1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-19 19:12:56 +02:00

Add dump of Win64 EH unwind data.

The new command line option -unwind-info dumps the Win64 EH unwind
data to the console. This is a nice feature if you need to debug
generated EH data (e.g. from LLVM). Includes a test case.

Initial patch by João Matos, extensions and rework by Kai Nacke.

llvm-svn: 169415
This commit is contained in:
Michael J. Spencer 2012-12-05 20:12:35 +00:00
parent 649965f21b
commit 938a1f875b
5 changed files with 497 additions and 3 deletions

View File

@ -0,0 +1,106 @@
// This test checks that the unwind data is dumped by llvm-objdump.
// RUN: llvm-mc -triple x86_64-pc-win32 -filetype=obj %s | llvm-objdump -u - | FileCheck %s
// CHECK: Unwind info:
// CHECK: Function Table:
// CHECK-NEXT: Start Address: .text
// CHECK-NEXT: End Address: .text + 0x001b
// CHECK-NEXT: Unwind Info Address: .xdata
// CHECK-NEXT: Version: 1
// CHECK-NEXT: Flags: 1 UNW_ExceptionHandler
// CHECK-NEXT: Size of prolog: 18
// CHECK-NEXT: Number of Codes: 8
// CHECK-NEXT: Frame register: RBX
// CHECK-NEXT: Frame offset: 0
// CHECK-NEXT: Unwind Codes:
// CHECK-NEXT: 0x00: UOP_SetFPReg
// CHECK-NEXT: 0x0f: UOP_PushNonVol RBX
// CHECK-NEXT: 0x0e: UOP_SaveXMM128 XMM8 [0x0000]
// CHECK-NEXT: 0x09: UOP_SaveNonVol RSI [0x0010]
// CHECK-NEXT: 0x04: UOP_AllocSmall 24
// CHECK-NEXT: 0x00: UOP_PushMachFrame w/o error code
// CHECK: Function Table:
// CHECK-NEXT: Start Address: .text + 0x0012
// CHECK-NEXT: End Address: .text + 0x0012
// CHECK-NEXT: Unwind Info Address: .xdata + 0x001c
// CHECK-NEXT: Version: 1
// CHECK-NEXT: Flags: 4 UNW_ChainInfo
// CHECK-NEXT: Size of prolog: 0
// CHECK-NEXT: Number of Codes: 0
// CHECK-NEXT: No frame pointer used
// CHECK: Function Table:
// CHECK-NEXT: Start Address: .text + 0x001b
// CHECK-NEXT: End Address: .text + 0x001c
// CHECK-NEXT: Unwind Info Address: .xdata + 0x002c
// CHECK-NEXT: Version: 1
// CHECK-NEXT: Flags: 0
// CHECK-NEXT: Size of prolog: 0
// CHECK-NEXT: Number of Codes: 0
// CHECK-NEXT: No frame pointer used
// CHECK: Function Table:
// CHECK-NEXT: Start Address: .text + 0x001c
// CHECK-NEXT: End Address: .text + 0x0039
// CHECK-NEXT: Unwind Info Address: .xdata + 0x0034
// CHECK-NEXT: Version: 1
// CHECK-NEXT: Flags: 0
// CHECK-NEXT: Size of prolog: 14
// CHECK-NEXT: Number of Codes: 6
// CHECK-NEXT: No frame pointer used
// CHECK-NEXT: Unwind Codes:
// CHECK-NEXT: 0x0e: UOP_AllocLarge 8454128
// CHECK-NEXT: 0x07: UOP_AllocLarge 8190
// CHECK-NEXT: 0x00: UOP_PushMachFrame w/o error code
.text
.globl func
.def func; .scl 2; .type 32; .endef
.seh_proc func
func:
.seh_pushframe @code
subq $24, %rsp
.seh_stackalloc 24
movq %rsi, 16(%rsp)
.seh_savereg %rsi, 16
movups %xmm8, (%rsp)
.seh_savexmm %xmm8, 0
pushq %rbx
.seh_pushreg 3
mov %rsp, %rbx
.seh_setframe 3, 0
.seh_endprologue
.seh_handler __C_specific_handler, @except
.seh_handlerdata
.long 0
.text
.seh_startchained
.seh_endprologue
.seh_endchained
lea (%rbx), %rsp
pop %rbx
addq $24, %rsp
ret
.seh_endproc
// Test emission of small functions.
.globl smallFunc
.def smallFunc; .scl 2; .type 32; .endef
.seh_proc smallFunc
smallFunc:
ret
.seh_endproc
// Function with big stack allocation.
.globl smallFunc
.def allocFunc; .scl 2; .type 32; .endef
.seh_proc smallFunc
allocFunc:
.seh_pushframe @code
subq $65520, %rsp
.seh_stackalloc 65520
sub $8454128, %rsp
.seh_stackalloc 8454128
.seh_endprologue
add $8454128, %rsp
addq $65520, %rsp
ret
.seh_endproc

