UbiquitousOS
/
OpenCloudOS-Kernel
mirror of https://gitee.com/OpenCloudOS/OpenCloudOS-Kernel.git
11
0
Fork 0
Code Issues Projects Releases Wiki Activity
OpenCloudOS-Kernel/fs/cifs/smb2inode.c

844 lines
23 KiB
C
Raw Normal View History

cifs: use SPDX-Licence-Identifier Add SPDX license identifier and replace license boilerplate. Corrects various checkpatch errors with the older format for noting the LGPL license. Signed-off-by: Steve French <stfrench@microsoft.com>
2021-06-18 13:31:49 +08:00
// SPDX-License-Identifier: LGPL-2.1
CIFS: Query SMB2 inode info Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com>
2011-12-29 21:06:33 +08:00
/*
*
* Copyright (C) International Business Machines Corp., 2002, 2011
* Etersoft, 2012
* Author(s): Pavel Shilovsky (pshilovsky@samba.org),
* Steve French (sfrench@us.ibm.com)
*
*/
#include <linux/fs.h>
#include <linux/stat.h>
#include <linux/slab.h>
#include <linux/pagemap.h>
#include <asm/div64.h>
#include "cifsfs.h"
#include "cifspdu.h"
#include "cifsglob.h"
#include "cifsproto.h"
#include "cifs_debug.h"
#include "cifs_fs_sb.h"
#include "cifs_unicode.h"
#include "fscache.h"
#include "smb2glob.h"
#include "smb2pdu.h"
#include "smb2proto.h"
cifs: Move cached-dir functions into a separate file Also rename crfid to cfid to have consistent naming for this variable. This commit does not change any logic. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-08-11 11:00:08 +08:00
#include "cached_dir.h"
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
#include "smb2status.h"
CIFS: Query SMB2 inode info Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com>
2011-12-29 21:06:33 +08:00
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
static void
free_set_inf_compound(struct smb_rqst *rqst)
{
if (rqst[1].rq_iov)
SMB2_set_info_free(&rqst[1]);
if (rqst[2].rq_iov)
SMB2_close_free(&rqst[2]);
}
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
struct cop_vars {
struct cifs_open_parms oparms;
struct kvec rsp_iov[3];
struct smb_rqst rqst[3];
struct kvec open_iov[SMB2_CREATE_IOV_SIZE];
struct kvec qi_iov[1];
struct kvec si_iov[SMB2_SET_INFO_IOV_SIZE];
struct kvec close_iov[1];
struct smb2_file_rename_info rename_info;
struct smb2_file_link_info link_info;
};
cifs: for compound requests, use open handle if possible For smb2_compound_op, it is possible to pass a ref to an already open file. We should be passing it whenever possible. i.e. if a matching handle is already kept open. If we don't do that, we will end up breaking leases for files kept open on the same client. Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-07-30 14:43:09 +08:00
/*
* note: If cfile is passed, the reference to it is dropped here.
* So make sure that you do not reuse cfile after return from this func.
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
*
* If passing @err_iov and @err_buftype, ensure to make them both large enough (>= 3) to hold all
* error responses. Caller is also responsible for freeing them up.
cifs: for compound requests, use open handle if possible For smb2_compound_op, it is possible to pass a ref to an already open file. We should be passing it whenever possible. i.e. if a matching handle is already kept open. If we don't do that, we will end up breaking leases for files kept open on the same client. Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-07-30 14:43:09 +08:00
*/
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
static int smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon,
struct cifs_sb_info *cifs_sb, const char *full_path,
__u32 desired_access, __u32 create_disposition, __u32 create_options,
umode_t mode, void *ptr, int command, struct cifsFileInfo *cfile,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
__u8 **extbuf, size_t *extbuflen,
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
struct kvec *err_iov, int *err_buftype)
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
{
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
struct cop_vars *vars = NULL;
struct kvec *rsp_iov;
struct smb_rqst *rqst;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
int rc;
__le16 *utf16_path = NULL;
__u8 oplock = SMB2_OPLOCK_LEVEL_NONE;
struct cifs_fid fid;
struct cifs_ses *ses = tcon->ses;
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
struct TCP_Server_Info *server;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
int num_rqst = 0;
int resp_buftype[3];
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
struct smb2_query_info_rsp *qi_rsp = NULL;
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
struct cifs_open_info_data *idata;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
int flags = 0;
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
__u8 delete_pending[8] = {1, 0, 0, 0, 0, 0, 0, 0};
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
unsigned int size[2];
void *data[2];
int len;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
vars = kzalloc(sizeof(*vars), GFP_ATOMIC);
if (vars == NULL)
return -ENOMEM;
