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.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15302
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Ralph Böhme <slow@samba.org>
Autobuild-Date(master): Sat Feb 11 08:48:05 UTC 2023 on atb-devel-224
Note: this is a behaviour change in the non-default case when the user
has disabled "getinfo ask sharemode".
Pair-Programmed-With: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
readlink_talloc() deals exactly the same way with a NULL relname
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Andreas Schneider <asn@samba.org>
Autobuild-User(master): Volker Lendecke <vl@samba.org>
Autobuild-Date(master): Mon Dec 5 16:06:51 UTC 2022 on sn-devel-184
Probably doesn't do much in compiled code, but looks cleaner to me
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This gives a nice speed up...
The following test with 256 commections all looping with open/close
on the same inode (share root) is improved drastically:
smbtorture //127.0.0.1/m -Uroot%test smb2.bench.path-contention-shared \
--option='torture:bench_path=' \
--option="torture:timelimit=60" \
--option="torture:nprocs=256" \
--option="torture:qdepth=1"
From something like this:
open[num/s=11536,avslat=0.011450,minlat=0.000039,maxlat=0.052707]
close[num/s=11534,avslat=0.010878,minlat=0.000022,maxlat=0.052342]
(only this commit with the close part reverted) to:
open[num/s=12722,avslat=0.009548,minlat=0.000051,maxlat=0.054338]
close[num/s=12720,avslat=0.010701,minlat=0.000033,maxlat=0.054372]
(with both patches) to:
open[num/s=37680,avslat=0.003471,minlat=0.000040,maxlat=0.061411]
close[num/s=37678,avslat=0.003440,minlat=0.000022,maxlat=0.051536]
So we are finally perform similar like we did in Samba 4.12,
which resulted in:
open[num/s=36846,avslat=0.003574,minlat=0.000043,maxlat=0.020378]
close[num/s=36844,avslat=0.003552,minlat=0.000026,maxlat=0.020321]
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Remember SMB2 Create is the only was to upgrade a lease.
The strategy is that opening of a file will always result
in storing the total lease bits.
But we're lazy clearing the flags on close.
We'll only clear them by traversing all entries when
we break a NONE or when opening a new handle.
We don't do any decision on SHARE_MODE_LEASE_{HANDLE,WRITE},
maybe we'll do in future, but at least it should be much more
sane for debugging now!
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This shows that the code in open_file_ntcreate() and
open_directory() is basically the same now, which
simplifies things a lot.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
It means we call open_mode_check() now only via handle_share_mode_lease()
and the fact that we never grant any directory leases (yet), means
that delay_for_oplocks() avoids the share_mode_forall_entries() loop.
This is a way into supporting directory leases, but that's not
the point for this commit, the point is that.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We don't support directory leases yet, so it should be
an noop for now.
The point is that we want to call
delay_for_oplock(oplock_request=NO_OPLOCK)
for directories soon.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The important part is the call to get a kernel oplock is deferred
until after set_share_mode(). The goal is to get the code
between get_share_mode_lock() and set_share_mode() free of any
blocking operation.
As we were optimistic to get the oplock that was asked for,
we need to remove_share_oplock() in order to set NO_OPLOCK
also in the share_mode entry.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This means we don't have to call remove_lease_if_stale() if
set_share_mode() fails. It's easier to cleanup the share mode entry.
And it makes the code flow easier to the following changes.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The aim is to call set_file_oplock() after set_share_mode(), so that we
only ask for kernel oplocks after set_share_mode().
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
It means delay_for_oplocks() is no longer asking for kernel oplocks.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We ignore the error from share_mode_forall_leases(), but
we still need to cleanup the share_mode_lock we are holding...
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
process_symlink_open() and check_reduced_name() are no longer used,
non_widelink_open() was the only user of both.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The previous implementation relied on recursion into
non_widelink_open() via process_symlink_open(). The latter used
readlink() to just make sure that the opened file is actually a
symlink.
