
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
auto *AMCI = cast<AtomicMemCpyInst>(CI);
|
|
|
|
AMCI->setDestAlignment(DstAlign);
|
|
|
|
AMCI->setSourceAlignment(SrcAlign);
|
2017-11-11 03:38:12 +08:00
|
|
|
|
2017-06-07 00:45:25 +08:00
|
|
|
// Set the TBAA info if present.
|
|
|
|
if (TBAATag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa, TBAATag);
|
|
|
|
|
|
|
|
// Set the TBAA Struct info if present.
|
|
|
|
if (TBAAStructTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa_struct, TBAAStructTag);
|
|
|
|
|
|
|
|
if (ScopeTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_alias_scope, ScopeTag);
|
|
|
|
|
|
|
|
if (NoAliasTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_noalias, NoAliasTag);
|
|
|
|
|
|
|
|
return CI;
|
|
|
|
}
|
|
|
|
|
2019-12-12 22:32:19 +08:00
|
|
|
CallInst *IRBuilderBase::CreateMemMove(Value *Dst, MaybeAlign DstAlign,
|
|
|
|
Value *Src, MaybeAlign SrcAlign,
|
|
|
|
Value *Size, bool isVolatile,
|
|
|
|
MDNode *TBAATag, MDNode *ScopeTag,
|
|
|
|
MDNode *NoAliasTag) {
|
2010-12-27 06:49:25 +08:00
|
|
|
Dst = getCastedInt8PtrValue(Dst);
|
|
|
|
Src = getCastedInt8PtrValue(Src);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
|
|
|
Value *Ops[] = {Dst, Src, Size, getInt1(isVolatile)};
|
2011-07-12 22:06:48 +08:00
|
|
|
Type *Tys[] = { Dst->getType(), Src->getType(), Size->getType() };
|
2010-12-27 06:49:25 +08:00
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn = Intrinsic::getDeclaration(M, Intrinsic::memmove, Tys);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
2011-07-15 16:37:34 +08:00
|
|
|
CallInst *CI = createCallHelper(TheFn, Ops, this);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
|
|
|
auto *MMI = cast<MemMoveInst>(CI);
|
2019-12-12 22:32:19 +08:00
|
|
|
if (DstAlign)
|
|
|
|
MMI->setDestAlignment(*DstAlign);
|
|
|
|
if (SrcAlign)
|
|
|
|
MMI->setSourceAlignment(*SrcAlign);
|
Remove alignment argument from memcpy/memmove/memset in favour of alignment attributes (Step 1)
Summary:
This is a resurrection of work first proposed and discussed in Aug 2015:
http://lists.llvm.org/pipermail/llvm-dev/2015-August/089384.html
and initially landed (but then backed out) in Nov 2015:
http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20151109/312083.html
The @llvm.memcpy/memmove/memset intrinsics currently have an explicit argument
which is required to be a constant integer. It represents the alignment of the
dest (and source), and so must be the minimum of the actual alignment of the
two.
This change is the first in a series that allows source and dest to each
have their own alignments by using the alignment attribute on their arguments.
In this change we:
1) Remove the alignment argument.
2) Add alignment attributes to the source & dest arguments. We, temporarily,
require that the alignments for source & dest be equal.
For example, code which used to read:
call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dest, i8* %src, i32 100, i32 4, i1 false)
will now read
call void @llvm.memcpy.p0i8.p0i8.i32(i8* align 4 %dest, i8* align 4 %src, i32 100, i1 false)
Downstream users may have to update their lit tests that check for
@llvm.memcpy/memmove/memset call/declaration patterns. The following extended sed script
may help with updating the majority of your tests, but it does not catch all possible
patterns so some manual checking and updating will be required.
s~declare void @llvm\.mem(set|cpy|move)\.p([^(]*)\((.*), i32, i1\)~declare void @llvm.mem\1.p\2(\3, i1)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* \3, i8 \4, i8 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* \3, i8 \4, i16 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* \3, i8 \4, i32 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* \3, i8 \4, i64 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* \3, i8 \4, i128 \5, i1 \6)~g
s~call void @llvm\.memset\.p([^(]*)i8\(i8([^*]*)\* (.*), i8 (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i8(i8\2* align \6 \3, i8 \4, i8 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i16\(i8([^*]*)\* (.*), i8 (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i16(i8\2* align \6 \3, i8 \4, i16 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i32\(i8([^*]*)\* (.*), i8 (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i32(i8\2* align \6 \3, i8 \4, i32 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i64\(i8([^*]*)\* (.*), i8 (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i64(i8\2* align \6 \3, i8 \4, i64 \5, i1 \7)~g
s~call void @llvm\.memset\.p([^(]*)i128\(i8([^*]*)\* (.*), i8 (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.memset.p\1i128(i8\2* align \6 \3, i8 \4, i128 \5, i1 \7)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* \4, i8\5* \6, i8 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* \4, i8\5* \6, i16 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* \4, i8\5* \6, i32 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* \4, i8\5* \6, i64 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 [01], i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* \4, i8\5* \6, i128 \7, i1 \8)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i8\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i8 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i8(i8\3* align \8 \4, i8\5* align \8 \6, i8 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i16\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i16 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i16(i8\3* align \8 \4, i8\5* align \8 \6, i16 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i32\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i32 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i32(i8\3* align \8 \4, i8\5* align \8 \6, i32 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i64\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i64 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i64(i8\3* align \8 \4, i8\5* align \8 \6, i64 \7, i1 \9)~g
s~call void @llvm\.mem(cpy|move)\.p([^(]*)i128\(i8([^*]*)\* (.*), i8([^*]*)\* (.*), i128 (.*), i32 ([0-9]*), i1 ([^)]*)\)~call void @llvm.mem\1.p\2i128(i8\3* align \8 \4, i8\5* align \8 \6, i128 \7, i1 \9)~g
The remaining changes in the series will:
Step 2) Expand the IRBuilder API to allow creation of memcpy/memmove with differing
source and dest alignments.
Step 3) Update Clang to use the new IRBuilder API.
Step 4) Update Polly to use the new IRBuilder API.
Step 5) Update LLVM passes that create memcpy/memmove calls to use the new IRBuilder API,
and those that use use MemIntrinsicInst::[get|set]Alignment() to use
getDestAlignment() and getSourceAlignment() instead.
