linux-stable.git/fs/fuse/dir.c, branch linux-3.4.y

fuse: hotfix truncate_pagecache() issue

2014-03-11T23:10:04+00:00

commit 06a7c3c2781409af95000c60a5df743fd4e2f8b4 upstream.

The way how fuse calls truncate_pagecache() from fuse_change_attributes()
is completely wrong. Because, w/o i_mutex held, we never sure whether
'oldsize' and 'attr->size' are valid by the time of execution of
truncate_pagecache(inode, oldsize, attr->size). In fact, as soon as we
released fc->lock in the middle of fuse_change_attributes(), we completely
loose control of actions which may happen with given inode until we reach
truncate_pagecache. The list of potentially dangerous actions includes
mmap-ed reads and writes, ftruncate(2) and write(2) extending file size.

The typical outcome of doing truncate_pagecache() with outdated arguments
is data corruption from user point of view. This is (in some sense)
acceptable in cases when the issue is triggered by a change of the file on
the server (i.e. externally wrt fuse operation), but it is absolutely
intolerable in scenarios when a single fuse client modifies a file without
any external intervention. A real life case I discovered by fsx-linux
looked like this:

1. Shrinking ftruncate(2) comes to fuse_do_setattr(). The latter sends
FUSE_SETATTR to the server synchronously, but before getting fc->lock ...
2. fuse_dentry_revalidate() is asynchronously called. It sends FUSE_LOOKUP
to the server synchronously, then calls fuse_change_attributes(). The
latter updates i_size, releases fc->lock, but before comparing oldsize vs
attr->size..
3. fuse_do_setattr() from the first step proceeds by acquiring fc->lock and
updating attributes and i_size, but now oldsize is equal to
outarg.attr.size because i_size has just been updated (step 2). Hence,
fuse_do_setattr() returns w/o calling truncate_pagecache().
4. As soon as ftruncate(2) completes, the user extends file size by
write(2) making a hole in the middle of file, then reads data from the hole
either by read(2) or mmap-ed read. The user expects to get zero data from
the hole, but gets stale data because truncate_pagecache() is not executed
yet.

The scenario above illustrates one side of the problem: not truncating the
page cache even though we should. Another side corresponds to truncating
page cache too late, when the state of inode changed significantly.
Theoretically, the following is possible:

1. As in the previous scenario fuse_dentry_revalidate() discovered that
i_size changed (due to our own fuse_do_setattr()) and is going to call
truncate_pagecache() for some 'new_size' it believes valid right now. But
by the time that particular truncate_pagecache() is called ...
2. fuse_do_setattr() returns (either having called truncate_pagecache() or
not -- it doesn't matter).
3. The file is extended either by write(2) or ftruncate(2) or fallocate(2).
4. mmap-ed write makes a page in the extended region dirty.

The result will be the lost of data user wrote on the fourth step.

The patch is a hotfix resolving the issue in a simplistic way: let's skip
dangerous i_size update and truncate_pagecache if an operation changing
file size is in progress. This simplistic approach looks correct for the
cases w/o external changes. And to handle them properly, more sophisticated
and intrusive techniques (e.g. NFS-like one) would be required. I'd like to
postpone it until the issue is well discussed on the mailing list(s).

Changed in v2:
 - improved patch description to cover both sides of the issue.

Signed-off-by: Maxim Patlasov 
Signed-off-by: Miklos Szeredi 
[bwh: Backported to 3.2: add the fuse_inode::state field which we didn't have]
Signed-off-by: Ben Hutchings 
Cc: Rui Xiang 
Signed-off-by: Greg Kroah-Hartman

fuse: readdir: check for slash in names

2014-03-11T23:10:04+00:00

commit efeb9e60d48f7778fdcad4a0f3ad9ea9b19e5dfd upstream.

Userspace can add names containing a slash character to the directory
listing.  Don't allow this as it could cause all sorts of trouble.

