From: "Guo, Gua" <gua.guo@intel.com>
To: "Wang, BruceX" <brucex.wang@intel.com>,
"devel@edk2.groups.io" <devel@edk2.groups.io>
Cc: "Dong, Guo" <guo.dong@intel.com>, "Ni, Ray" <ray.ni@intel.com>,
"Rhodes, Sean" <sean@starlabs.systems>,
"Lu, James" <james.lu@intel.com>
Subject: Re: [PATCH] * UefiPayloadPkg: Clang dependency removal
Date: Wed, 26 Apr 2023 07:17:29 +0000 [thread overview]
Message-ID: <BL1PR11MB54783B6AA1D4F254AEC3C4E1EF659@BL1PR11MB5478.namprd11.prod.outlook.com> (raw)
In-Reply-To: <0b1c996e78cddfa3a272187bd783f0c8b8e33492.1682492500.git.brucex.wang@intel.com>
@Wang, BruceX
Bugzilla please fill this link https://bugzilla.tianocore.org/show_bug.cgi?id=4426
Reviewed-by: Gua Guo <gua.guo@intel.com>
-----Original Message-----
From: Wang, BruceX <brucex.wang@intel.com>
Sent: Wednesday, April 26, 2023 3:12 PM
To: devel@edk2.groups.io
Cc: Wang, BruceX <brucex.wang@intel.com>; Dong, Guo <guo.dong@intel.com>; Ni, Ray <ray.ni@intel.com>; Rhodes, Sean <sean@starlabs.systems>; Lu, James <james.lu@intel.com>; Guo, Gua <gua.guo@intel.com>
Subject: [PATCH] * UefiPayloadPkg: Clang dependency removal
From: BruceX Wang <brucex.wang@intel.com>
REF: TBD
Use Python to replace llvm-objcopy in UniversalPayloadBuild.py.
1. AddSection32() and AddSection64(): Make a section named
<section> with the contents of <file>.
2. RemoveSection32() and RemoveSection64(): Remove <section>.
3. ReplaceFv (): remove the section before add the section
if the file exists.
Cc: Guo Dong <guo.dong@intel.com>
Cc: Ray Ni <ray.ni@intel.com>
Cc: Sean Rhodes <sean@starlabs.systems>
Cc: James Lu <james.lu@intel.com>
Cc: Gua Guo <gua.guo@intel.com>
Signed-off-by: BruceX Wang <brucex.wang@intel.com>
---
UefiPayloadPkg/Tools/ElfFv.py | 809 ++++++++++++++++++++++++
UefiPayloadPkg/Tools/__init__.py | 19 +
UefiPayloadPkg/UniversalPayloadBuild.py | 232 ++++---
3 files changed, 935 insertions(+), 125 deletions(-)
create mode 100644 UefiPayloadPkg/Tools/ElfFv.py
create mode 100644 UefiPayloadPkg/Tools/__init__.py
diff --git a/UefiPayloadPkg/Tools/ElfFv.py b/UefiPayloadPkg/Tools/ElfFv.py
new file mode 100644
index 0000000000..a3d5bf08b9
--- /dev/null
+++ b/UefiPayloadPkg/Tools/ElfFv.py
@@ -0,0 +1,809 @@
+## @file
+# OBJCOPY parser, it's used to replace FV
+#
+# Copyright (c) 2023, Intel Corporation. All rights reserved.<BR>
+# SPDX-License-Identifier: BSD-2-Clause-Patent
+##
+
+import argparse
+from ctypes import *
+import struct
+
+class ElfSectionHeader64:
+ def __init__(self, sh_name, sh_type, sh_flags, sh_addr, sh_offset, sh_size, sh_link, sh_info, sh_addralign, sh_entsize):
+ self.sh_name = sh_name
+ self.sh_type = sh_type
+ self.sh_flags = sh_flags
+ self.sh_addr = sh_addr
+ self.sh_offset = sh_offset
+ self.sh_size = sh_size
+ self.sh_link = sh_link
+ self.sh_info = sh_info
+ self.sh_addralign = sh_addralign
+ self.sh_entsize = sh_entsize
+
+ def pack(self):
+ return struct.pack('<IIQQQQIIQQ', self.sh_name, self.sh_type, self.sh_flags, self.sh_addr, self.sh_offset, self.sh_size, self.sh_link, self.sh_info, self.sh_addralign, self.sh_entsize)
+
+ @classmethod
+ def unpack(cls, data):
+ unpacked_data = struct.unpack('<IIQQQQIIQQ', data)
+ return cls(*unpacked_data)
+
+class ElfHeader64:
+ def __init__(self, data):
+ # Parse the ELF identification bytes
+ self.e_ident = struct.unpack('16s', data[:16])[0]
+ self.e_type = struct.unpack('H', data[16:18])[0]
+ self.e_machine = struct.unpack('H', data[18:20])[0]
+ self.e_version = struct.unpack('I', data[20:24])[0]
+ self.e_entry = struct.unpack('Q', data[24:32])[0]
+ self.e_phoff = struct.unpack('Q', data[32:40])[0]
+ self.e_shoff = struct.unpack('Q', data[40:48])[0]
+ self.e_flags = struct.unpack('I', data[48:52])[0]
+ self.e_ehsize = struct.unpack('H', data[52:54])[0]
+ self.e_phentsize = struct.unpack('H', data[54:56])[0]
+ self.e_phnum = struct.unpack('H', data[56:58])[0]
+ self.e_shentsize = struct.unpack('H', data[58:60])[0]
+ self.e_shnum = struct.unpack('H', data[60:62])[0]
+ self.e_shstrndx = struct.unpack('H', data[62:64])[0]
+
+ def pack(self):
+ # Pack the ELF header data into a binary string
+ data = b''
+ data += struct.pack('16s', self.e_ident)
+ data += struct.pack('H', self.e_type)
+ data += struct.pack('H', self.e_machine)
+ data += struct.pack('I', self.e_version)
+ data += struct.pack('Q', self.e_entry)
+ data += struct.pack('Q', self.e_phoff)
+ data += struct.pack('Q', self.e_shoff)
+ data += struct.pack('I', self.e_flags)
+ data += struct.pack('H', self.e_ehsize)
+ data += struct.pack('H', self.e_phentsize)
+ data += struct.pack('H', self.e_phnum)
+ data += struct.pack('H', self.e_shentsize)
+ data += struct.pack('H', self.e_shnum)
+ data += struct.pack('H', self.e_shstrndx)
+ return data
+
+class Elf64_Phdr:
+ def __init__(self, data):
+ self.p_type = struct.unpack("<L", data[0:4])[0]
+ self.p_flags = struct.unpack("<L", data[4:8])[0]
+ self.p_offset = struct.unpack("<Q", data[8:16])[0]
+ self.p_vaddr = struct.unpack("<Q", data[16:24])[0]
+ self.p_paddr = struct.unpack("<Q", data[24:32])[0]
+ self.p_filesz = struct.unpack("<Q", data[32:40])[0]
+ self.p_memsz = struct.unpack("<Q", data[40:48])[0]
+ self.p_align = struct.unpack("<Q", data[48:56])[0]
+
+ def pack(self):
+ # Pack the Program header table into a binary string
+ data = b''
+ data += struct.pack('<L', self.p_type)
+ data += struct.pack('<L', self.p_flags)
+ data += struct.pack('<Q', self.p_offset)
+ data += struct.pack('<Q', self.p_vaddr)
+ data += struct.pack('<Q', self.p_paddr)
+ data += struct.pack('<Q', self.p_filesz)
+ data += struct.pack('<Q', self.p_memsz)
+ data += struct.pack('<Q', self.p_align)
+ return data
+
+class ElfSectionHeader32:
+ def __init__(self, sh_name, sh_type, sh_flags, sh_addr, sh_offset, sh_size, sh_link, sh_info, sh_addralign, sh_entsize):
+ self.sh_name = sh_name
+ self.sh_type = sh_type
+ self.sh_flags = sh_flags
+ self.sh_addr = sh_addr
+ self.sh_offset = sh_offset
+ self.sh_size = sh_size
+ self.sh_link = sh_link
+ self.sh_info = sh_info
+ self.sh_addralign = sh_addralign
+ self.sh_entsize = sh_entsize
+
+ def pack(self):
+ return struct.pack('<IIIIIIIIII', self.sh_name, self.sh_type, self.sh_flags, self.sh_addr, self.sh_offset, self.sh_size, self.sh_link, self.sh_info, self.sh_addralign, self.sh_entsize)
