llvm-project/llvm/lib/Target/ARM/ARMTargetMachine.cpp

215 lines
7.6 KiB
C++
Raw Normal View History

//===-- ARMTargetMachine.cpp - Define TargetMachine for ARM ---------------===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
//
//===----------------------------------------------------------------------===//
#include "ARMTargetMachine.h"
#include "ARM.h"
#include "ARMFrameLowering.h"
#include "llvm/CodeGen/Passes.h"
#include "llvm/MC/MCAsmInfo.h"
#include "llvm/PassManager.h"
Making use of VFP / NEON floating point multiply-accumulate / subtraction is difficult on current ARM implementations for a few reasons. 1. Even though a single vmla has latency that is one cycle shorter than a pair of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause additional pipeline stall. So it's frequently better to single codegen vmul + vadd. 2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to stall for 4 cycles. We need to schedule them apart. 3. A vmla followed vmla is a special case. Obvious issuing back to back RAW vmla + vmla is very bad. But this isn't ideal either: vmul vadd vmla Instead, we want to expand the second vmla: vmla vmul vadd Even with the 4 cycle vmul stall, the second sequence is still 2 cycles faster. Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough but it isn't the optimial solution. This patch attempts to make it possible to use vmla / vmls in cases where it is profitable. A. Add missing isel predicates which cause vmla to be codegen'ed. B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to compute a fmul and a fmla. C. Add additional isel checks for vmla, avoid cases where vmla is feeding into fp instructions (except for the #3 exceptional case). D. Add ARM hazard recognizer to model the vmla / vmls hazards. E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the vmla / vmls will trigger one of the special hazards. Work in progress, only A+B are enabled. llvm-svn: 120960
2010-12-06 06:04:16 +08:00
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/FormattedStream.h"
#include "llvm/Support/TargetRegistry.h"
#include "llvm/Target/TargetOptions.h"
#include "llvm/Transforms/Scalar.h"
using namespace llvm;
static cl::opt<bool>
EnableGlobalMerge("global-merge", cl::Hidden,
cl::desc("Enable global merge pass"),
cl::init(true));
extern "C" void LLVMInitializeARMTarget() {
// Register the target.
RegisterTargetMachine<ARMTargetMachine> X(TheARMTarget);
RegisterTargetMachine<ThumbTargetMachine> Y(TheThumbTarget);
}
/// TargetMachine ctor - Create an ARM architecture model.
///
ARMBaseTargetMachine::ARMBaseTargetMachine(const Target &T, StringRef TT,
StringRef CPU, StringRef FS,
const TargetOptions &Options,
Reloc::Model RM, CodeModel::Model CM,
CodeGenOpt::Level OL)
: LLVMTargetMachine(T, TT, CPU, FS, Options, RM, CM, OL),
Subtarget(TT, CPU, FS),
JITInfo(),
2011-04-07 06:35:47 +08:00
InstrItins(Subtarget.getInstrItineraryData()) {
// Default to soft float ABI
if (Options.FloatABIType == FloatABI::Default)
this->Options.FloatABIType = FloatABI::Soft;
}
Switch TargetTransformInfo from an immutable analysis pass that requires a TargetMachine to construct (and thus isn't always available), to an analysis group that supports layered implementations much like AliasAnalysis does. This is a pretty massive change, with a few parts that I was unable to easily separate (sorry), so I'll walk through it. The first step of this conversion was to make TargetTransformInfo an analysis group, and to sink the nonce implementations in ScalarTargetTransformInfo and VectorTargetTranformInfo into a NoTargetTransformInfo pass. This allows other passes to add a hard requirement on TTI, and assume they will always get at least on implementation. The TargetTransformInfo analysis group leverages the delegation chaining trick that AliasAnalysis uses, where the base class for the analysis group delegates to the previous analysis *pass*, allowing all but tho NoFoo analysis passes to only implement the parts of the interfaces they support. It also introduces a new trick where each pass in the group retains a pointer to the top-most pass that has been initialized. This allows passes to