When cross-compiling, by default qemu_datadir is 'c:\Program
Files\QEMU', which is not recognized as being an absolute path, and
meson will end up adding the prefix again.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826110419.528931-6-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
An assertion failure is raised during request processing if
unshare(CLONE_FS) fails. Implement a probe at startup so the problem can
be detected right away.
Unfortunately Docker/Moby does not include unshare in the seccomp.json
list unless CAP_SYS_ADMIN is given. Other seccomp.json lists always
include unshare (e.g. podman is unaffected):
https://raw.githubusercontent.com/seccomp/containers-golang/master/seccomp.json
Use "docker run --security-opt seccomp=path/to/seccomp.json ..." if the
default seccomp.json is missing unshare.
Cc: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20200727190223.422280-4-stefanha@redhat.com>
Reviewed-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
virtiofsd does not need CAP_DAC_READ_SEARCH because it already has
the more powerful CAP_DAC_OVERRIDE. Drop it from the list of
capabilities.
This is important because container runtimes may not include
CAP_DAC_READ_SEARCH by default. This patch allows virtiofsd to reduce
its capabilities when running inside a Docker container.
Note that CAP_DAC_READ_SEARCH may be necessary again in the future if
virtiofsd starts using open_by_handle_at(2).
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200727190223.422280-2-stefanha@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Commit 93bb3d8d4c ("virtiofsd: remove symlink fallbacks") removed
the implementation of the "norace" option, so remove it from the
cmdline help and the documentation too.
Signed-off-by: Sergio Lopez <slp@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20200717121110.50580-1-slp@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Right now we enable remote posix locks by default. That means when guest
does a posix lock it sends request to server (virtiofsd). But currently
we only support non-blocking posix lock and return -EOPNOTSUPP for
blocking version.
This means that existing applications which are doing blocking posix
locks get -EOPNOTSUPP and fail. To avoid this, people have been
running virtiosd with option "-o no_posix_lock". For new users it
is still a surprise and trial and error takes them to this option.
Given posix lock implementation is not complete in virtiofsd, disable
it by default. This means that posix locks will work with-in applications
in a guest but not across guests. Anyway we don't support sharing
filesystem among different guests yet in virtiofs so this should
not lead to any kind of surprise or regression and will make life
little easier for virtiofs users.
Reported-by: Aa Aa <jimbothom@yandex.com>
Suggested-by: Miklos Szeredi <mszeredi@redhat.com>
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Allow capabilities to be added or removed from the allowed set for the
daemon; e.g.
default:
CapPrm: 00000000880000df
CapEff: 00000000880000df
-o modcaps=+sys_admin
CapPrm: 00000000882000df
CapEff: 00000000882000df
-o modcaps=+sys_admin:-chown
CapPrm: 00000000882000de
CapEff: 00000000882000de
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200629115420.98443-4-dgilbert@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Check the capability calls worked.
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Message-Id: <20200629115420.98443-3-dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
capng_updatev is a varargs function that needs a -1 to terminate it,
but it was missing.
In practice what seems to have been happening is that it's added the
capabilities we asked for, then runs into junk on the stack, so if
we're unlucky it might be adding some more, but in reality it's
failing - but after adding the capabilities we asked for.
Fixes: a59feb483b ("virtiofsd: only retain file system capabilities")
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Message-Id: <20200629115420.98443-2-dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
lo_setattr() invokes fchmod() in a rarely used code path, so it should
be whitelisted or virtiofsd will crash with EBADSYS.
Said code path can be triggered for example as follows:
On the host, in the shared directory, create a file with the sticky bit
set and a security.capability xattr:
(1) # touch foo
(2) # chmod u+s foo
(3) # setcap '' foo
Then in the guest let some process truncate that file after it has
dropped all of its capabilities (at least CAP_FSETID):
int main(int argc, char *argv[])
{
capng_setpid(getpid());
capng_clear(CAPNG_SELECT_BOTH);
capng_updatev(CAPNG_ADD, CAPNG_PERMITTED | CAPNG_EFFECTIVE, 0);
capng_apply(CAPNG_SELECT_BOTH);
ftruncate(open(argv[1], O_RDWR), 0);
}
This will cause the guest kernel to drop the sticky bit (i.e. perform a
mode change) as part of the truncate (where FATTR_FH is set), and that
will cause virtiofsd to invoke fchmod() instead of fchmodat().