+
+ @classmethod
+ def unpack(cls, data):
+ unpacked_data = struct.unpack('<IIIIIIIIII', data)
+ return cls(*unpacked_data)
+
+class ElfHeader32:
+ def __init__(self, data):
+ # Parse the ELF identification bytes
+ self.e_ident = struct.unpack('16s', data[:16])[0]
+ self.e_type = struct.unpack('H', data[16:18])[0]
+ self.e_machine = struct.unpack('H', data[18:20])[0]
+ self.e_version = struct.unpack('I', data[20:24])[0]
+ self.e_entry = struct.unpack('I', data[24:28])[0]
+ self.e_phoff = struct.unpack('I', data[28:32])[0]
+ self.e_shoff = struct.unpack('I', data[32:36])[0]
+ self.e_flags = struct.unpack('I', data[36:40])[0]
+ self.e_ehsize = struct.unpack('H', data[40:42])[0]
+ self.e_phentsize = struct.unpack('H', data[42:44])[0]
+ self.e_phnum = struct.unpack('H', data[44:46])[0]
+ self.e_shentsize = struct.unpack('H', data[46:48])[0]
+ self.e_shnum = struct.unpack('H', data[48:50])[0]
+ self.e_shstrndx = struct.unpack('H', data[50:52])[0]
+
+ def pack(self):
+ # Pack the ELF header data into a binary string
+ data = b''
+ data += struct.pack('16s', self.e_ident)
+ data += struct.pack('H', self.e_type)
+ data += struct.pack('H', self.e_machine)
+ data += struct.pack('I', self.e_version)
+ data += struct.pack('I', self.e_entry)
+ data += struct.pack('I', self.e_phoff)
+ data += struct.pack('I', self.e_shoff)
+ data += struct.pack('I', self.e_flags)
+ data += struct.pack('H', self.e_ehsize)
+ data += struct.pack('H', self.e_phentsize)
+ data += struct.pack('H', self.e_phnum)
+ data += struct.pack('H', self.e_shentsize)
+ data += struct.pack('H', self.e_shnum)
+ data += struct.pack('H', self.e_shstrndx)
+ return data
+
+class Elf32_Phdr:
+ def __init__(self, data):
+ self.p_type = struct.unpack("<L", data[0:4])[0]
+ self.p_offset = struct.unpack("<L", data[4:8])[0]
+ self.p_vaddr = struct.unpack("<L", data[8:12])[0]
+ self.p_paddr = struct.unpack("<L", data[12:16])[0]
+ self.p_filesz = struct.unpack("<L", data[16:20])[0]
+ self.p_memsz = struct.unpack("<L", data[20:24])[0]
+ self.p_flags = struct.unpack("<L", data[24:28])[0]
+ self.p_align = struct.unpack("<L", data[28:32])[0]
+
+ def pack(self):
+ # Pack the Program header table into a binary string
+ data = b''
+ data += struct.pack('<L', self.p_type)
+ data += struct.pack('<L', self.p_offset)
+ data += struct.pack('<L', self.p_vaddr)
+ data += struct.pack('<L', self.p_paddr)
+ data += struct.pack('<L', self.p_filesz)
+ data += struct.pack('<L', self.p_memsz)
+ data += struct.pack('<L', self.p_flags)
+ data += struct.pack('<L', self.p_align)
+ return data
+
+def SectionAlignment(NewUPLEntry, AlignmentIndex):
+ # Section entry Alignment
+ # Alignment is transfer to integer if AlignmentIndex is string.
+ if isinstance(AlignmentIndex, str):
+ int_num = int(AlignmentIndex, 16)
+ int_num = 10 * (int_num//16) + int_num % 16
+ else:
+ int_num = AlignmentIndex
+ if (int_num != 0 or int_num != 1):
+ if ((len(NewUPLEntry) % int_num) != 0):
+ AlignNumber = int_num - (len(NewUPLEntry) % int_num)
+ if (AlignNumber != 0):
+ for x in range(AlignNumber):
+ NewUPLEntry = NewUPLEntry + bytearray(b'\0')
+ return NewUPLEntry
+
+def SectionEntryFill(SectionEntry, Alignment, Value, Offset):
+ # Alignment
+ n = 0
+ if (len (Value) < Alignment):
+ Value = Value.zfill(Alignment)
+ for x in range(0, (Alignment//2)):
+ Index = '0x' + Value[n] + Value[n + 1]
+ SectionEntry[Offset - x] = int(Index,16)
+ n += 2
+ return SectionEntry
+
+def ElfHeaderParser(UPLEntry):
+ # Read EI_CLASS, it stores information that elf with 32-bit or 64-bit architectures.
+ EI_CLASS = UPLEntry[4]
+ # If Elf is 64-bit objects.
+ if (EI_CLASS == 2):
+ # Elf header is stored at 0x0-0x40 in 64-bits objects
+ ElfHeaderData = UPLEntry[:64]
+ # If Elf is 32-bit objects.
+ else:
+ # Elf header is stored at 0x0-0x34 in 32-bits objects
+ ElfHeaderData = UPLEntry[:53]
+ # If Elf is 64-bit objects.
+ if (EI_CLASS == 2):
+ elf_header = ElfHeader64(ElfHeaderData)
+ ElfHeaderOffset = elf_header.e_shoff
+ SectionHeaderEntryNumber = elf_header.e_shnum
+ StringIndexNumber = elf_header.e_shstrndx
+ SectionHeaderEntrySize = elf_header.e_shentsize
+ StringIndexEntryOffset = ElfHeaderOffset + (StringIndexNumber * SectionHeaderEntrySize)
+ unpacked_header = ElfSectionHeader64.unpack(UPLEntry[StringIndexEntryOffset: (StringIndexEntryOffset + SectionHeaderEntrySize)])
+ StringIndexSize = unpacked_header.sh_size
+ StringIndexOffset = unpacked_header.sh_offset
+ # If elf is 32-bit objects.
+ else:
+ elf_header = ElfHeader32(ElfHeaderData)
+ ElfHeaderOffset = elf_header.e_shoff
+ SectionHeaderEntryNumber = elf_header.e_shnum
+ StringIndexNumber = elf_header.e_shstrndx
+ SectionHeaderEntrySize = elf_header.e_shentsize
+ StringIndexEntryOffset = ElfHeaderOffset + (StringIndexNumber * SectionHeaderEntrySize)
+ unpacked_header = ElfSectionHeader32.unpack(UPLEntry[StringIndexEntryOffset: (StringIndexEntryOffset + SectionHeaderEntrySize)])
+ StringIndexSize = unpacked_header.sh_size
+ StringIndexOffset = unpacked_header.sh_offset
+ return ElfHeaderOffset, SectionHeaderEntryNumber, StringIndexNumber, StringIndexEntryOffset, StringIndexSize, SectionHeaderEntrySize, StringIndexOffset, EI_CLASS
+
+def FindSection(UPLEntry, SectionName):
+ ElfHeaderOffset, SectionHeaderEntryNumber, StringIndexNumber, _, StringIndexSize, SectionHeaderEntrySize, StringIndexOffset, EI_CLASS = ElfHeaderParser(UPLEntry)
+ # StringIndex is String Index section
+ StringIndex = UPLEntry[StringIndexOffset:StringIndexOffset+StringIndexSize]
+ # Section header isn't exist if SectionNameOffset = -1.
+ StringIndex = StringIndex.decode('utf-8', errors='ignore')
+ SectionNameOffset = StringIndex.find(SectionName)
+ return SectionNameOffset, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, StringIndexOffset, StringIndexNumber, EI_CLASS
+
+def AddNewSectionEntry64(LastUPLEntrylen, StringIndexValue, SectionSize, Alignment):
+ # If elf is 64-bit objects.
+ NewSectionEntry = ElfSectionHeader64 (StringIndexValue, 1, 0, 0, LastUPLEntrylen, SectionSize, 0, 0, Alignment, 0)
+ sh_bytes = NewSectionEntry.pack()
+ return sh_bytes
+
+def AddNewSectionEntry32(LastUPLEntrylen, StringIndexValue, SectionSize, Alignment):
+ # If elf is 32-bit objects.