implement one API in terms of another API and benefit when some other pass above them in the stack has more precise results for the second API. The second step of this conversion is to create a pass that implements the TargetTransformInfo analysis using the target-independent abstractions in the code generator. This replaces the ScalarTargetTransformImpl and VectorTargetTransformImpl classes in lib/Target with a single pass in lib/CodeGen called BasicTargetTransformInfo. This class actually provides most of the TTI functionality, basing it upon the TargetLowering abstraction and other information in the target independent code generator. The third step of the conversion adds support to all TargetMachines to register custom analysis passes. This allows building those passes with access to TargetLowering or other target-specific classes, and it also allows each target to customize the set of analysis passes desired in the pass manager. The baseline LLVMTargetMachine implements this interface to add the BasicTTI pass to the pass manager, and all of the tools that want to support target-aware TTI passes call this routine on whatever target machine they end up with to add the appropriate passes. The fourth step of the conversion created target-specific TTI analysis passes for the X86 and ARM backends. These passes contain the custom logic that was previously in their extensions of the ScalarTargetTransformInfo and VectorTargetTransformInfo interfaces. I separated them into their own file, as now all of the interface bits are private and they just expose a function to create the pass itself. Then I extended these target machines to set up a custom set of analysis passes, first adding BasicTTI as a fallback, and then adding their customized TTI implementations. The fourth step required logic that was shared between the target independent layer and the specific targets to move to a different interface, as they no longer derive from each other. As a consequence, a helper functions were added to TargetLowering representing the common logic needed both in the target implementation and the codegen implementation of the TTI pass. While technically this is the only change that could have been committed separately, it would have been a nightmare to extract. The final step of the conversion was just to delete all the old boilerplate. This got rid of the ScalarTargetTransformInfo and VectorTargetTransformInfo classes, all of the support in all of the targets for producing instances of them, and all of the support in the tools for manually constructing a pass based around them. Now that TTI is a relatively normal analysis group, two things become straightforward. First, we can sink it into lib/Analysis which is a more natural layer for it to live. Second, clients of this interface can depend on it *always* being available which will simplify their code and behavior. These (and other) simplifications will follow in subsequent commits, this one is clearly big enough. Finally, I'm very aware that much of the comments and documentation needs to be updated. As soon as I had this working, and plausibly well commented, I wanted to get it committed and in front of the build bots. I'll be doing a few passes over documentation later if it sticks. Commits to update DragonEgg and Clang will be made presently. llvm-svn: 171681
2013-01-07 09:37:14 +08:00
void ARMBaseTargetMachine::addAnalysisPasses(PassManagerBase &PM) {
// Add first the target-independent BasicTTI pass, then our X86 pass. This
// allows the X86 pass to delegate to the target independent layer when
// appropriate.
PM.add(createBasicTargetTransformInfoPass(getTargetLowering()));
PM.add(createARMTargetTransformInfoPass(this));
}
void ARMTargetMachine::anchor() { }
ARMTargetMachine::ARMTargetMachine(const Target &T, StringRef TT,
StringRef CPU, StringRef FS,
const TargetOptions &Options,
Reloc::Model RM, CodeModel::Model CM,
CodeGenOpt::Level OL)
: ARMBaseTargetMachine(T, TT, CPU, FS, Options, RM, CM, OL),
InstrInfo(Subtarget),
DL(Subtarget.isAPCS_ABI() ?
std::string("e-p:32:32-f64:32:64-i64:32:64-"
"v128:32:128-v64:32:64-n32-S32") :
Subtarget.isAAPCS_ABI() ?
