llvm-project/llvm/lib/IR/DiagnosticInfo.cpp

382 lines
15 KiB
C++
Raw Normal View History

//===- llvm/Support/DiagnosticInfo.cpp - Diagnostic Definitions -*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
//
// This file defines the different classes involved in low level diagnostics.
//
// Diagnostics reporting is still done as part of the LLVMContext.
//===----------------------------------------------------------------------===//
2016-09-28 04:55:01 +08:00
#include "llvm/IR/DiagnosticInfo.h"
#include "LLVMContextImpl.h"
Output optimization remarks in YAML (Re-committed after moving the template specialization under the yaml namespace. GCC was complaining about this.) This allows various presentation of this data using an external tool. This was first recommended here[1]. As an example, consider this module: 1 int foo(); 2 int bar(); 3 4 int baz() { 5 return foo() + bar(); 6 } The inliner generates these missed-optimization remarks today (the hotness information is pulled from PGO): remark: /tmp/s.c:5:10: foo will not be inlined into baz (hotness: 30) remark: /tmp/s.c:5:18: bar will not be inlined into baz (hotness: 30) Now with -pass-remarks-output=<yaml-file>, we generate this YAML file: --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 10 } Function: baz Hotness: 30 Args: - Callee: foo - String: will not be inlined into - Caller: baz ... --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 18 } Function: baz Hotness: 30 Args: - Callee: bar - String: will not be inlined into - Caller: baz ... This is a summary of the high-level decisions: * There is a new streaming interface to emit optimization remarks. E.g. for the inliner remark above: ORE.emit(DiagnosticInfoOptimizationRemarkMissed( DEBUG_TYPE, "NotInlined", &I) << NV("Callee", Callee) << " will not be inlined into " << NV("Caller", CS.getCaller()) << setIsVerbose()); NV stands for named value and allows the YAML client to process a remark using its name (NotInlined) and the named arguments (Callee and Caller) without parsing the text of the message. Subsequent patches will update ORE users to use the new streaming API. * I am using YAML I/O for writing the YAML file. YAML I/O requires you to specify reading and writing at once but reading is highly non-trivial for some of the more complex LLVM types. Since it's not clear that we (ever) want to use LLVM to parse this YAML file, the code supports and asserts that we're writing only. On the other hand, I did experiment that the class hierarchy starting at DiagnosticInfoOptimizationBase can be mapped back from YAML generated here (see D24479). * The YAML stream is stored in the LLVM context. * In the example, we can probably further specify the IR value used, i.e. print "Function" rather than "Value". * As before hotness is computed in the analysis pass instead of DiganosticInfo. This avoids the layering problem since BFI is in Analysis while DiagnosticInfo is in IR. [1] https://reviews.llvm.org/D19678#419445 Differential Revision: https://reviews.llvm.org/D24587 llvm-svn: 282539
2016-09-28 04:55:07 +08:00
#include "llvm/ADT/StringExtras.h"
#include "llvm/ADT/Twine.h"
#include "llvm/IR/Constants.h"
#include "llvm/IR/DebugInfo.h"
#include "llvm/IR/DiagnosticPrinter.h"
#include "llvm/IR/Function.h"
#include "llvm/IR/Instruction.h"
#include "llvm/IR/Metadata.h"
#include "llvm/IR/Module.h"
#include "llvm/Support/CommandLine.h"
#include "llvm/Support/Regex.h"
#include <atomic>
#include <string>
using namespace llvm;
namespace {
/// \brief Regular expression corresponding to the value given in one of the
/// -pass-remarks* command line flags. Passes whose name matches this regexp
/// will emit a diagnostic when calling the associated diagnostic function
/// (emitOptimizationRemark, emitOptimizationRemarkMissed or
/// emitOptimizationRemarkAnalysis).
struct PassRemarksOpt {
std::shared_ptr<Regex> Pattern;
void operator=(const std::string &Val) {
// Create a regexp object to match pass names for emitOptimizationRemark.