+ NewSectionEntry = ElfSectionHeader32 (StringIndexValue, 1, 0, 0, LastUPLEntrylen, SectionSize, 0, 0, Alignment, 0)
+ sh_bytes = NewSectionEntry.pack()
+ return sh_bytes
+
+def AddSectionHeader64(SHentry, NewUPLEntrylen, SectionHeaderEntrySize, Index, RemoveNameOffset, SectionName, StringIndexNumber):
+ SHentry = bytearray(SHentry)
+ unpacked_header = ElfSectionHeader64.unpack(SHentry[(Index * SectionHeaderEntrySize):((Index * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ # Section header of section 0 shows 0. It don't modify any offset.
+ if (Index != 0):
+ # read section offset.
+ unpacked_header.sh_offset = NewUPLEntrylen
+ # Modify offset of name in section entry
+ # if RemoveNameOffset != 0 that is remove function.
+ if (RemoveNameOffset != 0):
+ if (unpacked_header.sh_name > RemoveNameOffset):
+ unpacked_header.sh_name -= RemoveNameOffset
+ # Modify size of name string section entry in section entry.
+ if (Index == StringIndexNumber):
+ unpacked_header.sh_size -= len (SectionName)
+ # added section
+ else :
+ if (Index == StringIndexNumber):
+ unpacked_header.sh_size += len (SectionName)
+ NewSHentry = ElfSectionHeader64 (
+ unpacked_header.sh_name,
+ unpacked_header.sh_type,
+ unpacked_header.sh_flags,
+ unpacked_header.sh_addr,
+ unpacked_header.sh_offset,
+ unpacked_header.sh_size,
+ unpacked_header.sh_link,
+ unpacked_header.sh_info,
+ unpacked_header.sh_addralign,
+ unpacked_header.sh_entsize).pack()
+ return NewSHentry
+
+def AddSectionHeader32(SHentry, NewUPLEntrylen, SectionHeaderEntrySize, Index, RemoveNameOffset, SectionName, StringIndexNumber):
+ SHentry = bytearray(SHentry)
+ unpacked_header = ElfSectionHeader32.unpack(SHentry[(Index * SectionHeaderEntrySize):((Index * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ if (Index != 0):
+ NewSHentry = SHentry[(Index * SectionHeaderEntrySize):((Index * SectionHeaderEntrySize) + SectionHeaderEntrySize)]
+ unpacked_header.sh_offset = NewUPLEntrylen
+ # Modify offset of name in section entry
+ # if RemoveNameOffset != 0 that is remove function.
+ if (RemoveNameOffset != 0):
+ if (unpacked_header.sh_name > RemoveNameOffset):
+ unpacked_header.sh_name -= RemoveNameOffset
+ # Modify size of name string section entry in section entry.
+ if (Index == StringIndexNumber):
+ unpacked_header.sh_size -= len (SectionName)
+ # added section
+ else :
+ if (Index == StringIndexNumber):
+ unpacked_header.sh_size += len (SectionName)
+ NewSHentry = ElfSectionHeader32 (
+ unpacked_header.sh_name,
+ unpacked_header.sh_type,
+ unpacked_header.sh_flags,
+ unpacked_header.sh_addr,
+ unpacked_header.sh_offset,
+ unpacked_header.sh_size,
+ unpacked_header.sh_link,
+ unpacked_header.sh_info,
+ unpacked_header.sh_addralign,
+ unpacked_header.sh_entsize).pack()
+ return NewSHentry
+
+def ModifyPHSegmentOffset64(NewUPLEntry, ElfHeaderOffset, PHSegmentName):
+ # Modify offset and address of program header tables.
+ elf_header = ElfHeader64(NewUPLEntry[:64])
+ SHentry = NewUPLEntry[ElfHeaderOffset:]
+ # Elf program header tables start from 0x40 in 64-bits objects
+ PHentry = NewUPLEntry[64: 64 + (elf_header.e_phnum * elf_header.e_phentsize)]
+ PHdrs = []
+ SHdrs = []
+ for i in range(elf_header.e_shnum):
+ SHData = SHentry[(i * elf_header.e_shentsize): (i * elf_header.e_shentsize) + elf_header.e_shentsize]
+ unpacked_SectionHeader = ElfSectionHeader64.unpack(SHData)
+ SHdrs.append(unpacked_SectionHeader)
+ for i in range(elf_header.e_phnum):
+ PHData = PHentry[(i * elf_header.e_phentsize): (i * elf_header.e_phentsize) + elf_header.e_phentsize]
+ unpacked_ProgramHeader = Elf64_Phdr(PHData)
+ PHdrs.append(unpacked_ProgramHeader)
+ if (PHSegmentName == '.text'):
+ PHdrs[0].p_offset = SHdrs[1].sh_offset
+ PHdrs[0].p_paddr = SHdrs[1].sh_addr
+ PHdrs[4].p_offset = SHdrs[1].sh_offset
+ PHdrs[4].p_paddr = SHdrs[1].sh_addr
+ elif (PHSegmentName == '.dynamic'):
+ PHdrs[1].p_offset = SHdrs[2].sh_offset
+ PHdrs[1].p_paddr = SHdrs[2].sh_addr
+ PHdrs[3].p_offset = SHdrs[2].sh_offset
+ PHdrs[3].p_paddr = SHdrs[2].sh_addr
+ elif (PHSegmentName == '.data'):
+ PHdrs[2].p_offset = SHdrs[3].sh_offset
+ PHdrs[2].p_paddr = SHdrs[3].sh_addr
+ packed_PHData = b''
+ for phdr in PHdrs:
+ packed_PHData += phdr.pack()
+ NewUPLEntry = bytearray(NewUPLEntry)
+ NewUPLEntry[64: 64 + (elf_header.e_phnum * elf_header.e_phentsize)] = packed_PHData
+ return NewUPLEntry
+
+def ModifyPHSegmentOffset32(NewUPLEntry, ElfHeaderOffset, PHSegmentName):
+ # Modify offset and address of program header tables.
+ # Elf header is stored at 0x0-0x34 in 32-bits objects
+ elf_header = ElfHeader32(NewUPLEntry[:52])
+ SHentry = NewUPLEntry[ElfHeaderOffset:]
+ # Elf program header tables start from 0x34 in 32-bits objects
+ PHentry = NewUPLEntry[52: 52 + (elf_header.e_phnum * elf_header.e_phentsize)]
+ PHdrs = []
+ SHdrs = []
+ for i in range(elf_header.e_shnum):
+ SHData = SHentry[(i * elf_header.e_shentsize): (i * elf_header.e_shentsize) + elf_header.e_shentsize]
+ unpacked_SectionHeader = ElfSectionHeader32.unpack(SHData)
+ SHdrs.append(unpacked_SectionHeader)
+ for i in range(elf_header.e_phnum):
+ PHData = PHentry[(i * elf_header.e_phentsize): (i * elf_header.e_phentsize) + elf_header.e_phentsize]
+ unpacked_ProgramHeader = Elf32_Phdr(PHData)
+ PHdrs.append(unpacked_ProgramHeader)
+ if (PHSegmentName == '.text'):
+ PHdrs[0].p_offset = SHdrs[1].sh_offset
+ PHdrs[0].p_paddr = SHdrs[1].sh_addr
+ PHdrs[0].p_vaddr = SHdrs[1].sh_addr
+ PHdrs[2].p_offset = SHdrs[1].sh_offset
+ PHdrs[2].p_paddr = SHdrs[1].sh_addr
+ PHdrs[0].p_vaddr = SHdrs[1].sh_addr
+ elif (PHSegmentName == '.data'):
+ PHdrs[1].p_offset = SHdrs[2].sh_offset
+ PHdrs[1].p_paddr = SHdrs[2].sh_addr
+ PHdrs[1].p_vaddr = SHdrs[2].sh_addr
+ packed_PHData = b''
+ for phdr in PHdrs:
+ packed_PHData += phdr.pack()
+ NewUPLEntry = bytearray(NewUPLEntry)
+ NewUPLEntry[52: 52 + (elf_header.e_phnum * elf_header.e_phentsize)] = packed_PHData
+ return NewUPLEntry
+
+def RemoveSection64(UniversalPayloadEntry, RemoveSectionName):
+ # If elf is 64-bit objects.
+ # Get offsets as follows:
+ # 1. Section name which will remove in section name string.
+ # 2. Section which will remove.
+ # 3. Section header which will remove.