std::string("e-p:32:32-f64:64:64-i64:64:64-"
"v128:64:128-v64:64:64-n32-S64") :
std::string("e-p:32:32-f64:64:64-i64:64:64-"
"v128:64:128-v64:64:64-n32-S32")),
TLInfo(*this),
TSInfo(*this),
Switch TargetTransformInfo from an immutable analysis pass that requires a TargetMachine to construct (and thus isn't always available), to an analysis group that supports layered implementations much like AliasAnalysis does. This is a pretty massive change, with a few parts that I was unable to easily separate (sorry), so I'll walk through it. The first step of this conversion was to make TargetTransformInfo an analysis group, and to sink the nonce implementations in ScalarTargetTransformInfo and VectorTargetTranformInfo into a NoTargetTransformInfo pass. This allows other passes to add a hard requirement on TTI, and assume they will always get at least on implementation. The TargetTransformInfo analysis group leverages the delegation chaining trick that AliasAnalysis uses, where the base class for the analysis group delegates to the previous analysis *pass*, allowing all but tho NoFoo analysis passes to only implement the parts of the interfaces they support. It also introduces a new trick where each pass in the group retains a pointer to the top-most pass that has been initialized. This allows passes to implement one API in terms of another API and benefit when some other pass above them in the stack has more precise results for the second API. The second step of this conversion is to create a pass that implements the TargetTransformInfo analysis using the target-independent abstractions in the code generator. This replaces the ScalarTargetTransformImpl and VectorTargetTransformImpl classes in lib/Target with a single pass in lib/CodeGen called BasicTargetTransformInfo. This class actually provides most of the TTI functionality, basing it upon the TargetLowering abstraction and other information in the target independent code generator. The third step of the conversion adds support to all TargetMachines to register custom analysis passes. This allows building those passes with access to TargetLowering or other target-specific classes, and it also allows each target to customize the set of analysis passes desired in the pass manager. The baseline LLVMTargetMachine implements this interface to add the BasicTTI pass to the pass manager, and all of the tools that want to support target-aware TTI passes call this routine on whatever target machine they end up with to add the appropriate passes. The fourth step of the conversion created target-specific TTI analysis passes for the X86 and ARM backends. These passes contain the custom logic that was previously in their extensions of the ScalarTargetTransformInfo and VectorTargetTransformInfo interfaces. I separated them into their own file, as now all of the interface bits are private and they just expose a function to create the pass itself. Then I extended these target machines to set up a custom set of analysis passes, first adding BasicTTI as a fallback, and then adding their customized TTI implementations. The fourth step required logic that was shared between the target independent layer and the specific targets to move to a different interface, as they no longer derive from each other. As a consequence, a helper functions were added to TargetLowering representing the common logic needed both in the target implementation and the codegen implementation of the TTI pass. While technically this is the only change that could have been committed separately, it would have been a nightmare to extract. The final step of the conversion was just to delete all the old boilerplate. This got rid of the ScalarTargetTransformInfo and VectorTargetTransformInfo classes, all of the support in all of the targets for producing instances of them, and all of the support in the tools for manually constructing a pass based around them. Now that TTI is a relatively normal analysis group, two things become straightforward. First, we can sink it into lib/Analysis which is a more natural layer for it to live. Second, clients of this interface can depend on it *always* being available which will simplify their code and behavior. These (and other) simplifications will follow in subsequent commits, this one is clearly big enough. Finally, I'm very aware that much of the comments and documentation needs to be updated. As soon as I had this working, and plausibly well commented, I wanted to get it committed and in front of the build bots. I'll be doing a few passes over documentation later if it sticks. Commits to update DragonEgg and Clang will be made presently. llvm-svn: 171681
2013-01-07 09:37:14 +08:00
FrameLowering(Subtarget) {
if (!Subtarget.hasARMOps())
report_fatal_error("CPU: '" + Subtarget.getCPUString() + "' does not "
"support ARM mode execution!");
}
void ThumbTargetMachine::anchor() { }
ThumbTargetMachine::ThumbTargetMachine(const Target &T, StringRef TT,
StringRef CPU, StringRef FS,
const TargetOptions &Options,
Reloc::Model RM, CodeModel::Model CM,
CodeGenOpt::Level OL)
: ARMBaseTargetMachine(T, TT, CPU, FS, Options, RM, CM, OL),
InstrInfo(Subtarget.hasThumb2()
? ((ARMBaseInstrInfo*)new Thumb2InstrInfo(Subtarget))
: ((ARMBaseInstrInfo*)new Thumb1InstrInfo(Subtarget))),
DL(Subtarget.isAPCS_ABI() ?