Step 6) Remove the single-alignment IRBuilder API for memcpy/memmove, and the
MemIntrinsicInst::[get|set]Alignment() methods.
Reviewers: pete, hfinkel, lhames, reames, bollu
Reviewed By: reames
Subscribers: niosHD, reames, jholewinski, qcolombet, jfb, sanjoy, arsenm, dschuff, dylanmckay, mehdi_amini, sdardis, nemanjai, david2050, nhaehnle, javed.absar, sbc100, jgravelle-google, eraman, aheejin, kbarton, JDevlieghere, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, llvm-commits
Differential Revision: https://reviews.llvm.org/D41675
llvm-svn: 322965
2018-01-20 01:13:12 +08:00
|
|
|
|
2010-12-27 06:49:25 +08:00
|
|
|
// Set the TBAA info if present.
|
|
|
|
if (TBAATag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa, TBAATag);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
Add scoped-noalias metadata
This commit adds scoped noalias metadata. The primary motivations for this
feature are:
1. To preserve noalias function attribute information when inlining
2. To provide the ability to model block-scope C99 restrict pointers
Neither of these two abilities are added here, only the necessary
infrastructure. In fact, there should be no change to existing functionality,
only the addition of new features. The logic that converts noalias function
parameters into this metadata during inlining will come in a follow-up commit.
What is added here is the ability to generally specify noalias memory-access
sets. Regarding the metadata, alias-analysis scopes are defined similar to TBAA
nodes:
!scope0 = metadata !{ metadata !"scope of foo()" }
!scope1 = metadata !{ metadata !"scope 1", metadata !scope0 }
!scope2 = metadata !{ metadata !"scope 2", metadata !scope0 }
!scope3 = metadata !{ metadata !"scope 2.1", metadata !scope2 }
!scope4 = metadata !{ metadata !"scope 2.2", metadata !scope2 }
Loads and stores can be tagged with an alias-analysis scope, and also, with a
noalias tag for a specific scope:
... = load %ptr1, !alias.scope !{ !scope1 }
... = load %ptr2, !alias.scope !{ !scope1, !scope2 }, !noalias !{ !scope1 }
When evaluating an aliasing query, if one of the instructions is associated
with an alias.scope id that is identical to the noalias scope associated with
the other instruction, or is a descendant (in the scope hierarchy) of the
noalias scope associated with the other instruction, then the two memory
accesses are assumed not to alias.
Note that is the first element of the scope metadata is a string, then it can
be combined accross functions and translation units. The string can be replaced
by a self-reference to create globally unqiue scope identifiers.
[Note: This overview is slightly stylized, since the metadata nodes really need
to just be numbers (!0 instead of !scope0), and the scope lists are also global
unnamed metadata.]
Existing noalias metadata in a callee is "cloned" for use by the inlined code.
This is necessary because the aliasing scopes are unique to each call site
(because of possible control dependencies on the aliasing properties). For
example, consider a function: foo(noalias a, noalias b) { *a = *b; } that gets
inlined into bar() { ... if (...) foo(a1, b1); ... if (...) foo(a2, b2); } --
now just because we know that a1 does not alias with b1 at the first call site,
and a2 does not alias with b2 at the second call site, we cannot let inlining
these functons have the metadata imply that a1 does not alias with b2.
llvm-svn: 213864
2014-07-24 22:25:39 +08:00
|
|
|
if (ScopeTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_alias_scope, ScopeTag);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
Add scoped-noalias metadata
This commit adds scoped noalias metadata. The primary motivations for this
feature are:
1. To preserve noalias function attribute information when inlining
2. To provide the ability to model block-scope C99 restrict pointers
Neither of these two abilities are added here, only the necessary
infrastructure. In fact, there should be no change to existing functionality,
only the addition of new features. The logic that converts noalias function
parameters into this metadata during inlining will come in a follow-up commit.
What is added here is the ability to generally specify noalias memory-access
sets. Regarding the metadata, alias-analysis scopes are defined similar to TBAA
nodes:
!scope0 = metadata !{ metadata !"scope of foo()" }
!scope1 = metadata !{ metadata !"scope 1", metadata !scope0 }
!scope2 = metadata !{ metadata !"scope 2", metadata !scope0 }
!scope3 = metadata !{ metadata !"scope 2.1", metadata !scope2 }
!scope4 = metadata !{ metadata !"scope 2.2", metadata !scope2 }
Loads and stores can be tagged with an alias-analysis scope, and also, with a
noalias tag for a specific scope:
... = load %ptr1, !alias.scope !{ !scope1 }
... = load %ptr2, !alias.scope !{ !scope1, !scope2 }, !noalias !{ !scope1 }
When evaluating an aliasing query, if one of the instructions is associated
with an alias.scope id that is identical to the noalias scope associated with
the other instruction, or is a descendant (in the scope hierarchy) of the
noalias scope associated with the other instruction, then the two memory
accesses are assumed not to alias.
Note that is the first element of the scope metadata is a string, then it can
be combined accross functions and translation units. The string can be replaced
by a self-reference to create globally unqiue scope identifiers.
[Note: This overview is slightly stylized, since the metadata nodes really need
to just be numbers (!0 instead of !scope0), and the scope lists are also global
unnamed metadata.]
Existing noalias metadata in a callee is "cloned" for use by the inlined code.
This is necessary because the aliasing scopes are unique to each call site
(because of possible control dependencies on the aliasing properties). For
example, consider a function: foo(noalias a, noalias b) { *a = *b; } that gets
inlined into bar() { ... if (...) foo(a1, b1); ... if (...) foo(a2, b2); } --
now just because we know that a1 does not alias with b1 at the first call site,
and a2 does not alias with b2 at the second call site, we cannot let inlining
these functons have the metadata imply that a1 does not alias with b2.