if (!Val.empty()) {
Pattern = std::make_shared<Regex>(Val);
std::string RegexError;
if (!Pattern->isValid(RegexError))
report_fatal_error("Invalid regular expression '" + Val +
"' in -pass-remarks: " + RegexError,
false);
}
}
};
static PassRemarksOpt PassRemarksOptLoc;
static PassRemarksOpt PassRemarksMissedOptLoc;
static PassRemarksOpt PassRemarksAnalysisOptLoc;
// -pass-remarks
// Command line flag to enable emitOptimizationRemark()
static cl::opt<PassRemarksOpt, true, cl::parser<std::string>>
PassRemarks("pass-remarks", cl::value_desc("pattern"),
cl::desc("Enable optimization remarks from passes whose name match "
"the given regular expression"),
cl::Hidden, cl::location(PassRemarksOptLoc), cl::ValueRequired,
cl::ZeroOrMore);
// -pass-remarks-missed
// Command line flag to enable emitOptimizationRemarkMissed()
static cl::opt<PassRemarksOpt, true, cl::parser<std::string>> PassRemarksMissed(
"pass-remarks-missed", cl::value_desc("pattern"),
cl::desc("Enable missed optimization remarks from passes whose name match "
"the given regular expression"),
cl::Hidden, cl::location(PassRemarksMissedOptLoc), cl::ValueRequired,
cl::ZeroOrMore);
// -pass-remarks-analysis
// Command line flag to enable emitOptimizationRemarkAnalysis()
static cl::opt<PassRemarksOpt, true, cl::parser<std::string>>
PassRemarksAnalysis(
"pass-remarks-analysis", cl::value_desc("pattern"),
cl::desc(
"Enable optimization analysis remarks from passes whose name match "
"the given regular expression"),
cl::Hidden, cl::location(PassRemarksAnalysisOptLoc), cl::ValueRequired,
cl::ZeroOrMore);
}
int llvm::getNextAvailablePluginDiagnosticKind() {
static std::atomic<int> PluginKindID(DK_FirstPluginKind);
return ++PluginKindID;
}
const char *OptimizationRemarkAnalysis::AlwaysPrint = "";
DiagnosticInfoInlineAsm::DiagnosticInfoInlineAsm(const Instruction &I,
const Twine &MsgStr,
DiagnosticSeverity Severity)
: DiagnosticInfo(DK_InlineAsm, Severity), LocCookie(0), MsgStr(MsgStr),
Instr(&I) {
if (const MDNode *SrcLoc = I.getMetadata("srcloc")) {
if (SrcLoc->getNumOperands() != 0)
IR: Split Metadata from Value Split `Metadata` away from the `Value` class hierarchy, as part of PR21532. Assembly and bitcode changes are in the wings, but this is the bulk of the change for the IR C++ API. I have a follow-up patch prepared for `clang`. If this breaks other sub-projects, I apologize in advance :(. Help me compile it on Darwin I'll try to fix it. FWIW, the errors should be easy to fix, so it may be simpler to just fix it yourself. This breaks the build for all metadata-related code that's out-of-tree. Rest assured the transition is mechanical and the compiler should catch almost all of the problems. Here's a quick guide for updating your code: - `Metadata` is the root of a class hierarchy with three main classes: `MDNode`, `MDString`, and `ValueAsMetadata`. It is distinct from the `Value` class hierarchy. It is typeless -- i.e., instances do *not* have a `Type`. - `MDNode`'s operands are all `Metadata *` (instead of `Value *`). - `TrackingVH<MDNode>` and `WeakVH` referring to metadata can be replaced with `TrackingMDNodeRef` and `TrackingMDRef`, respectively. If you're referring solely to resolved `MDNode`s -- post graph construction -- just use `MDNode*`. - `MDNode` (and the rest of `Metadata`) have only limited support for `replaceAllUsesWith()`. As long as an `MDNode` is pointing at a forward declaration -- the result of `MDNode::getTemporary()` -- it maintains a side map of its uses and can RAUW itself. Once the forward declarations are fully resolved RAUW support is dropped on the ground. This means that uniquing collisions on changing operands cause nodes to become "distinct". (This already happened fairly commonly, whenever an operand went to null.) If you're constructing complex (non self-reference) `MDNode` cycles, you need to call `MDNode::resolveCycles()` on each node (or on a top-level node that somehow references all of the nodes). Also, don't do that. Metadata cycles (and the RAUW machinery needed to construct them) are expensive. - An `MDNode` can only refer to a `Constant` through a bridge called `ConstantAsMetadata` (one of the subclasses of `ValueAsMetadata`). As a side effect, accessing an operand of an `MDNode` that is known to be, e.g., `ConstantInt`, takes