rqst = &vars->rqst[0];
rsp_iov = &vars->rsp_iov[0];
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
server = cifs_pick_channel(ses);
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
if (smb3_encryption_required(tcon))
flags |= CIFS_TRANSFORM_REQ;
resp_buftype[0] = resp_buftype[1] = resp_buftype[2] = CIFS_NO_BUFFER;
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
/* We already have a handle so we can skip the open */
if (cfile)
goto after_open;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
/* Open */
utf16_path = cifs_convert_path_to_utf16(full_path, cifs_sb);
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
if (!utf16_path) {
rc = -ENOMEM;
goto finished;
}
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
cifs: Fix uninitialized memory reads for oparms.mode Use a struct assignment with implicit member initialization Signed-off-by: Volker Lendecke <vl@samba.org> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
2023-01-11 19:37:58 +08:00
vars->oparms = (struct cifs_open_parms) {
.tcon = tcon,
.desired_access = desired_access,
.disposition = create_disposition,
.create_options = cifs_create_options(cifs_sb, create_options),
.fid = &fid,
.mode = mode,
.cifs_sb = cifs_sb,
};
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
rqst[num_rqst].rq_iov = &vars->open_iov[0];
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
rqst[num_rqst].rq_nvec = SMB2_CREATE_IOV_SIZE;
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_open_init(tcon, server,
&rqst[num_rqst], &oplock, &vars->oparms,
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
utf16_path);
kfree(utf16_path);
if (rc)
goto finished;
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
smb2_set_next_command(tcon, &rqst[num_rqst]);
after_open:
num_rqst++;
rc = 0;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
/* Operation */
switch (command) {
case SMB2_OP_QUERY_INFO:
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
rqst[num_rqst].rq_iov = &vars->qi_iov[0];
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
rqst[num_rqst].rq_nvec = 1;
cifs: add a helper to find an existing readable handle to a file and convert smb2_query_path_info() to use it. This will eliminate the need for a SMB2_Create when we already have an open handle that can be used. This will also prevent a oplock break in case the other handle holds a lease. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-09-09 13:30:00 +08:00
if (cfile)
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_query_info_init(tcon, server,
&rqst[num_rqst],
cifs: add a helper to find an existing readable handle to a file and convert smb2_query_path_info() to use it. This will eliminate the need for a SMB2_Create when we already have an open handle that can be used. This will also prevent a oplock break in case the other handle holds a lease. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-09-09 13:30:00 +08:00
cfile->fid.persistent_fid,
cfile->fid.volatile_fid,
FILE_ALL_INFORMATION,
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
SMB2_O_INFO_FILE, 0,
sizeof(struct smb2_file_all_info) +
cifs: add IOCTL for QUERY_INFO passthrough to userspace This allows userspace tools to query the raw info levels for cifs files and process the response in userspace. In particular this is useful for many of those data where there is no corresponding native data structure in linux. For example querying the security descriptor for a file and extract the SIDs. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-10-08 08:19:58 +08:00
PATH_MAX * 2, 0, NULL);
cifs: add a helper to find an existing readable handle to a file and convert smb2_query_path_info() to use it. This will eliminate the need for a SMB2_Create when we already have an open handle that can be used. This will also prevent a oplock break in case the other handle holds a lease. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-09-09 13:30:00 +08:00
else {
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_query_info_init(tcon, server,
&rqst[num_rqst],
cifs: add a helper to find an existing readable handle to a file and convert smb2_query_path_info() to use it. This will eliminate the need for a SMB2_Create when we already have an open handle that can be used. This will also prevent a oplock break in case the other handle holds a lease. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-09-09 13:30:00 +08:00
COMPOUND_FID,
COMPOUND_FID,
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
FILE_ALL_INFORMATION,
cifs: add a helper to find an existing readable handle to a file and convert smb2_query_path_info() to use it. This will eliminate the need for a SMB2_Create when we already have an open handle that can be used. This will also prevent a oplock break in case the other handle holds a lease. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-09-09 13:30:00 +08:00
SMB2_O_INFO_FILE, 0,
sizeof(struct smb2_file_all_info) +
PATH_MAX * 2, 0, NULL);
if (!rc) {
smb2_set_next_command(tcon, &rqst[num_rqst]);
smb2_set_related(&rqst[num_rqst]);