llvm-svn: 213864
2014-07-24 22:25:39 +08:00
|
|
|
if (NoAliasTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_noalias, NoAliasTag);
|
2018-07-03 20:39:52 +08:00
|
|
|
|
|
|
|
return CI;
|
2010-12-27 06:49:25 +08:00
|
|
|
}
|
2011-05-22 07:14:36 +08:00
|
|
|
|
2018-05-31 04:02:56 +08:00
|
|
|
CallInst *IRBuilderBase::CreateElementUnorderedAtomicMemMove(
|
|
|
|
Value *Dst, unsigned DstAlign, Value *Src, unsigned SrcAlign, Value *Size,
|
|
|
|
uint32_t ElementSize, MDNode *TBAATag, MDNode *TBAAStructTag,
|
|
|
|
MDNode *ScopeTag, MDNode *NoAliasTag) {
|
|
|
|
assert(DstAlign >= ElementSize &&
|
|
|
|
"Pointer alignment must be at least element size");
|
|
|
|
assert(SrcAlign >= ElementSize &&
|
|
|
|
"Pointer alignment must be at least element size");
|
|
|
|
Dst = getCastedInt8PtrValue(Dst);
|
|
|
|
Src = getCastedInt8PtrValue(Src);
|
|
|
|
|
|
|
|
Value *Ops[] = {Dst, Src, Size, getInt32(ElementSize)};
|
|
|
|
Type *Tys[] = {Dst->getType(), Src->getType(), Size->getType()};
|
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn = Intrinsic::getDeclaration(
|
2018-05-31 04:02:56 +08:00
|
|
|
M, Intrinsic::memmove_element_unordered_atomic, Tys);
|
|
|
|
|
|
|
|
CallInst *CI = createCallHelper(TheFn, Ops, this);
|
|
|
|
|
|
|
|
// Set the alignment of the pointer args.
|
2019-10-15 20:56:24 +08:00
|
|
|
CI->addParamAttr(
|
|
|
|
0, Attribute::getWithAlignment(CI->getContext(), Align(DstAlign)));
|
|
|
|
CI->addParamAttr(
|
|
|
|
1, Attribute::getWithAlignment(CI->getContext(), Align(SrcAlign)));
|
2018-05-31 04:02:56 +08:00
|
|
|
|
|
|
|
// Set the TBAA info if present.
|
|
|
|
if (TBAATag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa, TBAATag);
|
|
|
|
|
|
|
|
// Set the TBAA Struct info if present.
|
|
|
|
if (TBAAStructTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_tbaa_struct, TBAAStructTag);
|
|
|
|
|
|
|
|
if (ScopeTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_alias_scope, ScopeTag);
|
|
|
|
|
|
|
|
if (NoAliasTag)
|
|
|
|
CI->setMetadata(LLVMContext::MD_noalias, NoAliasTag);
|
|
|
|
|
|
|
|
return CI;
|
|
|
|
}
|
|
|
|
|
2017-05-09 18:43:25 +08:00
|
|
|
static CallInst *getReductionIntrinsic(IRBuilderBase *Builder, Intrinsic::ID ID,
|
|
|
|
Value *Src) {
|
|
|
|
Module *M = Builder->GetInsertBlock()->getParent()->getParent();
|
|
|
|
Value *Ops[] = {Src};
|
2019-06-13 17:37:38 +08:00
|
|
|
Type *Tys[] = { Src->getType() };
|
2017-05-09 18:43:25 +08:00
|
|
|
auto Decl = Intrinsic::getDeclaration(M, ID, Tys);
|
|
|
|
return createCallHelper(Decl, Ops, Builder);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateFAddReduce(Value *Acc, Value *Src) {
|
|
|
|
Module *M = GetInsertBlock()->getParent()->getParent();
|
|
|
|
Value *Ops[] = {Acc, Src};
|
2019-05-20 17:54:06 +08:00
|
|
|
Type *Tys[] = {Acc->getType(), Src->getType()};
|
2017-05-09 18:43:25 +08:00
|
|
|
auto Decl = Intrinsic::getDeclaration(
|
2019-06-11 16:22:10 +08:00
|
|
|
M, Intrinsic::experimental_vector_reduce_v2_fadd, Tys);
|
2017-05-09 18:43:25 +08:00
|
|
|
return createCallHelper(Decl, Ops, this);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateFMulReduce(Value *Acc, Value *Src) {
|
|
|
|
Module *M = GetInsertBlock()->getParent()->getParent();
|
|
|
|
Value *Ops[] = {Acc, Src};
|
2019-05-20 17:54:06 +08:00
|
|
|
Type *Tys[] = {Acc->getType(), Src->getType()};
|
2017-05-09 18:43:25 +08:00
|
|
|
auto Decl = Intrinsic::getDeclaration(
|
2019-06-11 16:22:10 +08:00
|
|
|
M, Intrinsic::experimental_vector_reduce_v2_fmul, Tys);
|
2017-05-09 18:43:25 +08:00
|
|
|
return createCallHelper(Decl, Ops, this);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateAddReduce(Value *Src) {
|
|
|
|
return getReductionIntrinsic(this, Intrinsic::experimental_vector_reduce_add,
|
|
|
|
Src);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateMulReduce(Value *Src) {
|
|
|
|
return getReductionIntrinsic(this, Intrinsic::experimental_vector_reduce_mul,
|
|
|
|
Src);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateAndReduce(Value *Src) {
|
|
|
|
return getReductionIntrinsic(this, Intrinsic::experimental_vector_reduce_and,
|
|
|
|
Src);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateOrReduce(Value *Src) {
|
|
|
|
return getReductionIntrinsic(this, Intrinsic::experimental_vector_reduce_or,
|
|
|
|
Src);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateXorReduce(Value *Src) {
|
|
|
|
return getReductionIntrinsic(this, Intrinsic::experimental_vector_reduce_xor,
|
|
|
|
Src);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateIntMaxReduce(Value *Src, bool IsSigned) {
|
|
|
|
auto ID = IsSigned ? Intrinsic::experimental_vector_reduce_smax
|
|
|
|
: Intrinsic::experimental_vector_reduce_umax;
|
|
|
|
return getReductionIntrinsic(this, ID, Src);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateIntMinReduce(Value *Src, bool IsSigned) {
|
|
|
|
auto ID = IsSigned ? Intrinsic::experimental_vector_reduce_smin
|
|
|
|
: Intrinsic::experimental_vector_reduce_umin;
|
|
|
|
return getReductionIntrinsic(this, ID, Src);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateFPMaxReduce(Value *Src, bool NoNaN) {
