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.
Avoid an if-statement when just assigning the result of the
if-condition to a boolean.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Guenther Deschner <gd@samba.org>
‘tm’ must be initialized prior to calling strptime().
Signed-off-by: Joseph Sutton <josephsutton@catalyst.net.nz>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
I don't think this is an effective optimization at all anymore. It was
intended to speed up non-wildcard readdirs after we found the correct
entry. Nowadays we do the non-wildcard readdirs by a direct fstatat,
and after we successfully found the entry dptr_ReadDirName()
immediately returns without any further action. So my very strong
guess is that this never really kicked in anymore. Not using this flag
can't be *that* bad, smb2 never used it.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We must always do SMB3+POSIX operations on fsp's opened with a posix create context.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
This bases File-Ids on the inode numbers again. The whole stuff was
added because at that time Apple clients
1. would be upset by inode number reusage and
2. had a client side bug in their fallback implemetentation that
assigns File-Ids on the client side in case the server provides
File-Ids of 0.
After discussion with folks at Apple it should be safe these days to
rely on the Mac to generate its own File-Ids and let Samba return 0
File-Ids.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
For now no change in behaviour as all callers still pass conn->cwd_fsp. This
just prepared fd_openat() to deal with real dirfsp's pass by callers later on
when adding calls to fd_openat(dirfspm ...) in the directory enumeration loop.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
To me this is then easier to figure out what is defined there, and
where it's exactly used.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
get_original_lcomp() does all the name canonicalization required for the mask.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
This reverts commit e593c38f5a8602be497a29e145ccdd4f9cc595e7.
As stated in the original commit message, "When canonicalize_snapshot_path()
starts actually stripping off @GMT tokens, we can remove this again."
canonicalize_snapshot_path() now strips off @GMT tokens from
any SMB1 names, so this is now not needed anymore.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue May 5 20:55:42 UTC 2020 on sn-devel-184
This ensure unix_convert() calls canonicalize_snapshot_path() on the path to
process the @GMT token, so smb_fname->twrp is filled in with the converted time
value of the @GMT token.
When canonicalize_snapshot_path() starts actually stripping off @GMT tokens, we
can remove this again.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Use get_original_lcomp() call. Removes one use of smb_fname->original_lcomp.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
This implements two core changes:
* use NTTIME instead of struct timespec at the database layer
* use struct timespec { .tv_nsec = SAMBA_UTIME_OMIT } as special sentinel
value in smbd when processing timestamps
Using NTTIME at the database layer is only done to avoid storing the special
struct timespec sentinel values on disk. Instead, with NTTIME the sentinel value
for an "unset" timestamp is just 0 on-disk.
The NTTIME value of 0 gets translated by nt_time_to_full_timespec() to the
struct timespec sentinel value { .tv_nsec = SAMBA_UTIME_OMIT }.
The function is_omit_timespec() can be used to check this.
Beside nt_time_to_full_timespec(), there are various other new time conversion
functions with *full* in their name that can be used to safely convert between
different types with the changed sentinel value.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=7771
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The whole logic only applies to files.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Wed Sep 11 00:33:26 UTC 2019 on sn-devel-184
Now that dptr_create() is handle based, we can simply used the fsp name.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is not strictly required, as we ne never trigger additional VFS
requests via this codepath. But for safety reasons ensure we're running
in the correct impersonation state.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Mon Jan 14 22:30:24 CET 2019 on sn-devel-144
This is needed as the callback might be called in an arbitrary
impersonation state.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The additional overhead for the async processing should be avoided when
we know the lower level has to fallback to sync processing anyway.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
The ev argument is also a raw ev.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
This reverts commit 894e5001c747ce765dad5517778dda55d7d1f4d9.
See the discussion in
https://lists.samba.org/archive/samba-technical/2018-December/131731.html
for the reasoning behind this revert.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
This is basically a revert of commit
2be7518ae5a3c046f5fca04ecc83f9f7044eac74.
See the discussion in
https://lists.samba.org/archive/samba-technical/2018-December/131731.html
for the reasoning behind this revert.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
SMB_VFS_GETXATTRAT_SEND() gets passed a raw event context and the
default implementation uses that as well a raw threadpool. Impersonation
is done explicitly instead of by the tevent and pthreadpool wrappers.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Parametric options have a performance impact, use the normal options
added in the previous commit.