Signed-off-by: Miklos Szeredi 
[bwh: Backported to 3.2: drop changes to parse_dirplusfile() which we
 don't have]
Signed-off-by: Ben Hutchings 
Cc: Rui Xiang 
Signed-off-by: Greg Kroah-Hartman

fuse: invalidate inode attributes on xattr modification

2013-09-27T00:15:52+00:00

commit d331a415aef98717393dda0be69b7947da08eba3 upstream.

Calls like setxattr and removexattr result in updation of ctime.
Therefore invalidate inode attributes to force a refresh.

Signed-off-by: Anand Avati 
Reviewed-by: Brian Foster 
Signed-off-by: Miklos Szeredi 
Signed-off-by: Greg Kroah-Hartman

fuse: don't WARN when nlink is zero

2013-03-03T22:06:43+00:00

commit dfca7cebc2679f3d129f8e680a8f199a7ad16e38 upstream.

drop_nlink() warns if nlink is already zero.  This is triggerable by a buggy
userspace filesystem.  The cure, I think, is worse than the disease so disable
the warning.

Reported-by: Tero Roponen 
Signed-off-by: Miklos Szeredi 
Signed-off-by: Greg Kroah-Hartman

fuse: fix stat call on 32 bit platforms

2012-06-17T18:21:29+00:00

commit 45c72cd73c788dd18c8113d4a404d6b4a01decf1 upstream.

Now we store attr->ino at inode->i_ino, return attr->ino at the
first time and then return inode->i_ino if the attribute timeout
isn't expired. That's wrong on 32 bit platforms because attr->ino
is 64 bit and inode->i_ino is 32 bit in this case.

Fix this by saving 64 bit ino in fuse_inode structure and returning
it every time we call getattr. Also squash attr->ino into inode->i_ino
explicitly.

Signed-off-by: Pavel Shilovsky 
Signed-off-by: Miklos Szeredi 
Signed-off-by: Greg Kroah-Hartman

fuse: O_DIRECT support for files

2012-03-05T14:48:11+00:00

Implement ->direct_IO() method in aops. The ->direct_IO() method combines
the existing fuse_direct_read/fuse_direct_write methods to implement
O_DIRECT functionality.

Reaching ->direct_IO() in the read path via generic_file_aio_read ensures
proper synchronization with page cache with its existing framework.

Reaching ->direct_IO() in the write path via fuse_file_aio_write is made
to come via generic_file_direct_write() which makes it play nice with
the page cache w.r.t other mmap pages etc.

On files marked 'direct_io' by the filesystem server, IO always follows
the fuse_direct_read/write path. There is no effect of fcntl(O_DIRECT)
and it always succeeds.

On files not marked with 'direct_io' by the filesystem server, the IO
path depends on O_DIRECT flag by the application. This can be passed
at the time of open() as well as via fcntl().

Note that asynchronous O_DIRECT iocb jobs are completed synchronously
always (this has been the case with FUSE even before this patch)

Signed-off-by: Anand Avati 
Reviewed-by: Jeff Moyer 
Signed-off-by: Miklos Szeredi

fuse: fix nlink after unlink

2012-03-05T14:48:11+00:00

Anand Avati reports that the following sequence of system calls fail on a fuse
filesystem:


 	create("filename") => 0
 	link("filename", "linkname") => 0
 	unlink("filename") => 0
 	link("linkname", "filename") => -ENOENT ### BUG ###

vfs_link() fails with ENOENT if i_nlink is zero, this is done to prevent
resurrecting already deleted files.

Fuse clears i_nlink on unlink even if there are other links pointing to the
file.

Reported-by: Anand Avati 
Signed-off-by: Miklos Szeredi

Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse

2012-01-12T20:39:21+00:00

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
  FUSE: Notifying the kernel of deletion.
  fuse: support ioctl on directories
  fuse: Use kcalloc instead of kzalloc to allocate array
  fuse: llseek optimize SEEK_CUR and SEEK_SET

fuse: propagate umode_t

2012-01-04T03:55:07+00:00

Signed-off-by: Al Viro

switch ->mknod() to umode_t

2012-01-04T03:54:54+00:00

Signed-off-by: Al Viro