+ with open(UniversalPayloadEntry,'rb') as f:
+ UPLEntry = f.read()
+ RemoveSectionNameOffset, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, _, StringIndexNumber, _ = FindSection(UPLEntry, RemoveSectionName)
+ if (RemoveSectionNameOffset == -1):
+ raise argparse.ArgumentTypeError ('Section: {} not found.'.format (RemoveSectionNameOffset))
+ # Read section header entry
+ SHentry = UPLEntry[ElfHeaderOffset:]
+ # find deleted fv section offset.
+ # Elf header is stored at 0x0-0x40 in 64-bits objects
+ elf_header = ElfHeader64(UPLEntry[:64])
+ Counter = 0
+ RemoveIndex = 0
+ RemoveNameOffset = 0
+ for Index in range(0, elf_header.e_shnum):
+ # Read Index of section header.
+ unpacked_SectionHeader = ElfSectionHeader64.unpack(SHentry[(Index * elf_header.e_shentsize):((Index * elf_header.e_shentsize) + elf_header.e_shentsize)])
+ # Find offset of section name which is removed.
+ if (unpacked_SectionHeader.sh_name == RemoveSectionNameOffset):
+ RemoveIndex = Counter
+ Counter += 1
+ else:
+ Counter += 1
+ # Elf header is recombined.
+ # Elf header and program header table in front of first section are reserved.
+ # Elf header size is 0x40 with 64-bit object.
+ ElfHeaderSize = 64
+ ElfHandPH = ElfHeaderSize + (elf_header.e_phnum * elf_header.e_phentsize)
+ NewUPLEntry = UPLEntry[:ElfHandPH]
+ # Keep Section header and program header table, RemoveSection64() only recombined section and section header.
+ NewUPLEntry = bytearray(NewUPLEntry)
+ # Sections is recombined.
+ # 1. name of deleted section is removed in name string section.
+ # 2. deleted section is removed in dll file.
+ # 3. re-align sections before and after deleted section.
+ NewUPLEntrylen = []
+ for Index in range(0, (SectionHeaderEntryNumber)):
+ unpacked_SectionHeader = ElfSectionHeader64.unpack(SHentry[(Index * SectionHeaderEntrySize):((Index * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntrylen.append(len(NewUPLEntry))
+ if (Index == 0):
+ # Address alignment, section will align with alignment of next section.
+ AlignmentIndex = 8
+ if (SectionHeaderEntryNumber > 2):
+ unpacked_NextSectionHeader = ElfSectionHeader64.unpack(SHentry[((Index + 1) * SectionHeaderEntrySize):(((Index + 1) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ # Section in front of removed section
+ elif (Index + 1 == RemoveIndex):
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Read section address alignment
+ # If section that will be removed in .dll is not first and last one .
+ # Address alignment, section will align with alignment of section after deleted section.
+ # Check next and the section after next are not end of section.
+ if ((Index + 2) < (SectionHeaderEntryNumber - 1)):
+ unpacked_Next2SectionHeader = ElfSectionHeader64.unpack(SHentry[((Index + 2) * SectionHeaderEntrySize):(((Index + 2) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_Next2SectionHeader.sh_addralign)
+ else:
+ # It is align 8 bytes if next section or the section after next is last one.
+ AlignmentIndex = 8
+ NewUPLEntry = SectionAlignment(NewUPLEntry, AlignmentIndex)
+ # section is Deleted section
+ elif (Index == RemoveIndex):
+ # Don't add removed section to elf.
+ # Find offset of section name.
+ RemoveNameOffset = unpacked_SectionHeader.sh_name
+ # section is name string section.
+ elif (Index == StringIndexNumber):
+ # StringIndex is String Index section
+ StringIndex = UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Remove name of removed section in name string section.
+ # Section header isn't exist if RemoveSectionNameOffset equal to -1.
+ StringIndex = bytearray(StringIndex)
+ RemoveSectionName = bytearray(RemoveSectionName, encoding='utf-8')
+ RemoveSectionName = RemoveSectionName + bytes('\0', encoding='utf-8')
+ StringIndex = StringIndex.replace(RemoveSectionName,b'')
+ NewUPLEntry += StringIndex
+ # other sections.
+ else:
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Address alignment, section will align with alignment of next section.
+ if (Index < (SectionHeaderEntryNumber - 1)):
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ else:
+ # If section is last one.
+ AlignmentIndex = 8
+ NewUPLEntry = SectionAlignment(NewUPLEntry, AlignmentIndex)
+ SectionHeaderOffset = len(NewUPLEntry)
+ # Add section header
+ for Number in range(0, (SectionHeaderEntryNumber)):
+ if (Number != RemoveIndex):
+ NewSHentry = AddSectionHeader64(SHentry, NewUPLEntrylen[Number], SectionHeaderEntrySize, Number, RemoveNameOffset, RemoveSectionName, StringIndexNumber)
+ NewUPLEntry += NewSHentry
+ # Modify number of sections and offset of section header in Elf header.
+ elf_header.e_shoff = SectionHeaderOffset
+ elf_header.e_shnum -= 1
+ NewUPLEntry = elf_header.pack() + NewUPLEntry[64:]
+ # write to Elf.
+ with open(UniversalPayloadEntry,'wb') as f:
+ f.write(NewUPLEntry)
+
+def RemoveSection32(UniversalPayloadEntry, RemoveSectionName):
+ # If elf is 32-bit objects.
+ # Get offsets as follows:
+ # 1. Section name which will remove in section name string.
+ # 2. Section which will remove.
+ # 3. Section header which will remove.
+ with open(UniversalPayloadEntry,'rb') as f:
+ UPLEntry = f.read()
+ RemoveSectionNameOffset, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, _, StringIndexNumber, EI_CLASS = FindSection(UPLEntry, RemoveSectionName)
+ if (RemoveSectionNameOffset == -1):
+ raise argparse.ArgumentTypeError ('Section: {} not found.'.format (RemoveSectionNameOffset))
+ # Read section header entry
+ SHentry = UPLEntry[ElfHeaderOffset:]
+ # find deleted fv section offset.
+ # Elf header is stored at 0x0-0x34 in 32-bits objects
+ elf_header = ElfHeader32(UPLEntry[:52])
+ Counter = 0
+ RemoveIndex = 0
+ RemoveNameOffset = 0
+ for Index in range(0, elf_header.e_shnum):
+ # Read Index of section header.
+ unpacked_SectionHeader = ElfSectionHeader32.unpack(SHentry[(Index * elf_header.e_shentsize):((Index * elf_header.e_shentsize) + elf_header.e_shentsize)])
+ # Find offset of section name which is removed.
+ if (unpacked_SectionHeader.sh_name == RemoveSectionNameOffset):
+ RemoveIndex = Counter
+ Counter += 1
+ else:
+ Counter += 1
+ # Elf header is recombined.
+ # Elf header and program header table in front of first section are reserved.
+ # Elf header size is 0x34 with 32-bit object.
+ ElfHeaderSize = 52
+ ElfHandPH = ElfHeaderSize + (elf_header.e_phnum * elf_header.e_phentsize)
+ NewUPLEntry = UPLEntry[:ElfHandPH]
+ # Keep Section header and program header table, RemoveSection32() only recombined section and section header.
+ NewUPLEntry = bytearray(NewUPLEntry)
+ # Sections is recombined.
+ # 1. name of deleted section is removed in name string section.
+ # 2. deleted section is removed in dll file.
+ # 3. re-align sections before and after deleted section.
+ NewUPLEntrylen = []
+ for Index in range(0, (SectionHeaderEntryNumber)):
+ unpacked_SectionHeader = ElfSectionHeader32.unpack(SHentry[(Index * SectionHeaderEntrySize):((Index * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntrylen.append(len(NewUPLEntry))
+ if (Index == 0):
+ # Address alignment, section will align with alignment of next section.
+ AlignmentIndex = 8
+ if (SectionHeaderEntryNumber > 2):
+ unpacked_NextSectionHeader = ElfSectionHeader32.unpack(SHentry[((Index + 1) * SectionHeaderEntrySize):(((Index + 1) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ # Section in front of removed section
+ elif (Index + 1 == RemoveIndex):
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Read section address alignment
+ # If section that will be removed in .dll is not first and last one .
+ # Address alignment, section will align with alignment of section after deleted section.
+ # Check next and the section after next are not end of section.
+ if ((Index + 2) < (SectionHeaderEntryNumber - 1)):
+ unpacked_Next2SectionHeader = ElfSectionHeader32.unpack(SHentry[((Index + 2) * SectionHeaderEntrySize):(((Index + 2) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_Next2SectionHeader.sh_addralign)
+ else:
+ # It is align 8 bytes if next section or the section after next is last one.