}
}
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
if (rc)
goto finished;
num_rqst++;
trace_smb3_query_info_compound_enter(xid, ses->Suid, tcon->tid,
full_path);
break;
case SMB2_OP_POSIX_QUERY_INFO:
rqst[num_rqst].rq_iov = &vars->qi_iov[0];
rqst[num_rqst].rq_nvec = 1;
if (cfile)
rc = SMB2_query_info_init(tcon, server,
&rqst[num_rqst],
cfile->fid.persistent_fid,
cfile->fid.volatile_fid,
SMB_FIND_FILE_POSIX_INFO,
SMB2_O_INFO_FILE, 0,
/* TBD: fix following to allow for longer SIDs */
sizeof(struct smb311_posix_qinfo *) + (PATH_MAX * 2) +
(sizeof(struct cifs_sid) * 2), 0, NULL);
else {
rc = SMB2_query_info_init(tcon, server,
&rqst[num_rqst],
COMPOUND_FID,
COMPOUND_FID,
SMB_FIND_FILE_POSIX_INFO,
SMB2_O_INFO_FILE, 0,
sizeof(struct smb311_posix_qinfo *) + (PATH_MAX * 2) +
(sizeof(struct cifs_sid) * 2), 0, NULL);
if (!rc) {
smb2_set_next_command(tcon, &rqst[num_rqst]);
smb2_set_related(&rqst[num_rqst]);
}
}
cifs: fix crash in smb2_compound_op()/smb2_set_next_command() RHBZ: 1722704 In low memory situations the various SMB2_*_init() functions can fail to allocate a request PDU and thus leave the request iovector as NULL. If we don't check the return code for failure we end up calling smb2_set_next_command() with a NULL iovector causing a crash when it tries to dereference it. CC: Stable <stable@vger.kernel.org> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-07-16 08:41:46 +08:00
if (rc)
goto finished;
cifs: add a helper to find an existing readable handle to a file and convert smb2_query_path_info() to use it. This will eliminate the need for a SMB2_Create when we already have an open handle that can be used. This will also prevent a oplock break in case the other handle holds a lease. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-09-09 13:30:00 +08:00
num_rqst++;
smb311: Add tracepoints for new compound posix query info Add dynamic tracepoints for new SMB3.1.1. posix extensions query info level (100) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 12:02:33 +08:00
trace_smb3_posix_query_info_compound_enter(xid, ses->Suid, tcon->tid, full_path);
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
break;
cifs: change unlink to use a compound This,and previous patches, drops the number of roundtrips from five to two for unlink() Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:43 +08:00
case SMB2_OP_DELETE:
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
trace_smb3_delete_enter(xid, ses->Suid, tcon->tid, full_path);
cifs: change unlink to use a compound This,and previous patches, drops the number of roundtrips from five to two for unlink() Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:43 +08:00
break;
cifs: change mkdir to use a compound This with the previous patch changes mkdir() from needing 6 roundtrips to just 3. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:42 +08:00
case SMB2_OP_MKDIR:
/*
* Directories are created through parameters in the
* SMB2_open() call.
*/
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
trace_smb3_mkdir_enter(xid, ses->Suid, tcon->tid, full_path);
cifs: change mkdir to use a compound This with the previous patch changes mkdir() from needing 6 roundtrips to just 3. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:42 +08:00
break;
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
case SMB2_OP_RMDIR:
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
rqst[num_rqst].rq_iov = &vars->si_iov[0];
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
rqst[num_rqst].rq_nvec = 1;
smb3: Fix rmdir compounding regression to strict servers Some servers require that the setinfo matches the exact size, and in this case compounding changes introduced by commit c2e0fe3f5aae ("cifs: make rmdir() use compounding") caused us to send 8 bytes (padded length) instead of 1 byte (the size of the structure). See MS-FSCC section 2.4.11. Fixing this when we send a SET_INFO command for delete file disposition, then ends up as an iov of a single byte but this causes problems with SMB3 and encryption. To avoid this, instead of creating a one byte iov for the disposition value and then appending an additional iov with a 7 byte padding we now handle this as a single 8 byte iov containing both the disposition byte as well as the padding in one single buffer. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com> Acked-by: Paulo Alcantara <palcantara@suse.de>
2018-12-19 07:49:05 +08:00
size[0] = 1; /* sizeof __u8 See MS-FSCC section 2.4.11 */
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
data[0] = &delete_pending[0];
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_set_info_init(tcon, server,
&rqst[num_rqst], COMPOUND_FID,
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
COMPOUND_FID, current->tgid,
FILE_DISPOSITION_INFORMATION,
SMB2_O_INFO_FILE, 0, data, size);
