1 //===- llvm-profgen.cpp - LLVM SPGO profile generation tool -----*- C++ -*-===//
2 //
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 //
7 //===----------------------------------------------------------------------===//
8 //
9 // llvm-profgen generates SPGO profiles from perf script ouput.
10 //
11 //===----------------------------------------------------------------------===//
12 
13 #include "ErrorHandling.h"
14 #include "PerfReader.h"
15 #include "ProfileGenerator.h"
16 #include "ProfiledBinary.h"
17 #include "llvm/DebugInfo/Symbolize/SymbolizableModule.h"
18 #include "llvm/Support/CommandLine.h"
19 #include "llvm/Support/FileSystem.h"
20 #include "llvm/Support/InitLLVM.h"
21 #include "llvm/Support/TargetSelect.h"
22 
23 static cl::OptionCategory ProfGenCategory("ProfGen Options");
24 
25 static cl::opt<std::string> PerfScriptFilename(
26     "perfscript", cl::value_desc("perfscript"),
27     llvm::cl::MiscFlags::CommaSeparated,
28     cl::desc("Path of perf-script trace created by Linux perf tool with "
29              "`script` command(the raw perf.data should be profiled with -b)"),
30     cl::cat(ProfGenCategory));
31 static cl::alias PSA("ps", cl::desc("Alias for --perfscript"),
32                      cl::aliasopt(PerfScriptFilename));
33 
34 static cl::opt<std::string> PerfDataFilename(
35     "perfdata", cl::value_desc("perfdata"), llvm::cl::MiscFlags::CommaSeparated,
36     cl::desc("Path of raw perf data created by Linux perf tool (it should be "
37              "profiled with -b)"),
38     cl::cat(ProfGenCategory));
39 static cl::alias PDA("pd", cl::desc("Alias for --perfdata"),
40                      cl::aliasopt(PerfDataFilename));
41 
42 static cl::opt<std::string> UnsymbolizedProfFilename(
43     "unsymbolized-profile", cl::value_desc("unsymbolized profile"),
44     llvm::cl::MiscFlags::CommaSeparated,
45     cl::desc("Path of the unsymbolized profile created by "
46              "`llvm-profgen` with `--skip-symbolization`"),
47     cl::cat(ProfGenCategory));
48 static cl::alias UPA("up", cl::desc("Alias for --unsymbolized-profile"),
49                      cl::aliasopt(UnsymbolizedProfFilename));
50 
51 static cl::opt<std::string> SampleProfFilename(
52     "llvm-sample-profile", cl::value_desc("llvm sample profile"),
53     cl::desc("Path of the LLVM sample profile"), cl::cat(ProfGenCategory));
54 
55 static cl::opt<std::string>
56     BinaryPath("binary", cl::value_desc("binary"), cl::Required,
57                cl::desc("Path of profiled executable binary."),
58                cl::cat(ProfGenCategory));
59 
60 static cl::opt<uint32_t>
61     ProcessId("pid", cl::value_desc("process Id"), cl::init(0),
62               cl::desc("Process Id for the profiled executable binary."),
63               cl::cat(ProfGenCategory));
64 
65 static cl::opt<std::string> DebugBinPath(
66     "debug-binary", cl::value_desc("debug-binary"),
67     cl::desc("Path of debug info binary, llvm-profgen will load the DWARF info "
68              "from it instead of the executable binary."),
69     cl::cat(ProfGenCategory));
70 
71 extern cl::opt<bool> ShowDisassemblyOnly;
72 extern cl::opt<bool> ShowSourceLocations;
73 extern cl::opt<bool> SkipSymbolization;
74 
75 using namespace llvm;
76 using namespace sampleprof;
77 
78 // Validate the command line input.
validateCommandLine()79 static void validateCommandLine() {
80   // Allow the missing perfscript if we only use to show binary disassembly.
81   if (!ShowDisassemblyOnly) {
82     // Validate input profile is provided only once
83     uint16_t HasPerfData = PerfDataFilename.getNumOccurrences();
84     uint16_t HasPerfScript = PerfScriptFilename.getNumOccurrences();
85     uint16_t HasUnsymbolizedProfile =
86         UnsymbolizedProfFilename.getNumOccurrences();
87     uint16_t HasSampleProfile = SampleProfFilename.getNumOccurrences();
88     uint16_t S =
89         HasPerfData + HasPerfScript + HasUnsymbolizedProfile + HasSampleProfile;
90     if (S != 1) {
91       std::string Msg =
92           S > 1
93               ? "`--perfscript`, `--perfdata` and `--unsymbolized-profile` "
94                 "cannot be used together."
