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.
This restores behaviour of previous versions. The proper fix would be for the
ioctl() to work on O_PATH handles.
BUG: https://bugzilla.samba.org/show_bug.cgi?id=15004
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): Tue Jun 14 18:27:43 UTC 2022 on sn-devel-184
Make it explicit. When we add POSIX handles to SMB2 we will only
look at the handle type. lp_posix_cifsu_locktype() already does this,
but hidden inside init_strict_lock_struct() makes it hard to see.
No logic change.
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Noel Power <npower@samba.org>
We missed these values which follow from MS-FSCC 2.3.80 “FSCTL_OFFLOAD_READ
Reply”:
Flags (4 bytes):
A 32-bit unsigned integer that indicates which flags were returned for this
operation. Possible values for the flags follow. All unused bits are reserved
for future use, SHOULD be set to 0, and MUST be ignored.
OFFLOAD_READ_FLAG_ALL_ZERO_BEYOND_CURRENT_RANGE (0x00000001)
=> The data beyond the current range is logically equivalent to zero.
TransferLength (8 bytes):
A 64-bit unsigned integer that contains the amount, in bytes, of data that the
Token logically represents. This value indicates a contiguous region of the
file from the beginning of the requested offset in the FileOffset field in the
FSCTL_OFFLOAD_READ_INPUT data element (section 2.3.79). This value can be
smaller than the CopyLength field specified in the FSCTL_OFFLOAD_READ_INPUT
data element, which indicates that less data was logically
represented (logically read) with the Token than was requested. The value of
this field MUST be greater than 0x0000000000000000 and MUST be aligned to a
logical sector boundary on the volume.
As we currently only implement COPY_CHUNK over the OFFLOAD VFS interface, the
VFS COPY_CHUNK backend in vfs_default just sets both values to 0 and they are
unused in the SMB frontend.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
BUG: https://bugzilla.samba.org/show_bug.cgi?id=14790
RB: vfs_btrfs compression support broken
Reported-by: noel.kuntze@thermi.consulting
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): Fri Sep 10 18:16:18 UTC 2021 on sn-devel-184
Now that handle based fdos_mode() is used everywhere we can be sure that we're
also always getting a handle in SMB_VFS_GET_COMPRESSION() so we can now safely
remove the path parameter. :)
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
In all places where we access or modify a file or it's associated metadata, we
use fsp_get_io_fd() to fetch the low-level fd from the fsp. This ensures we
don't accidentally use a pathref fsp where the fd would be opened as root on
systems lacking O_PATH.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We should make the behavior change (that gives up some protection)
more obvious, by changing the function names.
At least some OEMs have patches relying on the 4.9/4.10 behaviour
and we want them to detect that they have to do more work when they
need to change directories.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
We can use become_user_by_fsp()/unbecome_user() as it spans only parts of
the btrfs_offload_write_send() function and never goes async in between.
This may matter if at least one share uses "force user".
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
This can be a local variable as used in most of our tevent_req based
code.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
This matches our naming conventions used for tevent_req based functions.
Signed-off-by: Stefan Metzmacher <metze@samba.org>
Reviewed-by: Ralph Boehme <slow@samba.org>
Signed-off-by: Christof Schmitt <cs@samba.org>
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
Autobuild-User(master): Andrew Bartlett <abartlet@samba.org>
Autobuild-Date(master): Mon Dec 18 13:32:00 CET 2017 on sn-devel-144
As per MS-SMB2 and MS-FSA and our SMB_VFS_STRICT_LOCK implementation,
we're merely testing for locks, not setting any.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=12887
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): Tue Jul 11 03:37:44 CEST 2017 on sn-devel-144
It's just a noop, so let's remove it. SMB_VFS_STRICT_LOCK doesn't set
logs, it just checks for the presence of incompatible locks.
Bug: https://bugzilla.samba.org/show_bug.cgi?id=12887
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
...and instead use the fsctl to infer required behaviour in the VFS
backends.
Note that this removes the check from vfs_default because there we only
handle FSCTL_SRV_COPYCHUNK(_WRITE) and must always perform the lock
checks.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Remove the source fsp argument and instead pass the offload token
generated with SMB_VFS_OFFLOAD_READ_SEND/RECV.
An actual offload fsctl is not implemented yet, neither in the VFS nor
at the SMB ioctl layer, and returns NT_STATUS_NOT_IMPLEMENTED
With these changes we now pass the copy-chunk-across-shares test.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
No change in behaviour, just a rename in preperation of more changes to
SMB_VFS_OFFLOAD_WRITE_SEND. It helps keeping the diff of the actual
changes smaller.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
Add SMB_VFS_OFFLOAD_READ_SEND an SMB_VFS_OFFLOAD_READ_RECV.
This paves the way for supporting server-side copy-chunk with source and
destination file-handles on different shares. It can be used to
implement copy offload fsctl in the future, but for now this will be
used as a mere copy-chunk replacement.
SMB_VFS_OFFLOAD_READ generates a token that associates an fsp with the
token and stores the fsp in a in-memory db.
Initially only a copy-chunk resume key fsctl is supported. In the future
this can be enhanced to support real offload fsctl.
Signed-off-by: Ralph Boehme <slow@samba.org>
Reviewed-by: Stefan Metzmacher <metze@samba.org>
As confirmed by the Microsoft Protocol Open Specifications Team: Windows
Server 2016 (ReFS) ignores locks for FSCTL_DUPLICATE_EXTENTS_TO_FILE...
From: Jeff McCashland
To: David Disseldorp
CC: "cifs-protocol@lists.samba.org" <cifs-protocol@lists.samba.org>
Subject: RE: [116100414754619] FSCTL_DUPLICATE_EXTENTS_TO_FILE appears
to completely bypass file locks
Date: Mon, 12 Dec 2016 20:44:08 +0000
Hi David,
We have made the following spec changes for the next doc release:
In section 2.1.5.9.4 FSCTL_DUPLICATE_EXTENTS_TO_FILE behavior notes have
been added to the following paragraphs.
Before:
§ The object store MUST check for byte range lock conflicts on
Open.Stream using the algorithm described in section 2.1.4.10 with
ByteOffset set to InputBuffer.TargetFileOffset, Length set to
InputBuffer.ByteCount, IsExclusive set to TRUE, LockIntent set to FALSE,
and Open set to Open. If a conflict is detected, the operation MUST be
failed with STATUS_FILE_LOCK_CONFLICT.
§ The object store MUST check for byte range lock conflicts on Source
using the algorithm described in section 2.1.4.10 with ByteOffset set to
InputBuffer.SourceFileOffset, Length set to InputBuffer.ByteCount,
IsExclusive set to FALSE, LockIntent set to FALSE, and Open set to
InputBuffer.FileHandle. If a conflict is detected, the operation MUST be
failed with STATUS_FILE_LOCK_CONFLICT.
After:
§ The object store SHOULD<WBN1> check for byte range lock conflicts on
Open.Stream using the algorithm described in section 2.1.4.10 with
ByteOffset set to InputBuffer.TargetFileOffset, Length set to
InputBuffer.ByteCount, IsExclusive set to TRUE, LockIntent set to FALSE,
and Open set to Open. If a conflict is detected, the operation MUST be
failed with STATUS_FILE_LOCK_CONFLICT.
§ The object store SHOULD<WBN2> check for byte range lock conflicts on
Source using the algorithm described in section 2.1.4.10 with ByteOffset
set to InputBuffer.SourceFileOffset, Length set to
InputBuffer.ByteCount, IsExclusive set to FALSE, LockIntent set to
FALSE, and Open set to InputBuffer.FileHandle. If a conflict is
detected, the operation MUST be failed with STATUS_FILE_LOCK_CONFLICT.
WBN1: The ReFS file system in Windows Server 2016 does not check for
byte range lock conflicts on Open.Stream.
WBN2: The ReFS file system in Windows Server 2016 does not check for
byte range lock conflicts on Source.
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Ralph Böhme <slow@samba.org>
FSCTL_DUPLICATE_EXTENTS_TO_FILE must be handled as a COW clone. Add a
copy-chunk flags parameter to the VFS to handle this.
Signed-off-by: Aurelien Aptel <aaptel@suse.com>
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Ralph Böhme <slow@samba.org>
Not currently used - no logic changes inside.
This will make it possible to pass down a long-lived talloc
context from the loading function for modules to use instead
of having them internally all use talloc_autofree_context()
which is a hidden global.
Updated all known module interface numbers, and added a
WHATSNEW.
Signed-off-by: Jeremy Allison <jra@samba.org>
Signed-off-by: Ralph Böhme <slow@samba.org>
Autobuild-User(master): Jeremy Allison <jra@samba.org>
Autobuild-Date(master): Sat Apr 22 01:17:00 CEST 2017 on sn-devel-144
The `unused` member being set to zero is an uint64_t array. We want the
byte size here.
Signed-off-by: Aurelien Aptel <aaptel@suse.com>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
The "btrfs: manipulate snapshots" smb.conf parameter is disabled by
default, to encourage use of, and pass requests through to, the
vfs_snapper module.
When enabled, issue BTRFS_IOC_SNAP_CREATE_V2 and BTRFS_IOC_SNAP_DESTROY
ioctls accordingly. The ioctls are issued as root, so rely on permission
checks in the calling FSRVP server process.
Base share paths must exist as btrfs subvolumes in order to
be supported for snapshot operations.
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Closing a directory handle file descriptor via close() is undefined,
according to:
http://pubs.opengroup.org/onlinepubs/9699919799/functions/dirfd.html
Signed-off-by: Noel Power <noel.power@suse.com>
Reviewed-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Autobuild-User(master): David Disseldorp <ddiss@samba.org>
Autobuild-Date(master): Tue Nov 4 20:51:02 CET 2014 on sn-devel-104
In changes to come this will be possible for an INTERNAL_OPEN_ONLY.
The protection was already in place for some code paths, this
makes the coverage compete.
Bug 10564 - Lock order violation and file lost
https://bugzilla.samba.org/show_bug.cgi?id=10564
Signed-off-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Volker Lendecke <vl@samba.org>
Never map copy-chunk(len=0) requests to BTRFS_IOC_CLONE_RANGE ioctls. A
BTRFS_IOC_CLONE_RANGE with @src_length=0 results in a clone of all data
from @src_offset->EOF!
BUG: https://bugzilla.samba.org/show_bug.cgi?id=10424
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Plumb into the get_fs_capabilities code path to advertise the
FILE_FILE_COMPRESSION attribute.
Also, document file compression support in the VFS btrfs man page.
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Translate such requests into FS_IOC_GETFLAGS and FS_IOC_SETFLAGS ioctls
respectively. The module makes no distinction between compression types,
only whether or not compression is enabled.
Signed-off-by: David Disseldorp <ddiss@samba.org>
Reviewed-by: Jeremy Allison <jra@samba.org>
Reviewed-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Autobuild-User(master): Rusty Russell <rusty@rustcorp.com.au>
Autobuild-Date(master): Sun Mar 24 06:17:55 CET 2013 on sn-devel-104
uint64_t are not unsigned longs on 32-bit platforms:
[3265/3996] Compiling source3/modules/vfs_btrfs.c
../source3/modules/vfs_btrfs.c: In function ‘btrfs_copy_chunk_send’:
../source3/modules/vfs_btrfs.c:118:3: error: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 3 has type ‘uint64_t’ [-Werror=format]
../source3/modules/vfs_btrfs.c:118:3: error: format ‘%ld’ expects argument of type ‘long int’, but argument 4 has type ‘int64_t’ [-Werror=format]
../source3/modules/vfs_btrfs.c:118:3: error: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 5 has type ‘uint64_t’ [-Werror=format]
../source3/modules/vfs_btrfs.c:118:3: error: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 7 has type ‘uint64_t’ [-Werror=format]
../source3/modules/vfs_btrfs.c: In function ‘btrfs_copy_chunk_recv’:
../source3/modules/vfs_btrfs.c:180:2: error: format ‘%lu’ expects argument of type ‘long unsigned int’, but argument 2 has type ‘off_t’ [-Werror=format]
cc1: some warnings being treated as errors
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Currently it only plumbs itself into the copy_chunk call path,
translating such requests into BTRFS_IOC_CLONE_RANGE calls.
Reviewed-by: Andrew Bartlett <abartlet@samba.org>
