Implement sys::fs::copy_file using the macOS copyfile(3) API to support APFS clones

[aprantl]

Summary

This patch adds a Darwin-specific implementation of llvm::sys::fs::copy_file() that uses the macOS copyfile(3) API to support APFS copy-on-write clones, which should be faster and much more space efficient.

https://developer.apple.com/library/archive/documentation/FileManagement/Conceptual/APFS_Guide/ToolsandAPIs/ToolsandAPIs.html

(I recently started using copy_file in dsymutil: https://reviews.llvm.org/D60642)

llvm/trunk/lib/Support/Path.cpp

if (Parent.empty())
     return EC;
 
   if ((EC = create_directories(Parent, IgnoreExisting, Perms)))
       return EC;
 
   return create_directory(P, IgnoreExisting, Perms);
 }
 
+#ifndef __APPLE__
 static std::error_code copy_file_internal(int ReadFD, int WriteFD) {
   const size_t BufSize = 4096;
   char *Buf = new char[BufSize];
   int BytesRead = 0, BytesWritten = 0;
   for (;;) {
     BytesRead = read(ReadFD, Buf, BufSize);
     if (BytesRead <= 0)
       break;
if (std::error_code EC = openFileForRead(From, ReadFD, OF_None))
     return EC;
 
   std::error_code EC = copy_file_internal(ReadFD, ToFD);
 
   close(ReadFD);
 
   return EC;
 }
+#endif
 
 ErrorOr<MD5::MD5Result> md5_contents(int FD) {
   MD5 Hash;
 
   constexpr size_t BufSize = 4096;
   std::vector<uint8_t> Buf(BufSize);
   int BytesRead = 0;
   for (;;) {

llvm/trunk/lib/Support/Unix/Path.inc

 #endif
 
 #include <dirent.h>
 #include <pwd.h>
 
 #ifdef __APPLE__
 #include <mach-o/dyld.h>
 #include <sys/attr.h>
+#include <copyfile.h>
 #elif defined(__DragonFly__)
 #include <sys/mount.h>
 #endif
 
 // Both stdio.h and cstdio are included via different paths and
 // stdcxx's cstdio doesn't include stdio.h, so it doesn't #undef the macros
 // either.
 #undef ferror
if (getDarwinConfDir(ErasedOnReboot, Result))
     return;
 
   const char *RequestedDir = getDefaultTempDir(ErasedOnReboot);
   Result.append(RequestedDir, RequestedDir + strlen(RequestedDir));
 }
 
 } // end namespace path
 
+namespace fs {
+
+#ifdef __APPLE__
+/// This implementation tries to perform an APFS CoW clone of the file,
+/// which can be much faster and uses less space.
+std::error_code copy_file(const Twine &From, const Twine &To) {
+  uint32_t Flag = COPYFILE_DATA;
+  bool IsSymlink;
+  if (std::error_code Error = is_symlink_file(From, IsSymlink))
+    return Error;
+
+  if (!IsSymlink)
+    if (__builtin_available(macos 10.12, *))
+      Flag = COPYFILE_CLONE;
+
+  int Status =
+      copyfile(From.str().c_str(), To.str().c_str(), /* State */ NULL, Flag);
+
+  if (Status == 0)
+    return std::error_code();
+  return std::error_code(errno, std::generic_category());
+}
+
+/// This implementation tries to perform an APFS CoW clone of the file,
+/// which can be much faster and uses less space.
+std::error_code copy_file(const Twine &From, int ToFD) {
+  uint32_t Flag = COPYFILE_DATA;
+  bool IsSymlink;
+  if (std::error_code Error = is_symlink_file(From, IsSymlink))
+    return Error;
+
+  int ReadFD;
+  if (std::error_code EC = openFileForRead(From, ReadFD, OF_None))
+    return EC;
+
+  if (!IsSymlink)
+    if (__builtin_available(macos 10.12, *))
+      Flag = COPYFILE_CLONE;
+
+  int Status = fcopyfile(ReadFD, ToFD, /*State*/ NULL, Flag);
+
+  close(ReadFD);
+  if (Status == 0)
+    return std::error_code();
+  return std::error_code(errno, std::generic_category());
+}
+#endif
+
+} // end namespace fs
+
 } // end namespace sys
 } // end namespace llvm