[BOLT] Detect Linux kernel version if the binary is a Linux kernel

Move LinuxKernelVersion to LinuxKernelRewriter

Author: FLZ101

bolt/include/bolt/Core/BinaryContext.h

@@ -180,34 +180,6 @@ struct JournalingStreams {
 Error createNonFatalBOLTError(const Twine &S);
 Error createFatalBOLTError(const Twine &S);
 
-/// Linux kernel version
-struct LKVersion {
-  LKVersion() {}
-  LKVersion(unsigned Major, unsigned Minor, unsigned Rev)
-      : Major(Major), Minor(Minor), Rev(Rev) {}
-
-  bool operator<(const LKVersion &Other) const {
-    return std::make_tuple(Major, Minor, Rev) <
-           std::make_tuple(Other.Major, Other.Minor, Other.Rev);
-  }
-
-  bool operator>(const LKVersion &Other) const { return Other < *this; }
-
-  bool operator<=(const LKVersion &Other) const { return !(*this > Other); }
-
-  bool operator>=(const LKVersion &Other) const { return !(*this < Other); }
-
-  bool operator==(const LKVersion &Other) const {
-    return Major == Other.Major && Minor == Other.Minor && Rev == Other.Rev;
-  }
-
-  bool operator!=(const LKVersion &Other) const { return !(*this == Other); }
-
-  unsigned Major{0};
-  unsigned Minor{0};
-  unsigned Rev{0};
-};
-
 class BinaryContext {
   BinaryContext() = delete;
 
@@ -698,8 +670,6 @@ class BinaryContext {
   /// Indicates if the binary is Linux kernel.
   bool IsLinuxKernel{false};
 
-  LKVersion LinuxKernelVersion;
-
   /// Indicates if relocations are available for usage.
   bool HasRelocations{false};
 

bolt/include/bolt/Core/BinaryData.h

@@ -169,6 +169,11 @@ class BinaryData {
     return Parent && (Parent == BD || Parent->isAncestorOf(BD));
   }
 
+  void updateSize(uint64_t N) {
+    if (N > Size)
+      Size = N;
+  }
+
   void setIsMoveable(bool Flag) { IsMoveable = Flag; }
   void setSection(BinarySection &NewSection);
   void setOutputSection(BinarySection &NewSection) {

bolt/lib/Core/BinaryContext.cpp

@@ -1076,6 +1076,7 @@ MCSymbol *BinaryContext::registerNameAtAddress(StringRef Name, uint64_t Address,
     BD = GAI->second;
     if (!BD->hasName(Name)) {
       GlobalSymbols[Name] = BD;
+      BD->updateSize(Size);
       BD->Symbols.push_back(Symbol);
     }
   }

bolt/lib/Rewrite/LinuxKernelRewriter.cpp

@@ -21,6 +21,8 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/Errc.h"
+#include "llvm/Support/ErrorOr.h"
+#include <regex>
 
 #define DEBUG_TYPE "bolt-linux"
 
@@ -89,6 +91,34 @@ static cl::opt<bool>
 
 } // namespace opts
 
+/// Linux kernel version
+struct LKVersion {
+  LKVersion() {}
+  LKVersion(unsigned Major, unsigned Minor, unsigned Rev)
+      : Major(Major), Minor(Minor), Rev(Rev) {}
+
+  bool operator<(const LKVersion &Other) const {
+    return std::make_tuple(Major, Minor, Rev) <
+           std::make_tuple(Other.Major, Other.Minor, Other.Rev);
+  }
+
+  bool operator>(const LKVersion &Other) const { return Other < *this; }
+
+  bool operator<=(const LKVersion &Other) const { return !(*this > Other); }
+
+  bool operator>=(const LKVersion &Other) const { return !(*this < Other); }
+
+  bool operator==(const LKVersion &Other) const {
+    return Major == Other.Major && Minor == Other.Minor && Rev == Other.Rev;
+  }
+
+  bool operator!=(const LKVersion &Other) const { return !(*this == Other); }
+
+  unsigned Major{0};
+  unsigned Minor{0};
+  unsigned Rev{0};
+};
+
 /// Linux Kernel supports stack unwinding using ORC (oops rewind capability).
 /// ORC state at every IP can be described by the following data structure.
 struct ORCState {
@@ -124,6 +154,8 @@ inline raw_ostream &operator<<(raw_ostream &OS, const ORCState &E) {
 namespace {
 
 class LinuxKernelRewriter final : public MetadataRewriter {
+  LKVersion LinuxKernelVersion;
+
   /// Information required for updating metadata referencing an instruction.
   struct InstructionFixup {
     BinarySection &Section; // Section referencing the instruction.
@@ -225,6 +257,8 @@ class LinuxKernelRewriter final : public MetadataRewriter {
   ErrorOr<BinarySection &> PCIFixupSection = std::errc::bad_address;
   static constexpr size_t PCI_FIXUP_ENTRY_SIZE = 16;
 
+  Error detectLinuxKernelVersion();
+
   /// Process linux kernel special sections and their relocations.
   void processLKSections();
 
@@ -290,6 +324,9 @@ class LinuxKernelRewriter final : public MetadataRewriter {
       : MetadataRewriter("linux-kernel-rewriter", BC) {}
 
   Error preCFGInitializer() override {
+    if (Error E = detectLinuxKernelVersion())
+      return E;
+
     processLKSections();
 
     if (Error E = processSMPLocks())
@@ -370,6 +407,28 @@ class LinuxKernelRewriter final : public MetadataRewriter {
   }
 };
 