+ AlignmentIndex = 8
+ NewUPLEntry = SectionAlignment(NewUPLEntry, AlignmentIndex)
+ # section is Deleted section
+ elif (Index == RemoveIndex):
+ # Don't add removed section to elf.
+ # Find offset of section name.
+ RemoveNameOffset = unpacked_SectionHeader.sh_name
+ # section is name string section.
+ elif (Index == StringIndexNumber):
+ # StringIndex is String Index section
+ StringIndex = UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Remove name of removed section in name string section.
+ # Section header isn't exist if RemoveSectionNameOffset equal to -1.
+ StringIndex = bytearray(StringIndex)
+ RemoveSectionName = bytearray(RemoveSectionName, encoding='utf-8')
+ RemoveSectionName = RemoveSectionName + bytes('\0', encoding='utf-8')
+ StringIndex = StringIndex.replace(RemoveSectionName,b'')
+ NewUPLEntry += StringIndex
+ # other sections.
+ else:
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Address alignment, section will align with alignment of next section.
+ if (Index < (SectionHeaderEntryNumber - 1)):
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ else:
+ # If section is last one.
+ AlignmentIndex = 8
+ NewUPLEntry = SectionAlignment(NewUPLEntry, AlignmentIndex)
+ SectionHeaderOffset = len(NewUPLEntry)
+ # Add section header
+ for Number in range(0, (SectionHeaderEntryNumber)):
+ if (Number != RemoveIndex):
+ NewSHentry = AddSectionHeader32(SHentry, NewUPLEntrylen[Number], SectionHeaderEntrySize, Number, RemoveNameOffset, RemoveSectionName, StringIndexNumber)
+ NewUPLEntry += NewSHentry
+ # Modify number of sections and offset of section header in Elf header.
+ elf_header.e_shoff = SectionHeaderOffset
+ elf_header.e_shnum -= 1
+ NewUPLEntry = elf_header.pack() + NewUPLEntry[52:]
+ # write to Elf.
+ with open(UniversalPayloadEntry,'wb') as f:
+ f.write(NewUPLEntry)
+
+def AddSection64(UniversalPayloadEntry, AddSectionName, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, StringIndexNumber, FileBinary, Alignment):
+ with open(UniversalPayloadEntry,'rb+') as f:
+ UPLEntry = f.read()
+ fFileBinary = open(FileBinary, 'rb')
+ Binary_File = fFileBinary.read()
+ ElfHeaderOffset, SectionHeaderEntryNumber, StringIndexNumber, _, _, SectionHeaderEntrySize, _, _ = ElfHeaderParser(UPLEntry)
+ # Read section header entry
+ SHentry = UPLEntry[ElfHeaderOffset:]
+ # Elf header is recombined.
+ # Elf header and program header table in front of first section are reserved.
+ # Elf header is stored at 0x0-0x40 in 64-bits objects
+ elf_header = ElfHeader64(UPLEntry[:64])
+ # Elf header size is 0x40 with 64-bit object.
+ ElfHeaderSize = 64
+ ElfHandPH = ElfHeaderSize + (elf_header.e_phnum * elf_header.e_phentsize)
+ NewUPLEntry = UPLEntry[:ElfHandPH]
+ # Keep Section header and program header table, AddSection64() only recombined section and section header.
+ NewUPLEntry = bytearray(NewUPLEntry)
+ # Sections is recombined.
+ # 1. name of added section is added in name string section.
+ # 2. added section is added in dll file.
+ # 3. re-align sections before and after added section.
+ NewUPLEntrylen = []
+ StringIndexValue = 0
+ for Index in range(0, SectionHeaderEntryNumber):
+ NewUPLEntrylen.append(len(NewUPLEntry))
+ unpacked_SectionHeader = ElfSectionHeader64.unpack(SHentry[(Index * SectionHeaderEntrySize):((Index * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ # Sections is recombined.
+ if (Index == 0):
+ # Address alignment, section will align with alignment of next section.
+ AlignmentIndex = 8
+ if (SectionHeaderEntryNumber > 2):
+ unpacked_NextSectionHeader = ElfSectionHeader64.unpack(SHentry[((Index + 1) * SectionHeaderEntrySize):(((Index + 1) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ # Section is last one.
+ elif (Index == (SectionHeaderEntryNumber - 1)):
+ # Add new section at the end.
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ NewUPLEntry = SectionAlignment(NewUPLEntry, Alignment)
+ LastUPLEntrylen = len(NewUPLEntry)
+ NewUPLEntry += Binary_File
+ # Address alignment, section will align with alignment of next section.
+ AlignmentIndex = 8
+ NewUPLEntry = SectionAlignment(NewUPLEntry, AlignmentIndex)
+ # section is name string section.
+ elif (Index == StringIndexNumber):
+ # StringIndex is String Index section
+ StringIndex = UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Read name of added Section after StringIndex is transform into string.
+ StringIndex = bytearray(StringIndex)
+ StringIndexValue = len(StringIndex)
+ AddSectionName = bytearray(AddSectionName, encoding='utf-8') + bytes('\0', encoding='utf-8')
+ StringIndex += AddSectionName
+ NewUPLEntry += StringIndex
+ # section after name string section but not last one.
+ elif ((Index > StringIndexNumber) and (Index < (SectionHeaderEntryNumber - 1))):
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Address alignment, section will align with alignment of next section.
+ unpacked_NextSectionHeader = ElfSectionHeader64.unpack(SHentry[((Index + 1) * SectionHeaderEntrySize):(((Index + 1) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ # Section before name string section.
+ else:
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Address alignment, section will align with alignment of next section.
+ if (Index < (SectionHeaderEntryNumber - 1)):
+ unpacked_NextSectionHeader = ElfSectionHeader64.unpack(SHentry[((Index + 1) * SectionHeaderEntrySize):(((Index + 1) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ SectionHeaderOffset = len(NewUPLEntry)
+ RemoveNameOffset = 0
+ # Add section header
+ for Number in range(0, (SectionHeaderEntryNumber)):
+ NewSHentry = AddSectionHeader64(SHentry, NewUPLEntrylen[Number], SectionHeaderEntrySize, Number, RemoveNameOffset, AddSectionName, StringIndexNumber)
+ NewUPLEntry += NewSHentry
+ NewUPLEntry += bytearray(AddNewSectionEntry64(LastUPLEntrylen, StringIndexValue, len(Binary_File), Alignment))
+ # Modify number of sections and offset of section header in Elf header.
+ # Modify offset in in Elf header.
+ elf_header.e_shoff = SectionHeaderOffset
+ elf_header.e_shnum += 1
+ elf_header = elf_header.pack()
+ UPLEntryBin = elf_header + NewUPLEntry[64:]
+ # Modify offsets and address of program header table in elf.
+ PHSegmentName = '.text'
+ _, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, _, StringIndexNumber, _ = FindSection(UPLEntryBin, PHSegmentName)
+ UPLEntryBin = ModifyPHSegmentOffset64(UPLEntryBin, ElfHeaderOffset, PHSegmentName)
+ # Modify offsets and address of program header table in elf.
+ PHSegmentName = '.dynamic'
+ _, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, _, StringIndexNumber, _ = FindSection(UPLEntryBin, PHSegmentName)
+ UPLEntryBin = ModifyPHSegmentOffset64(UPLEntryBin, ElfHeaderOffset, PHSegmentName)
+ # Modify offsets and address of program header table in elf.
+ PHSegmentName = '.data'
+ _, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, _, StringIndexNumber, _ = FindSection(UPLEntryBin, PHSegmentName)
+ UPLEntryBin = ModifyPHSegmentOffset64(UPLEntryBin, ElfHeaderOffset, PHSegmentName)
+ fFileBinary.close()
+ return UPLEntryBin
+
+def AddSection32(UniversalPayloadEntry, AddSectionName, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, StringIndexNumber, FileBinary, Alignment):
+ with open(UniversalPayloadEntry,'rb+') as f:
+ # Read Elf and binary which will be write to elf.
+ UPLEntry = f.read()
+ fFileBinary = open(FileBinary, 'rb')
+ Binary_File = fFileBinary.read()
+ ElfHeaderOffset, SectionHeaderEntryNumber, StringIndexNumber, _, _, SectionHeaderEntrySize, _, _ = ElfHeaderParser(UPLEntry)
+ # Read section header entry
+ SHentry = UPLEntry[ElfHeaderOffset:]
+ # Elf header is recombined.