(A similar configuration exists further below with futimens() vs.
utimensat(), but the former is not a syscall but just a wrapper for the
latter, so no further whitelisting is required.)
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1842667
Reported-by: Qian Cai <caiqian@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200608093111.14942-1-mreitz@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Path lookup in the kernel has special rules for looking up magic symlinks
under /proc. If a filesystem operation is instructed to follow symlinks
(e.g. via AT_SYMLINK_FOLLOW or lack of AT_SYMLINK_NOFOLLOW), and the final
component is such a proc symlink, then the target of the magic symlink is
used for the operation, even if the target itself is a symlink. I.e. path
lookup is always terminated after following a final magic symlink.
I was erronously assuming that in the above case the target symlink would
also be followed, and so workarounds were added for a couple of operations
to handle the symlink case. Since the symlink can be handled simply by
following the proc symlink, these workardouds are not needed.
Also remove the "norace" option, which disabled the workarounds.
Commit bdfd667883 ("virtiofsd: Fix xattr operations") already dealt with
the same issue for xattr operations.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Message-Id: <20200514140736.20561-1-mszeredi@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
All this process does is wait for its child. No capabilities are
needed.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
virtiofsd runs as root but only needs a subset of root's Linux
capabilities(7). As a file server its purpose is to create and access
files on behalf of a client. It needs to be able to access files with
arbitrary uid/gid owners. It also needs to be create device nodes.
Introduce a Linux capabilities(7) whitelist and drop all capabilities
that we don't need, making the virtiofsd process less powerful than a
regular uid root process.
# cat /proc/PID/status
...
Before After
CapInh: 0000000000000000 0000000000000000
CapPrm: 0000003fffffffff 00000000880000df
CapEff: 0000003fffffffff 00000000880000df
CapBnd: 0000003fffffffff 0000000000000000
CapAmb: 0000000000000000 0000000000000000
Note that file capabilities cannot be used to achieve the same effect on
the virtiofsd executable because mount is used during sandbox setup.
Therefore we drop capabilities programmatically at the right point
during startup.
This patch only affects the sandboxed child process. The parent process
that sits in waitpid(2) still has full root capabilities and will be
addressed in the next patch.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20200416164907.244868-2-stefanha@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Currently, setup_mounts() bind-mounts the shared directory without
MS_REC. This makes all submounts disappear.
Pass MS_REC so that the guest can see submounts again.
Fixes: 5baa3b8e95
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20200424133516.73077-1-mreitz@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Changed Fixes to point to the commit with the problem rather than
the commit that turned it on
While it's not possible to escape the proc filesystem through
lo->proc_self_fd, it is possible to escape to the root of the proc
filesystem itself through "../..".
Use a temporary mount for opening lo->proc_self_fd, that has it's root at
/proc/self/fd/, preventing access to the ancestor directories.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Message-Id: <20200429124733.22488-1-mszeredi@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
The system-wide fs.file-max sysctl value determines how many files can
be open. It defaults to a value calculated based on the machine's RAM
size. Previously virtiofsd would try to set RLIMIT_NOFILE to 1,000,000
and this allowed the FUSE client to exhaust the number of open files
system-wide on Linux hosts with less than 10 GB of RAM!
Take fs.file-max into account when choosing the default RLIMIT_NOFILE
value.
Fixes: CVE-2020-10717
Reported-by: Yuval Avrahami <yavrahami@paloaltonetworks.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200501140644.220940-3-stefanha@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Make it possible to specify the RLIMIT_NOFILE on the command-line.
Users running multiple virtiofsd processes should allocate a certain
number to each process so that the system-wide limit can never be
exhausted.
