diff options
Diffstat (limited to 'xpcom/glue/FileUtils.cpp')
-rw-r--r-- | xpcom/glue/FileUtils.cpp | 582 |
1 files changed, 582 insertions, 0 deletions
diff --git a/xpcom/glue/FileUtils.cpp b/xpcom/glue/FileUtils.cpp new file mode 100644 index 0000000000..d017d9cea7 --- /dev/null +++ b/xpcom/glue/FileUtils.cpp @@ -0,0 +1,582 @@ +/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */ +/* vim: set ts=8 sts=2 et sw=2 tw=80: */ +/* This Source Code Form is subject to the terms of the Mozilla Public + * License, v. 2.0. If a copy of the MPL was not distributed with this + * file, You can obtain one at http://mozilla.org/MPL/2.0/. */ + +#include "mozilla/FileUtils.h" + +#include <errno.h> +#include <stdio.h> +#include <inttypes.h> + +#include "nscore.h" +#include "private/pprio.h" +#include "prmem.h" +#include "mozilla/Assertions.h" +#include "mozilla/MemUtils.h" +#include "mozilla/BaseProfilerMarkers.h" + +#if defined(XP_MACOSX) +# include <fcntl.h> +# include <unistd.h> +# include <mach/machine.h> +# include <mach-o/fat.h> +# include <mach-o/loader.h> +# include <sys/mman.h> +# include <sys/stat.h> +# include <limits.h> +#elif defined(XP_UNIX) +# include <fcntl.h> +# include <unistd.h> +# if defined(LINUX) +# include <elf.h> +# endif +# include <sys/types.h> +# include <sys/stat.h> +#elif defined(XP_WIN) +# include <nsWindowsHelpers.h> +# include <mozilla/NativeNt.h> +# include <mozilla/ScopeExit.h> +#endif + +// Functions that are not to be used in standalone glue must be implemented +// within this #if block +#if defined(MOZILLA_INTERNAL_API) + +# include "nsString.h" + +bool mozilla::fallocate(PRFileDesc* aFD, int64_t aLength) { +# if defined(HAVE_POSIX_FALLOCATE) + return posix_fallocate(PR_FileDesc2NativeHandle(aFD), 0, aLength) == 0; +# elif defined(XP_WIN) + int64_t oldpos = PR_Seek64(aFD, 0, PR_SEEK_CUR); + if (oldpos == -1) { + return false; + } + + if (PR_Seek64(aFD, aLength, PR_SEEK_SET) != aLength) { + return false; + } + + bool retval = (0 != SetEndOfFile((HANDLE)PR_FileDesc2NativeHandle(aFD))); + + PR_Seek64(aFD, oldpos, PR_SEEK_SET); + return retval; +# elif defined(XP_MACOSX) + int fd = PR_FileDesc2NativeHandle(aFD); + fstore_t store = {F_ALLOCATECONTIG, F_PEOFPOSMODE, 0, aLength}; + // Try to get a continous chunk of disk space + int ret = fcntl(fd, F_PREALLOCATE, &store); + if (ret == -1) { + // OK, perhaps we are too fragmented, allocate non-continuous + store.fst_flags = F_ALLOCATEALL; + ret = fcntl(fd, F_PREALLOCATE, &store); + if (ret == -1) { + return false; + } + } + return ftruncate(fd, aLength) == 0; +# elif defined(XP_UNIX) + // The following is copied from fcntlSizeHint in sqlite + /* If the OS does not have posix_fallocate(), fake it. First use + ** ftruncate() to set the file size, then write a single byte to + ** the last byte in each block within the extended region. This + ** is the same technique used by glibc to implement posix_fallocate() + ** on systems that do not have a real fallocate() system call. + */ + int64_t oldpos = PR_Seek64(aFD, 0, PR_SEEK_CUR); + if (oldpos == -1) { + return false; + } + + struct