+ # Elf header and program header table in front of first section are reserved.
+ # Elf header is stored at 0x0-0x34 in 32-bits objects
+ elf_header = ElfHeader32(UPLEntry[:52])
+ # Elf header size is 0x34 with 32-bit object.
+ ElfHeaderSize = 52
+ ElfHandPH = ElfHeaderSize + (elf_header.e_phnum * elf_header.e_phentsize)
+ NewUPLEntry = UPLEntry[:ElfHandPH]
+ # Keep Section header and program header table, AddSection64() only recombined section and section header.
+ NewUPLEntry = bytearray(NewUPLEntry)
+ # Sections is recombined.
+ # 1. name of added section is added in name string section.
+ # 2. added section is added in dll file.
+ # 3. re-align sections before and after added section.
+ NewUPLEntrylen = []
+ StringIndexValue = 0
+ for Index in range(0, SectionHeaderEntryNumber):
+ NewUPLEntrylen.append(len(NewUPLEntry))
+ unpacked_SectionHeader = ElfSectionHeader32.unpack(SHentry[(Index * SectionHeaderEntrySize):((Index * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ # Sections is recombined.
+ if (Index == 0):
+ # Address alignment, section will align with alignment of next section.
+ AlignmentIndex = 8
+ if (SectionHeaderEntryNumber > 2):
+ unpacked_NextSectionHeader = ElfSectionHeader32.unpack(SHentry[((Index + 1) * SectionHeaderEntrySize):(((Index + 1) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ # Section is last one.
+ elif (Index == (SectionHeaderEntryNumber - 1)):
+ # Add new section at the end.
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ NewUPLEntry = SectionAlignment(NewUPLEntry, Alignment)
+ LastUPLEntrylen = len(NewUPLEntry)
+ NewUPLEntry += Binary_File
+ # Address alignment, section will align with alignment of next section.
+ AlignmentIndex = 8
+ NewUPLEntry = SectionAlignment(NewUPLEntry, AlignmentIndex)
+ # section is name string section.
+ elif (Index == StringIndexNumber):
+ # StringIndex is String Index section
+ StringIndex = UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Read name of added Section after StringIndex is transform into string.
+ StringIndex = bytearray(StringIndex)
+ StringIndexValue = len(StringIndex)
+ AddSectionName = bytearray(AddSectionName, encoding='utf-8') + bytes('\0', encoding='utf-8')
+ StringIndex += AddSectionName
+ NewUPLEntry += StringIndex
+ # section after name string section but not last one.
+ elif ((Index > StringIndexNumber) and (Index < (SectionHeaderEntryNumber - 1))):
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Address alignment, section will align with alignment of next section.
+ unpacked_NextSectionHeader = ElfSectionHeader32.unpack(SHentry[((Index + 1) * SectionHeaderEntrySize):(((Index + 1) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ # Section before name string section.
+ else:
+ NewUPLEntry += UPLEntry[unpacked_SectionHeader.sh_offset:(unpacked_SectionHeader.sh_offset + unpacked_SectionHeader.sh_size)]
+ # Address alignment, section will align with alignment of next section.
+ if (Index < (SectionHeaderEntryNumber - 1)):
+ unpacked_NextSectionHeader = ElfSectionHeader32.unpack(SHentry[((Index + 1) * SectionHeaderEntrySize):(((Index + 1) * SectionHeaderEntrySize) + SectionHeaderEntrySize)])
+ NewUPLEntry = SectionAlignment(NewUPLEntry, unpacked_NextSectionHeader.sh_addralign)
+ SectionHeaderOffset = len(NewUPLEntry)
+ RemoveNameOffset = 0
+ # Add section header
+ for Number in range(0, (SectionHeaderEntryNumber)):
+ NewSHentry = AddSectionHeader32(SHentry, NewUPLEntrylen[Number], SectionHeaderEntrySize, Number, RemoveNameOffset, AddSectionName, StringIndexNumber)
+ NewUPLEntry += NewSHentry
+ NewUPLEntry += bytearray(AddNewSectionEntry32(LastUPLEntrylen, StringIndexValue, len(Binary_File), Alignment))
+ # Modify number of sections and offset of section header in Elf header.
+ # Modify offset in in Elf header.
+ elf_header.e_shoff = SectionHeaderOffset
+ elf_header.e_shnum += 1
+ PHTableSize = elf_header.e_phentsize
+ elf_header = elf_header.pack()
+ UPLEntryBin = elf_header + NewUPLEntry[52:]
+ # Modify offsets and address of program header table in elf.
+ PHSegmentName = '.text'
+ _, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, _, StringIndexNumber, _ = FindSection(UPLEntryBin, PHSegmentName)
+ UPLEntryBin = ModifyPHSegmentOffset32(UPLEntryBin, ElfHeaderOffset, PHSegmentName)
+ # Modify offsets and address of program header table in elf. Its are stored at 0x08-0x0F and 0x10-0x17
+ PHSegmentName = '.data'
+ _, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, _, StringIndexNumber, _ = FindSection(UPLEntryBin, PHSegmentName)
+ UPLEntryBin = ModifyPHSegmentOffset32(UPLEntryBin, ElfHeaderOffset, PHSegmentName)
+ fFileBinary.close()
+ return UPLEntryBin
+
+def ReplaceFv (UniversalPayloadEntry, FileBinary, AddSectionName, Alignment = 16):
+ with open(UniversalPayloadEntry,'rb+') as f:
+ UPLEntry = f.read()
+ SectionNameOffset, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, _, StringIndexNumber, EI_CLASS = FindSection(UPLEntry, AddSectionName)
+ # If elf is 64-bit objects.
+ if (EI_CLASS == 2):
+ # Remove section if it exists.
+ if (SectionNameOffset != -1):
+ RemoveSection64(UniversalPayloadEntry, AddSectionName)
+ # Add section.
+ NewUPLEntry = AddSection64(UniversalPayloadEntry, AddSectionName, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, StringIndexNumber, FileBinary, Alignment)
+ # If elf is 32-bit objects.
+ else:
+ # Remove section if it exists.
+ if (SectionNameOffset != -1):
+ RemoveSection32(UniversalPayloadEntry, AddSectionName)
+ # Add section.
+ NewUPLEntry = AddSection32(UniversalPayloadEntry, AddSectionName, ElfHeaderOffset, SectionHeaderEntrySize, SectionHeaderEntryNumber, StringIndexNumber, FileBinary, Alignment)
+ with open(UniversalPayloadEntry,'wb') as f:
+ f.write(NewUPLEntry)
+ return 0
diff --git a/UefiPayloadPkg/Tools/__init__.py b/UefiPayloadPkg/Tools/__init__.py
new file mode 100644
index 0000000000..de57d1b907
--- /dev/null
+++ b/UefiPayloadPkg/Tools/__init__.py
@@ -0,0 +1,19 @@
+## @file
+# Python 'Tools' package initialization file.
+#
+# @copyright
+# INTEL CONFIDENTIAL
+#
+# Copyright (C) 2023 Intel Corporation
+#
+# This software and the related documents are Intel copyrighted materials, and your
+# use of them is governed by the express license under which they were provided to
+# you ("License"). Unless the License provides otherwise, you may not use, modify,
+# copy, publish, distribute, disclose or transmit this software or the related
+# documents without Intel's prior written permission.
+#
+# This software and the related documents are provided as is, with no express or
+# implied warranties, other than those that are expressly stated in the License.
+#
+# @par Specification
+##
diff --git a/UefiPayloadPkg/UniversalPayloadBuild.py b/UefiPayloadPkg/UniversalPayloadBuild.py
index 83e0de95d8..416946a431 100644
--- a/UefiPayloadPkg/UniversalPayloadBuild.py
+++ b/UefiPayloadPkg/UniversalPayloadBuild.py
@@ -11,7 +11,7 @@ import os
import shutil
import sys
from ctypes import *
-
+from Tools.ElfFv import ReplaceFv
sys.dont_write_bytecode = True
class UPLD_INFO_HEADER(LittleEndianStructure):
@@ -36,106 +36,56 @@ class UPLD_INFO_HEADER(LittleEndianStructure):
self.ImageId = b'UEFI'
self.ProducerId = b'INTEL'
-def GenSpecRevision (Argument):
- try:
- (MajorStr, MinorStr) = Argument.split('.')