When this option is set to 0 the rlimit is left at its current value.
This is useful when a management tool wants to configure the rlimit
itself.
The default behavior remains unchanged: try to set the limit to
1,000,000 file descriptors if the current rlimit is lower.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200501140644.220940-2-stefanha@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
On success, the fdopendir() call closes fd. Later on the error
path we try to close an already-closed fd. This can lead to
use-after-free. Fix by only closing the fd if the fdopendir()
call failed.
Cc: qemu-stable@nongnu.org
Fixes: b39bce121b (add dirp_map to hide lo_dirp pointers)
Reported-by: Coverity (CID 1421933 USE_AFTER_FREE)
Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200321120654.7985-1-philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Current virtiofsd has problems about xattr operations and
they does not work properly for directory/symlink/special file.
The fundamental cause is that virtiofsd uses openat() + f...xattr()
systemcalls for xattr operation but we should not open symlink/special
file in the daemon. Therefore the function is restricted.
Fix this problem by:
1. during setup of each thread, call unshare(CLONE_FS)
2. in xattr operations (i.e. lo_getxattr), if inode is not a regular
file or directory, use fchdir(proc_loot_fd) + ...xattr() +
fchdir(root.fd) instead of openat() + f...xattr()
(Note: for a regular file/directory openat() + f...xattr()
is still used for performance reason)
With this patch, xfstests generic/062 passes on virtiofs.
This fix is suggested by Miklos Szeredi and Stefan Hajnoczi.
The original discussion can be found here:
https://www.redhat.com/archives/virtio-fs/2019-October/msg00046.html
Signed-off-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Message-Id: <20200227055927.24566-3-misono.tomohiro@jp.fujitsu.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
This is a cleanup patch to simplify the following xattr fix and
there is no functional changes.
- Move memory allocation to head of the function
- Unify fgetxattr/flistxattr call for both size == 0 and
size != 0 case
- Remove redundant lo_inode_put call in error path
(Note: second call is ignored now since @inode is already NULL)
Signed-off-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Message-Id: <20200227055927.24566-2-misono.tomohiro@jp.fujitsu.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
All code in fuse.h and struct fuse_module are not used by virtiofsd
so removing them is safe.
Signed-off-by: Xiao Yang <yangx.jy@cn.fujitsu.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Fix warning reported by Clang static code analyzer:
CC tools/virtiofsd/fuse_lowlevel.o
tools/virtiofsd/fuse_lowlevel.c:195:9: warning: Value stored to 'error' is never read
error = -ERANGE;
^ ~~~~~~~
Fixes: 3db2876
Reported-by: Clang Static Analyzer
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Fix warning reported by Clang static code analyzer:
CC tools/virtiofsd/passthrough_ll.o
tools/virtiofsd/passthrough_ll.c:925:9: warning: Value stored to 'newfd' is never read
newfd = -1;
^ ~~
tools/virtiofsd/passthrough_ll.c:942:9: warning: Value stored to 'newfd' is never read
newfd = -1;
^ ~~
Fixes: 7c6b66027
Reported-by: Clang Static Analyzer
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Fix warning reported by Clang static code analyzer:
CC tools/virtiofsd/passthrough_ll.o
tools/virtiofsd/passthrough_ll.c:1083:5: warning: Value stored to 'saverr' is never read
saverr = ENOMEM;
^ ~~~~~~
Fixes: 7c6b66027
Reported-by: Clang Static Analyzer
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Missing a NULL check if the argument fetch fails.
Fixes: Coverity CID 1413119
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Missing unlock in error path.
Fixes: Covertiy CID 1413123
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
If we fail when bringing up the socket we can leak the listen_fd;
in practice the daemon will exit so it's not really a problem.
Fixes: Coverity CID 1413121
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Remove fuse_req_getgroups that's unused in virtiofsd; it came in
from libfuse but we don't actually use it. It was called from
fuse_getgroups which we previously removed (but had left it's header
in).
