diff options
-rw-r--r-- | CMakeLists.txt | 1 | ||||
-rw-r--r-- | docs/yaml2obj.rst | 222 | ||||
-rw-r--r-- | include/llvm/Support/COFF.h | 8 | ||||
-rw-r--r-- | test/Object/COFF/i386.yaml | 83 | ||||
-rw-r--r-- | test/Object/COFF/x86-64.yaml | 83 | ||||
-rw-r--r-- | test/Object/Inputs/trivial-object-test.coff-i386 | bin | 346 -> 346 bytes | |||
-rw-r--r-- | test/Object/nm-trivial-object.test | 8 | ||||
-rw-r--r-- | test/Object/objdump-file-header.test | 3 | ||||
-rw-r--r-- | test/Object/objdump-section-content.test | 5 | ||||
-rw-r--r-- | test/Object/objdump-symbol-table.test | 6 | ||||
-rw-r--r-- | utils/Makefile | 2 | ||||
-rw-r--r-- | utils/yaml2obj/CMakeLists.txt | 5 | ||||
-rw-r--r-- | utils/yaml2obj/Makefile | 20 | ||||
-rw-r--r-- | utils/yaml2obj/yaml2obj.cpp | 879 |
14 files changed, 1311 insertions, 14 deletions
diff --git a/CMakeLists.txt b/CMakeLists.txt index bb64db91bc..ed230aee5b 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -402,6 +402,7 @@ add_subdirectory(utils/not) add_subdirectory(utils/llvm-lit) add_subdirectory(utils/yaml-bench) add_subdirectory(utils/obj2yaml) +add_subdirectory(utils/yaml2obj) add_subdirectory(projects) diff --git a/docs/yaml2obj.rst b/docs/yaml2obj.rst new file mode 100644 index 0000000000..cb59162e5a --- /dev/null +++ b/docs/yaml2obj.rst @@ -0,0 +1,222 @@ +.. _yaml2obj: + +yaml2obj +======== + +yaml2obj takes a YAML description of an object file and converts it to a binary +file. + + $ yaml2py input-file + +.. program:: yaml2py + +Outputs the binary to stdout. + +COFF Syntax +----------- + +Here's a sample COFF file. + +.. code-block:: yaml + + header: + Machine: IMAGE_FILE_MACHINE_I386 # (0x14C) + + sections: + - Name: .text + Characteristics: [ IMAGE_SCN_CNT_CODE + , IMAGE_SCN_ALIGN_16BYTES + , IMAGE_SCN_MEM_EXECUTE + , IMAGE_SCN_MEM_READ + ] # 0x60500020 + SectionData: + "\x83\xEC\x0C\xC7\x44\x24\x08\x00\x00\x00\x00\xC7\x04\x24\x00\x00\x00\x00\xE8\x00\x00\x00\x00\xE8\x00\x00\x00\x00\x8B\x44\x24\x08\x83\xC4\x0C\xC3" # |....D$.......$...............D$.....| + + symbols: + - Name: .text + Value: 0 + SectionNumber: 1 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_STATIC # (3) + NumberOfAuxSymbols: 1 + AuxillaryData: + "\x24\x00\x00\x00\x03\x00\x00\x00\x00\x00\x00\x00\x01\x00\x00\x00\x00\x00" # |$.................| + + - Name: _main + Value: 0 + SectionNumber: 1 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_EXTERNAL # (2) + +Here's a simplified Kwalify_ schema with an extension to allow alternate types. + +.. _Kwalify: http://www.kuwata-lab.com/kwalify/ruby/users-guide.html + +.. code-block:: yaml + + type: map + mapping: + header: + type: map + mapping: + Machine: [ {type: str, enum: + [ IMAGE_FILE_MACHINE_UNKNOWN + , IMAGE_FILE_MACHINE_AM33 + , IMAGE_FILE_MACHINE_AMD64 + , IMAGE_FILE_MACHINE_ARM + , IMAGE_FILE_MACHINE_ARMV7 + , IMAGE_FILE_MACHINE_EBC + , IMAGE_FILE_MACHINE_I386 + , IMAGE_FILE_MACHINE_IA64 + , IMAGE_FILE_MACHINE_M32R + , IMAGE_FILE_MACHINE_MIPS16 + , IMAGE_FILE_MACHINE_MIPSFPU + , IMAGE_FILE_MACHINE_MIPSFPU16 + , IMAGE_FILE_MACHINE_POWERPC + , IMAGE_FILE_MACHINE_POWERPCFP + , IMAGE_FILE_MACHINE_R4000 + , IMAGE_FILE_MACHINE_SH3 + , IMAGE_FILE_MACHINE_SH3DSP + , IMAGE_FILE_MACHINE_SH4 + , IMAGE_FILE_MACHINE_SH5 + , IMAGE_FILE_MACHINE_THUMB + , IMAGE_FILE_MACHINE_WCEMIPSV2 + ]} + , {type: int} + ] + Characteristics: + - type: seq + sequence: + - type: str + enum: [ IMAGE_FILE_RELOCS_STRIPPED + , IMAGE_FILE_EXECUTABLE_IMAGE + , IMAGE_FILE_LINE_NUMS_STRIPPED + , IMAGE_FILE_LOCAL_SYMS_STRIPPED + , IMAGE_FILE_AGGRESSIVE_WS_TRIM + , IMAGE_FILE_LARGE_ADDRESS_AWARE + , IMAGE_FILE_BYTES_REVERSED_LO + , IMAGE_FILE_32BIT_MACHINE + , IMAGE_FILE_DEBUG_STRIPPED + , IMAGE_FILE_REMOVABLE_RUN_FROM_SWAP + , IMAGE_FILE_NET_RUN_FROM_SWAP + , IMAGE_FILE_SYSTEM + , IMAGE_FILE_DLL + , IMAGE_FILE_UP_SYSTEM_ONLY + , IMAGE_FILE_BYTES_REVERSED_HI + ] + - type: int + sections: + type: seq + sequence: + - type: map + mapping: + Name: {type: str} + Characteristics: + - type: seq + sequence: + - type: str + enum: [ IMAGE_SCN_TYPE_NO_PAD + , IMAGE_SCN_CNT_CODE + , IMAGE_SCN_CNT_INITIALIZED_DATA + , IMAGE_SCN_CNT_UNINITIALIZED_DATA + , IMAGE_SCN_LNK_OTHER + , IMAGE_SCN_LNK_INFO + , IMAGE_SCN_LNK_REMOVE + , IMAGE_SCN_LNK_COMDAT + , IMAGE_SCN_GPREL + , IMAGE_SCN_MEM_PURGEABLE + , IMAGE_SCN_MEM_16BIT + , IMAGE_SCN_MEM_LOCKED + , IMAGE_SCN_MEM_PRELOAD + , IMAGE_SCN_ALIGN_1BYTES + , IMAGE_SCN_ALIGN_2BYTES + , IMAGE_SCN_ALIGN_4BYTES + , IMAGE_SCN_ALIGN_8BYTES + , IMAGE_SCN_ALIGN_16BYTES + , IMAGE_SCN_ALIGN_32BYTES + , IMAGE_SCN_ALIGN_64BYTES + , IMAGE_SCN_ALIGN_128BYTES + , IMAGE_SCN_ALIGN_256BYTES + , IMAGE_SCN_ALIGN_512BYTES + , IMAGE_SCN_ALIGN_1024BYTES + , IMAGE_SCN_ALIGN_2048BYTES + , IMAGE_SCN_ALIGN_4096BYTES + , IMAGE_SCN_ALIGN_8192BYTES + , IMAGE_SCN_LNK_NRELOC_OVFL + , IMAGE_SCN_MEM_DISCARDABLE + , IMAGE_SCN_MEM_NOT_CACHED + , IMAGE_SCN_MEM_NOT_PAGED + , IMAGE_SCN_MEM_SHARED + , IMAGE_SCN_MEM_EXECUTE + , IMAGE_SCN_MEM_READ + , IMAGE_SCN_MEM_WRITE + ] + - type: int + SectionData: {type: str} + symbols: + type: seq + sequence: + - type: map + mapping: + Name: {type: str} + Value: {type: int} + SectionNumber: {type: int} + SimpleType: [ {type: str, enum: [ IMAGE_SYM_TYPE_NULL + , IMAGE_SYM_TYPE_VOID + , IMAGE_SYM_TYPE_CHAR + , IMAGE_SYM_TYPE_SHORT + , IMAGE_SYM_TYPE_INT + , IMAGE_SYM_TYPE_LONG + , IMAGE_SYM_TYPE_FLOAT + , IMAGE_SYM_TYPE_DOUBLE + , IMAGE_SYM_TYPE_STRUCT + , IMAGE_SYM_TYPE_UNION + , IMAGE_SYM_TYPE_ENUM + , IMAGE_SYM_TYPE_MOE + , IMAGE_SYM_TYPE_BYTE + , IMAGE_SYM_TYPE_WORD + , IMAGE_SYM_TYPE_UINT + , IMAGE_SYM_TYPE_DWORD + ]} + , {type: int} + ] + ComplexType: [ {type: str, enum: [ IMAGE_SYM_DTYPE_NULL + , IMAGE_SYM_DTYPE_POINTER + , IMAGE_SYM_DTYPE_FUNCTION + , IMAGE_SYM_DTYPE_ARRAY + ]} + , {type: int} + ] + StorageClass: [ {type: str, enum: + [ IMAGE_SYM_CLASS_END_OF_FUNCTION + , IMAGE_SYM_CLASS_NULL + , IMAGE_SYM_CLASS_AUTOMATIC + , IMAGE_SYM_CLASS_EXTERNAL + , IMAGE_SYM_CLASS_STATIC + , IMAGE_SYM_CLASS_REGISTER + , IMAGE_SYM_CLASS_EXTERNAL_DEF + , IMAGE_SYM_CLASS_LABEL + , IMAGE_SYM_CLASS_UNDEFINED_LABEL + , IMAGE_SYM_CLASS_MEMBER_OF_STRUCT + , IMAGE_SYM_CLASS_ARGUMENT + , IMAGE_SYM_CLASS_STRUCT_TAG + , IMAGE_SYM_CLASS_MEMBER_OF_UNION + , IMAGE_SYM_CLASS_UNION_TAG + , IMAGE_SYM_CLASS_TYPE_DEFINITION + , IMAGE_SYM_CLASS_UNDEFINED_STATIC + , IMAGE_SYM_CLASS_ENUM_TAG + , IMAGE_SYM_CLASS_MEMBER_OF_ENUM + , IMAGE_SYM_CLASS_REGISTER_PARAM + , IMAGE_SYM_CLASS_BIT_FIELD + , IMAGE_SYM_CLASS_BLOCK + , IMAGE_SYM_CLASS_FUNCTION + , IMAGE_SYM_CLASS_END_OF_STRUCT + , IMAGE_SYM_CLASS_FILE + , IMAGE_SYM_CLASS_SECTION + , IMAGE_SYM_CLASS_WEAK_EXTERNAL + , IMAGE_SYM_CLASS_CLR_TOKEN + ]} + , {type: int} + ] diff --git a/include/llvm/Support/COFF.h b/include/llvm/Support/COFF.h index b18ba2b550..6c2ee08756 100644 --- a/include/llvm/Support/COFF.h +++ b/include/llvm/Support/COFF.h @@ -50,6 +50,8 @@ namespace COFF { }; enum MachineTypes { + MT_Invalid = -1, + IMAGE_FILE_MACHINE_UNKNOWN = 0x0, IMAGE_FILE_MACHINE_AM33 = 0x13, IMAGE_FILE_MACHINE_AMD64 = 0x8664, @@ -74,6 +76,8 @@ namespace COFF { }; enum Characteristics { + C_Invalid = 0, + /// The file does not contain base relocations and must be loaded at its /// preferred base. If this cannot be done, the loader will error. IMAGE_FILE_RELOCS_STRIPPED = 0x0001, @@ -138,6 +142,8 @@ namespace COFF { /// Storage class tells where and what the symbol represents enum SymbolStorageClass { + SSC_Invalid = -1, + IMAGE_SYM_CLASS_END_OF_FUNCTION = -1, ///< Physical end of function IMAGE_SYM_CLASS_NULL = 0, ///< No symbol IMAGE_SYM_CLASS_AUTOMATIC = 1, ///< Stack variable @@ -214,6 +220,8 @@ namespace COFF { }; enum SectionCharacteristics { + SC_Invalid = -1, + IMAGE_SCN_TYPE_NO_PAD = 0x00000008, IMAGE_SCN_CNT_CODE = 0x00000020, IMAGE_SCN_CNT_INITIALIZED_DATA = 0x00000040, diff --git a/test/Object/COFF/i386.yaml b/test/Object/COFF/i386.yaml new file mode 100644 index 0000000000..ca902220c1 --- /dev/null +++ b/test/Object/COFF/i386.yaml @@ -0,0 +1,83 @@ +header: !Header + Machine: IMAGE_FILE_MACHINE_I386 # (0x14c) + +sections: + - !Section + Name: .text + Characteristics: [IMAGE_SCN_CNT_CODE, IMAGE_SCN_ALIGN_16BYTES, IMAGE_SCN_MEM_EXECUTE, IMAGE_SCN_MEM_READ, ] # 0x60500020 + SectionData: !hex "83EC0CC744240800000000C7042400000000E800000000E8000000008B44240883C40CC3" # |....D$.......$...............D$.....| + + Relocations: + - !Relocation + VirtualAddress: 0xe + SymbolTableIndex: 5 + Type: IMAGE_REL_I386_DIR32 + + - !Relocation + VirtualAddress: 0x13 + SymbolTableIndex: 6 + Type: IMAGE_REL_I386_REL32 + + - !Relocation + VirtualAddress: 0x18 + SymbolTableIndex: 7 + Type: IMAGE_REL_I386_REL32 + + - !Section + Name: .data + Characteristics: [IMAGE_SCN_CNT_INITIALIZED_DATA, IMAGE_SCN_ALIGN_1BYTES, IMAGE_SCN_MEM_READ, IMAGE_SCN_MEM_WRITE, ] # 0xc0100040 + SectionData: !hex "48656C6C6F20576F726C642100" # |Hello World!.| + +symbols: + - !Symbol + Name: .text + Value: 0 + SectionNumber: 1 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_STATIC # (3) + NumberOfAuxSymbols: 1 + AuxillaryData: !hex "240000000300000000000000010000000000" # |$.................| + + - !Symbol + Name: .data + Value: 0 + SectionNumber: 2 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_STATIC # (3) + NumberOfAuxSymbols: 1 + AuxillaryData: !hex "0D0000000000000000000000020000000000" # |..................| + + - !Symbol + Name: _main + Value: 0 + SectionNumber: 1 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_FUNCTION # (2) + StorageClass: IMAGE_SYM_CLASS_EXTERNAL # (2) + + - !Symbol + Name: L_.str + Value: 0 + SectionNumber: 2 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_STATIC # (3) + + - !Symbol + Name: _puts + Value: 0 + SectionNumber: 0 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_EXTERNAL # (2) + + - !Symbol + Name: _SomeOtherFunction + Value: 0 + SectionNumber: 0 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_EXTERNAL # (2) + diff --git a/test/Object/COFF/x86-64.yaml b/test/Object/COFF/x86-64.yaml new file mode 100644 index 0000000000..0b1265f6f5 --- /dev/null +++ b/test/Object/COFF/x86-64.yaml @@ -0,0 +1,83 @@ +header: !Header + Machine: IMAGE_FILE_MACHINE_AMD64 # (0x8664) + +sections: + - !Section + Name: .text + Characteristics: [IMAGE_SCN_CNT_CODE, IMAGE_SCN_ALIGN_16BYTES, IMAGE_SCN_MEM_EXECUTE, IMAGE_SCN_MEM_READ, ] # 0x60500020 + SectionData: !hex "4883EC28C744242400000000488D0D00000000E800000000E8000000008B4424244883C428C3" # |H..(.D$$....H.................D$$H..(.| + + Relocations: + - !Relocation + VirtualAddress: 0xf + SymbolTableIndex: 5 + Type: IMAGE_REL_AMD64_REL32 + + - !Relocation + VirtualAddress: 0x14 + SymbolTableIndex: 6 + Type: IMAGE_REL_AMD64_REL32 + + - !Relocation + VirtualAddress: 0x19 + SymbolTableIndex: 7 + Type: IMAGE_REL_AMD64_REL32 + + - !Section + Name: .data + Characteristics: [IMAGE_SCN_CNT_INITIALIZED_DATA, IMAGE_SCN_ALIGN_1BYTES, IMAGE_SCN_MEM_READ, IMAGE_SCN_MEM_WRITE, ] # 0xc0100040 + SectionData: !hex "48656C6C6F20576F726C642100" # |Hello World!.| + +symbols: + - !Symbol + Name: .text + Value: 0 + SectionNumber: 1 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_STATIC # (3) + NumberOfAuxSymbols: 1 + AuxillaryData: !hex "260000000300000000000000010000000000" # |&.................| + + - !Symbol + Name: .data + Value: 0 + SectionNumber: 2 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_STATIC # (3) + NumberOfAuxSymbols: 1 + AuxillaryData: !hex "0D0000000000000000000000020000000000" # |..................| + + - !Symbol + Name: main + Value: 0 + SectionNumber: 1 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_FUNCTION # (2) + StorageClass: IMAGE_SYM_CLASS_EXTERNAL # (2) + + - !Symbol + Name: L.str + Value: 0 + SectionNumber: 2 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_STATIC # (3) + + - !Symbol + Name: puts + Value: 0 + SectionNumber: 0 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_EXTERNAL # (2) + + - !Symbol + Name: SomeOtherFunction + Value: 0 + SectionNumber: 0 + SimpleType: IMAGE_SYM_TYPE_NULL # (0) + ComplexType: IMAGE_SYM_DTYPE_NULL # (0) + StorageClass: IMAGE_SYM_CLASS_EXTERNAL # (2) + diff --git a/test/Object/Inputs/trivial-object-test.coff-i386 b/test/Object/Inputs/trivial-object-test.coff-i386 Binary files differindex 8cfd9949b1..d4ab63b70d 100644 --- a/test/Object/Inputs/trivial-object-test.coff-i386 +++ b/test/Object/Inputs/trivial-object-test.coff-i386 diff --git a/test/Object/nm-trivial-object.test b/test/Object/nm-trivial-object.test index e5635ab475..b6f70a02a7 100644 --- a/test/Object/nm-trivial-object.test +++ b/test/Object/nm-trivial-object.test @@ -1,7 +1,5 @@ -RUN: llvm-nm %p/Inputs/trivial-object-test.coff-i386 \ -RUN: | FileCheck %s -check-prefix COFF -RUN: llvm-nm %p/Inputs/trivial-object-test.coff-x86-64 \ -RUN: | FileCheck %s -check-prefix COFF +RUN: yaml2obj %p/COFF/i386.yaml | llvm-nm | FileCheck %s -check-prefix COFF +RUN: yaml2obj %p/COFF/x86-64.yaml | llvm-nm | FileCheck %s -check-prefix COFF RUN: llvm-nm %p/Inputs/trivial-object-test.elf-i386 \ RUN: | FileCheck %s -check-prefix ELF RUN: llvm-nm %p/Inputs/trivial-object-test.elf-x86-64 \ @@ -30,4 +28,4 @@ macho: 00000000 U _puts macho64: 00000028 s L_.str macho64: 00000000 u _SomeOtherFunction macho64: 00000000 s _main -macho64: 00000000 u _puts