three steps: first, cast from `Metadata` to `ConstantAsMetadata`; second, extract the `Constant`; third, cast down to `ConstantInt`. The eventual goal is to introduce `MDInt`/`MDFloat`/etc. and have metadata schema owners transition away from using `Constant`s when the type isn't important (and they don't care about referring to `GlobalValue`s). In the meantime, I've added transitional API to the `mdconst` namespace that matches semantics with the old code, in order to avoid adding the error-prone three-step equivalent to every call site. If your old code was: MDNode *N = foo(); bar(isa <ConstantInt>(N->getOperand(0))); baz(cast <ConstantInt>(N->getOperand(1))); bak(cast_or_null <ConstantInt>(N->getOperand(2))); bat(dyn_cast <ConstantInt>(N->getOperand(3))); bay(dyn_cast_or_null<ConstantInt>(N->getOperand(4))); you can trivially match its semantics with: MDNode *N = foo(); bar(mdconst::hasa <ConstantInt>(N->getOperand(0))); baz(mdconst::extract <ConstantInt>(N->getOperand(1))); bak(mdconst::extract_or_null <ConstantInt>(N->getOperand(2))); bat(mdconst::dyn_extract <ConstantInt>(N->getOperand(3))); bay(mdconst::dyn_extract_or_null<ConstantInt>(N->getOperand(4))); and when you transition your metadata schema to `MDInt`: MDNode *N = foo(); bar(isa <MDInt>(N->getOperand(0))); baz(cast <MDInt>(N->getOperand(1))); bak(cast_or_null <MDInt>(N->getOperand(2))); bat(dyn_cast <MDInt>(N->getOperand(3))); bay(dyn_cast_or_null<MDInt>(N->getOperand(4))); - A `CallInst` -- specifically, intrinsic instructions -- can refer to metadata through a bridge called `MetadataAsValue`. This is a subclass of `Value` where `getType()->isMetadataTy()`. `MetadataAsValue` is the *only* class that can legally refer to a `LocalAsMetadata`, which is a bridged form of non-`Constant` values like `Argument` and `Instruction`. It can also refer to any other `Metadata` subclass. (I'll break all your testcases in a follow-up commit, when I propagate this change to assembly.) llvm-svn: 223802
2014-12-10 02:38:53 +08:00
if (const auto *CI =
mdconst::dyn_extract<ConstantInt>(SrcLoc->getOperand(0)))
LocCookie = CI->getZExtValue();
}
}
void DiagnosticInfoInlineAsm::print(DiagnosticPrinter &DP) const {
DP << getMsgStr();
if (getLocCookie())
DP << " at line " << getLocCookie();
}
void DiagnosticInfoResourceLimit::print(DiagnosticPrinter &DP) const {
DP << getResourceName() << " limit";
if (getResourceLimit() != 0)
DP << " of " << getResourceLimit();
DP << " exceeded (" << getResourceSize() << ") in " << getFunction();
}
void DiagnosticInfoDebugMetadataVersion::print(DiagnosticPrinter &DP) const {
DP << "ignoring debug info with an invalid version (" << getMetadataVersion()
<< ") in " << getModule();
}
void DiagnosticInfoIgnoringInvalidDebugMetadata::print(
DiagnosticPrinter &DP) const {
DP << "ignoring invalid debug info in " << getModule().getModuleIdentifier();
}
void DiagnosticInfoSampleProfile::print(DiagnosticPrinter &DP) const {
if (!FileName.empty()) {
DP << getFileName();
if (LineNum > 0)
DP << ":" << getLineNum();
DP << ": ";
}
DP << getMsg();
}
void DiagnosticInfoPGOProfile::print(DiagnosticPrinter &DP) const {
if (getFileName())
DP << getFileName() << ": ";
DP << getMsg();
}
bool DiagnosticInfoWithDebugLocBase::isLocationAvailable() const {
return getDebugLoc();
}
void DiagnosticInfoWithDebugLocBase::getLocation(StringRef *Filename,
unsigned *Line,
unsigned *Column) const {
DILocation *L = getDebugLoc();
assert(L != nullptr && "debug location is invalid");
*Filename = L->getFilename();
*Line = L->getLine();
*Column = L->getColumn();
}
const std::string DiagnosticInfoWithDebugLocBase::getLocationStr() const {
StringRef Filename("<unknown>");
unsigned Line = 0;
unsigned Column = 0;
if (isLocationAvailable())
getLocation(&Filename, &Line, &Column);
return (Filename + ":" + Twine(Line) + ":" + Twine(Column)).str();
}
DiagnosticInfoOptimizationBase::Argument::Argument(StringRef Key, const Value *V)
: Key(Key) {
if (auto *F = dyn_cast<Function>(V)) {
if (DISubprogram *SP = F->getSubprogram())
DLoc = DebugLoc::get(SP->getScopeLine(), 0, SP);
}
else if (auto *I = dyn_cast<Instruction>(V))
DLoc = I->getDebugLoc();
// Only include names that correspond to user variables. FIXME: we should use
// debug info if available to get the name of the user variable.
