On the OpenIndiana Hipster distribution, compiling with GCC 4.9 would
fail because __BYTE_ORDER__ was defined but not to any of the values
assumed to be associated with this define (__LITTLE_ENDIAN__ or
__BIG_ENDIAN__). Instead it was defined to either
__ORDER_LITTLE_ENDIAN__ or __ORDER_BIG_ENDIAN__. This caused
compilation to fail.
Fixed by checking that all referenced defines are in fact defined
before using them and adding an additional #elif clause for this newly
discovered condition.
This fixes compiler warnings emitted when you compare an le32 value with
e.g. 'const_cpu_to_le32(-1)' on a little-endian system, because
previously the expansion of the macro expression 'const_cpu_to_le32(-1)'
would be '(-1)' on a little-endian system but '(u32)((((u32)(-1) &
0xff000000u) >> 24) | (((u32)(-1) & 0x00ff0000u) >> 8) | (((u32)(-1) &
0x0000ff00u) << 8) | (((u32)(-1) & 0x000000ffu) << 24))' on a
big-endian system, i.e. the type of the expanded expression would be
'int' (signed) in the little-endian case but 'u32' (unsigned) in the
big-endian case.
With this commit the type of the expanded expression will be 'le32' in
both the little-endian and the big-endian case.
Strict checking of endian-specific types mean that types that have a
fixed endianness in the data representation of the value are now defined
as complex types, enabling the compiler to catch mixed usage of these
types with native-endian types. This allows us to catch most issues
relating to usage on big-endian systems since we cannot anymore assign a
fixed-endian value to a native-endian variable and vice-versa without a
compiler error.
The downside is that we aren't able to apply simple binary operators to
the fixed-endian types anymore since they are complex... so all
combining fixed-endian constants and values with |, &, etc. and
comparison with ==, !=, <=, etc. must be replaced with a macro which
unpacks the wrapped value and performs the operation. Lots of changes,
lots of work but in the interest of good code quality it's justified.
The new compression formats used by Windows 10 uses reparse data, and
a new reparse tag which it is useful to define even though these formats
is not yet supported by ntfs-3g.
On-disk struct definitions used native types (u16/u32/u64/s16/s32/s64),
which doesn't say anything about the intended interpretation of the
data. The intention of having little-endian-specific types and
big-endian-specific types must have been to clarify interpretation of
data and intentions in the code. Therefore it seems reasonable to use
these types in struct definitions to clarify what data represention is
used to encode field data.
Because some struct members in layout.h are big-endian, this change also
means moving the duplicated definitions for big-endian byteswapping
macros and big-endian types found in acls.h and security.h to the
appropriate locations in endians.h and types.h respectively in order to
make them available for the struct definitions in layout.h.
The fuse cache does not handle properly hard-linked files, so ntfs-3g
traditionally disables it by using a null time-out. With an upgrade of
the fuse kernel module on OpenIndiana, cacheing of non hard-linked files
is now possible, so use it.
ntfs_valid_sid() required that the subauthority count be between 1 and 8
inclusively. However, Windows permits more than 8 subauthorities as well
as 0 subauthorities:
- The install.wim file for the latest Windows 10 build contains a file
whose DACL contains a SID with 10 subauthorities.
ntfs_set_ntfs_acl() was failing on this file.
- The IsValidSid() function on Windows returns true for subauthority
less than or equal to 15, including 0.
There was actually already a another SID validation function that had the
Windows-compatible behavior, so I merged the two together.
Contributed by Eric Biggers
Newer versions of Windows use more recent definitions of upper-case
table defined by the Unicode consortium. Now using the same table as
Windows 7, windows 8 and Windows 10. This only has an effect on file
systems newly created by mkntfs.
The support for ioctls has been added to fuse when using protocol 7.18,
and an equivalent upgrade has been done in fuse lite with commit [ae9aee].
For old kernels, a fall back to protocol 7.8 was implemented, but this
appears not to be supported in not-so-old kernels (e.g. 2.6.35).
With this patch, the fall back protocol is set to 7.12 or to the highest
level supported by the kernel.
fstrim(8) discards unused blocks on a mounted filesystem. It is useful for
solid-state drives (SSDs) and thinly-provisioned storage.
Only trimming the full device (with no option) is supported.
Contributed by Richard W.M. Jones
This is backporting code from the full FUSE library in order to support
ioctls. The fuse protocol level negociated is now 7.18 instead of 7.8.
A fallback protocol to 7.8 is provided for compatibility with older kernels.
32-bit ioctls are not supported by a 64-bit library
When Posix ACLs are used, the umask is ignored and the initial permissions
of created files are taken for the parent directory. However the umask
should still be used when the Posix ACLs are not enabled in the mount
options.
chkdsk deletes the ACLs when they are bad or when they are not used any
more. This fixes inserting a new ACL after the previously last ACL (or
even all of them) was deleted.
Windows applies legacy restrictions to file names, so when the option
windows_names is applied, reject the same reserved names, which are
CON, PRN, AUX, NUL, COM1..COM9, and LPT1..LPT9
When the runlist of the data attribute of MFT has to be split across
several extents, the location of each extent has to be known from the
runlist present in previous extents. So, force the first extent into
record 15 to avoid a bad layout.
Erik Larsson
Jean-Pierre André