View File

@ -9,6 +9,7 @@ set(LLVM_LINK_COMPONENTS
add_llvm_tool(llvm-objdump
llvm-objdump.cpp
COFFDump.cpp
MachODump.cpp
MCFunction.cpp
)

View File

@ -0,0 +1,355 @@
//===-- COFFDump.cpp - COFF-specific dumper ---------------------*- C++ -*-===//
//
// The LLVM Compiler Infrastructure
//
// This file is distributed under the University of Illinois Open Source
// License. See LICENSE.TXT for details.
//
//===----------------------------------------------------------------------===//
///
/// \file
/// \brief This file implements the COFF-specific dumper for llvm-objdump.
/// It outputs the Win64 EH data structures as plain text.
/// The encoding of the unwind codes is decribed in MSDN:
/// http://msdn.microsoft.com/en-us/library/ck9asaa9.aspx
///
//===----------------------------------------------------------------------===//
#include "llvm-objdump.h"
#include "llvm/Object/COFF.h"
#include "llvm/Object/ObjectFile.h"
#include "llvm/Support/Format.h"
#include "llvm/Support/SourceMgr.h"
#include "llvm/Support/raw_ostream.h"
#include "llvm/Support/system_error.h"
#include "llvm/Support/Win64EH.h"
#include <algorithm>
#include <cstring>
using namespace llvm;
using namespace object;
using namespace llvm::Win64EH;
// Returns the name of the unwind code.
static StringRef getUnwindCodeTypeName(uint8_t Code) {
switch(Code) {
default: llvm_unreachable("Invalid unwind code");
case UOP_PushNonVol: return "UOP_PushNonVol";
case UOP_AllocLarge: return "UOP_AllocLarge";
case UOP_AllocSmall: return "UOP_AllocSmall";
case UOP_SetFPReg: return "UOP_SetFPReg";
case UOP_SaveNonVol: return "UOP_SaveNonVol";
case UOP_SaveNonVolBig: return "UOP_SaveNonVolBig";
case UOP_SaveXMM128: return "UOP_SaveXMM128";
case UOP_SaveXMM128Big: return "UOP_SaveXMM128Big";
case UOP_PushMachFrame: return "UOP_PushMachFrame";
}
}
// Returns the name of a referenced register.
static StringRef getUnwindRegisterName(uint8_t Reg) {
switch(Reg) {
default: llvm_unreachable("Invalid register");
case 0: return "RAX";
case 1: return "RCX";
case 2: return "RDX";
case 3: return "RBX";
case 4: return "RSP";
case 5: return "RBP";
case 6: return "RSI";
case 7: return "RDI";
case 8: return "R8";
case 9: return "R9";
case 10: return "R10";
case 11: return "R11";
case 12: return "R12";
case 13: return "R13";
case 14: return "R14";
case 15: return "R15";
}
}
// Calculates the number of array slots required for the unwind code.
static unsigned getNumUsedSlots(const UnwindCode &UnwindCode) {
switch (UnwindCode.getUnwindOp()) {
default: llvm_unreachable("Invalid unwind code");
case UOP_PushNonVol:
case UOP_AllocSmall:
case UOP_SetFPReg:
case UOP_PushMachFrame:
return 1;
case UOP_SaveNonVol:
case UOP_SaveXMM128:
return 2;
case UOP_SaveNonVolBig:
case UOP_SaveXMM128Big:
return 3;
case UOP_AllocLarge:
return (UnwindCode.getOpInfo() == 0) ? 2 : 3;
}
}
// Prints one unwind code. Because an unwind code can occupy up to 3 slots in
// the unwind codes array, this function requires that the correct number of
// slots is provided.
static void printUnwindCode(ArrayRef<UnwindCode> UCs) {
assert(UCs.size() >= getNumUsedSlots(UCs[0]));
outs() << format(" 0x%02x: ", unsigned(UCs[0].u.CodeOffset))
<< getUnwindCodeTypeName(UCs[0].getUnwindOp());
switch (UCs[0].getUnwindOp()) {
case UOP_PushNonVol:
outs() << " " << getUnwindRegisterName(UCs[0].getOpInfo());
break;
case UOP_AllocLarge:
if (UCs[0].getOpInfo() == 0) {
outs() << " " << UCs[1].FrameOffset;
} else {
outs() << " " << UCs[1].FrameOffset
+ (static_cast<uint32_t>(UCs[2].FrameOffset) << 16);
}
break;
case UOP_AllocSmall:
outs() << " " << ((UCs[0].getOpInfo() + 1) * 8);
break;
case UOP_SetFPReg:
outs() << " ";