stat buf; + int fd = PR_FileDesc2NativeHandle(aFD); + if (fstat(fd, &buf)) { + return false; + } + + if (buf.st_size >= aLength) { + return false; + } + + const int nBlk = buf.st_blksize; + + if (!nBlk) { + return false; + } + + if (ftruncate(fd, aLength)) { + return false; + } + + int nWrite; // Return value from write() + int64_t iWrite = ((buf.st_size + 2 * nBlk - 1) / nBlk) * nBlk - + 1; // Next offset to write to + while (iWrite < aLength) { + nWrite = 0; + if (PR_Seek64(aFD, iWrite, PR_SEEK_SET) == iWrite) { + nWrite = PR_Write(aFD, "", 1); + } + if (nWrite != 1) { + break; + } + iWrite += nBlk; + } + + PR_Seek64(aFD, oldpos, PR_SEEK_SET); + return nWrite == 1; +# endif + return false; +} + +void mozilla::ReadAheadLib(nsIFile* aFile) { +# if defined(XP_WIN) + nsAutoString path; + if (!aFile || NS_FAILED(aFile->GetPath(path))) { + return; + } + ReadAheadLib(path.get()); +# elif defined(LINUX) && !defined(ANDROID) || defined(XP_MACOSX) + nsAutoCString nativePath; + if (!aFile || NS_FAILED(aFile->GetNativePath(nativePath))) { + return; + } + ReadAheadLib(nativePath.get()); +# endif +} + +void mozilla::ReadAheadFile(nsIFile* aFile, const size_t aOffset, + const size_t aCount, mozilla::filedesc_t* aOutFd) { +# if defined(XP_WIN) + nsAutoString path; + if (!aFile || NS_FAILED(aFile->GetPath(path))) { + return; + } + ReadAheadFile(path.get(), aOffset, aCount, aOutFd); +# elif defined(LINUX) && !defined(ANDROID) || defined(XP_MACOSX) + nsAutoCString nativePath; + if (!aFile || NS_FAILED(aFile->GetNativePath(nativePath))) { + return; + } + ReadAheadFile(nativePath.get(), aOffset, aCount, aOutFd); +# endif +} + +mozilla::PathString mozilla::GetLibraryName(mozilla::pathstr_t aDirectory, + const char* aLib) { +# ifdef XP_WIN + nsAutoString fullName; + if (aDirectory) { + fullName.Assign(aDirectory); + fullName.Append('\\'); + } + AppendUTF8toUTF16(MakeStringSpan(aLib), fullName); + if (!strstr(aLib, ".dll")) { + fullName.AppendLiteral(".dll"); + } + return std::move(fullName); +# else + char* temp = PR_GetLibraryName(aDirectory, aLib); + if (!temp) { + return ""_ns; + } + nsAutoCString libname(temp); + PR_FreeLibraryName(temp); + return std::move(libname); +# endif +} + +mozilla::PathString mozilla::GetLibraryFilePathname(mozilla::pathstr_t aName, + PRFuncPtr aAddr) { +# ifdef XP_WIN + HMODULE handle = GetModuleHandleW(char16ptr_t(aName)); + if (!handle) { + return u""_ns; + } + + nsAutoString path; + path.SetLength(MAX_PATH); + DWORD len = GetModuleFileNameW(handle, char16ptr_t(path.BeginWriting()), + path.Length()); + if (!len) { + return u""_ns; + } + + path.SetLength(len); + return std::move(path); +# else + char* temp = PR_GetLibraryFilePathname(aName, aAddr); + if (!temp) { + return ""_ns; + } + nsAutoCString path(temp); + PR_Free(temp); // PR_GetLibraryFilePathname() uses PR_Malloc(). + return std::move(path); +# endif +} + +#endif // defined(MOZILLA_INTERNAL_API) + +#if defined(LINUX) && !defined(ANDROID) + +static const unsigned int bufsize = 4096; + +# ifdef __LP64__ +typedef Elf64_Ehdr Elf_Ehdr; +typedef Elf64_Phdr Elf_Phdr; +static