Reuse the old "readdir" callback, but add a flags argument, that has
FUSE_READDIR_PLUS in case this is a "plus" version. Filesystems can safely
ignore this flag, but if they want they can add optimizations based on it:
i.e. only retrieve the full attributes in PLUS mode.
The filler function is also given a flags argument and the filesystem can
set FUSE_FILL_DIR_PLUS if all the attributes in "stat" are valid.
Distribute ulockmgr separately. It is not needed for the building of
libfuse, only fusexmp_fh. Check ulockmgr library in ./configure and if not
disable remote-lock suport in fusexmp_fh.
Allow 2.X and 3.X to coexist. Includes are now stored under
/usr/include/fuse3 and library is named libfuse3.*. Invoke pkg-config with
"fuse3" as the first argument to build with version 3 of the library.
- fuse_kern_unmount closes handle (e.g. 19)
- a thread in my process opens a file - the OS assigns newly freed
handle (i.e. 19)
- fuse_kern_chan_destroy closes the same handle (i.e. 19)
- a thread in my process opens another file - the OS assigns newly
freed handle (i.e. 19)
- * MAYHEM *
Reported by Dan Greenfield
- fuse_kern_unmount closes handle (e.g. 19)
- a thread in my process opens a file - the OS assigns newly freed
handle (i.e. 19)
- fuse_kern_chan_destroy closes the same handle (i.e. 19)
- a thread in my process opens another file - the OS assigns newly
freed handle (i.e. 19)
- * MAYHEM *
Reported by Dan Greenfield
FUSE_CFLAGS defines -D_FILE_OFFSET_BITS=64. There are three problems
with this:
(1) A larger program using libfuse might have modules compiled with
and without FUSE_CFLAGS, which, if LFS is not enabled and the platform
is 32 bit, would result in a fatal mix of 32 and 64 bit off_t. (This
would, of course, be a bug, but I think there is a better way to
detect this -- see below)
(2) Programs may need to be adjusted to support LFS. It's the
intention of the LFS standard that the _programmer_ enables LFS once
the program has been checked/adjusted.
(3) _FILE_OFFSET_BITS does not need to be defined at all on 64 bit
Linux. 64 bit off_t is the default there.
So I think it's better not to force -D_FILE_OFFSET_BITS=64, and
because of (3) I also think you shouldn't test for it.
However off_t must still be 64 bits, so how to enforce that? C1X will
define static assertions[1], and these can be used to check the size
of off_t.
Not all compilers support static assertions yet, although several do.
Therefore I have surrounded the static assertion with a conservative
check that the compiler is GCC >= 4.6. In the long run, this test can
be removed and you can just use 'static_assert'.
Mount can be used with an "-o context=" option in order to specify a
mountpoint-wide SELinux security context different from the default context
provided by the active SELinux policy.
This is useful in order to enable users to mount multiple sshfs targets under
distinct contexts, which is my main motivation for getting this patch mainlined.
This patch implements readdirplus support in FUSE usersapce. It adds
a new fuse lowlevel operations fuse_lowleve_ops::readdir_plus,
corespoding mount options and helper functions to maintain buffer.
[From: Eric Wong <normalperson@yhbt.net>]
This makes our terminology consistent with NFS and
our kernel module, as well as reducing user/developer
confusion in the command-line.
Note: I'm keeping "fuse_add_direntry_plus" since that is
less standardized in its use than "readdirplus" for now.
Signed-off-by: Feng Shuo <steve.shuo.feng@gmail.com>
The FUSE library may be used from any number of programs which
may also fork() + execve(), so set the close-on-exec flag to
avoid inadvertant leakage of pipe file descriptors.
While we're at it, attempt to use pipe2() since this is within a
(currently) Linux-only code path and pipe2() offers thread-safety.
fuse_interrupted(), fuse_get_context(), etc... crashed in non-fuse threads.
Instead return false, NULL or error depending on the function.
Reported by Michael Berlin
Apparently new versions of util-linux umount return with an exit status of 1
even if the "--fake" option is recognised (despite being documented otherwise).
Check this condition and don't complain about util-linux version.
The lru list was not initialized for the "/" path. This resulted in
remove_node_lru() crashing on LOOKUP-DOTDOT.
Patch by Madan Valluri.
--
ChangeLog | 4 ++++
lib/fuse.c | 4 ++++
2 files changed, 8 insertions(+)
Patch by Ratna Manoj.
queue_element_unlock() should set ->first_locked and ->second_locked to false.
Discovered with 'fs_racer'. The assert(wnode->treelock == TREELOCK_WRITE) in
unlock_path() was hit within minutes.
Miklos: simplified patch
Due to an oversight, splice will never actually be used for i/o. Someone forgot
to #include "config.h" in lib/buffer.c (in fact almost no files include that
header). As a result, even though configure detects splice support and puts
HAVE_SPLICE in config.h, buffer.c is always compiled as if there is no splice
support.
Also add #include "config.h" to fuse.c and fuse_lowlevel.c. These currently
include it indirectly through fuse_misc.h, but we don't want to depend on that.
Reported by Matthew Gabeler-Lee
Commit 4dc7e675bb (Don't unhash name in FORGET) broke the forget logic in a
subtle way, resulting in "fuse internal error: node NNN not found" and causing
the filesystem daemon to abort.
Fix by incrementing the node refcount if nlookup goes from zero to one.
Reported by Kyle Lippincott
Eric Wong