break;
case UOP_SaveNonVol:
outs() << " " << getUnwindRegisterName(UCs[0].getOpInfo())
<< format(" [0x%04x]", 8 * UCs[1].FrameOffset);
break;
case UOP_SaveNonVolBig:
outs() << " " << getUnwindRegisterName(UCs[0].getOpInfo())
<< format(" [0x%08x]", UCs[1].FrameOffset
+ (static_cast<uint32_t>(UCs[2].FrameOffset) << 16));
break;
case UOP_SaveXMM128:
outs() << " XMM" << static_cast<uint32_t>(UCs[0].getOpInfo())
<< format(" [0x%04x]", 16 * UCs[1].FrameOffset);
break;
case UOP_SaveXMM128Big:
outs() << " XMM" << UCs[0].getOpInfo()
<< format(" [0x%08x]", UCs[1].FrameOffset
+ (static_cast<uint32_t>(UCs[2].FrameOffset) << 16));
break;
case UOP_PushMachFrame:
outs() << " " << (UCs[0].getOpInfo() ? "w/o" : "w")
<< " error code";
break;
}
outs() << "\n";
}
static void printAllUnwindCodes(ArrayRef<UnwindCode> UCs) {
for (const UnwindCode *I = UCs.begin(), *E = UCs.end(); I < E; ) {
unsigned UsedSlots = getNumUsedSlots(*I);
if (UsedSlots > UCs.size()) {
outs() << "Unwind data corrupted: Encountered unwind op "
<< getUnwindCodeTypeName((*I).getUnwindOp())
<< " which requires " << UsedSlots
<< " slots, but only " << UCs.size()
<< " remaining in buffer";
return ;
}
printUnwindCode(ArrayRef<UnwindCode>(I, E));
I += UsedSlots;
}
}
// Given a symbol sym this functions returns the address and section of it.
static error_code resolveSectionAndAddress(const COFFObjectFile *Obj,
const SymbolRef &Sym,
const coff_section *&ResolvedSection,
uint64_t &ResolvedAddr) {
if (error_code ec = Sym.getAddress(ResolvedAddr)) return ec;
section_iterator iter(Obj->begin_sections());
if (error_code ec = Sym.getSection(iter)) return ec;
ResolvedSection = Obj->getCOFFSection(iter);
return object_error::success;
}
// Given a vector of relocations for a section and an offset into this section
// the function returns the symbol used for the relocation at the offset.
static error_code resolveSymbol(const std::vector<RelocationRef> &Rels,
uint64_t Offset, SymbolRef &Sym) {
for (std::vector<RelocationRef>::const_iterator I = Rels.begin(),
E = Rels.end();
I != E; ++I) {
uint64_t Ofs;
if (error_code ec = I->getOffset(Ofs)) return ec;
if (Ofs == Offset) {
if (error_code ec = I->getSymbol(Sym)) return ec;
break;
}
}
return object_error::success;
}
// Given a vector of relocations for a section and an offset into this section
// the function resolves the symbol used for the relocation at the offset and
// returns the section content and the address inside the content pointed to
// by the symbol.
static error_code getSectionContents(const COFFObjectFile *Obj,
const std::vector<RelocationRef> &Rels,
uint64_t Offset,
ArrayRef<uint8_t> &Contents,
uint64_t &Addr) {
SymbolRef Sym;
if (error_code ec = resolveSymbol(Rels, Offset, Sym)) return ec;
const coff_section *Section;
if (error_code ec = resolveSectionAndAddress(Obj, Sym, Section, Addr))
return ec;
if (error_code ec = Obj->getSectionContents(Section, Contents)) return ec;
return object_error::success;
}
// Given a vector of relocations for a section and an offset into this section
// the function returns the name of the symbol used for the relocation at the
// offset.
static error_code resolveSymbolName(const std::vector<RelocationRef> &Rels,
uint64_t Offset, StringRef &Name) {
SymbolRef Sym;
if (error_code ec = resolveSymbol(Rels, Offset, Sym)) return ec;
if (error_code ec = Sym.getName(Name)) return ec;
return object_error::success;
}
static void printCOFFSymbolAddress(llvm::raw_ostream &Out,
const std::vector<RelocationRef> &Rels,
uint64_t Offset, uint32_t Disp) {
StringRef Sym;
if (error_code ec = resolveSymbolName(Rels, Offset, Sym)) {
error(ec);
return ;
}
Out << Sym;