const unsigned char ELFCLASS = ELFCLASS64; +typedef Elf64_Off Elf_Off; +# else +typedef Elf32_Ehdr Elf_Ehdr; +typedef Elf32_Phdr Elf_Phdr; +static const unsigned char ELFCLASS = ELFCLASS32; +typedef Elf32_Off Elf_Off; +# endif + +#elif defined(XP_MACOSX) + +# if defined(__i386__) +static const uint32_t CPU_TYPE = CPU_TYPE_X86; +# elif defined(__x86_64__) +static const uint32_t CPU_TYPE = CPU_TYPE_X86_64; +# elif defined(__ppc__) +static const uint32_t CPU_TYPE = CPU_TYPE_POWERPC; +# elif defined(__ppc64__) +static const uint32_t CPU_TYPE = CPU_TYPE_POWERPC64; +# elif defined(__aarch64__) +static const uint32_t CPU_TYPE = CPU_TYPE_ARM64; +# else +# error Unsupported CPU type +# endif + +# ifdef __LP64__ +# undef LC_SEGMENT +# define LC_SEGMENT LC_SEGMENT_64 +# undef MH_MAGIC +# define MH_MAGIC MH_MAGIC_64 +# define cpu_mach_header mach_header_64 +# define segment_command segment_command_64 +# else +# define cpu_mach_header mach_header +# endif + +class ScopedMMap { + public: + explicit ScopedMMap(const char* aFilePath) : buf(nullptr) { + fd = open(aFilePath, O_RDONLY); + if (fd < 0) { + return; + } + struct stat st; + if (fstat(fd, &st) < 0) { + return; + } + size = st.st_size; + buf = (char*)mmap(nullptr, size, PROT_READ, MAP_PRIVATE, fd, 0); + } + ~ScopedMMap() { + if (buf) { + munmap(buf, size); + } + if (fd >= 0) { + close(fd); + } + } + operator char*() { return buf; } + int getFd() { return fd; } + + private: + int fd; + char* buf; + size_t size; +}; +#endif + +void mozilla::ReadAhead(mozilla::filedesc_t aFd, const size_t aOffset, + const size_t aCount) { +#if defined(XP_WIN) + + LARGE_INTEGER fpOriginal; + LARGE_INTEGER fpOffset; +# if defined(HAVE_LONG_LONG) + fpOffset.QuadPart = 0; +# else + fpOffset.u.LowPart = 0; + fpOffset.u.HighPart = 0; +# endif + + // Get the current file pointer so that we can restore it. This isn't + // really necessary other than to provide the same semantics regarding the + // file pointer that other platforms do + if (!SetFilePointerEx(aFd, fpOffset, &fpOriginal, FILE_CURRENT)) { + return; + } + + if (aOffset) { +# if defined(HAVE_LONG_LONG) + fpOffset.QuadPart = static_cast<LONGLONG>(aOffset); +# else + fpOffset.u.LowPart = aOffset; + fpOffset.u.HighPart = 0; +# endif + + if (!SetFilePointerEx(aFd, fpOffset, nullptr, FILE_BEGIN)) { + return; + } + } + + char buf[64 * 1024]; + size_t totalBytesRead = 0; + DWORD dwBytesRead; + // Do dummy reads to trigger kernel-side readhead via + // FILE_FLAG_SEQUENTIAL_SCAN. Abort when underfilling because during testing + // the buffers are read fully A buffer that's not keeping up would imply that + // readahead isn't working right + while (totalBytesRead < aCount && + ReadFile(aFd, buf, sizeof(buf), &dwBytesRead, nullptr) && + dwBytesRead == sizeof(buf)) { + totalBytesRead += dwBytesRead; + } + + // Restore the file pointer + SetFilePointerEx(aFd, fpOriginal, nullptr, FILE_BEGIN); + +#elif defined(LINUX) && !defined(ANDROID) + + readahead(aFd, aOffset, aCount); + +#elif defined(XP_MACOSX) + + struct radvisory ra; + ra.ra_offset = aOffset; + ra.ra_count = aCount; + // The