if (isa<llvm::Argument>(V) || isa<GlobalValue>(V))
Val = GlobalValue::getRealLinkageName(V->getName());
else if (isa<Constant>(V)) {
raw_string_ostream OS(Val);
V->printAsOperand(OS, /*PrintType=*/false);
} else if (auto *I = dyn_cast<Instruction>(V))
Val = I->getOpcodeName();
}
DiagnosticInfoOptimizationBase::Argument::Argument(StringRef Key, const Type *T)
: Key(Key) {
raw_string_ostream OS(Val);
OS << *T;
}
Output optimization remarks in YAML (Re-committed after moving the template specialization under the yaml namespace. GCC was complaining about this.) This allows various presentation of this data using an external tool. This was first recommended here[1]. As an example, consider this module: 1 int foo(); 2 int bar(); 3 4 int baz() { 5 return foo() + bar(); 6 } The inliner generates these missed-optimization remarks today (the hotness information is pulled from PGO): remark: /tmp/s.c:5:10: foo will not be inlined into baz (hotness: 30) remark: /tmp/s.c:5:18: bar will not be inlined into baz (hotness: 30) Now with -pass-remarks-output=<yaml-file>, we generate this YAML file: --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 10 } Function: baz Hotness: 30 Args: - Callee: foo - String: will not be inlined into - Caller: baz ... --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 18 } Function: baz Hotness: 30 Args: - Callee: bar - String: will not be inlined into - Caller: baz ... This is a summary of the high-level decisions: * There is a new streaming interface to emit optimization remarks. E.g. for the inliner remark above: ORE.emit(DiagnosticInfoOptimizationRemarkMissed( DEBUG_TYPE, "NotInlined", &I) << NV("Callee", Callee) << " will not be inlined into " << NV("Caller", CS.getCaller()) << setIsVerbose()); NV stands for named value and allows the YAML client to process a remark using its name (NotInlined) and the named arguments (Callee and Caller) without parsing the text of the message. Subsequent patches will update ORE users to use the new streaming API. * I am using YAML I/O for writing the YAML file. YAML I/O requires you to specify reading and writing at once but reading is highly non-trivial for some of the more complex LLVM types. Since it's not clear that we (ever) want to use LLVM to parse this YAML file, the code supports and asserts that we're writing only. On the other hand, I did experiment that the class hierarchy starting at DiagnosticInfoOptimizationBase can be mapped back from YAML generated here (see D24479). * The YAML stream is stored in the LLVM context. * In the example, we can probably further specify the IR value used, i.e. print "Function" rather than "Value". * As before hotness is computed in the analysis pass instead of DiganosticInfo. This avoids the layering problem since BFI is in Analysis while DiagnosticInfo is in IR. [1] https://reviews.llvm.org/D19678#419445 Differential Revision: https://reviews.llvm.org/D24587 llvm-svn: 282539
2016-09-28 04:55:07 +08:00
DiagnosticInfoOptimizationBase::Argument::Argument(StringRef Key, int N)
: Key(Key), Val(itostr(N)) {}
DiagnosticInfoOptimizationBase::Argument::Argument(StringRef Key, unsigned N)
: Key(Key), Val(utostr(N)) {}
void DiagnosticInfoOptimizationBase::print(DiagnosticPrinter &DP) const {
DP << getLocationStr() << ": " << getMsg();