if (Disp > 0)
Out << format(" + 0x%04x", Disp);
}
void llvm::printCOFFUnwindInfo(const COFFObjectFile *Obj) {
const coff_file_header *Header;
if (error(Obj->getHeader(Header))) return;
if (Header->Machine != COFF::IMAGE_FILE_MACHINE_AMD64) {
errs() << "Unsupported image machine type "
"(currently only AMD64 is supported).\n";
return;
}
const coff_section *Pdata = 0;
error_code ec;
for (section_iterator SI = Obj->begin_sections(),
SE = Obj->end_sections();
SI != SE; SI.increment(ec)) {
if (error(ec)) return;
StringRef Name;
if (error(SI->getName(Name))) continue;
if (Name != ".pdata") continue;
Pdata = Obj->getCOFFSection(SI);
std::vector<RelocationRef> Rels;
for (relocation_iterator RI = SI->begin_relocations(),
RE = SI->end_relocations();
RI != RE; RI.increment(ec)) {
if (error(ec)) break;
Rels.push_back(*RI);
}
// Sort relocations by address.
std::sort(Rels.begin(), Rels.end(), RelocAddressLess);
ArrayRef<uint8_t> Contents;
if (error(Obj->getSectionContents(Pdata, Contents))) continue;
if (Contents.empty()) continue;
ArrayRef<RuntimeFunction> RFs(
reinterpret_cast<const RuntimeFunction *>(Contents.data()),
Contents.size() / sizeof(RuntimeFunction));
for (const RuntimeFunction *I = RFs.begin(), *E = RFs.end(); I < E; ++I) {
const uint64_t SectionOffset = std::distance(RFs.begin(), I)
* sizeof(RuntimeFunction);
outs() << "Function Table:\n";
outs() << " Start Address: ";
printCOFFSymbolAddress(outs(), Rels, SectionOffset +
offsetof(RuntimeFunction, StartAddress),
I->StartAddress);
outs() << "\n";
outs() << " End Address: ";
printCOFFSymbolAddress(outs(), Rels, SectionOffset +
offsetof(RuntimeFunction, EndAddress),
I->EndAddress);
outs() << "\n";
outs() << " Unwind Info Address: ";
printCOFFSymbolAddress(outs(), Rels, SectionOffset +
offsetof(RuntimeFunction, UnwindInfoOffset),
I->UnwindInfoOffset);
outs() << "\n";
ArrayRef<uint8_t> XContents;
uint64_t UnwindInfoOffset = 0;
if (error(getSectionContents(Obj, Rels, SectionOffset +
offsetof(RuntimeFunction, UnwindInfoOffset),
XContents, UnwindInfoOffset))) continue;
if (XContents.empty()) continue;
UnwindInfoOffset += I->UnwindInfoOffset;
if (UnwindInfoOffset > XContents.size()) continue;
const Win64EH::UnwindInfo *UI =
reinterpret_cast<const Win64EH::UnwindInfo *>
(XContents.data() + UnwindInfoOffset);
// The casts to int are required in order to output the value as number.
// Without the casts the value would be interpreted as char data (which
// results in garbage output).
outs() << " Version: " << static_cast<int>(UI->getVersion()) << "\n";
outs() << " Flags: " << static_cast<int>(UI->getFlags());
if (UI->getFlags()) {
if (UI->getFlags() & UNW_ExceptionHandler)
outs() << " UNW_ExceptionHandler";
if (UI->getFlags() & UNW_TerminateHandler)
outs() << " UNW_TerminateHandler";
if (UI->getFlags() & UNW_ChainInfo)
outs() << " UNW_ChainInfo";
}
outs() << "\n";
outs() << " Size of prolog: "
<< static_cast<int>(UI->PrologSize) << "\n";
outs() << " Number of Codes: "
<< static_cast<int>(UI->NumCodes) << "\n";
// Maybe this should move to output of UOP_SetFPReg?
if (UI->getFrameRegister()) {
outs() << " Frame register: "
<< getUnwindRegisterName(UI->getFrameRegister())
<< "\n";
outs() << " Frame offset: "
<< 16 * UI->getFrameOffset()
<< "\n";
} else {
outs() << " No frame pointer used\n";
}
if (UI->getFlags() & (UNW_ExceptionHandler | UNW_TerminateHandler)) {
// FIXME: Output exception handler data
} else if (UI->getFlags() & UNW_ChainInfo) {
// FIXME: Output chained unwind info
}
if (UI->NumCodes)
outs() << " Unwind Codes:\n";
printAllUnwindCodes(ArrayRef<UnwindCode>(&UI->UnwindCodes[0],
UI->NumCodes));
outs() << "\n\n";
outs().flush();
}
}
}