|
|
|
|
auto Rdx = getReductionIntrinsic(
|
|
|
|
this, Intrinsic::experimental_vector_reduce_fmax, Src);
|
|
|
|
if (NoNaN) {
|
|
|
|
FastMathFlags FMF;
|
|
|
|
FMF.setNoNaNs();
|
|
|
|
Rdx->setFastMathFlags(FMF);
|
|
|
|
}
|
|
|
|
return Rdx;
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateFPMinReduce(Value *Src, bool NoNaN) {
|
|
|
|
auto Rdx = getReductionIntrinsic(
|
|
|
|
this, Intrinsic::experimental_vector_reduce_fmin, Src);
|
|
|
|
if (NoNaN) {
|
|
|
|
FastMathFlags FMF;
|
|
|
|
FMF.setNoNaNs();
|
|
|
|
Rdx->setFastMathFlags(FMF);
|
|
|
|
}
|
|
|
|
return Rdx;
|
|
|
|
}
|
|
|
|
|
2011-05-22 07:14:36 +08:00
|
|
|
CallInst *IRBuilderBase::CreateLifetimeStart(Value *Ptr, ConstantInt *Size) {
|
|
|
|
assert(isa<PointerType>(Ptr->getType()) &&
|
2012-07-19 08:11:40 +08:00
|
|
|
"lifetime.start only applies to pointers.");
|
2011-05-22 07:14:36 +08:00
|
|
|
Ptr = getCastedInt8PtrValue(Ptr);
|
|
|
|
if (!Size)
|
|
|
|
Size = getInt64(-1);
|
|
|
|
else
|
|
|
|
assert(Size->getType() == getInt64Ty() &&
|
2012-07-19 08:11:40 +08:00
|
|
|
"lifetime.start requires the size to be an i64");
|
2011-05-22 07:14:36 +08:00
|
|
|
Value *Ops[] = { Size, Ptr };
|
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn =
|
|
|
|
Intrinsic::getDeclaration(M, Intrinsic::lifetime_start, {Ptr->getType()});
|
2011-07-15 16:37:34 +08:00
|
|
|
return createCallHelper(TheFn, Ops, this);
|
2011-05-22 07:14:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateLifetimeEnd(Value *Ptr, ConstantInt *Size) {
|
|
|
|
assert(isa<PointerType>(Ptr->getType()) &&
|
2012-07-19 08:11:40 +08:00
|
|
|
"lifetime.end only applies to pointers.");
|
2011-05-22 07:14:36 +08:00
|
|
|
Ptr = getCastedInt8PtrValue(Ptr);
|
|
|
|
if (!Size)
|
|
|
|
Size = getInt64(-1);
|
|
|
|
else
|
|
|
|
assert(Size->getType() == getInt64Ty() &&
|
2012-07-19 08:11:40 +08:00
|
|
|
"lifetime.end requires the size to be an i64");
|
2011-05-22 07:14:36 +08:00
|
|
|
Value *Ops[] = { Size, Ptr };
|
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn =
|
|
|
|
Intrinsic::getDeclaration(M, Intrinsic::lifetime_end, {Ptr->getType()});
|
2011-07-15 16:37:34 +08:00
|
|
|
return createCallHelper(TheFn, Ops, this);
|
2011-05-22 07:14:36 +08:00
|
|
|
}
|
2014-10-16 07:44:22 +08:00
|
|
|
|
2016-07-23 04:57:23 +08:00
|
|
|
CallInst *IRBuilderBase::CreateInvariantStart(Value *Ptr, ConstantInt *Size) {
|
|
|
|
|
|
|
|
assert(isa<PointerType>(Ptr->getType()) &&
|
|
|
|
"invariant.start only applies to pointers.");
|
|
|
|
Ptr = getCastedInt8PtrValue(Ptr);
|
|
|
|
if (!Size)
|
|
|
|
Size = getInt64(-1);
|
|
|
|
else
|
|
|
|
assert(Size->getType() == getInt64Ty() &&
|
|
|
|
"invariant.start requires the size to be an i64");
|
|
|
|
|
|
|
|
Value *Ops[] = {Size, Ptr};
|
|
|
|
// Fill in the single overloaded type: memory object type.
|
|
|
|
Type *ObjectPtr[1] = {Ptr->getType()};
|
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn =
|
2016-07-23 04:57:23 +08:00
|
|
|
Intrinsic::getDeclaration(M, Intrinsic::invariant_start, ObjectPtr);
|
|
|
|
return createCallHelper(TheFn, Ops, this);
|
|
|
|
}
|
|
|
|
|
2014-10-16 07:44:22 +08:00
|
|
|
CallInst *IRBuilderBase::CreateAssumption(Value *Cond) {
|
|
|
|
assert(Cond->getType() == getInt1Ty() &&
|
|
|
|
"an assumption condition must be of type i1");
|
|
|
|
|
|
|
|
Value *Ops[] = { Cond };
|
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *FnAssume = Intrinsic::getDeclaration(M, Intrinsic::assume);
|
2014-10-16 07:44:22 +08:00
|
|
|
return createCallHelper(FnAssume, Ops, this);
|
|
|
|
}
|
|
|
|
|
2018-05-01 23:54:18 +08:00
|
|
|
/// Create a call to a Masked Load intrinsic.
|
2016-02-18 03:23:04 +08:00
|
|
|
/// \p Ptr - base pointer for the load
|
|
|
|
/// \p Align - alignment of the source location
|
|
|
|
/// \p Mask - vector of booleans which indicates what vector lanes should
|
|
|
|
/// be accessed in memory
|
|
|
|
/// \p PassThru - pass-through value that is used to fill the masked-off lanes
|
|
|
|
/// of the result
|
|
|
|
/// \p Name - name of the result variable
|
2014-12-30 22:28:14 +08:00
|
|
|
CallInst *IRBuilderBase::CreateMaskedLoad(Value *Ptr, unsigned Align,
|
|
|
|
Value *Mask, Value *PassThru,
|
|
|
|
const Twine &Name) {
|
2018-03-30 08:47:31 +08:00
|
|
|
auto *PtrTy = cast<PointerType>(Ptr->getType());
|
2016-06-29 02:27:25 +08:00
|
|
|
Type *DataTy = PtrTy->getElementType();
|
2014-12-30 22:28:14 +08:00
|
|
|
assert(DataTy->isVectorTy() && "Ptr should point to a vector");
|
2017-07-31 21:21:42 +08:00
|
|
|
assert(Mask && "Mask should not be all-ones (null)");
|
2014-12-30 22:28:14 +08:00
|
|
|
if (!PassThru)
|
|
|
|
PassThru = UndefValue::get(DataTy);
|
2016-06-29 02:27:25 +08:00
|
|
|
Type *OverloadedTypes[] = { DataTy, PtrTy };
|
2014-12-30 22:28:14 +08:00
|
|
|
Value *Ops[] = { Ptr, getInt32(Align), Mask, PassThru};
|
2016-06-29 02:27:25 +08:00
|
|
|
return CreateMaskedIntrinsic(Intrinsic::masked_load, Ops,
|
|
|
|
OverloadedTypes, Name);
|
2014-12-04 17:40:44 +08:00
|
|
|
}
|
|
|
|
|
2018-05-01 23:54:18 +08:00
|
|
|
/// Create a call to a Masked Store intrinsic.