This implementation now relies on a fstat/fstatat on failure to open a
file, removing a little complexity deciphering error codes
correctly. It also relies on reading the symlink in user space,
turning the recursion into a loop.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Avoid the "is_share_root" boolean: One special case less to take care
of further down and in callers: Sanitize the relative name so that it
can never contain a path separator
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
So far we only call CONNECTPATH on full paths. In the future, we'll
have a call that will not have converted a relative path to absolute
just for efficiency reasons. To give shadow_copy2 the chance to still
find the snapshot directory, pass the dirfsp down to it.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is what a Windows server returns. Tested with a share residing on a FAT
formatted drive, a Windows filesystem that doesn't support streams.
Combinations tested:
file::$DATA
file:stream
file:stream:$DATA
All three fail with NT_STATUS_OBJECT_NAME_INVALID.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15126
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15161
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14215
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 Aug 29 18:20:20 UTC 2022 on sn-devel-184
Commit d71ef1365cdde47aeb3465699181656b0655fa04 caused a regression where the
creation date on streams wasn't updated anymore on the stream fsp.
By adding a simple wrapper vfs_fget_dos_attributes() that takes care of
- passing only the base_fsp to the VFS, so the VFS can be completely agnostic of
all the streams related complexity like fake fds,
- propagating any updated btime from the base_fsp->fsp_name to the
stream_fsp->fsp_name
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15126
MR: https://gitlab.com/samba-team/samba/-/merge_requests/2643
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
get_lease_type() will just call leases_db_get() again for leases,
so only call it for oplocks.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Fri Aug 19 19:39:18 UTC 2022 on sn-devel-184
get_lease_type() will just call leases_db_get() again...
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We should avoid accessing fsp->lease if possible.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
This is really expensive as share_mode_forall_entries() is currently
doing a talloc_memdup() of the whole record...
This is mainly used to avoid regressions, so only
use smbd:validate_oplock_types = yes in make test,
but skip it for production.
This improves the following test:
time smbtorture //127.0.0.1/m -Uroot%test \
smb2.create.bench-path-contention-shared \
--option='torture:bench_path=file.dat' \
--option="torture:timelimit=60" \
--option="torture:nprocs=256" \
--option="torture:qdepth=1"
From:
open[num/s=8852,avslat=0.014999,minlat=0.000042,maxlat=0.054600]
close[num/s=8850,avslat=0.014136,minlat=0.000025,maxlat=0.054537]
to:
open[num/s=11377,avslat=0.012075,minlat=0.000041,maxlat=0.054107]
close[num/s=11375,avslat=0.010594,minlat=0.000023,maxlat=0.053620]
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15125
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We currently only ever pass SMB2_LEASE_READ and both
have the same value of 0x1, so for now it's only cosmetic,
but that will change soon.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15148
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
This is inspired by 0d4cb5a641,
which avoids SMB_VFS_FGET_NT_ACL() for the root user again.
Opens with just FILE_READ_ATTRIBUTES are very common, so it's worth
optimizing for it.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
For streams, access is already checked in create_file_unixpath() by
check_base_file_access().
We already skip the access check in this function when doing an IO open of a
file, see above in open_file(), also skip it for "stat opens".
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15126
MR: https://gitlab.com/samba-team/samba/-/merge_requests/2643
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
One reference to conn->cwd_fsp less, makes "mkdir" look less ugly in
strace.
Signed-off-by: Volker Lendecke <vl@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
process_symlink_open goes with it
Pair-programmed-with: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The Linux prototype for openat2 looks like this:
long openat2(int dirfd, const char *pathname,
struct open_how *how, size_t size);
where "struct open_how" is defined in "linux/openat2.h". It is
designed to be extensible with further flags.
The "size" parameter is required because there is no type checking
between userland and kernelspace, so the way for Linux to find which
version of open_how is being passed in is looking at the size:
"open_how" is expected to only every grow with additional fields,
should a change be necessary in the future.
Samba does not have this problem, we can typecheck the struct and
pointers, we expect all VFS modules to be compiled against the current
vfs.h.
For now this adds no functionality, but it will make further patches
much smaller.
Pair-programmed-with: Stefan Metzmacher <metze@samba.org>
Signed-off-by: Volker Lendecke <vl@samba.org>
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
commit 8e3798dd22 actually came with a
change in behavior..., as SMB_VFS_GET_NT_ACL_AT() (at the time) and
now SMB_VFS_FGET_NT_ACL() is always called even if it's not needed.
E.g. access by root.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Tue Jul 26 00:31:29 UTC 2022 on sn-devel-184