std::string("e-p:32:32-f64:32:64-i64:32:64-"
"i16:16:32-i8:8:32-i1:8:32-"
"v128:32:128-v64:32:64-a:0:32-n32-S32") :
Subtarget.isAAPCS_ABI() ?
std::string("e-p:32:32-f64:64:64-i64:64:64-"
"i16:16:32-i8:8:32-i1:8:32-"
"v128:64:128-v64:64:64-a:0:32-n32-S64") :
std::string("e-p:32:32-f64:64:64-i64:64:64-"
"i16:16:32-i8:8:32-i1:8:32-"
"v128:64:128-v64:64:64-a:0:32-n32-S32")),
TLInfo(*this),
TSInfo(*this),
FrameLowering(Subtarget.hasThumb2()
? new ARMFrameLowering(Subtarget)
Switch TargetTransformInfo from an immutable analysis pass that requires a TargetMachine to construct (and thus isn't always available), to an analysis group that supports layered implementations much like AliasAnalysis does. This is a pretty massive change, with a few parts that I was unable to easily separate (sorry), so I'll walk through it. The first step of this conversion was to make TargetTransformInfo an analysis group, and to sink the nonce implementations in ScalarTargetTransformInfo and VectorTargetTranformInfo into a NoTargetTransformInfo pass. This allows other passes to add a hard requirement on TTI, and assume they will always get at least on implementation. The TargetTransformInfo analysis group leverages the delegation chaining trick that AliasAnalysis uses, where the base class for the analysis group delegates to the previous analysis *pass*, allowing all but tho NoFoo analysis passes to only implement the parts of the interfaces they support. It also introduces a new trick where each pass in the group retains a pointer to the top-most pass that has been initialized. This allows passes to implement one API in terms of another API and benefit when some other pass above them in the stack has more precise results for the second API. The second step of this conversion is to create a pass that implements the TargetTransformInfo analysis using the target-independent abstractions in the code generator. This replaces the ScalarTargetTransformImpl and VectorTargetTransformImpl classes in lib/Target with a single pass in lib/CodeGen called BasicTargetTransformInfo. This class actually provides most of the TTI functionality, basing it upon the TargetLowering abstraction and other information in the target independent code generator. The third step of the conversion adds support to all TargetMachines to register custom analysis passes. This allows building those passes with access to TargetLowering or other target-specific classes, and it also allows each target to customize the set of analysis passes desired in the pass manager. The baseline LLVMTargetMachine implements this interface to add the BasicTTI pass to the pass manager, and all of the tools that want to support target-aware TTI passes call this routine on whatever target machine they end up with to add the appropriate passes. The fourth step of the conversion created target-specific TTI analysis passes for the X86 and ARM backends. These passes contain the custom logic that was previously in their extensions of the ScalarTargetTransformInfo and VectorTargetTransformInfo interfaces. I separated them into their own file, as now all of the interface bits are private and they just expose a function to create the pass itself. Then I extended these target machines to set up a custom set of analysis passes, first adding BasicTTI as a fallback, and then adding their customized TTI implementations. The fourth step required logic that was shared between the target independent layer and the specific targets to move to a different interface, as they no longer derive from each other. As a consequence, a helper functions were added to TargetLowering representing the common logic needed both in the target implementation and the codegen implementation of the TTI pass. While technically this is the only change that could have been committed separately, it would have been a nightmare to extract. The final step of the conversion was just to delete all the old boilerplate. This got rid of the ScalarTargetTransformInfo and VectorTargetTransformInfo classes, all of the support in all of the targets for producing instances of them, and all of the support in the tools for manually constructing a pass based around them. Now that TTI is a relatively normal analysis group, two things become straightforward. First, we can sink it into lib/Analysis which is a more natural layer for it to live. Second, clients of this interface can depend on it *always* being available which will simplify their code and behavior. These (and other) simplifications will follow in subsequent commits, this one is clearly big enough. Finally, I'm very aware that much of the comments and documentation needs to be updated. As soon as I had this working, and plausibly well commented, I wanted to get it committed and in front of the build bots. I'll be doing a few passes over documentation later if it sticks. Commits to update DragonEgg and Clang will be made presently. llvm-svn: 171681
2013-01-07 09:37:14 +08:00
: (ARMFrameLowering*)new Thumb1FrameLowering(Subtarget)) {
}
namespace {
/// ARM Code Generator Pass Configuration Options.