Coverity had complained about null termination in it, but removing
it is the easiest answer.
Fixes: Coverity CID: 1413117 (String not null terminated)
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Add following options to the help message:
- cache
- flock|no_flock
- norace
- posix_lock|no_posix_lock
- readdirplus|no_readdirplus
- timeout
- writeback|no_writeback
- xattr|no_xattr
Signed-off-by: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
dgilbert: Split cache, norace, posix_lock, readdirplus off
into our own earlier patches that added the options
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
On guest graceful shutdown, virtiofsd receives VHOST_USER_GET_VRING_BASE
request from VMM and shuts down virtqueues by calling fv_set_started(),
which joins fv_queue_thread() threads. So when virtio_loop() returns,
there should be no thread is still accessing data in fuse session and/or
virtio dev.
But on abnormal exit, e.g. guest got killed for whatever reason,
vhost-user socket is closed and virtio_loop() breaks out the main loop
and returns to main(). But it's possible fv_queue_worker()s are still
working and accessing fuse session and virtio dev, which results in
crash or use-after-free.
Fix it by stopping fv_queue_thread()s before virtio_loop() returns,
to make sure there's no-one could access fuse session and virtio dev.
Reported-by: Qingming Su <qingming.su@linux.alibaba.com>
Signed-off-by: Eryu Guan <eguan@linux.alibaba.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
lo_copy_file_range() passes -errno to fuse_reply_err() and then fuse_reply_err()
changes it to errno again, so that subsequent fuse_send_reply_iov_nofree() catches
the wrong errno.(i.e. reports "fuse: bad error value: ...").
Make fuse_send_reply_iov_nofree() accept the correct -errno by passing errno
directly in lo_copy_file_range().
Signed-off-by: Xiao Yang <yangx.jy@cn.fujitsu.com>
Reviewed-by: Eryu Guan <eguan@linux.alibaba.com>
dgilbert: Sent upstream and now Merged as aa1185e153f774f1df65
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
lo_destroy was relying on some implicit knowledge of the locking;
we can avoid this if we create an unref_inode that doesn't take
the lock and then grab it for the whole of the lo_destroy.
Suggested-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Add an option to control the size of the thread pool. Requests are now
processed in parallel by default.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Now that lo_destroy() is serialized we can call unref_inode() so that
all inode resources are freed.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
When running with multiple threads it can be tricky to handle
FUSE_INIT/FUSE_DESTROY in parallel with other request types or in
parallel with themselves. Serialize FUSE_INIT and FUSE_DESTROY so that
malicious clients cannot trigger race conditions.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Introduce a thread pool so that fv_queue_thread() just pops
VuVirtqElements and hands them to the thread pool. For the time being
only one worker thread is allowed since passthrough_ll.c is not
thread-safe yet. Future patches will lift this restriction so that
multiple FUSE requests can be processed in parallel.
The main new concept is struct FVRequest, which contains both
VuVirtqElement and struct fuse_chan. We now have fv_VuDev for a device,
fv_QueueInfo for a virtqueue, and FVRequest for a request. Some of
fv_QueueInfo's fields are moved into FVRequest because they are
per-request. The name FVRequest conforms to QEMU coding style and I
expect the struct fv_* types will be renamed in a future refactoring.
This patch series is not optimal. fbuf reuse is dropped so each request
does malloc(se->bufsize), but there is no clean and cheap way to keep
this with a thread pool. The vq_lock mutex is held for longer than
necessary, especially during the eventfd_write() syscall. Performance
can be improved in the future.
prctl(2) had to be added to the seccomp whitelist because glib invokes
it.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
fuse_buf_writev() only handles the normal write in which src is buffer
and dest is fd. Specially if src buffer represents guest physical
address that can't be mapped by the daemon process, IO must be bounced
back to the VMM to do it by fuse_buf_copy().
Signed-off-by: Jun Piao <piaojun@huawei.com>
Suggested-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Suggested-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Define fuse_buf_writev() which use pwritev and writev to improve io
bandwidth. Especially, the src bufs with 0 size should be skipped as
their mems are not *block_size* aligned which will cause writev failed
in direct io mode.