"aio max threads" can only be calculated at run time and requires a
handle to a pthreadpool_tevent which loadparm will never have.
Because of that lp_smbd_max_async_dosmode() will always return 0 as
default and it's up to us to calculate "aio max threads * 2" if
lp_smbd_max_async_dosmode() returns 0.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Parametric options have a performance impact, use the normal options
added in the previous commit.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Parametric options have a performance impact, use the normal options
added in the previous commit.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
All existing callers pass NULL, no change in behaviour.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=13455
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
'return req' should do the same as 'return true' for a bool function,
it's implicitly expanded as 'return (req!=NULL)?true:false.
There's no point in that as 'req' is always a valid pointer.
This was most likely just a copy and paste bug.
So we make this explicit now and avoid that Coverity reports this:
CID 1438158: Null pointer dereferences (REVERSE_INULL)
Null-checking "req" suggests that it may be null, but it has already
been dereferenced on all paths leading to the check.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Tue Jul 31 14:20:49 CEST 2018 on sn-devel-144
Finally: use the new dos_mode_at_send() in the directory enumeration
loop. This means that fetching the DOS attributes for directory entries
is done asynchronously with regard to the enumeration loop.
As the DOS attribute is typically read from an extended attribute in the
filesytem, this avoids sequentially blocking on IO. If the IO subsystem
is slow servicing these request, enabling async processing can result in
performance improvements.
A parametric option
smbd:async dosmode = true | false (default: false)
can be used to enable the new async processing.
Simulating slow IO with usleep(5000) in the synchronous and asynchronous
versions of SMB_VFS_GET_DOS_ATTRIBUTES(), the results of enumerating a
directory with 10,000 files are:
smbd:async dosmode = no:
$ time bin/smbclient -U slow%x //localhost/test -c "ls dir\*" > /dev/null
real 0m59.597s
user 0m0.024s
sys 0m0.012s
smbd:async dosmode = yes:
$ time bin/smbclient -U slow%x //localhost/test -c "ls dir\*" > /dev/null
real 0m0.698s
user 0m0.038s
sys 0m0.025s
Performance gains in real world workloads depends on whether the actual
IO requests can be merged and parallelized by the kernel. Without such
wins at the IO layer, the async processing may even be slower then the
sync processing due to the additional overhead.
The following parameters can be used to adapt async processing behaviour
for specific workloads and systems:
aio max threads = X (default: 100)
smbd:max async dosmode = Y (default: "aio max threads" * 2)
By default we have at most twice the number of async requests in flight
as threads provided by the underlying threadpool. This ensures a worker
thread that finishes a job can directly pick up a new one without going
to sleep.
It may be advisable to reduce the number of threads to avoid scheduling
overhead while also increasing "smbd:max async dosmode".
Note that we disable async processing for certain VFS modules in the VFS
connect function to avoid the overhead of triggering the sync fallback
in dos_mode_at_send(). This is done for VFS modules that implement the
sync SMB_VFS_GET_DOS_ATTRIBUTES(), but not the async version (gpfs), and
for VFS modules that don't share a real filesystem where fchdir() can be
used (ceph, gluster). It is disabled for catia, because we realized that
the catia name translation macros used on
fsps (CATIA_FETCH_FSP_[PRE|POST]_NEXT) have a bug (#13547).
We use threadpool = smb_vfs_ev_glue_tp_chdir_safe() and then
pthreadpool_tevent_max_threads(threadpool) to get the number of maximum
worker threads which matches the pool used by the low level
SMB_VFS_GETXATTRAT_[SEND|RECV] implementation in vfs_default.
This is a terrible abstraction leak that should be removed in the future
by maybe making it possible to ask a VFS function which threadpool it
uses, internally suporting chaining so VFS function FOO that internally
uses BAR can forward the question to BAR.
On a hyphotetical system that had a getxattrat(dirfd, path, ...)
syscall and at the same time doesn't support per-thread current working
directories (eg FreeBSD doesn't have the latter) but has support for
per-thread-credentials, pthreadpool_tevent_max_threads() on the
tp_chdir_safe threadpool returns 1.
So when hooking the hyphotetical getxattrat() into the async
SMB_VFS_GETXATTRAT_[SEND|RECV] implementation in an VFS module, the
implementation could use the tp_path_safe threadpool, but the SMB2
layer would use the wrong threadpool in the call to
pthreadpool_tevent_max_threads(), resulting in no parallelism.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