class ARMPassConfig : public TargetPassConfig {
public:
ARMPassConfig(ARMBaseTargetMachine *TM, PassManagerBase &PM)
: TargetPassConfig(TM, PM) {}
ARMBaseTargetMachine &getARMTargetMachine() const {
return getTM<ARMBaseTargetMachine>();
}
const ARMSubtarget &getARMSubtarget() const {
return *getARMTargetMachine().getSubtargetImpl();
}
virtual bool addPreISel();
virtual bool addInstSelector();
virtual bool addPreRegAlloc();
virtual bool addPreSched2();
virtual bool addPreEmitPass();
};
} // namespace
TargetPassConfig *ARMBaseTargetMachine::createPassConfig(PassManagerBase &PM) {
return new ARMPassConfig(this, PM);
}
bool ARMPassConfig::addPreISel() {
if (TM->getOptLevel() != CodeGenOpt::None && EnableGlobalMerge)
addPass(createGlobalMergePass(TM->getTargetLowering()));
return false;
}
bool ARMPassConfig::addInstSelector() {
addPass(createARMISelDag(getARMTargetMachine(), getOptLevel()));
const ARMSubtarget *Subtarget = &getARMSubtarget();
if (Subtarget->isTargetELF() && !Subtarget->isThumb1Only() &&
TM->Options.EnableFastISel)
addPass(createARMGlobalBaseRegPass());
return false;
}
bool ARMPassConfig::addPreRegAlloc() {
// FIXME: temporarily disabling load / store optimization pass for Thumb1.
if (getOptLevel() != CodeGenOpt::None && !getARMSubtarget().isThumb1Only())
addPass(createARMLoadStoreOptimizationPass(true));
if (getOptLevel() != CodeGenOpt::None && getARMSubtarget().isLikeA9())
addPass(createMLxExpansionPass());
return true;
}
bool ARMPassConfig::addPreSched2() {
// FIXME: temporarily disabling load / store optimization pass for Thumb1.
if (getOptLevel() != CodeGenOpt::None) {
if (!getARMSubtarget().isThumb1Only()) {
addPass(createARMLoadStoreOptimizationPass());
printAndVerify("After ARM load / store optimizer");
}
if (getARMSubtarget().hasNEON())
addPass(createExecutionDependencyFixPass(&ARM::DPRRegClass));
}
// Expand some pseudo instructions into multiple instructions to allow
// proper scheduling.
addPass(createARMExpandPseudoPass());
if (getOptLevel() != CodeGenOpt::None) {
if (!getARMSubtarget().isThumb1Only())
addPass(&IfConverterID);
}
if (getARMSubtarget().isThumb2())
addPass(createThumb2ITBlockPass());
return true;
}
bool ARMPassConfig::addPreEmitPass() {
if (getARMSubtarget().isThumb2()) {
if (!getARMSubtarget().prefers32BitThumb())
addPass(createThumb2SizeReductionPass());
// Constant island pass work on unbundled instructions.
addPass(&UnpackMachineBundlesID);
}
addPass(createARMConstantIslandPass());
return true;
}
2012-04-07 07:43:50 +08:00
bool ARMBaseTargetMachine::addCodeEmitter(PassManagerBase &PM,
JITCodeEmitter &JCE) {
// Machine code emitter pass for ARM.
PM.add(createARMJITCodeEmitterPass(*this, JCE));
return false;
}