F_RDADVISE fcntl is equivalent to Linux' readahead() system call. + fcntl(aFd, F_RDADVISE, &ra); + +#endif +} + +void mozilla::ReadAheadLib(mozilla::pathstr_t aFilePath) { + if (!aFilePath) { + return; + } + +#ifdef MOZ_GECKO_PROFILER +# ifdef XP_WIN + auto WideToUTF8 = [](const wchar_t* aStr) -> std::string { + std::string s; + // Determine the number of output bytes (including null terminator). + const int numConv = ::WideCharToMultiByte(CP_UTF8, 0, aStr, -1, nullptr, 0, + nullptr, nullptr); + if (numConv == 0) { + return s; + } + s.resize(numConv); + const int numConvd = ::WideCharToMultiByte(CP_UTF8, 0, aStr, -1, s.data(), + numConv, nullptr, nullptr); + if (numConvd != numConv) { + // Error during conversion, remove any temporary data. + s.clear(); + } + return s; + }; +# endif + + AUTO_BASE_PROFILER_MARKER_TEXT("ReadAheadLib", OTHER, {}, +# ifdef XP_WIN + WideToUTF8(aFilePath) +# else + aFilePath +# endif + ); +#endif + +#if defined(XP_WIN) + if (!CanPrefetchMemory()) { + ReadAheadFile(aFilePath); + return; + } + nsAutoHandle fd(CreateFileW(aFilePath, GENERIC_READ | GENERIC_EXECUTE, + FILE_SHARE_READ, nullptr, OPEN_EXISTING, + FILE_FLAG_SEQUENTIAL_SCAN, nullptr)); + if (!fd) { + return; + } + + nsAutoHandle mapping(CreateFileMapping( + fd, nullptr, SEC_IMAGE | PAGE_EXECUTE_READ, 0, 0, nullptr)); + if (!mapping) { + return; + } + + PVOID data = MapViewOfFile( + mapping, FILE_MAP_READ | FILE_MAP_EXECUTE | SEC_IMAGE, 0, 0, 0); + if (!data) { + return; + } + auto guard = MakeScopeExit([=]() { UnmapViewOfFile(data); }); + mozilla::nt::PEHeaders headers(data); + Maybe<Span<const uint8_t>> bounds = headers.GetBounds(); + if (!bounds) { + return; + } + + PrefetchMemory((uint8_t*)data, bounds->Length()); + +#elif defined(LINUX) && !defined(ANDROID) + int fd = open(aFilePath, O_RDONLY); + if (fd < 0) { + return; + } + + union { + char buf[bufsize]; + Elf_Ehdr ehdr; + } elf; + // Read ELF header (ehdr) and program header table (phdr). + // We check that the ELF magic is found, that the ELF class matches + // our own, and that the program header table as defined in the ELF + // headers fits in the buffer we read. + if ((read(fd, elf.buf, bufsize) <= 0) || (memcmp(elf.buf, ELFMAG, 4)) || + (elf.ehdr.e_ident[EI_CLASS] != ELFCLASS) || + // Upcast e_phentsize so the multiplication is done in the same precision + // as the subsequent addition, to satisfy static analyzers and avoid + // issues with abnormally large program header tables. + (elf.ehdr.e_phoff + + (static_cast<Elf_Off>(elf.ehdr.e_phentsize) * elf.ehdr.e_phnum) >= + bufsize)) { + close(fd); + return; + } + // The program header table contains segment definitions. One such + // segment type is PT_LOAD, which describes how the dynamic loader + // is going to map the file in memory. We use that information to + // find the biggest offset from the library that will be mapped in + // memory. + Elf_Phdr* phdr = (Elf_Phdr*)&elf.buf[elf.ehdr.e_phoff]; + Elf_Off end = 0; + for (int phnum = elf.ehdr.e_phnum; phnum; phdr++, phnum--) { + if ((phdr->p_type == PT_LOAD) && (end < phdr->p_offset + phdr->p_filesz)) { + end = phdr->p_offset + phdr->p_filesz; + } + } + // Let the kernel read ahead what the dynamic loader is going to + // map in memory soon after. + if (end > 0) { + ReadAhead(fd, 0, end); + } + close(fd); +#elif defined(XP_MACOSX) + ScopedMMap buf(aFilePath); + char* base = buf; + if (!base) { + return; + } + + // An OSX binary might either be a fat (universal) binary or a + // Mach-O binary. A fat binary actually embeds several Mach-O + // binaries. If we have a fat binary, find the offset where the + // Mach-O binary for our CPU type can be found. + struct fat_header* fh = (struct fat_header*)base; + + if (OSSwapBigToHostInt32(fh->magic) == FAT_MAGIC) { + uint32_t nfat_arch = OSSwapBigToHostInt32(fh->nfat_arch); + struct fat_arch* arch = (struct fat_arch*)&buf[sizeof(struct fat_header)]; + for (; nfat_arch; arch++, nfat_arch--) { + if (OSSwapBigToHostInt32(arch->cputype) == CPU_TYPE) { + base += OSSwapBigToHostInt32(arch->offset); + break; + } + } + if (base == buf) { + return; + } + } + + // Check Mach-O magic in the Mach header + struct cpu_mach_header* mh = (struct cpu_mach_header*)base; + if (mh->magic != MH_MAGIC) { + return; + } + + // The Mach header is followed by a sequence of load commands. + // Each command has a header containing the command type and the + // command size. LD_SEGMENT commands describes how the dynamic + // loader is going to map the file in memory. We use that + // information to find the biggest offset from the library that + // will be mapped in memory. + char* cmd = &base[sizeof(struct cpu_mach_header)]; + uint32_t end = 0; + for (uint32_t ncmds = mh->ncmds; ncmds; ncmds--) { + struct segment_command* sh = (struct segment_command*)cmd; + if (sh->cmd != LC_SEGMENT) { + continue; + } + if (end < sh->fileoff + sh->filesize) { + end = sh->fileoff + sh->filesize; + } + cmd += sh->cmdsize; + } + // Let the kernel read ahead what the dynamic loader is going to + // map in memory soon after. + if (end > 0) { + ReadAhead(buf.getFd(), base - buf, end); + } +#endif +} + +void mozilla::ReadAheadFile(mozilla::pathstr_t aFilePath, const size_t aOffset, + const size_t aCount, mozilla::filedesc_t* aOutFd) { +#if defined(XP_WIN) + if (!aFilePath) { + if (aOutFd) { + *aOutFd = INVALID_HANDLE_VALUE; + } + return; + } + HANDLE fd = CreateFileW(aFilePath, GENERIC_READ, FILE_SHARE_READ, nullptr, + OPEN_EXISTING, FILE_FLAG_SEQUENTIAL_SCAN, nullptr); + if (aOutFd) { + *aOutFd = fd; + } + if (fd == INVALID_HANDLE_VALUE) { + return; + } + ReadAhead(fd, aOffset, aCount); + if (!aOutFd) { + CloseHandle(fd); + } +#elif defined(LINUX) && !defined(ANDROID) || defined(XP_MACOSX) + if (!aFilePath) { + if (aOutFd) { + *aOutFd = -1; + } + return; + } + int fd = open(aFilePath, O_RDONLY); + if (aOutFd) { + *aOutFd = fd; + } + if (fd < 0) { + return; + } + size_t count; + if (aCount == SIZE_MAX) { + struct stat st; + if (fstat(fd, &st) < 0) { + if (!aOutFd) { + close(fd); + } + return; + } + count = st.st_size; + } else { + count = aCount; + } + ReadAhead(fd, aOffset, count); + if (!aOutFd) { + close(fd); + } +#endif +} |