- except:
- raise argparse.ArgumentTypeError ('{} is not a valid SpecRevision format (Major[8-bits].Minor[8-bits]).'.format (Argument))
- #
- # Spec Revision Bits 15 : 8 - Major Version. Bits 7 : 0 - Minor Version.
- #
- if len(MinorStr) > 0 and len(MinorStr) < 3:
- try:
- Minor = int(MinorStr, 16) if len(MinorStr) == 2 else (int(MinorStr, 16) << 4)
- except:
- raise argparse.ArgumentTypeError ('{} Minor version of SpecRevision is not a valid integer value.'.format (Argument))
- else:
- raise argparse.ArgumentTypeError ('{} is not a valid SpecRevision format (Major[8-bits].Minor[8-bits]).'.format (Argument))
+def BuildUniversalPayload(Args):
+ def RunCommand(cmd):
+ print(cmd)
+ p = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT,cwd=os.environ['WORKSPACE'])
+ while True:
+ line = p.stdout.readline()
+ if not line:
+ break
+ print(line.strip().decode(errors='ignore'))
+
+ p.communicate()
+ if p.returncode != 0:
+ print("- Failed - error happened when run command: %s"%cmd)
+ raise Exception("ERROR: when run command: %s"%cmd)
- if len(MajorStr) > 0 and len(MajorStr) < 3:
- try:
- Major = int(MajorStr, 16)
- except:
- raise argparse.ArgumentTypeError ('{} Major version of SpecRevision is not a valid integer value.'.format (Argument))
- else:
- raise argparse.ArgumentTypeError ('{} is not a valid SpecRevision format (Major[8-bits].Minor[8-bits]).'.format (Argument))
-
- return int('0x{0:02x}{1:02x}'.format(Major, Minor), 0)
-
-def Validate32BitInteger (Argument):
- try:
- Value = int (Argument, 0)
- except:
- raise argparse.ArgumentTypeError ('{} is not a valid integer value.'.format (Argument))
- if Value < 0:
- raise argparse.ArgumentTypeError ('{} is a negative value.'.format (Argument))
- if Value > 0xffffffff:
- raise argparse.ArgumentTypeError ('{} is larger than 32-bits.'.format (Argument))
- return Value
-
-def RunCommand(cmd):
- print(cmd)
- p = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.STDOUT,cwd=os.environ['WORKSPACE'])
- while True:
- line = p.stdout.readline()
- if not line:
- break
- print(line.strip().decode(errors='ignore'))
-
- p.communicate()
- if p.returncode != 0:
- print("- Failed - error happened when run command: %s"%cmd)
- raise Exception("ERROR: when run command: %s"%cmd)
-
-def BuildUniversalPayload(Args, MacroList):
BuildTarget = Args.Target
ToolChain = Args.ToolChain
Quiet = "--quiet" if Args.Quiet else ""
ElfToolChain = 'CLANGDWARF'
BuildDir = os.path.join(os.environ['WORKSPACE'], os.path.normpath("Build/UefiPayloadPkgX64"))
+ BuildModule = ""
+ BuildArch = ""
if Args.Arch == 'X64':
BuildArch = "X64"
- ObjCopyFlag = "elf64-x86-64"
EntryOutputDir = os.path.join(BuildDir, "{}_{}".format (BuildTarget, ElfToolChain), os.path.normpath("X64/UefiPayloadPkg/UefiPayloadEntry/UniversalPayloadEntry/DEBUG/UniversalPayloadEntry.dll"))
else:
BuildArch = "IA32 -a X64"
- ObjCopyFlag = "elf32-i386"
EntryOutputDir = os.path.join(BuildDir, "{}_{}".format (BuildTarget, ElfToolChain), os.path.normpath("IA32/UefiPayloadPkg/UefiPayloadEntry/UniversalPayloadEntry/DEBUG/UniversalPayloadEntry.dll"))
if Args.PreBuildUplBinary is not None:
EntryOutputDir = os.path.abspath(Args.PreBuildUplBinary)
- EntryModuleInf = os.path.normpath("UefiPayloadPkg/UefiPayloadEntry/UniversalPayloadEntry.inf")
DscPath = os.path.normpath("UefiPayloadPkg/UefiPayloadPkg.dsc")
- DxeFvOutputDir = os.path.join(BuildDir, "{}_{}".format (BuildTarget, ToolChain), os.path.normpath("FV/DXEFV.Fv"))
- BdsFvOutputDir = os.path.join(BuildDir, "{}_{}".format (BuildTarget, ToolChain), os.path.normpath("FV/BDSFV.Fv"))
- NetworkFvOutputDir = os.path.join(BuildDir, "{}_{}".format (BuildTarget, ToolChain), os.path.normpath("FV/NETWORKFV.Fv"))
- PayloadReportPath = os.path.join(BuildDir, "UefiUniversalPayload.txt")
ModuleReportPath = os.path.join(BuildDir, "UefiUniversalPayloadEntry.txt")
UpldInfoFile = os.path.join(BuildDir, "UniversalPayloadInfo.bin")
- if "CLANG_BIN" in os.environ:
- LlvmObjcopyPath = os.path.join(os.environ["CLANG_BIN"], "llvm-objcopy")
- else:
- LlvmObjcopyPath = "llvm-objcopy"
- try:
- RunCommand('"%s" --version'%LlvmObjcopyPath)
- except:
- print("- Failed - Please check if LLVM is installed or if CLANG_BIN is set correctly")
- sys.exit(1)
-
Pcds = ""
if (Args.pcd != None):
for PcdItem in Args.pcd:
Pcds += " --pcd {}".format (PcdItem)
Defines = ""
- for key in MacroList:
- Defines +=" -D {0}={1}".format(key, MacroList[key])
+ if (Args.Macro != None):
+ for MacroItem in Args.Macro:
+ Defines += " -D {}".format (MacroItem)
#
# Building DXE core and DXE drivers as DXEFV.
#
if Args.BuildEntryOnly == False:
+ PayloadReportPath = os.path.join(BuildDir, "UefiUniversalPayload.txt")
BuildPayload = "build -p {} -b {} -a X64 -t {} -y {} {}".format (DscPath, BuildTarget, ToolChain, PayloadReportPath, Quiet)
BuildPayload += Pcds
BuildPayload += Defines
@@ -144,6 +94,7 @@ def BuildUniversalPayload(Args, MacroList):
# Building Universal Payload entry.
#
if Args.PreBuildUplBinary is None:
+ EntryModuleInf = os.path.normpath("UefiPayloadPkg/UefiPayloadEntry/UniversalPayloadEntry.inf")
BuildModule = "build -p {} -b {} -a {} -m {} -t {} -y {} {}".format (DscPath, BuildTarget, BuildArch, EntryModuleInf, ElfToolChain, ModuleReportPath, Quiet)
BuildModule += Pcds
BuildModule += Defines
@@ -151,59 +102,80 @@ def BuildUniversalPayload(Args, MacroList):
#
# Buid Universal Payload Information Section ".upld_info"
#
- upld_info_hdr = UPLD_INFO_HEADER()
+ upld_info_hdr = UPLD_INFO_HEADER()
upld_info_hdr.SpecRevision = Args.SpecRevision
- upld_info_hdr.Revision = Args.Revision
- upld_info_hdr.ProducerId = Args.ProducerId.encode()[:16]
- upld_info_hdr.ImageId = Args.ImageId.encode()[:16]
- upld_info_hdr.Attribute |= 1 if BuildTarget == "DEBUG" else 0
+ upld_info_hdr.Revision = Args.Revision
+ upld_info_hdr.ProducerId = Args.ProducerId.encode()[:16]
+ upld_info_hdr.ImageId = Args.ImageId.encode()[:16]
+ upld_info_hdr.Attribute |= 1 if BuildTarget == "DEBUG" else 0
fp = open(UpldInfoFile, 'wb')
fp.write(bytearray(upld_info_hdr))
fp.close()
+ MultiFvList = []
if Args.BuildEntryOnly == False:
- #
- # Copy the DXEFV as a section in elf format Universal Payload entry.