|
2016-02-18 03:23:04 +08:00
|
|
|
/// \p Val - data to be stored,
|
|
|
|
/// \p Ptr - base pointer for the store
|
|
|
|
/// \p Align - alignment of the destination location
|
|
|
|
/// \p Mask - vector of booleans which indicates what vector lanes should
|
|
|
|
/// be accessed in memory
|
2014-12-30 22:28:14 +08:00
|
|
|
CallInst *IRBuilderBase::CreateMaskedStore(Value *Val, Value *Ptr,
|
|
|
|
unsigned Align, Value *Mask) {
|
2018-03-30 08:47:31 +08:00
|
|
|
auto *PtrTy = cast<PointerType>(Ptr->getType());
|
2016-06-29 02:27:25 +08:00
|
|
|
Type *DataTy = PtrTy->getElementType();
|
|
|
|
assert(DataTy->isVectorTy() && "Ptr should point to a vector");
|
2017-07-31 21:21:42 +08:00
|
|
|
assert(Mask && "Mask should not be all-ones (null)");
|
2016-06-29 02:27:25 +08:00
|
|
|
Type *OverloadedTypes[] = { DataTy, PtrTy };
|
2014-12-30 22:28:14 +08:00
|
|
|
Value *Ops[] = { Val, Ptr, getInt32(Align), Mask };
|
2016-06-29 02:27:25 +08:00
|
|
|
return CreateMaskedIntrinsic(Intrinsic::masked_store, Ops, OverloadedTypes);
|
2014-12-04 17:40:44 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/// Create a call to a Masked intrinsic, with given intrinsic Id,
|
2016-06-29 02:27:25 +08:00
|
|
|
/// an array of operands - Ops, and an array of overloaded types -
|
|
|
|
/// OverloadedTypes.
|
2015-05-21 01:16:39 +08:00
|
|
|
CallInst *IRBuilderBase::CreateMaskedIntrinsic(Intrinsic::ID Id,
|
2014-12-04 17:40:44 +08:00
|
|
|
ArrayRef<Value *> Ops,
|
2016-06-29 02:27:25 +08:00
|
|
|
ArrayRef<Type *> OverloadedTypes,
|
2014-12-30 22:28:14 +08:00
|
|
|
const Twine &Name) {
|
2014-12-04 17:40:44 +08:00
|
|
|
Module *M = BB->getParent()->getParent();
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *TheFn = Intrinsic::getDeclaration(M, Id, OverloadedTypes);
|
2014-12-30 22:28:14 +08:00
|
|
|
return createCallHelper(TheFn, Ops, this, Name);
|
2014-12-04 17:40:44 +08:00
|
|
|
}
|
2014-12-30 13:55:58 +08:00
|
|
|
|
2018-05-01 23:54:18 +08:00
|
|
|
/// Create a call to a Masked Gather intrinsic.
|
2016-02-18 03:23:04 +08:00
|
|
|
/// \p Ptrs - vector of pointers for loading
|
|
|
|
/// \p Align - alignment for one element
|
|
|
|
/// \p Mask - vector of booleans which indicates what vector lanes should
|
|
|
|
/// be accessed in memory
|
|
|
|
/// \p PassThru - pass-through value that is used to fill the masked-off lanes
|
|
|
|
/// of the result
|
|
|
|
/// \p Name - name of the result variable
|
|
|
|
CallInst *IRBuilderBase::CreateMaskedGather(Value *Ptrs, unsigned Align,
|
|
|
|
Value *Mask, Value *PassThru,
|
|
|
|
const Twine& Name) {
|
|
|
|
auto PtrsTy = cast<VectorType>(Ptrs->getType());
|
|
|
|
auto PtrTy = cast<PointerType>(PtrsTy->getElementType());
|
|
|
|
unsigned NumElts = PtrsTy->getVectorNumElements();
|
|
|
|
Type *DataTy = VectorType::get(PtrTy->getElementType(), NumElts);
|
|
|
|
|
|
|
|
if (!Mask)
|
|
|
|
Mask = Constant::getAllOnesValue(VectorType::get(Type::getInt1Ty(Context),
|
|
|
|
NumElts));
|
|
|
|
|
2017-05-19 18:40:18 +08:00
|
|
|
if (!PassThru)
|
|
|
|
PassThru = UndefValue::get(DataTy);
|
|
|
|
|
2017-05-03 20:28:54 +08:00
|
|
|
Type *OverloadedTypes[] = {DataTy, PtrsTy};
|
2017-05-19 18:40:18 +08:00
|
|
|
Value * Ops[] = {Ptrs, getInt32(Align), Mask, PassThru};
|
2016-02-18 03:23:04 +08:00
|
|
|
|
|
|
|
// We specify only one type when we create this intrinsic. Types of other
|
|
|
|
// arguments are derived from this type.
|
2017-05-03 20:28:54 +08:00
|
|
|
return CreateMaskedIntrinsic(Intrinsic::masked_gather, Ops, OverloadedTypes,
|
|
|
|
Name);
|
2016-02-18 03:23:04 +08:00
|
|
|
}
|
|
|
|
|
2018-05-01 23:54:18 +08:00
|
|
|
/// Create a call to a Masked Scatter intrinsic.