cifs: fix crash in smb2_compound_op()/smb2_set_next_command() RHBZ: 1722704 In low memory situations the various SMB2_*_init() functions can fail to allocate a request PDU and thus leave the request iovector as NULL. If we don't check the return code for failure we end up calling smb2_set_next_command() with a NULL iovector causing a crash when it tries to dereference it. CC: Stable <stable@vger.kernel.org> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-07-16 08:41:46 +08:00
if (rc)
goto finished;
cifs: we can not use small padding iovs together with encryption We can not append small padding buffers as separate iovs when encryption is used. For this case we must flatten the request into a single buffer containing both the data from all the iovs as well as the padding bytes. This is at least needed for 4.20 as well due to compounding changes. CC: Stable <stable@vger.kernel.org> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-12-31 11:43:40 +08:00
smb2_set_next_command(tcon, &rqst[num_rqst]);
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
smb2_set_related(&rqst[num_rqst++]);
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
trace_smb3_rmdir_enter(xid, ses->Suid, tcon->tid, full_path);
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
break;
cifs: change SMB2_OP_SET_EOF to use compounding This changes SMB2_OP_SET_EOF to use compounding in some situations. This is part of the path based API to truncate a file. Most of the time this will however not be invoked for SMB2 since cifs_set_file_size() will as far as I can tell almost always just open the file synchronously and switch to the handle based truncate code path, thus bypassing the compounding we add here. Rewriting cifs_set_file_size() and make that whole pile of code more compounding friendly, and also easier to read and understand, is a different project though and not for this patch. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:46 +08:00
case SMB2_OP_SET_EOF:
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
rqst[num_rqst].rq_iov = &vars->si_iov[0];
cifs: change SMB2_OP_SET_EOF to use compounding This changes SMB2_OP_SET_EOF to use compounding in some situations. This is part of the path based API to truncate a file. Most of the time this will however not be invoked for SMB2 since cifs_set_file_size() will as far as I can tell almost always just open the file synchronously and switch to the handle based truncate code path, thus bypassing the compounding we add here. Rewriting cifs_set_file_size() and make that whole pile of code more compounding friendly, and also easier to read and understand, is a different project though and not for this patch. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:46 +08:00
rqst[num_rqst].rq_nvec = 1;
size[0] = 8; /* sizeof __le64 */
data[0] = ptr;
cifs: Fix smb2_set_path_size() If cifs_get_writable_path() finds a writable file, smb2_compound_op() must use that file's FID and not the COMPOUND_FID. Cc: stable@vger.kernel.org Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2023-03-13 23:09:54 +08:00
if (cfile) {
rc = SMB2_set_info_init(tcon, server,
&rqst[num_rqst],
cfile->fid.persistent_fid,
cfile->fid.volatile_fid,
current->tgid,
FILE_END_OF_FILE_INFORMATION,
SMB2_O_INFO_FILE, 0,
data, size);
} else {
rc = SMB2_set_info_init(tcon, server,
&rqst[num_rqst],
COMPOUND_FID,
COMPOUND_FID,
current->tgid,
FILE_END_OF_FILE_INFORMATION,
SMB2_O_INFO_FILE, 0,
data, size);
if (!rc) {
smb2_set_next_command(tcon, &rqst[num_rqst]);
smb2_set_related(&rqst[num_rqst]);
}
}
cifs: fix crash in smb2_compound_op()/smb2_set_next_command() RHBZ: 1722704 In low memory situations the various SMB2_*_init() functions can fail to allocate a request PDU and thus leave the request iovector as NULL. If we don't check the return code for failure we end up calling smb2_set_next_command() with a NULL iovector causing a crash when it tries to dereference it. CC: Stable <stable@vger.kernel.org> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-07-16 08:41:46 +08:00
if (rc)
goto finished;
cifs: Fix smb2_set_path_size() If cifs_get_writable_path() finds a writable file, smb2_compound_op() must use that file's FID and not the COMPOUND_FID. Cc: stable@vger.kernel.org Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2023-03-13 23:09:54 +08:00
num_rqst++;
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
trace_smb3_set_eof_enter(xid, ses->Suid, tcon->tid, full_path);
cifs: change SMB2_OP_SET_EOF to use compounding This changes SMB2_OP_SET_EOF to use compounding in some situations. This is part of the path based API to truncate a file. Most of the time this will however not be invoked for SMB2 since cifs_set_file_size() will as far as I can tell almost always just open the file synchronously and switch to the handle based truncate code path, thus bypassing the compounding we add here. Rewriting cifs_set_file_size() and make that whole pile of code more compounding friendly, and also easier to read and understand, is a different project though and not for this patch. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:46 +08:00
break;