- #
- remove_section = '"{}" -I {} -O {} --remove-section .upld_info --remove-section .upld.uefi_fv --remove-section .upld.bds_fv {}'.format (
- LlvmObjcopyPath,
- ObjCopyFlag,
- ObjCopyFlag,
- EntryOutputDir
- )
- add_section = '"{}" -I {} -O {} --add-section .upld_info={} --add-section .upld.uefi_fv={} --add-section .upld.bds_fv={} {}'.format (
- LlvmObjcopyPath,
- ObjCopyFlag,
- ObjCopyFlag,
- UpldInfoFile,
- DxeFvOutputDir,
- BdsFvOutputDir,
- EntryOutputDir
- )
- set_section = '"{}" -I {} -O {} --set-section-alignment .upld_info=4 --set-section-alignment .upld.uefi_fv=16 --set-section-alignment .upld.bds_fv=16 {}'.format (
- LlvmObjcopyPath,
- ObjCopyFlag,
- ObjCopyFlag,
- EntryOutputDir
- )
- #
- # Append network fv to sections if exists
- #
- if os.path.isfile(NetworkFvOutputDir):
- index = remove_section.find(EntryOutputDir)
- remove_section = remove_section[:index] + '--remove-section .upld.network_fv ' + remove_section[index:]
- index = add_section.find(EntryOutputDir)
- add_section = add_section[:index] + '--add-section .upld.network_fv=' + NetworkFvOutputDir + ' ' + add_section[index:]
- index = set_section.find(EntryOutputDir)
- set_section = set_section[:index] + '--set-section-alignment .upld.network_fv=16 ' + set_section[index:]
-
- RunCommand(remove_section)
- RunCommand(add_section)
- RunCommand(set_section)
+ MultiFvList = [
+ ['uefi_fv', os.path.join(BuildDir, "{}_{}".format (BuildTarget, ToolChain), os.path.normpath("FV/DXEFV.Fv")) ],
+ ['bds_fv', os.path.join(BuildDir, "{}_{}".format (BuildTarget, ToolChain), os.path.normpath("FV/BDSFV.Fv")) ],
+ ['network_fv', os.path.join(BuildDir, "{}_{}".format (BuildTarget, ToolChain), os.path.normpath("FV/NETWORKFV.Fv")) ],
+ ]
+ AddSectionName = '.upld_info'
+ ReplaceFv (EntryOutputDir, UpldInfoFile, AddSectionName, Alignment = 4)
shutil.copy (EntryOutputDir, os.path.join(BuildDir, 'UniversalPayload.elf'))
+ return MultiFvList, os.path.join(BuildDir, 'UniversalPayload.elf')
+
def main():
+ def ValidateSpecRevision (Argument):
+ try:
+ (MajorStr, MinorStr) = Argument.split('.')
+ except:
+ raise argparse.ArgumentTypeError ('{} is not a valid SpecRevision format (Major[8-bits].Minor[8-bits]).'.format (Argument))
+ #
+ # Spec Revision Bits 15 : 8 - Major Version. Bits 7 : 0 - Minor Version.
+ #
+ if len(MinorStr) > 0 and len(MinorStr) < 3:
+ try:
+ Minor = int(MinorStr, 16) if len(MinorStr) == 2 else (int(MinorStr, 16) << 4)
+ except:
+ raise argparse.ArgumentTypeError ('{} Minor version of SpecRevision is not a valid integer value.'.format (Argument))
+ else:
+ raise argparse.ArgumentTypeError ('{} is not a valid SpecRevision format (Major[8-bits].Minor[8-bits]).'.format (Argument))
+
+ if len(MajorStr) > 0 and len(MajorStr) < 3:
+ try:
+ Major = int(MajorStr, 16)
+ except:
+ raise argparse.ArgumentTypeError ('{} Major version of SpecRevision is not a valid integer value.'.format (Argument))
+ else:
+ raise argparse.ArgumentTypeError ('{} is not a valid SpecRevision format (Major[8-bits].Minor[8-bits]).'.format (Argument))
+
+ return int('0x{0:02x}{1:02x}'.format(Major, Minor), 0)
+
+ def Validate32BitInteger (Argument):
+ try:
+ Value = int (Argument, 0)
+ except:
+ raise argparse.ArgumentTypeError ('{} is not a valid integer value.'.format (Argument))
+ if Value < 0:
+ raise argparse.ArgumentTypeError ('{} is a negative value.'.format (Argument))
+ if Value > 0xffffffff:
+ raise argparse.ArgumentTypeError ('{} is larger than 32-bits.'.format (Argument))
+ return Value
+
+ def ValidateAddFv (Argument):
+ Value = Argument.split ("=")
+ if len (Value) != 2:
+ raise argparse.ArgumentTypeError ('{} is incorrect format with "xxx_fv=xxx.fv"'.format (Argument))
+ if Value[0][-3:] != "_fv":
+ raise argparse.ArgumentTypeError ('{} is incorrect format with "xxx_fv=xxx.fv"'.format (Argument))
+ if Value[1][-3:].lower () != ".fv":
+ raise argparse.ArgumentTypeError ('{} is incorrect format with "xxx_fv=xxx.fv"'.format (Argument))
+ if os.path.exists (Value[1]) == False:
+ raise argparse.ArgumentTypeError ('File {} is not found.'.format (Value[1]))
+ return Value
+
parser = argparse.ArgumentParser(description='For building Universal Payload')
parser.add_argument('-t', '--ToolChain')
parser.add_argument('-b', '--Target', default='DEBUG')
@@ -212,22 +184,32 @@ def main():
parser.add_argument('-i', '--ImageId', type=str, help='Specify payload ID (16 bytes maximal).', default ='UEFI')
parser.add_argument('-q', '--Quiet', action='store_true', help='Disable all build messages except FATAL ERRORS.')
parser.add_argument("-p", "--pcd", action="append")
- parser.add_argument("-s", "--SpecRevision", type=GenSpecRevision, default ='0.7', help='Indicates compliance with a revision of this specification in the BCD format.')
+ parser.add_argument("-s", "--SpecRevision", type=ValidateSpecRevision, default ='0.7', help='Indicates compliance with a revision of this specification in the BCD format.')
parser.add_argument("-r", "--Revision", type=Validate32BitInteger, default ='0x0000010105', help='Revision of the Payload binary. Major.Minor.Revision.Build')
parser.add_argument("-o", "--ProducerId", default ='INTEL', help='A null-terminated OEM-supplied string that identifies the payload producer (16 bytes maximal).')
parser.add_argument("-e", "--BuildEntryOnly", action='store_true', help='Build UniversalPayload Entry file')
parser.add_argument("-pb", "--PreBuildUplBinary", default=None, help='Specify the UniversalPayload file')
- MacroList = {}
+ parser.add_argument("-sk", "--SkipBuild", action='store_true', help='Skip UniversalPayload build')
+ parser.add_argument("-af", "--AddFv", type=ValidateAddFv, action='append', help='Add or replace specific FV into payload, Ex: uefi_fv=XXX.fv')
args = parser.parse_args()
- if args.Macro is not None:
- for Argument in args.Macro:
- if Argument.count('=') != 1:
- print("Unknown variable passed in: %s"%Argument)
- raise Exception("ERROR: Unknown variable passed in: %s"%Argument)
- tokens = Argument.strip().split('=')
- MacroList[tokens[0].upper()] = tokens[1]
- BuildUniversalPayload(args, MacroList)
- print ("Successfully build Universal Payload")
+
+ MultiFvList = []
+ UniversalPayloadBinary = args.PreBuildUplBinary
+ if (args.SkipBuild == False):
+ MultiFvList, UniversalPayloadBinary = BuildUniversalPayload(args)
+
+ if (args.AddFv != None):
+ for (SectionName, SectionFvFile) in args.AddFv:
+ MultiFvList.append ([SectionName, SectionFvFile])
+
+ if (UniversalPayloadBinary != None):
+ for (SectionName, SectionFvFile) in MultiFvList:
+ if os.path.exists (SectionFvFile) == False:
+ continue
+ print ("Patch {}={} into {}".format (SectionName, SectionFvFile, UniversalPayloadBinary))
+ ReplaceFv (UniversalPayloadBinary, SectionFvFile, '.upld.{}'.format (SectionName))
+
+ print ("\nSuccessfully build Universal Payload")
if __name__ == '__main__':
main()
--
2.39.1.windows.1
prev parent reply other threads:[~2023-04-26 7:17 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-26 7:11 [PATCH] * UefiPayloadPkg: Clang dependency removal brucex.wang
2023-04-26 7:17 ` Guo, Gua [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-list from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=BL1PR11MB54783B6AA1D4F254AEC3C4E1EF659@BL1PR11MB5478.namprd11.prod.outlook.com \
--to=devel@edk2.groups.io \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox