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.
Pair-Programmed-With: Volker Lendecke <vl@samba.org>
Autobuild-User(master): Michael Adam <obnox@samba.org>
Autobuild-Date(master): Fri Jun 8 18:08:09 CEST 2012 on sn-devel-104
Now that we always require a 64 bit off_t, we no longer need SMB_OFF_T.
Andrew Bartlett
Autobuild-User: Andrew Bartlett <abartlet@samba.org>
Autobuild-Date: Fri Apr 6 01:47:43 CEST 2012 on sn-devel-104
Several places want "microseconds from current time", and several were
simply handing "usecs" values which could be over a million.
Using a helper to do this is safer and more readable.
I didn't replace any obviously correct callers (ie. constants).
I also renamed wait_nsec in source3/lib/util_sock.c; it's actually
microseconds not nanoseconds (introduced with this code in Volker's
19b783cc Async wrapper for open_socket_out_send/recv).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This has been a wrapper around server_event_context() for some time
now, and removing this from dummmysmbd.c assists with library
dependencies.
Andrew Bartlett
strict allocation on sparse files. Files opened as POSIX opens are always
sparse.
Autobuild-User: Jeremy Allison <jra@samba.org>
Autobuild-Date: Tue Dec 21 04:12:22 CET 2010 on sn-devel-104
When a write cache is empty and we write one 0, there is no point in actually
writing into the write cache, which would trigger a potentially superfluous
write cache flush. We can rely on posix semantics to give us 0's when
subsequent read calls want to read from file areas not written yet.
Jeremy, please check!
In case of the one-byte allocating writes we don't work work optimally because
we start the write cache at the current offset. This patch tries to avoid this
case.
"Normal" non truncate writes always cause the timestamp to
be set on close. Once a close is done on a handle this can
reset the sticky write time to current time also.
Updated smbtorture4 confirms this.
Jeremy.
SMB_VFS_CHFLAGS isn't actually getting the smb_filename struct for now
since it only operates on the basefile. This is the strategy for all
path-based operations that will never actually operate on a stream.
By clarifying the meaning of path based operations that don't take an
smb_filename struct, modules that implement streams such as vfs_onefs
no longer need to implement SMB_VFS_CHFLAGS to ensure it's only called
on the base_name.
This patch introduces two new temporary helper functions
vfs_stat_smb_fname and vfs_lstat_smb_fname. They basically allowed me
to call the new smb_filename version of stat, while avoiding plumbing
it through callers that are still too inconvenient. As the conversion
moves along, I will be able to remove callers of this, with the goal
being to remove all callers.
There was also a bug in create_synthetic_smb_fname_split (also a
temporary utility function) that caused it to incorrectly handle
filenames with ':'s in them when in posix mode. This is now fixed.
This patch introduces
struct stat_ex {
dev_t st_ex_dev;
ino_t st_ex_ino;
mode_t st_ex_mode;
nlink_t st_ex_nlink;
uid_t st_ex_uid;
gid_t st_ex_gid;
dev_t st_ex_rdev;
off_t st_ex_size;
struct timespec st_ex_atime;
struct timespec st_ex_mtime;
struct timespec st_ex_ctime;
struct timespec st_ex_btime; /* birthtime */
blksize_t st_ex_blksize;
blkcnt_t st_ex_blocks;
};
typedef struct stat_ex SMB_STRUCT_STAT;
It is really large because due to the friendly libc headers playing macro
tricks with fields like st_ino, so I renamed them to st_ex_xxx.
Why this change? To support birthtime, we already have quite a few #ifdef's at
places where it does not really belong. With a stat struct that we control, we
can consolidate the nanosecond timestamps and the birthtime deep in the VFS
stat calls.
At this moment it is triggered by a request to support the birthtime field for
GPFS. GPFS does not extend the system level struct stat, but instead has a
separate call that gets us the additional information beyond posix. Without
being able to do that within the VFS stat calls, that support would have to be
scattered around the main smbd code.
It will very likely break all the onefs modules, but I think the changes will
be reasonably easy to do.
Why?? :-)
Another one of the little micro-optimizations that I just came across: If you
allocate a variable in a sub-block like the "fstring sharename" in
write_file(), gcc even with -O3 will allocate this variable unconditionally on
the stack at the beginning of the routine. So with eliminating this fstring we
cut 256 bytes of stack in a very hot code path writing to a file. It might make
us a bit more cache-friendly.
This would probably not be worth a second look if it involved larger code
changes, but this one was just too simple to let it pass :-)
This replaces release_level2_oplocks_on_change with
contend_level2_oplock_begin/end in order to contend level2 oplocks
throughout an operation rather than just at the begining. This is
necessary for some kernel oplock implementations, and also lays the
groundwork for better correctness in Samba's standard level2 oplock
handling. The next step for non-kernel oplocks is to add additional
state to the share mode lock struct that prevents any new opens from
granting oplocks while a contending operation is in progress.
All operations that contend level 2 oplocks are now correctly spanned
except for aio and synchronous writes. The two write paths both have
non-trivial error paths that need extra care to get right.
RAW-OPLOCK and the rest of 'make test' are still passing with this
change.
- only the first non truncating write causes
the write time update with 2 seconds delay.
It's not enough to check for an existing update event
as it will be NULL after the event was triggered.
- SMBwrite truncates always update the write time
unless the sticky write time is set.
- SMBwrite truncates don't trigger a write time update on close.
metze
(This used to be commit 3d17089b6d)
Ok, here's the fix for the write times breakage
with the new tests in S4 smbtorture.
The key is keeping in the share mode struct
the "old_file_time" as the real write time,
set by all the write and allocation calls,
and the "changed_write_time" as the "sticky"
write time - set by the SET_FILE_TIME calls.
We can set them independently (although I
kept the optimization of not setting the
"old_file_time" is a "changed_write_time"
was already set, as we'll never see it.
This allows us to update the write time
immediately on the SMBwrite truncate case,
SET_END_OF_FILE and SET_ALLOCATION_SIZE calls,
whilst still have the 2 second delay on the
"normal" SMBwrite, SMBwriteX calls.
I think in a subsequent patch I'd like to
change the name of these from "old_file_time"
to "write_time" and "changed_write_time" to
"sticky_write_time" to make this clearer.
I think I also fixed a bug in Metze's original
code in that once a write timestamp had been
set from a "normal" SMBwriteX call the fsp->update_write_time_triggered
variable was set and then never reset - thus
meaning the write timestamp would never get
updated again on subsequent SMBwriteX's.
The new code checks the update_write_time_event
event instead, and doesn't update is there's
an event already scheduled.
Metze especially, please check this over for
your understanding.
Jeremy.
(This used to be commit 6f20585419)
We now never call file_ntimes() directly, every update
is done via smb_set_file_time().
This let samba3 pass the BASE-DELAYWRITE test.
The write time is only updated 2 seconds after the
first write() on any open handle to the current time
(not the time of the first write).
Each handle which had write requests updates the write
time to the current time on close().
If the write time is set explicit via setfileinfo or setpathinfo
the write time is visible directly and a following close
on the same handle doesn't update the write time.
metze
(This used to be commit 2eab212ea2)