\ No newline at end of file +macho64: 00000000 u _puts diff --git a/test/Object/objdump-file-header.test b/test/Object/objdump-file-header.test index 3fce3f4071..f35a8a1fa2 100644 --- a/test/Object/objdump-file-header.test +++ b/test/Object/objdump-file-header.test @@ -1,5 +1,4 @@ -RUN: llvm-objdump -f %p/Inputs/trivial-object-test.coff-i386 \ -RUN: | FileCheck %s -check-prefix COFF-i386 +RUN: yaml2obj %p/COFF/i386.yaml | llvm-objdump -f - | FileCheck %s -check-prefix COFF-i386 RUN: llvm-objdump -f %p/Inputs/trivial-object-test.elf-i386 \ RUN: | FileCheck %s -check-prefix ELF-i386 diff --git a/test/Object/objdump-section-content.test b/test/Object/objdump-section-content.test index 581e75eb5a..56b7989a60 100644 --- a/test/Object/objdump-section-content.test +++ b/test/Object/objdump-section-content.test @@ -1,9 +1,8 @@ -RUN: llvm-objdump -s %p/Inputs/trivial-object-test.coff-i386 \ -RUN: | FileCheck %s -check-prefix COFF-i386 +RUN: yaml2obj %p/COFF/i386.yaml | llvm-objdump -s - | FileCheck %s -check-prefix COFF-i386 RUN: llvm-objdump -s %p/Inputs/trivial-object-test.elf-i386 \ RUN: | FileCheck %s -check-prefix ELF-i386 -COFF-i386: trivial-object-test.coff-i386: file format +COFF-i386: file format COFF-i386: Contents of section .text: COFF-i386: 0000 83ec0cc7 44240800 000000c7 04240000 ....D$.......$.. COFF-i386: 0010 0000e800 000000e8 00000000 8b442408 .............D$. diff --git a/test/Object/objdump-symbol-table.test b/test/Object/objdump-symbol-table.test index 8a0f44042d..a721171c09 100644 --- a/test/Object/objdump-symbol-table.test +++ b/test/Object/objdump-symbol-table.test @@ -1,17 +1,17 @@ -RUN: llvm-objdump -t %p/Inputs/trivial-object-test.coff-i386 \ +RUN: yaml2obj %p/COFF/i386.yaml | llvm-objdump -t - \ RUN: | FileCheck %s -check-prefix COFF-i386 RUN: llvm-objdump -t %p/Inputs/trivial-object-test.elf-i386 \ RUN: | FileCheck %s -check-prefix ELF-i386 RUN: llvm-objdump -t %p/Inputs/trivial-object-test.macho-i386 \ RUN: | FileCheck %s -check-prefix macho-i386 -COFF-i386: trivial-object-test.coff-i386: file format +COFF-i386: file format COFF-i386: SYMBOL TABLE: COFF-i386: [ 0](sec 1)(fl 0x00)(ty 0)(scl 3) (nx 1) 0x00000000 .text COFF-i386: AUX scnlen 0x24 nreloc 3 nlnno 0 checksum 0x0 assoc 1 comdat 0 COFF-i386: [ 2](sec 2)(fl 0x00)(ty 0)(scl 3) (nx 1) 0x00000000 .data COFF-i386: AUX scnlen 0xd nreloc 0 nlnno 0 checksum 0x0 assoc 2 comdat 0 -COFF-i386: [ 4](sec 1)(fl 0x00)(ty 200)(scl 2) (nx 0) 0x00000000 _main +COFF-i386: [ 4](sec 1)(fl 0x00)(ty 20)(scl 2) (nx 0) 0x00000000 _main COFF-i386: [ 5](sec 2)(fl 0x00)(ty 0)(scl 3) (nx 0) 0x00000000 L_.str COFF-i386: [ 6](sec 0)(fl 0x00)(ty 0)(scl 2) (nx 0) 0x00000000 _puts COFF-i386: [ 7](sec 0)(fl 0x00)(ty 0)(scl 2) (nx 0) 0x00000000 _SomeOtherFunction diff --git a/utils/Makefile b/utils/Makefile index ecb30bed7c..7a3c17d032 100644 --- a/utils/Makefile +++ b/utils/Makefile @@ -9,7 +9,7 @@ LEVEL = .. PARALLEL_DIRS := FileCheck FileUpdate TableGen PerfectShuffle \ - count fpcmp llvm-lit not unittest + count fpcmp llvm-lit not unittest yaml2obj EXTRA_DIST := check-each-file codegen-diff countloc.sh \ DSAclean.py DSAextract.py emacs findsym.pl GenLibDeps.pl \ diff --git a/utils/yaml2obj/CMakeLists.txt