|
2016-02-18 03:23:04 +08:00
|
|
|
/// \p Data - data to be stored,
|
|
|
|
/// \p Ptrs - the vector of pointers, where the \p Data elements should be
|
|
|
|
/// stored
|
|
|
|
/// \p Align - alignment for one element
|
|
|
|
/// \p Mask - vector of booleans which indicates what vector lanes should
|
|
|
|
/// be accessed in memory
|
|
|
|
CallInst *IRBuilderBase::CreateMaskedScatter(Value *Data, Value *Ptrs,
|
|
|
|
unsigned Align, Value *Mask) {
|
|
|
|
auto PtrsTy = cast<VectorType>(Ptrs->getType());
|
|
|
|
auto DataTy = cast<VectorType>(Data->getType());
|
|
|
|
unsigned NumElts = PtrsTy->getVectorNumElements();
|
|
|
|
|
2016-02-18 05:16:59 +08:00
|
|
|
#ifndef NDEBUG
|
|
|
|
auto PtrTy = cast<PointerType>(PtrsTy->getElementType());
|
2016-02-18 03:23:04 +08:00
|
|
|
assert(NumElts == DataTy->getVectorNumElements() &&
|
2016-02-18 05:16:59 +08:00
|
|
|
PtrTy->getElementType() == DataTy->getElementType() &&
|
|
|
|
"Incompatible pointer and data types");
|
|
|
|
#endif
|
2016-02-18 03:23:04 +08:00
|
|
|
|
|
|
|
if (!Mask)
|
|
|
|
Mask = Constant::getAllOnesValue(VectorType::get(Type::getInt1Ty(Context),
|
|
|
|
NumElts));
|
2017-05-03 20:28:54 +08:00
|
|
|
|
|
|
|
Type *OverloadedTypes[] = {DataTy, PtrsTy};
|
2016-02-18 03:23:04 +08:00
|
|
|
Value * Ops[] = {Data, Ptrs, getInt32(Align), Mask};
|
|
|
|
|
|
|
|
// We specify only one type when we create this intrinsic. Types of other
|
|
|
|
// arguments are derived from this type.
|
2017-05-03 20:28:54 +08:00
|
|
|
return CreateMaskedIntrinsic(Intrinsic::masked_scatter, Ops, OverloadedTypes);
|
2016-02-18 03:23:04 +08:00
|
|
|
}
|
|
|
|
|
2015-10-08 03:52:12 +08:00
|
|
|
template <typename T0, typename T1, typename T2, typename T3>
|
2015-05-13 07:52:24 +08:00
|
|
|
static std::vector<Value *>
|
|
|
|
getStatepointArgs(IRBuilderBase &B, uint64_t ID, uint32_t NumPatchBytes,
|
2015-10-09 07:18:33 +08:00
|
|
|
Value *ActualCallee, uint32_t Flags, ArrayRef<T0> CallArgs,
|
|
|
|
ArrayRef<T1> TransitionArgs, ArrayRef<T2> DeoptArgs,
|
|
|
|
ArrayRef<T3> GCArgs) {
|
2015-05-07 07:53:09 +08:00
|
|
|
std::vector<Value *> Args;
|
2015-05-13 07:52:24 +08:00
|
|
|
Args.push_back(B.getInt64(ID));
|
|
|
|
Args.push_back(B.getInt32(NumPatchBytes));
|
2015-05-07 07:53:09 +08:00
|
|
|
Args.push_back(ActualCallee);
|
|
|
|
Args.push_back(B.getInt32(CallArgs.size()));
|
2015-10-09 07:18:33 +08:00
|
|
|
Args.push_back(B.getInt32(Flags));
|
2015-05-07 07:53:09 +08:00
|
|
|
Args.insert(Args.end(), CallArgs.begin(), CallArgs.end());
|
2015-10-08 03:52:12 +08:00
|
|
|
Args.push_back(B.getInt32(TransitionArgs.size()));
|
|
|
|
Args.insert(Args.end(), TransitionArgs.begin(), TransitionArgs.end());
|
2015-05-07 07:53:09 +08:00
|
|
|
Args.push_back(B.getInt32(DeoptArgs.size()));
|
|
|
|
Args.insert(Args.end(), DeoptArgs.begin(), DeoptArgs.end());
|
|
|
|
Args.insert(Args.end(), GCArgs.begin(), GCArgs.end());
|
|
|
|
|
|
|
|
return Args;
|
|
|
|
}
|
|
|
|
|
2015-10-08 03:52:12 +08:00
|
|
|
template <typename T0, typename T1, typename T2, typename T3>
|
|
|
|
static CallInst *CreateGCStatepointCallCommon(
|
|
|
|
IRBuilderBase *Builder, uint64_t ID, uint32_t NumPatchBytes,
|
2015-10-09 07:18:33 +08:00
|
|
|
Value *ActualCallee, uint32_t Flags, ArrayRef<T0> CallArgs,
|
2015-10-08 03:52:12 +08:00
|
|
|
ArrayRef<T1> TransitionArgs, ArrayRef<T2> DeoptArgs, ArrayRef<T3> GCArgs,
|
|
|
|
const Twine &Name) {
|
2015-05-06 10:36:34 +08:00
|
|
|
// Extract out the type of the callee.