95               : "Perf input file is missing, please use one of `--perfscript`, "
96                 "`--perfdata` and `--unsymbolized-profile` for the input.";
97       exitWithError(Msg);
98     }
99 
100     auto CheckFileExists = [](bool H, StringRef File) {
101       if (H && !llvm::sys::fs::exists(File)) {
102         std::string Msg = "Input perf file(" + File.str() + ") doesn't exist.";
103         exitWithError(Msg);
104       }
105     };
106 
107     CheckFileExists(HasPerfData, PerfDataFilename);
108     CheckFileExists(HasPerfScript, PerfScriptFilename);
109     CheckFileExists(HasUnsymbolizedProfile, UnsymbolizedProfFilename);
110     CheckFileExists(HasSampleProfile, SampleProfFilename);
111   }
112 
113   if (!llvm::sys::fs::exists(BinaryPath)) {
114     std::string Msg = "Input binary(" + BinaryPath + ") doesn't exist.";
115     exitWithError(Msg);
116   }
117 
118   if (CSProfileGenerator::MaxCompressionSize < -1) {
119     exitWithError("Value of --compress-recursion should >= -1");
120   }
121   if (ShowSourceLocations && !ShowDisassemblyOnly) {
122     exitWithError("--show-source-locations should work together with "
123                   "--show-disassembly-only!");
124   }
125 }
126 
getPerfInputFile()127 static PerfInputFile getPerfInputFile() {
128   PerfInputFile File;
129   if (PerfDataFilename.getNumOccurrences()) {
130     File.InputFile = PerfDataFilename;
131     File.Format = PerfFormat::PerfData;
132   } else if (PerfScriptFilename.getNumOccurrences()) {
133     File.InputFile = PerfScriptFilename;
134     File.Format = PerfFormat::PerfScript;
135   } else if (UnsymbolizedProfFilename.getNumOccurrences()) {
136     File.InputFile = UnsymbolizedProfFilename;
137     File.Format = PerfFormat::UnsymbolizedProfile;
138   }
139   return File;
140 }
141 
main(int argc,const char * argv[])142 int main(int argc, const char *argv[]) {
143   InitLLVM X(argc, argv);
144 
145   // Initialize targets and assembly printers/parsers.
146   InitializeAllTargetInfos();
147   InitializeAllTargetMCs();
148   InitializeAllDisassemblers();
149 
150   cl::HideUnrelatedOptions({&ProfGenCategory, &getColorCategory()});
151   cl::ParseCommandLineOptions(argc, argv, "llvm SPGO profile generator\n");
152   validateCommandLine();
153 
154   // Load symbols and disassemble the code of a given binary.
155   std::unique_ptr<ProfiledBinary> Binary =
156       std::make_unique<ProfiledBinary>(BinaryPath, DebugBinPath);
157   if (ShowDisassemblyOnly)
158     return EXIT_SUCCESS;
159 
160   if (SampleProfFilename.getNumOccurrences()) {
161     LLVMContext Context;
162     auto ReaderOrErr = SampleProfileReader::create(SampleProfFilename, Context);
163     std::unique_ptr<sampleprof::SampleProfileReader> Reader =
164         std::move(ReaderOrErr.get());
165     Reader->read();
166     std::unique_ptr<ProfileGeneratorBase> Generator =
167         ProfileGeneratorBase::create(Binary.get(), Reader->getProfiles(),
168                                      Reader->profileIsCS());
169     Generator->generateProfile();
170     Generator->write();
171   } else {
172     Optional<uint32_t> PIDFilter;
173     if (ProcessId.getNumOccurrences())
174       PIDFilter = ProcessId;
175     PerfInputFile PerfFile = getPerfInputFile();
176     std::unique_ptr<PerfReaderBase> Reader =
177         PerfReaderBase::create(Binary.get(), PerfFile, PIDFilter);
178     // Parse perf events and samples
179     Reader->parsePerfTraces();
180 
181     if (SkipSymbolization)
182       return EXIT_SUCCESS;
183 
184     std::unique_ptr<ProfileGeneratorBase> Generator =
185         ProfileGeneratorBase::create(Binary.get(), &Reader->getSampleCounters(),
186                                      Reader->profileIsCS());
187     Generator->generateProfile();
188     Generator->write();
189   }
190 
191   return EXIT_SUCCESS;
192 }
193