b/utils/yaml2obj/CMakeLists.txt new file mode 100644 index 0000000000..f8b1197524 --- /dev/null +++ b/utils/yaml2obj/CMakeLists.txt @@ -0,0 +1,5 @@ +add_llvm_utility(yaml2obj + yaml2obj.cpp + ) + +target_link_libraries(yaml2obj LLVMSupport) diff --git a/utils/yaml2obj/Makefile b/utils/yaml2obj/Makefile new file mode 100644 index 0000000000..e746d85190 --- /dev/null +++ b/utils/yaml2obj/Makefile @@ -0,0 +1,20 @@ +##===- utils/yaml2obj/Makefile ----------------------------*- Makefile -*-===## +# +# The LLVM Compiler Infrastructure +# +# This file is distributed under the University of Illinois Open Source +# License. See LICENSE.TXT for details. +# +##===----------------------------------------------------------------------===## + +LEVEL = ../.. +TOOLNAME = yaml2obj +USEDLIBS = LLVMSupport.a + +# This tool has no plugins, optimize startup time. +TOOL_NO_EXPORTS = 1 + +# Don't install this utility +NO_INSTALL = 1 + +include $(LEVEL)/Makefile.common diff --git a/utils/yaml2obj/yaml2obj.cpp b/utils/yaml2obj/yaml2obj.cpp new file mode 100644 index 0000000000..d21336f5fd --- /dev/null +++ b/utils/yaml2obj/yaml2obj.cpp @@ -0,0 +1,879 @@ +//===- yaml2obj - Convert YAML to a binary object file --------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// +// +// This program takes a YAML description of an object file and outputs the +// binary equivalent. +// +// This is used for writing tests that require binary files. +// +//===----------------------------------------------------------------------===// + +#include "llvm/ADT/SmallString.h" +#include "llvm/ADT/StringExtras.h" +#include "llvm/ADT/StringMap.h" +#include "llvm/ADT/StringSwitch.h" +#include "llvm/Support/COFF.h" +#include "llvm/Support/Casting.h" +#include "llvm/Support/CommandLine.h" +#include "llvm/Support/Endian.h" +#include "llvm/Support/ManagedStatic.h" +#include "llvm/Support/MemoryBuffer.h" +#include "llvm/Support/PrettyStackTrace.h" +#include "llvm/Support/raw_ostream.h" +#include "llvm/Support/Signals.h" +#include "llvm/Support/SourceMgr.h" +#include "llvm/Support/system_error.h" +#include "llvm/Support/YAMLParser.h" + +#include <vector> + +using namespace llvm; + +static cl::opt<std::string> + Input(cl::Positional, cl::desc("<input>"), cl::init("-")); + +template<class T> +typename std::enable_if<std::numeric_limits<T>::is_integer, bool>::type +getAs(const llvm::yaml::ScalarNode *SN, T &Result) { + SmallString<4> Storage; + StringRef Value = SN->getValue(Storage); + if (Value.getAsInteger(0, Result)) + return false; + return true; +} + +// Given a container with begin and end with ::value_type of a character type. +// Iterate through pairs of characters in the the set of [a-fA-F0-9] ignoring +// all other characters. +struct hex_pair_iterator { + StringRef::const_iterator Current, End; + typedef SmallVector<char, 2> value_type; + value_type Pair; + bool IsDone; + + hex_pair_iterator(StringRef C) + : Current(C.begin()), End(C.end()), IsDone(false) { + // Initalize Pair. + ++*this; + } + + // End iterator. + hex_pair_iterator() : Current(), End(), IsDone(true) {} + + value_type operator *() const { + return Pair; + } + + hex_pair_iterator operator ++() { + // We're at the end of the input. + if (Current == End) { + IsDone = true; + return *this; + } + Pair = value_type(); + for (; Current != End && Pair.size() != 2; ++Current) { + // Is a valid hex digit. + if ((*Current >= '0' && *Current <= '9') || + (*Current >= 'a' && *Current <= 'f') || + (*Current >= 'A' && *Current <= 'F')) + Pair.push_back(*Current); + } + // Hit the end without getting 2 hex digits. Pair is invalid. + if (Pair.size() != 2) + IsDone = true; + return *this; + } + + bool operator ==(const hex_pair_iterator Other) { + return (IsDone == Other.IsDone == true) || + (Current == Other.Current && End == Other.End); + } + + bool operator !=(const hex_pair_iterator Other) { + return !(*this == Other); + } +}; + +template <class ContainerOut> +static bool hexStringToByteArray(StringRef Str, ContainerOut &Out) { + for (hex_pair_iterator I(Str), E; I != E; ++I) { + typename hex_pair_iterator::value_type Pair = *I; + typename ContainerOut::value_type Byte; + if (StringRef(Pair.data(), 2).getAsInteger(16, Byte)) + return false; + Out.push_back(Byte); + } + return true; +} + +/// This parses a yaml stream that represents a COFF object file. +/// See docs/yaml2obj for the yaml scheema. +struct COFFParser { + COFFParser(yaml::Stream &Input) : YS(Input) { + std::memset(&Header, 0, sizeof(Header)); + // A COFF string table always starts with a 4 byte size field. Offsets into + // it include this size, so allocate it now. + StringTable.append(4, 0); + } + + bool parseHeader(yaml::Node *HeaderN) { + yaml::MappingNode *MN = dyn_cast<yaml::MappingNode>(HeaderN); + if (!MN) { + YS.printError(HeaderN, "header's value must be a mapping node"); + return false; + } + for (yaml::MappingNode::iterator i = MN->begin(), e = MN->end(); + i != e; ++i) { + yaml::ScalarNode *Key = dyn_cast<yaml::ScalarNode>(i->getKey()); + if (!Key) { + YS.printError(i->getKey(), "Keys must be scalar values"); + return false; + } + SmallString<32> Storage; + StringRef KeyValue = Key->getValue(Storage); + if (KeyValue == "Characteristics") { + if (!parseHeaderCharacteristics(i->getValue())) + return false; + } else { + yaml::ScalarNode *Value = dyn_cast<yaml::ScalarNode>(i->getValue()); + if (!Value) { + YS.printError(Value, + Twine(KeyValue) + " must be a scalar value"); + return false; + } + if (KeyValue == "Machine") { + uint16_t Machine; + if (!getAs(Value, Machine)) { + // It's not a raw number, try matching the string. + StringRef ValueValue = Value->getValue(Storage); + Machine = StringSwitch<COFF::MachineTypes>(ValueValue) + .Case( "IMAGE_FILE_MACHINE_UNKNOWN" + , COFF::IMAGE_FILE_MACHINE_UNKNOWN) + .Case( "IMAGE_FILE_MACHINE_AM33" + , COFF::IMAGE_FILE_MACHINE_AM33) + .Case( "IMAGE_FILE_MACHINE_AMD64" + , COFF::IMAGE_FILE_MACHINE_AMD64) + .Case( "IMAGE_FILE_MACHINE_ARM" + , COFF::IMAGE_FILE_MACHINE_ARM) + .Case( "IMAGE_FILE_MACHINE_ARMV7" + , COFF::IMAGE_FILE_MACHINE_ARMV7) + .Case( "IMAGE_FILE_MACHINE_EBC" + , COFF::IMAGE_FILE_MACHINE_EBC) + .Case( "IMAGE_FILE_MACHINE_I386" + , COFF::IMAGE_FILE_MACHINE_I386) + .Case( "IMAGE_FILE_MACHINE_IA64" + , COFF::IMAGE_FILE_MACHINE_IA64) + .Case( "IMAGE_FILE_MACHINE_M32R" + , COFF::IMAGE_FILE_MACHINE_M32R) + .Case( "IMAGE_FILE_MACHINE_MIPS16" + , COFF::IMAGE_FILE_MACHINE_MIPS16) + .Case( "IMAGE_FILE_MACHINE_MIPSFPU" + , COFF::IMAGE_FILE_MACHINE_MIPSFPU) + .Case( "IMAGE_FILE_MACHINE_MIPSFPU16" + , COFF::IMAGE_FILE_MACHINE_MIPSFPU16) + .Case( "IMAGE_FILE_MACHINE_POWERPC" + , COFF::IMAGE_FILE_MACHINE_POWERPC) + .Case( "IMAGE_FILE_MACHINE_POWERPCFP" + , COFF::IMAGE_FILE_MACHINE_POWERPCFP) + .Case( "IMAGE_FILE_MACHINE_R4000" + , COFF::IMAGE_FILE_MACHINE_R4000) + .Case( "IMAGE_FILE_MACHINE_SH3" + , COFF::IMAGE_FILE_MACHINE_SH3) + .Case( "IMAGE_FILE_MACHINE_SH3DSP" + , COFF::IMAGE_FILE_MACHINE_SH3DSP) + .Case( "IMAGE_FILE_MACHINE_SH4" + , COFF::IMAGE_FILE_MACHINE_SH4) + .Case( "IMAGE_FILE_MACHINE_SH5" + , COFF::IMAGE_FILE_MACHINE_SH5) + .Case( "IMAGE_FILE_MACHINE_THUMB" + , COFF::IMAGE_FILE_MACHINE_THUMB) + .Case( "IMAGE_FILE_MACHINE_WCEMIPSV2" + , COFF::IMAGE_FILE_MACHINE_WCEMIPSV2) + .Default(COFF::MT_Invalid); + if (Machine == COFF::MT_Invalid) { + YS.printError(Value, "Invalid value for Machine"); + return false; + } + } + Header.Machine = Machine; + } else if (KeyValue == "NumberOfSections") { + if (!getAs(Value, Header.NumberOfSections)) { + YS.printError(Value, "Invalid value for NumberOfSections"); + return false; + } + } else if (KeyValue == "TimeDateStamp") { + if (!getAs(Value, Header.TimeDateStamp)) { + YS.printError(Value, "Invalid value for TimeDateStamp"); + return false; + } + } else if (KeyValue == "PointerToSymbolTable") { + if (!getAs(Value, Header.PointerToSymbolTable)) { + YS.printError(Value, "Invalid value for PointerToSymbolTable"); + return false; + } + } else if (KeyValue == "NumberOfSymbols") { + if (!getAs(Value, Header.NumberOfSymbols)) { + YS.printError(Value, "Invalid value for NumberOfSymbols"); + return false; + } + } else if (KeyValue == "SizeOfOptionalHeader") { + if (!getAs(Value, Header.SizeOfOptionalHeader)) { + YS.printError(Value, "Invalid value for SizeOfOptionalHeader"); + return false; + } + } else { + YS.printError(Key, "Unrecognized key in header"); + return false; + } + } + } + return true; + } + + bool parseHeaderCharacteristics(yaml::Node *Characteristics) { + yaml::ScalarNode *Value = dyn_cast<yaml::ScalarNode>(Characteristics); + yaml::SequenceNode *SeqValue + = dyn_cast<yaml::SequenceNode>(Characteristics); + if (!Value && !SeqValue) { + YS.printError(Characteristics, + "Characteristics must either be a number or sequence"); + return false; + } + if (Value) { + if (!getAs(Value, Header.Characteristics)) { + YS.printError(Value, "Invalid value for Characteristics"); + return false; + } + } else { + for (yaml::SequenceNode::iterator ci = SeqValue->begin(), + ce = SeqValue->end(); + ci != ce; ++ci) { + yaml::ScalarNode *CharValue = dyn_cast<yaml::ScalarNode>(&*ci); + if (!CharValue) { + YS.printError(CharValue, + "Characteristics must be scalar values"); + return false; + } + SmallString<32> Storage; + StringRef Char = CharValue->getValue(Storage); + uint16_t Characteristic = StringSwitch<COFF::Characteristics>(Char) + .Case( "IMAGE_FILE_RELOCS_STRIPPED" + , COFF::IMAGE_FILE_RELOCS_STRIPPED) + .Case( "IMAGE_FILE_EXECUTABLE_IMAGE" + , COFF::IMAGE_FILE_EXECUTABLE_IMAGE) + .Case( "IMAGE_FILE_LINE_NUMS_STRIPPED" + , COFF::IMAGE_FILE_LINE_NUMS_STRIPPED) + .Case( "IMAGE_FILE_LOCAL_SYMS_STRIPPED" + , COFF::IMAGE_FILE_LOCAL_SYMS_STRIPPED) + .Case( "IMAGE_FILE_AGGRESSIVE_WS_TRIM" + , COFF::IMAGE_FILE_AGGRESSIVE_WS_TRIM) + .Case( "IMAGE_FILE_LARGE_ADDRESS_AWARE" + , COFF::IMAGE_FILE_LARGE_ADDRESS_AWARE) + .Case( "IMAGE_FILE_BYTES_REVERSED_LO |