[OptRemark,LDist] RFC: Add hotness attribute Summary: This is the first set of changes implementing the RFC from http://thread.gmane.org/gmane.comp.compilers.llvm.devel/98334 This is a cross-sectional patch; rather than implementing the hotness attribute for all optimization remarks and all passes in a patch set, it implements it for the 'missed-optimization' remark for Loop Distribution. My goal is to shake out the design issues before scaling it up to other types and passes. Hotness is computed as an integer as the multiplication of the block frequency with the function entry count. It's only printed in opt currently since clang prints the diagnostic fields directly. E.g.: remark: /tmp/t.c:3:3: loop not distributed: use -Rpass-analysis=loop-distribute for more info (hotness: 300) A new API added is similar to emitOptimizationRemarkMissed. The difference is that it additionally takes a code region that the diagnostic corresponds to. From this, hotness is computed using BFI. The new API is exposed via an analysis pass so that it can be made dependent on LazyBFI. (Thanks to Hal for the analysis pass idea.) This feature can all be enabled by setDiagnosticHotnessRequested in the LLVM context. If this is off, LazyBFI is not calculated (D22141) so there should be no overhead. A new command-line option is added to turn this on in opt. My plan is to switch all user of emitOptimizationRemark* to use this module instead. Reviewers: hfinkel Subscribers: rcox2, mzolotukhin, llvm-commits Differential Revision: http://reviews.llvm.org/D21771 llvm-svn: 275583
2016-07-16 01:23:20 +08:00
if (Hotness)
DP << " (hotness: " << *Hotness << ")";
}
OptimizationRemark::OptimizationRemark(const char *PassName,
StringRef RemarkName,
const DebugLoc &DLoc, Value *CodeRegion)
: DiagnosticInfoIROptimization(
DK_OptimizationRemark, DS_Remark, PassName, RemarkName,
*cast<BasicBlock>(CodeRegion)->getParent(), DLoc, CodeRegion) {}
OptimizationRemark::OptimizationRemark(const char *PassName,
StringRef RemarkName, Instruction *Inst)
: DiagnosticInfoIROptimization(DK_OptimizationRemark, DS_Remark, PassName,
RemarkName, *Inst->getParent()->getParent(),
Inst->getDebugLoc(), Inst->getParent()) {}
bool OptimizationRemark::isEnabled(StringRef PassName) {
return PassRemarksOptLoc.Pattern &&
PassRemarksOptLoc.Pattern->match(PassName);
}
OptimizationRemarkMissed::OptimizationRemarkMissed(const char *PassName,
StringRef RemarkName,
const DebugLoc &DLoc,
Value *CodeRegion)
: DiagnosticInfoIROptimization(
DK_OptimizationRemarkMissed, DS_Remark, PassName, RemarkName,
*cast<BasicBlock>(CodeRegion)->getParent(), DLoc, CodeRegion) {}
OptimizationRemarkMissed::OptimizationRemarkMissed(const char *PassName,
StringRef RemarkName,
Instruction *Inst)
: DiagnosticInfoIROptimization(DK_OptimizationRemarkMissed, DS_Remark,
PassName, RemarkName,
*Inst->getParent()->getParent(),
Inst->getDebugLoc(), Inst->getParent()) {}
Output optimization remarks in YAML (Re-committed after moving the template specialization under the yaml namespace. GCC was complaining about this.) This allows various presentation of this data using an external tool. This was first recommended here[1]. As an example, consider this module: 1 int foo(); 2 int bar(); 3 4 int baz() { 5 return foo() + bar(); 6 } The inliner generates these missed-optimization remarks today (the hotness information is pulled from PGO): remark: /tmp/s.c:5:10: foo will not be inlined into baz (hotness: 30) remark: /tmp/s.c:5:18: bar will not be inlined into baz (hotness: 30) Now with -pass-remarks-output=<yaml-file>, we generate this YAML file: --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 10 } Function: baz Hotness: 30 Args: - Callee: foo - String: will not be inlined into - Caller: baz ... --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 18 } Function: baz Hotness: 30 Args: - Callee: bar - String: will not be inlined into - Caller: baz ... This is a summary of the high-level decisions: * There is a new streaming interface to emit optimization remarks. E.g. for the inliner remark above: ORE.emit(DiagnosticInfoOptimizationRemarkMissed( DEBUG_TYPE, "NotInlined", &I) << NV("Callee", Callee) << " will not be inlined into " << NV("Caller", CS.getCaller()) << setIsVerbose()); NV stands for named value and allows the YAML client to process a remark using its name (NotInlined) and the named arguments (Callee and