View File

@ -104,9 +104,16 @@ static cl::opt<bool>
NoShowRawInsn("no-show-raw-insn", cl::desc("When disassembling instructions, "
"do not print the instruction bytes."));
static cl::opt<bool>
UnwindInfo("unwind-info", cl::desc("Display unwind information"));
static cl::alias
UnwindInfoShort("u", cl::desc("Alias for --unwind-info"),
cl::aliasopt(UnwindInfo));
static StringRef ToolName;
static bool error(error_code ec) {
bool llvm::error(error_code ec) {
if (!ec) return false;
outs() << ToolName << ": error reading file: " << ec.message() << ".\n";
@ -165,7 +172,7 @@ void llvm::DumpBytes(StringRef bytes) {
outs() << output;
}
static bool RelocAddressLess(RelocationRef a, RelocationRef b) {
bool llvm::RelocAddressLess(RelocationRef a, RelocationRef b) {
uint64_t a_addr, b_addr;
if (error(a.getAddress(a_addr))) return false;
if (error(b.getAddress(b_addr))) return false;
@ -573,6 +580,19 @@ static void PrintSymbolTable(const ObjectFile *o) {
}
}
static void PrintUnwindInfo(const ObjectFile *o) {
outs() << "Unwind info:\n\n";
if (const COFFObjectFile *coff = dyn_cast<COFFObjectFile>(o)) {
printCOFFUnwindInfo(coff);
} else {
// TODO: Extract DWARF dump tool to objdump.
errs() << "This operation is only currently supported "
"for COFF object files.\n";
return;
}
}
static void DumpObject(const ObjectFile *o) {
outs() << '\n';
outs() << o->getFileName()
@ -588,6 +608,8 @@ static void DumpObject(const ObjectFile *o) {
PrintSectionContents(o);
if (SymbolTable)
PrintSymbolTable(o);
if (UnwindInfo)
PrintUnwindInfo(o);
}
/// @brief Dump each object file in \a a;
@ -666,7 +688,8 @@ int main(int argc, char **argv) {
&& !Relocations
&& !SectionHeaders
&& !SectionContents
&& !SymbolTable) {
&& !SymbolTable
&& !UnwindInfo) {
cl::PrintHelpMessage();
return 2;
}

View File

@ -17,12 +17,21 @@
namespace llvm {
namespace object {
class COFFObjectFile;
class RelocationRef;
}
class error_code;
extern cl::opt<std::string> TripleName;
extern cl::opt<std::string> ArchName;
// Various helper functions.
bool error(error_code ec);
bool RelocAddressLess(object::RelocationRef a, object::RelocationRef b);
void DumpBytes(StringRef bytes);
void DisassembleInputMachO(StringRef Filename);
void printCOFFUnwindInfo(const object::COFFObjectFile* o);
class StringRefMemoryObject : public MemoryObject {
virtual void anchor();