Signed-off-by: Jun Piao <piaojun@huawei.com>
Suggested-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Since keep_cache(FOPEN_KEEP_CACHE) has no effect for directory as
described in fuse_common.h, use cache_readdir(FOPNE_CACHE_DIR) for
diretory open when cache=always mode.
Signed-off-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
When writeback mode is enabled (-o writeback), O_APPEND handling is
done in kernel. Therefore virtiofsd clears O_APPEND flag when open.
Otherwise O_APPEND flag takes precedence over pwrite() and write
data may corrupt.
Currently clearing O_APPEND flag is done in lo_open(), but we also
need the same operation in lo_create(). So, factor out the flag
update operation in lo_open() to update_open_flags() and call it
in both lo_open() and lo_create().
This fixes the failure of xfstest generic/069 in writeback mode
(which tests O_APPEND write data integrity).
Signed-off-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
If an application wants to do direct IO and opens a file with O_DIRECT
in guest, that does not necessarily mean that we need to bypass page
cache on host as well. So reset this flag on host.
If somebody needs to bypass page cache on host as well (and it is safe to
do so), we can add a knob in daemon later to control this behavior.
I check virtio-9p and they do reset O_DIRECT flag.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Eryu Guan <eguan@linux.alibaba.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Right now we always enable it regardless of given commandlines.
Fix it by setting the flag relying on the lo->flock bit.
Signed-off-by: Peng Tao <tao.peng@linux.alibaba.com>
Reviewed-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Reviewed-by: Sergio Lopez <slp@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
If thread A is using an inode it must not be deleted by thread B when
processing a FUSE_FORGET request.
The FUSE protocol itself already has a counter called nlookup that is
used in FUSE_FORGET messages. We cannot trust this counter since the
untrusted client can manipulate it via FUSE_FORGET messages.
Introduce a new refcount to keep inodes alive for the required lifespan.
lo_inode_put() must be called to release a reference. FUSE's nlookup
counter holds exactly one reference so that the inode stays alive as
long as the client still wants to remember it.
Note that the lo_inode->is_symlink field is moved to avoid creating a
hole in the struct due to struct field alignment.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Reviewed-by: Sergio Lopez <slp@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Reviewed-by: Misono Tomohiro <misono.tomohiro@jp.fujitsu.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
This reference counter plays a specific role in the FUSE protocol. It's
not a generic object reference counter and the FUSE kernel code calls it
"nlookup".
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Introduce lo_dirp_put() so that FUSE_RELEASEDIR does not cause
use-after-free races with other threads that are accessing lo_dirp.
Also make lo_releasedir() atomic to prevent FUSE_RELEASEDIR racing with
itself. This prevents double-frees.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Hold the lock across both lo_map_get() and lo_map_remove() to prevent
races between two FUSE_RELEASE requests. In this case I don't see a
serious bug but it's safer to do things atomically.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
We call into libvhost-user from the virtqueue handler thread and the
vhost-user message processing thread without a lock. There is nothing
protecting the virtqueue handler thread if the vhost-user message
processing thread changes the virtqueue or memory table while it is
running.
This patch introduces a read-write lock. Virtqueue handler threads are
readers. The vhost-user message processing thread is a writer. This
will allow concurrency for multiqueue in the future while protecting
against fv_queue_thread() vs virtio_loop() races.
Note that the critical sections could be made smaller but it would be
more invasive and require libvhost-user changes. Let's start simple and
improve performance later, if necessary. Another option would be an
RCU-style approach with lighter-weight primitives.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Marc-André Lureau
Stefan Hajnoczi
Sergio Lopez
Vivek Goyal
Paolo Bonzini
Dr. David Alan Gilbert
Max Reitz
Miklos Szeredi
Philippe Mathieu-Daudé
Misono Tomohiro
Xiao Yang
Masayoshi Mizuma
Eryu Guan
piaojun
Peng Tao