Caller) without parsing the text of the message. Subsequent patches will update ORE users to use the new streaming API. * I am using YAML I/O for writing the YAML file. YAML I/O requires you to specify reading and writing at once but reading is highly non-trivial for some of the more complex LLVM types. Since it's not clear that we (ever) want to use LLVM to parse this YAML file, the code supports and asserts that we're writing only. On the other hand, I did experiment that the class hierarchy starting at DiagnosticInfoOptimizationBase can be mapped back from YAML generated here (see D24479). * The YAML stream is stored in the LLVM context. * In the example, we can probably further specify the IR value used, i.e. print "Function" rather than "Value". * As before hotness is computed in the analysis pass instead of DiganosticInfo. This avoids the layering problem since BFI is in Analysis while DiagnosticInfo is in IR. [1] https://reviews.llvm.org/D19678#419445 Differential Revision: https://reviews.llvm.org/D24587 llvm-svn: 282539
2016-09-28 04:55:07 +08:00
bool OptimizationRemarkMissed::isEnabled(StringRef PassName) {
return PassRemarksMissedOptLoc.Pattern &&
PassRemarksMissedOptLoc.Pattern->match(PassName);
}
OptimizationRemarkAnalysis::OptimizationRemarkAnalysis(const char *PassName,
StringRef RemarkName,
const DebugLoc &DLoc,
Value *CodeRegion)
: DiagnosticInfoIROptimization(
DK_OptimizationRemarkAnalysis, DS_Remark, PassName, RemarkName,
*cast<BasicBlock>(CodeRegion)->getParent(), DLoc, CodeRegion) {}
OptimizationRemarkAnalysis::OptimizationRemarkAnalysis(const char *PassName,
StringRef RemarkName,
Instruction *Inst)
: DiagnosticInfoIROptimization(DK_OptimizationRemarkAnalysis, DS_Remark,
PassName, RemarkName,
*Inst->getParent()->getParent(),
Inst->getDebugLoc(), Inst->getParent()) {}
OptimizationRemarkAnalysis::OptimizationRemarkAnalysis(enum DiagnosticKind Kind,
const char *PassName,
StringRef RemarkName,
const DebugLoc &DLoc,
Value *CodeRegion)
: DiagnosticInfoIROptimization(Kind, DS_Remark, PassName, RemarkName,
*cast<BasicBlock>(CodeRegion)->getParent(),
DLoc, CodeRegion) {}
bool OptimizationRemarkAnalysis::isEnabled(StringRef PassName) {
return PassRemarksAnalysisOptLoc.Pattern &&
PassRemarksAnalysisOptLoc.Pattern->match(PassName);
}
void DiagnosticInfoMIRParser::print(DiagnosticPrinter &DP) const {
DP << Diagnostic;
}
void llvm::emitOptimizationRemark(LLVMContext &Ctx, const char *PassName,
const Function &Fn, const DebugLoc &DLoc,
const Twine &Msg) {
Ctx.diagnose(OptimizationRemark(PassName, Fn, DLoc, Msg));
}
void llvm::emitOptimizationRemarkMissed(LLVMContext &Ctx, const char *PassName,
const Function &Fn,
const DebugLoc &DLoc,
const Twine &Msg) {
Ctx.diagnose(OptimizationRemarkMissed(PassName, Fn, DLoc, Msg));
}
void llvm::emitOptimizationRemarkAnalysis(LLVMContext &Ctx,
const char *PassName,
const Function &Fn,
const DebugLoc &DLoc,
const Twine &Msg) {
Ctx.diagnose(OptimizationRemarkAnalysis(PassName, Fn, DLoc, Msg));
}
void llvm::emitOptimizationRemarkAnalysisFPCommute(LLVMContext &Ctx,
const char *PassName,
const Function &Fn,
const DebugLoc &DLoc,
const Twine &Msg) {
Ctx.diagnose(OptimizationRemarkAnalysisFPCommute(PassName, Fn, DLoc, Msg));
}
void llvm::emitOptimizationRemarkAnalysisAliasing(LLVMContext &Ctx,
const char *PassName,
const Function &Fn,
const DebugLoc &DLoc,
const Twine &Msg) {
Ctx.diagnose(OptimizationRemarkAnalysisAliasing(PassName, Fn, DLoc, Msg));
}
DiagnosticInfoOptimizationFailure::DiagnosticInfoOptimizationFailure(
const char *PassName, StringRef RemarkName, const DebugLoc &DLoc,
Value *CodeRegion)
: DiagnosticInfoIROptimization(
DK_OptimizationFailure, DS_Warning, PassName, RemarkName,
*cast<BasicBlock>(CodeRegion)->getParent(), DLoc, CodeRegion) {}
bool DiagnosticInfoOptimizationFailure::isEnabled() const {
// Only print warnings.
