IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
The OneFS Samba implementation of change notify is modeled after the
usage of Linux's inotify kernel subsystem. A single call is made
into the onefs.so VFS module to initialize kernel tracking of certain
file change events. When these events occur a kernel notification is
sent to smbd and the notification event is translated and given to the
general Samba Change Notify layer through a callback function.
The most difficult aspect is converting an SMB CompletionFilter to
a matching ifs_event mask, and then back to an appropriate change
notify action. Currently, not all possible cases are handled by the
this module, but the most prevalent ones, which are tested by
smbtorture, are implemented.
As the NFSv4 ACL mapping code doesn't map write directory into the DELETE_CHILD
permission bit (which we require before allowing a delete) no one can delete
files without an explicit DELETE_CHILD bit set on the directory. Add this mapping.
Jeremy.
This extends the file_id struct to add an additional generic uint64_t
field: extid. For backwards compatibility with dev/inodes stored in
xattr_tdbs and acl_tdbs, the ext id is ignored for these databases.
This patch should cause no functional change on systems that don't use
SMB_VFS_FILE_ID_CREATE to set the extid.
Existing code that uses the smb_share_mode library will need to be
updated to be compatibile with the new extid.
Previously, we didn’t call SMB_VFS_OPEN_DIR from the streams module,
instead we called fdopendir(). As a result we failed to populate the
dir_state list in the readdirplus module. So when we tried to view the
stream data, we will always returned NULL.
To solve this I separated onefs_opendir() and the initialization of
the dir_state list. This is done by introducing a new utility function
“onefs_rdp_add_dir_state()”, which initializes the dir_state structure
and adds it to the dir_state list. This function is called from the
streams module before calling readdir().
* Much of the beginning should look familiar, as I re-used the OneFS oplock
callback record concept. This was necessary to keep our own state around - it
really only consists of a lock state, per asynchronous lock that is currently
unsatisfied. The onefs_cbrl_callback_records map to BLRs by the id.
* There are 4 states an async lock can be in. NONE means there is no async
currently out for the lock, as opposed to ASYNC. DONE means we've locked
*every* lock (keep in mind a request can ask for multiple locks at a time.)
ERROR is an error.
* onefs_cbrl_async_success: The lock_num is incremented, and the state changed,
so that when process_blocking_lock_queue is run, we will try the *next* lock,
rather than the same one again.
* onefs_brl_lock_windows() has some complicated logic:
* We do a no-op if we're passed a BLR and the matching state is ASYNC --
this means Samba is trying to get the same lock twice, and we just need
to wait longer, so we return an error.
* PENDING lock calls happen when the lock is being queued on the BLQ -- we
do async in this case.
* We also do async in the case that we're passed a BLR, but the lock is not
pending. This is an async lock being probed by process_blocking_lock_queue.
* We do a sync lock for any normal first request of a lock.
* Failure is returned, but it doesn't go to the client unless the lock has
actually timed out.
This patch adds 3 new VFS OPs for Windows byte range locking: BRL_LOCK_WINDOWS,
BRL_UNLOCK_WINDOWS and BRL_CANCEL_WINDOWS. Specifically:
* I renamed brl_lock_windows, brl_unlock_windows and brl_lock_cancel to
*_default as the default implementations of the VFS ops.
* The blocking_lock_record (BLR) is now passed into the brl_lock_windows and
brl_cancel_windows paths. The Onefs implementation uses it - future
implementations may find it useful too.
* Created brl_lock_cancel to do what brl_lock/brl_unlock do: set up a
lock_struct and call either the Posix or Windows lock function. These happen
to be the same for the default implementation.
* Added helper functions: increment_current_lock_count() and
decrement_current_lock_count().
* Minor spelling correction in brl_timeout_fn: brl -> blr.
* Changed blocking_lock_cancel() to return the BLR that it has cancelled. This
allows us to assert its the lock that we wanted to cancel. If this assert ever
fires, this path will need to take in the BLR to cancel, rather than choosing
on its own.
* Adds a small helper function: find_blocking_lock_record_by_id(). Used by the
OneFS implementation, but could be useful for others.
OneFS provides the bulk directory enumeration syscall readdirplus(). This
syscall has the same semantics as the NFSv3 READDIRPLUS command, returning
a batch of directory entries with prefetched stat information via one
syscall.
This commit wraps the readdirplus() call in the existing POSIX
readdir/seekdir VFS interface. By default a batch of 128 directory entries
are optimistically read from the kernel into a global cache, and fed to
iterative calls of VFS_OP_READDIR.
The global buffers could be avoided in the future by hanging connection
specific buffers off the conn struct.
Added new parameter "onefs:use readdirplus" which toggles usage of this
code on or off.
By default this VFS call is a NOOP, but the onefs vfs module takes advantage
of it to initialize direntry search caches at the beginning of each
TRANS2_FIND_FIRST, TRANS2_FIND_NEXT, SMBffirst, SMBsearch, and SMBunique
* this allows VFS implementations that prefetch stat information on
readdir to return it through one VFS call
* backwards compatibility is maintained by passing in NULL
* if the system readdir doesn't return stat info, the stat struct is
set to invalid
This ensures that getting/stting a security descriptor does not
contend an oplock. The correct access checks will be still be done in
the kernel on the get/set rather than the open.
A few functions in oplocks_onefs.c need to be accessed from the onefs
vfs module. It would be ideal if oplocks were implemented at the vfs
layer, but since they aren't yet, a new header is added to
source3/include to make these functions available to the onefs vfs
module. oplocks_onefs.o doesn't need to be linked into the onefs vfs
module explicitly, since it is already linked into smbd by default.
This checkin enables setting arbitrary timestamps on files matching
the pattern stored in smb.conf. This was a customer request for a
specific workflow.
Changes include:
1) configuration state machine to avoid tons of string comparisons on
each and every stat.
2) Code to adjust post-stat() times to match time now, or sloptime +
time now.