+Error LinuxKernelRewriter::detectLinuxKernelVersion() {
+  if (BinaryData *BD = BC.getBinaryDataByName("linux_banner")) {
+    const BinarySection &Section = BD->getSection();
+    const std::string S =
+        Section.getContents().substr(BD->getOffset(), BD->getSize()).str();
+
+    const std::regex Re(R"---(Linux version ((\d+)\.(\d+)(\.(\d+))?))---");
+    std::smatch Match;
+    if (std::regex_search(S, Match, Re)) {
+      const unsigned Major = std::stoi(Match[2].str());
+      const unsigned Minor = std::stoi(Match[3].str());
+      const unsigned Rev = Match.size() > 5 ? std::stoi(Match[5].str()) : 0;
+      LinuxKernelVersion = LKVersion(Major, Minor, Rev);
+      BC.outs() << "BOLT-INFO: Linux kernel version is " << Match[1].str()
+                << "\n";
+      return Error::success();
+    }
+  }
+  return createStringError(errc::executable_format_error,
+                           "Linux kernel version is unknown");
+}
+
 void LinuxKernelRewriter::processLKSections() {
   processLKKSymtab();
   processLKKSymtab(true);

bolt/lib/Rewrite/RewriteInstance.cpp

@@ -61,7 +61,6 @@
 #include <fstream>
 #include <memory>
 #include <optional>
-#include <regex>
 #include <system_error>
 
 #undef  DEBUG_TYPE
@@ -1031,25 +1030,6 @@ void RewriteInstance::discoverFileObjects() {
       continue;
     }
 
-    if (BC->IsLinuxKernel && SymName == "linux_banner") {
-      const StringRef SectionContents =
-          cantFail(Section->getContents(), "can not get section contents");
-      const std::string S =
-          SectionContents
-              .substr(SymbolAddress - Section->getAddress(), SymbolSize)
-              .str();
-
-      const std::regex Re(R"---(Linux version ((\d+)\.(\d+)(\.(\d+))?))---");
-      std::smatch Match;
-      if (std::regex_search(S, Match, Re)) {
-        unsigned Major = std::stoi(Match[2].str());
-        unsigned Minor = std::stoi(Match[3].str());
-        unsigned Rev = Match.size() > 5 ? std::stoi(Match[5].str()) : 0;
-        BC->LinuxKernelVersion = LKVersion(Major, Minor, Rev);
-        BC->outs() << "BOLT-INFO: Linux kernel version is " << Match[1].str();
-      }
-    }
-
     if (!Section->isText()) {
       assert(SymbolType != SymbolRef::ST_Function &&
              "unexpected function inside non-code section");
@@ -1225,9 +1205,6 @@ void RewriteInstance::discoverFileObjects() {
     PreviousFunction = BF;
   }
 
-  if (BC->IsLinuxKernel && !BC->LinuxKernelVersion.Major)
-    BC->errs() << "BOLT-WARNING: Linux kernel version is unknown\n";
-
   // Read dynamic relocation first as their presence affects the way we process
   // static relocations. E.g. we will ignore a static relocation at an address
   // that is a subject to dynamic relocation processing.

bolt/test/X86/linux-alt-instruction.s

@@ -142,6 +142,15 @@ _start:
   .section .orc_unwind_ip
   .long .L0 + 2 - .
 
+## Linux kernel version
+  .rodata
+  .align 16
+  .globl linux_banner
+  .type  linux_banner, @object
+linux_banner:
+  .string  "Linux version 6.6.61\n"
+  .size  linux_banner, . - linux_banner
+
 ## Fake Linux Kernel sections.
   .section __ksymtab,"a",@progbits
   .section __ksymtab_gpl,"a",@progbits

bolt/test/X86/linux-bug-table.s

@@ -56,6 +56,15 @@ _start:
   .long .L1 - .  # instruction
   .org 2b + 12
 
+## Linux kernel version
+  .rodata
+  .align 16
+  .globl linux_banner
+  .type  linux_banner, @object
+linux_banner:
+  .string  "Linux version 6.6.61\n"
+  .size  linux_banner, . - linux_banner
+
 ## Fake Linux Kernel sections.
   .section __ksymtab,"a",@progbits
   .section __ksymtab_gpl,"a",@progbits

bolt/test/X86/linux-exceptions.s

@@ -59,6 +59,15 @@ foo:
   .long .LF0 - . # fixup
   .long 0        # data
 
+## Linux kernel version
+  .rodata
+  .align 16
+  .globl linux_banner
+  .type  linux_banner, @object
+linux_banner:
+  .string  "Linux version 6.6.61\n"
+  .size  linux_banner, . - linux_banner
+
 ## Fake Linux Kernel sections.
   .section __ksymtab,"a",@progbits
   .section __ksymtab_gpl,"a",@progbits

bolt/test/X86/linux-orc.s

@@ -157,6 +157,15 @@ bar:
   .section .orc_unwind_ip
   .long .L4 - .
 
+## Linux kernel version
+  .rodata
+  .align 16
+  .globl linux_banner
+  .type  linux_banner, @object
+linux_banner:
+  .string  "Linux version 6.6.61\n"
+  .size  linux_banner, . - linux_banner
+
 ## Fake Linux Kernel sections.
   .section __ksymtab,"a",@progbits
   .section __ksymtab_gpl,"a",@progbits

bolt/test/X86/linux-parainstructions.s

@@ -49,6 +49,15 @@ _start:
   .byte 1        # type
   .byte 7        # length
 
+## Linux kernel version
+  .rodata
+  .align 16
+  .globl linux_banner
+  .type  linux_banner, @object
+linux_banner:
+  .string  "Linux version 6.6.61\n"
+  .size  linux_banner, . - linux_banner
+
 ## Fake Linux Kernel sections.
   .section __ksymtab,"a",@progbits
   .section __ksymtab_gpl,"a",@progbits