|
2018-03-30 08:47:31 +08:00
|
|
|
auto *FuncPtrType = cast<PointerType>(ActualCallee->getType());
|
2015-05-06 10:36:34 +08:00
|
|
|
assert(isa<FunctionType>(FuncPtrType->getElementType()) &&
|
|
|
|
"actual callee must be a callable value");
|
2014-12-30 13:55:58 +08:00
|
|
|
|
2015-10-08 03:52:12 +08:00
|
|
|
Module *M = Builder->GetInsertBlock()->getParent()->getParent();
|
2015-05-06 10:36:34 +08:00
|
|
|
// Fill in the one generic type'd argument (the function is also vararg)
|
|
|
|
Type *ArgTypes[] = { FuncPtrType };
|
|
|
|
Function *FnStatepoint =
|
|
|
|
Intrinsic::getDeclaration(M, Intrinsic::experimental_gc_statepoint,
|
|
|
|
ArgTypes);
|
2014-12-30 13:55:58 +08:00
|
|
|
|
2018-03-30 08:47:31 +08:00
|
|
|
std::vector<Value *> Args =
|
2015-10-08 03:52:12 +08:00
|
|
|
getStatepointArgs(*Builder, ID, NumPatchBytes, ActualCallee, Flags,
|
|
|
|
CallArgs, TransitionArgs, DeoptArgs, GCArgs);
|
|
|
|
return createCallHelper(FnStatepoint, Args, Builder, Name);
|
2014-12-30 13:55:58 +08:00
|
|
|
}
|
|
|
|
|
2015-05-13 07:52:24 +08:00
|
|
|
CallInst *IRBuilderBase::CreateGCStatepointCall(
|
|
|
|
uint64_t ID, uint32_t NumPatchBytes, Value *ActualCallee,
|
2015-10-08 03:52:12 +08:00
|
|
|
ArrayRef<Value *> CallArgs, ArrayRef<Value *> DeoptArgs,
|
2015-05-13 07:52:24 +08:00
|
|
|
ArrayRef<Value *> GCArgs, const Twine &Name) {
|
2015-10-08 03:52:12 +08:00
|
|
|
return CreateGCStatepointCallCommon<Value *, Value *, Value *, Value *>(
|
2015-10-09 07:18:33 +08:00
|
|
|
this, ID, NumPatchBytes, ActualCallee, uint32_t(StatepointFlags::None),
|
|
|
|
CallArgs, None /* No Transition Args */, DeoptArgs, GCArgs, Name);
|
2015-05-07 07:53:09 +08:00
|
|
|
}
|
|
|
|
|
2015-10-08 03:52:12 +08:00
|
|
|
CallInst *IRBuilderBase::CreateGCStatepointCall(
|
2015-10-09 07:18:33 +08:00
|
|
|
uint64_t ID, uint32_t NumPatchBytes, Value *ActualCallee, uint32_t Flags,
|
|
|
|
ArrayRef<Use> CallArgs, ArrayRef<Use> TransitionArgs,
|
2015-10-08 03:52:12 +08:00
|
|
|
ArrayRef<Use> DeoptArgs, ArrayRef<Value *> GCArgs, const Twine &Name) {
|
|
|
|
return CreateGCStatepointCallCommon<Use, Use, Use, Value *>(
|
|
|
|
this, ID, NumPatchBytes, ActualCallee, Flags, CallArgs, TransitionArgs,
|
|
|
|
DeoptArgs, GCArgs, Name);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateGCStatepointCall(
|
|
|
|
uint64_t ID, uint32_t NumPatchBytes, Value *ActualCallee,
|
|
|
|
ArrayRef<Use> CallArgs, ArrayRef<Value *> DeoptArgs,
|
2015-05-07 07:53:09 +08:00
|
|
|
ArrayRef<Value *> GCArgs, const Twine &Name) {
|
2015-10-08 03:52:12 +08:00
|
|
|
return CreateGCStatepointCallCommon<Use, Value *, Value *, Value *>(
|
2015-10-09 07:18:33 +08:00
|
|
|
this, ID, NumPatchBytes, ActualCallee, uint32_t(StatepointFlags::None),
|
|
|
|
CallArgs, None, DeoptArgs, GCArgs, Name);
|
2015-10-08 03:52:12 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
template <typename T0, typename T1, typename T2, typename T3>
|
|
|
|
static InvokeInst *CreateGCStatepointInvokeCommon(
|
|
|
|
IRBuilderBase *Builder, uint64_t ID, uint32_t NumPatchBytes,
|
|
|
|
Value *ActualInvokee, BasicBlock *NormalDest, BasicBlock *UnwindDest,
|
2015-10-09 07:18:33 +08:00
|
|
|
uint32_t Flags, ArrayRef<T0> InvokeArgs, ArrayRef<T1> TransitionArgs,
|
2015-10-08 03:52:12 +08:00
|
|
|
ArrayRef<T2> DeoptArgs, ArrayRef<T3> GCArgs, const Twine &Name) {
|
2015-05-07 07:53:09 +08:00
|
|
|
// Extract out the type of the callee.
|
2018-03-30 08:47:31 +08:00
|
|
|
auto *FuncPtrType = cast<PointerType>(ActualInvokee->getType());
|
2015-05-07 07:53:09 +08:00
|
|
|
assert(isa<FunctionType>(FuncPtrType->getElementType()) &&
|
|
|
|
"actual callee must be a callable value");
|
|
|
|
|
2015-10-08 03:52:12 +08:00
|
|
|
Module *M = Builder->GetInsertBlock()->getParent()->getParent();
|
2015-05-07 07:53:09 +08:00
|
|
|
// Fill in the one generic type'd argument (the function is also vararg)
|
|
|
|
Function *FnStatepoint = Intrinsic::getDeclaration(
|
|
|
|
M, Intrinsic::experimental_gc_statepoint, {FuncPtrType});
|
|
|
|
|
2018-03-30 08:47:31 +08:00
|
|
|
std::vector<Value *> Args =
|
2015-10-08 03:52:12 +08:00
|
|
|
getStatepointArgs(*Builder, ID, NumPatchBytes, ActualInvokee, Flags,
|
|
|
|
InvokeArgs, TransitionArgs, DeoptArgs, GCArgs);
|
|
|
|
return createInvokeHelper(FnStatepoint, NormalDest, UnwindDest, Args, Builder,
|
2015-05-07 07:53:09 +08:00
|
|
|
Name);
|
|
|
|
}
|
|
|
|
|
2015-10-08 03:52:12 +08:00
|
|
|
InvokeInst *IRBuilderBase::CreateGCStatepointInvoke(
|
|
|
|
uint64_t ID, uint32_t NumPatchBytes, Value *ActualInvokee,
|
|
|
|
BasicBlock *NormalDest, BasicBlock *UnwindDest,
|
|
|
|
ArrayRef<Value *> InvokeArgs, ArrayRef<Value *> DeoptArgs,
|
|
|
|
ArrayRef<Value *> GCArgs, const Twine &Name) {
|
|
|
|
return CreateGCStatepointInvokeCommon<Value *, Value *, Value *, Value *>(
|
|
|
|
this, ID, NumPatchBytes, ActualInvokee, NormalDest, UnwindDest,
|
2015-10-09 07:18:33 +08:00
|
|
|
uint32_t(StatepointFlags::None), InvokeArgs, None /* No Transition Args*/,
|
2015-10-08 03:52:12 +08:00
|
|
|
DeoptArgs, GCArgs, Name);
|
|
|
|
}
|
|
|
|
|
|
|
|
InvokeInst *IRBuilderBase::CreateGCStatepointInvoke(
|
|
|
|
uint64_t ID, uint32_t NumPatchBytes, Value *ActualInvokee,
|
2015-10-09 07:18:33 +08:00
|
|
|
BasicBlock *NormalDest, BasicBlock *UnwindDest, uint32_t Flags,
|
2015-10-08 03:52:12 +08:00
|
|
|
ArrayRef<Use> InvokeArgs, ArrayRef<Use> TransitionArgs,
|
|
|
|
ArrayRef<Use> DeoptArgs, ArrayRef<Value *> GCArgs, const Twine &Name) {
|
|
|
|
return CreateGCStatepointInvokeCommon<Use, Use, Use, Value *>(
|
|
|
|
this, ID, NumPatchBytes, ActualInvokee, NormalDest, UnwindDest, Flags,
|
|
|
|
InvokeArgs, TransitionArgs, DeoptArgs, GCArgs, Name);
|
|
|
|
}
|
|
|
|
|
2015-05-07 07:53:09 +08:00
|
|
|
InvokeInst *IRBuilderBase::CreateGCStatepointInvoke(
|
2015-05-13 07:52:24 +08:00
|
|
|
uint64_t ID, uint32_t NumPatchBytes, Value *ActualInvokee,
|
|
|
|
BasicBlock *NormalDest, BasicBlock *UnwindDest, ArrayRef<Use> InvokeArgs,
|
|
|
|
ArrayRef<Value *> DeoptArgs, ArrayRef<Value *> GCArgs, const Twine &Name) {
|
2015-10-08 03:52:12 +08:00
|
|
|
return CreateGCStatepointInvokeCommon<Use, Value *, Value *, Value *>(
|
|
|
|
this, ID, NumPatchBytes, ActualInvokee, NormalDest, UnwindDest,
|
2015-10-09 07:18:33 +08:00
|
|
|
uint32_t(StatepointFlags::None), InvokeArgs, None, DeoptArgs, GCArgs,
|
|
|
|
Name);
|
2015-02-26 08:35:56 +08:00
|
|
|
}
|
|
|
|
|
2014-12-30 13:55:58 +08:00
|
|
|
CallInst *IRBuilderBase::CreateGCResult(Instruction *Statepoint,
|
|
|
|
Type *ResultType,
|
|
|
|
const Twine &Name) {
|
2015-01-23 04:14:38 +08:00
|
|
|
Intrinsic::ID ID = Intrinsic::experimental_gc_result;
|
2014-12-30 13:55:58 +08:00
|
|
|
Module *M = BB->getParent()->getParent();
|
|
|
|
Type *Types[] = {ResultType};
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *FnGCResult = Intrinsic::getDeclaration(M, ID, Types);
|
2014-12-30 13:55:58 +08:00
|
|
|
|
|
|
|
Value *Args[] = {Statepoint};
|
|
|
|
return createCallHelper(FnGCResult, Args, this, Name);
|
|
|
|
}
|
|
|
|
|
|
|
|
CallInst *IRBuilderBase::CreateGCRelocate(Instruction *Statepoint,
|
|
|
|
int BaseOffset,
|
|
|
|
int DerivedOffset,
|
|
|
|
Type *ResultType,
|
|
|
|
const Twine &Name) {
|
|
|
|
Module *M = BB->getParent()->getParent();
|
|
|
|
Type *Types[] = {ResultType};
|
2019-02-02 04:43:25 +08:00
|
|
|
Function *FnGCRelocate =
|
|
|
|
Intrinsic::getDeclaration(M, Intrinsic::experimental_gc_relocate, Types);
|
2014-12-30 13:55:58 +08:00
|
|
|
|
|
|
|
Value *Args[] = {Statepoint,
|
|
|
|
getInt32(BaseOffset),
|
|
|
|
getInt32(DerivedOffset)};
|
|
|
|
return createCallHelper(FnGCRelocate, Args, this, Name);
|
|
|
|
}
|
2017-02-28 07:08:49 +08:00
|
|
|
|
2018-10-08 18:32:33 +08:00
|
|
|
CallInst *IRBuilderBase::CreateUnaryIntrinsic(Intrinsic::ID ID, Value *V,
|
|
|
|
Instruction *FMFSource,
|
|
|
|
const Twine &Name) {
|
2018-02-24 05:16:12 +08:00
|
|
|
Module *M = BB->getModule();
|
2018-10-08 18:32:33 +08:00
|
|
|
Function *Fn = Intrinsic::getDeclaration(M, ID, {V->getType()});
|
|
|
|
return createCallHelper(Fn, {V}, this, Name, FMFSource);
|
2017-02-28 07:08:49 +08:00
|
|
|
}
|
2018-02-24 05:16:12 +08:00
|
|
|
|
2018-10-08 18:32:33 +08:00
|
|
|
CallInst *IRBuilderBase::CreateBinaryIntrinsic(Intrinsic::ID ID, Value *LHS,
|
|
|
|
Value *RHS,
|
|
|
|
Instruction *FMFSource,
|
|
|
|
const Twine &Name) {
|
2018-05-30 02:06:50 +08:00
|
|
|
Module *M = BB->getModule();
|
2018-10-08 18:32:33 +08:00
|
|
|
Function *Fn = Intrinsic::getDeclaration(M, ID, { LHS->getType() });
|
|
|
|
return createCallHelper(Fn, {LHS, RHS}, this, Name, FMFSource);
|
2018-05-30 02:06:50 +08:00
|
|
|
}
|
|
|
|
|
2018-02-24 05:16:12 +08:00
|
|
|
CallInst *IRBuilderBase::CreateIntrinsic(Intrinsic::ID ID,
|
2018-10-08 18:32:33 +08:00
|
|
|
ArrayRef<Type *> Types,
|
2018-02-24 05:16:12 +08:00
|
|
|
ArrayRef<Value *> Args,
|
|
|
|
Instruction *FMFSource,
|
|
|
|
const Twine &Name) {
|
|
|
|
Module *M = BB->getModule();
|
2018-10-08 18:32:33 +08:00
|
|
|
Function *Fn = Intrinsic::getDeclaration(M, ID, Types);
|
2018-02-24 05:16:12 +08:00
|
|
|
return createCallHelper(Fn, Args, this, Name, FMFSource);
|
|
|
|
}
|