return getSeverity() == DS_Warning;
}
void DiagnosticInfoUnsupported::print(DiagnosticPrinter &DP) const {
std::string Str;
raw_string_ostream OS(Str);
OS << getLocationStr() << ": in function " << getFunction().getName() << ' '
<< *getFunction().getFunctionType() << ": " << Msg << '\n';
OS.flush();
DP << Str;
}
void DiagnosticInfoISelFallback::print(DiagnosticPrinter &DP) const {
DP << "Instruction selection used fallback path for " << getFunction();
}
Output optimization remarks in YAML (Re-committed after moving the template specialization under the yaml namespace. GCC was complaining about this.) This allows various presentation of this data using an external tool. This was first recommended here[1]. As an example, consider this module: 1 int foo(); 2 int bar(); 3 4 int baz() { 5 return foo() + bar(); 6 } The inliner generates these missed-optimization remarks today (the hotness information is pulled from PGO): remark: /tmp/s.c:5:10: foo will not be inlined into baz (hotness: 30) remark: /tmp/s.c:5:18: bar will not be inlined into baz (hotness: 30) Now with -pass-remarks-output=<yaml-file>, we generate this YAML file: --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 10 } Function: baz Hotness: 30 Args: - Callee: foo - String: will not be inlined into - Caller: baz ... --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 18 } Function: baz Hotness: 30 Args: - Callee: bar - String: will not be inlined into - Caller: baz ... This is a summary of the high-level decisions: * There is a new streaming interface to emit optimization remarks. E.g. for the inliner remark above: ORE.emit(DiagnosticInfoOptimizationRemarkMissed( DEBUG_TYPE, "NotInlined", &I) << NV("Callee", Callee) << " will not be inlined into " << NV("Caller", CS.getCaller()) << setIsVerbose()); NV stands for named value and allows the YAML client to process a remark using its name (NotInlined) and the named arguments (Callee and Caller) without parsing the text of the message. Subsequent patches will update ORE users to use the new streaming API. * I am using YAML I/O for writing the YAML file. YAML I/O requires you to specify reading and writing at once but reading is highly non-trivial for some of the more complex LLVM types. Since it's not clear that we (ever) want to use LLVM to parse this YAML file, the code supports and asserts that we're writing only. On the other hand, I did experiment that the class hierarchy starting at DiagnosticInfoOptimizationBase can be mapped back from YAML generated here (see D24479). * The YAML stream is stored in the LLVM context. * In the example, we can probably further specify the IR value used, i.e. print "Function" rather than "Value". * As before hotness is computed in the analysis pass instead of DiganosticInfo. This avoids the layering problem since BFI is in Analysis while DiagnosticInfo is in IR. [1] https://reviews.llvm.org/D19678#419445 Differential Revision: https://reviews.llvm.org/D24587 llvm-svn: 282539
2016-09-28 04:55:07 +08:00
DiagnosticInfoOptimizationBase &DiagnosticInfoOptimizationBase::
operator<<(StringRef S) {
Args.emplace_back(S);
return *this;
}
DiagnosticInfoOptimizationBase &DiagnosticInfoOptimizationBase::
operator<<(Argument A) {
Args.push_back(std::move(A));
return *this;
}
DiagnosticInfoOptimizationBase &DiagnosticInfoOptimizationBase::
operator<<(setIsVerbose V) {
IsVerbose = true;
return *this;
}
DiagnosticInfoOptimizationBase &DiagnosticInfoOptimizationBase::
operator<<(setExtraArgs EA) {
FirstExtraArgIndex = Args.size();
return *this;
}
Output optimization remarks in YAML (Re-committed after moving the template specialization under the yaml namespace. GCC was complaining about this.) This allows various presentation of this data using an external tool. This was first recommended here[1]. As an example, consider this module: 1 int foo(); 2 int bar(); 3 4 int baz() { 5 return foo() + bar(); 6 } The inliner generates these missed-optimization remarks today (the hotness information is pulled from PGO): remark: /tmp/s.c:5:10: foo will not be inlined into baz (hotness: 30) remark: /tmp/s.c:5:18: bar will not be inlined into baz (hotness: 30) Now with -pass-remarks-output=<yaml-file>, we generate this YAML file: --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 10 } Function: baz Hotness: 30 Args: - Callee: foo - String: will not be inlined into - Caller: baz ... --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 18 } Function: baz Hotness: 30 Args: - Callee: bar - String: will not be inlined into - Caller: baz ... This is a summary of the high-level decisions: * There is a new streaming interface to emit optimization remarks. E.g. for the inliner remark above: ORE.emit(DiagnosticInfoOptimizationRemarkMissed( DEBUG_TYPE, "NotInlined", &I) << NV("Callee", Callee) << " will not be inlined into " << NV("Caller", CS.getCaller()) << setIsVerbose()); NV stands for named value and allows the YAML client to process a remark using its name (NotInlined) and the named arguments (Callee and Caller) without parsing the text of the message. Subsequent patches will update ORE users to use the new streaming API. * I am using YAML I/O for writing the YAML file. YAML I/O requires you to specify reading and writing at once but reading is highly non-trivial for some of the more complex LLVM types. Since it's not clear that we (ever) want to use LLVM to parse this YAML file, the code supports and asserts that we're writing only. On the other hand, I did experiment that the class hierarchy starting at DiagnosticInfoOptimizationBase can be mapped back from YAML generated here (see D24479). * The YAML stream is stored in the LLVM context. * In the example, we can probably further specify the IR value used, i.e. print "Function" rather than "Value". * As before hotness is computed in the analysis pass instead of DiganosticInfo. This avoids the layering problem since BFI is in Analysis while DiagnosticInfo is in IR. [1] https://reviews.llvm.org/D19678#419445 Differential Revision: https://reviews.llvm.org/D24587 llvm-svn: 282539
2016-09-28 04:55:07 +08:00
std::string DiagnosticInfoOptimizationBase::getMsg() const {
std::string Str;
raw_string_ostream OS(Str);
for (const DiagnosticInfoOptimizationBase::Argument &Arg :
make_range(Args.begin(), FirstExtraArgIndex == -1
? Args.end()
: Args.begin() + FirstExtraArgIndex))
Output optimization remarks in YAML (Re-committed after moving the template specialization under the yaml namespace. GCC was complaining about this.) This allows various presentation of this data using an external tool. This was first recommended here[1]. As an example, consider this module: 1 int foo(); 2 int bar(); 3 4 int baz() { 5 return foo() + bar(); 6 } The inliner generates these missed-optimization remarks today (the hotness information is pulled from PGO): remark: /tmp/s.c:5:10: foo will not be inlined into baz (hotness: 30) remark: /tmp/s.c:5:18: bar will not be inlined into baz (hotness: 30) Now with -pass-remarks-output=<yaml-file>, we generate this YAML file: --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 10 } Function: baz Hotness: 30 Args: - Callee: foo - String: will not be inlined into - Caller: baz ... --- !Missed Pass: inline Name: NotInlined DebugLoc: { File: /tmp/s.c, Line: 5, Column: 18 } Function: baz Hotness: 30 Args: - Callee: bar - String: will not be inlined into - Caller: baz ... This is a summary of the high-level decisions: * There is a new streaming interface to emit optimization remarks. E.g. for the inliner remark above: ORE.emit(DiagnosticInfoOptimizationRemarkMissed( DEBUG_TYPE, "NotInlined", &I) << NV("Callee", Callee) << " will not be inlined into " << NV("Caller", CS.getCaller()) << setIsVerbose()); NV stands for named value and allows the YAML client to process a remark using its name (NotInlined) and the named arguments (Callee and Caller) without parsing the text of the message. Subsequent patches will update ORE users to use the new streaming API. * I am using YAML I/O for writing the YAML file. YAML I/O requires you to specify reading and writing at once but reading is highly non-trivial for some of the more complex LLVM types. Since it's not clear that we (ever) want to use LLVM to parse this YAML file, the code supports and asserts that we're writing only. On the other hand, I did experiment that the class hierarchy starting at DiagnosticInfoOptimizationBase can be mapped back from YAML generated here (see D24479). * The YAML stream is stored in the LLVM context. * In the example, we can probably further specify the IR value used, i.e. print "Function" rather than "Value". * As before hotness is computed in the analysis pass instead of DiganosticInfo. This avoids the layering problem since BFI is in Analysis while DiagnosticInfo is in IR. [1] https://reviews.llvm.org/D19678#419445 Differential Revision: https://reviews.llvm.org/D24587 llvm-svn: 282539
2016-09-28 04:55:07 +08:00
OS << Arg.Val;
return OS.str();
}