cifs: change SMB2_OP_SET_INFO to use compounding Cuts number of network roundtrips significantly for some common syscalls Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:48 +08:00
case SMB2_OP_SET_INFO:
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
rqst[num_rqst].rq_iov = &vars->si_iov[0];
cifs: change SMB2_OP_SET_INFO to use compounding Cuts number of network roundtrips significantly for some common syscalls Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:48 +08:00
rqst[num_rqst].rq_nvec = 1;
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
cifs: change SMB2_OP_SET_INFO to use compounding Cuts number of network roundtrips significantly for some common syscalls Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:48 +08:00
size[0] = sizeof(FILE_BASIC_INFO);
data[0] = ptr;
cifs: use existing handle for compound_op(OP_SET_INFO) when possible If we already have a writable handle for a path we want to set the attributes for then use that instead of a create/set-info/close compound. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 07:53:56 +08:00
if (cfile)
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_set_info_init(tcon, server,
&rqst[num_rqst],
cifs: use existing handle for compound_op(OP_SET_INFO) when possible If we already have a writable handle for a path we want to set the attributes for then use that instead of a create/set-info/close compound. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 07:53:56 +08:00
cfile->fid.persistent_fid,
cfile->fid.volatile_fid, current->tgid,
FILE_BASIC_INFORMATION,
SMB2_O_INFO_FILE, 0, data, size);
else {
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_set_info_init(tcon, server,
&rqst[num_rqst],
cifs: use existing handle for compound_op(OP_SET_INFO) when possible If we already have a writable handle for a path we want to set the attributes for then use that instead of a create/set-info/close compound. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 07:53:56 +08:00
COMPOUND_FID,
COMPOUND_FID, current->tgid,
FILE_BASIC_INFORMATION,
SMB2_O_INFO_FILE, 0, data, size);
if (!rc) {
smb2_set_next_command(tcon, &rqst[num_rqst]);
smb2_set_related(&rqst[num_rqst]);
}
}
cifs: fix crash in smb2_compound_op()/smb2_set_next_command() RHBZ: 1722704 In low memory situations the various SMB2_*_init() functions can fail to allocate a request PDU and thus leave the request iovector as NULL. If we don't check the return code for failure we end up calling smb2_set_next_command() with a NULL iovector causing a crash when it tries to dereference it. CC: Stable <stable@vger.kernel.org> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-07-16 08:41:46 +08:00
if (rc)
goto finished;
cifs: use existing handle for compound_op(OP_SET_INFO) when possible If we already have a writable handle for a path we want to set the attributes for then use that instead of a create/set-info/close compound. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 07:53:56 +08:00
num_rqst++;
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
trace_smb3_set_info_compound_enter(xid, ses->Suid, tcon->tid,
full_path);
cifs: change SMB2_OP_SET_INFO to use compounding Cuts number of network roundtrips significantly for some common syscalls Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:48 +08:00
break;
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
case SMB2_OP_RENAME:
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
rqst[num_rqst].rq_iov = &vars->si_iov[0];
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
rqst[num_rqst].rq_nvec = 2;
len = (2 * UniStrnlen((wchar_t *)ptr, PATH_MAX));
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
vars->rename_info.ReplaceIfExists = 1;
vars->rename_info.RootDirectory = 0;
vars->rename_info.FileNameLength = cpu_to_le32(len);
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
size[0] = sizeof(struct smb2_file_rename_info);
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
data[0] = &vars->rename_info;
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
size[1] = len + 2 /* null */;
data[1] = (__le16 *)ptr;
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
if (cfile)
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_set_info_init(tcon, server,
&rqst[num_rqst],
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
cfile->fid.persistent_fid,
cfile->fid.volatile_fid,
current->tgid, FILE_RENAME_INFORMATION,
SMB2_O_INFO_FILE, 0, data, size);
else {
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_set_info_init(tcon, server,
&rqst[num_rqst],
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
COMPOUND_FID, COMPOUND_FID,
current->tgid, FILE_RENAME_INFORMATION,
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
SMB2_O_INFO_FILE, 0, data, size);
cifs: use existing handle for compound_op(OP_SET_INFO) when possible If we already have a writable handle for a path we want to set the attributes for then use that instead of a create/set-info/close compound. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 07:53:56 +08:00
if (!rc) {
smb2_set_next_command(tcon, &rqst[num_rqst]);
smb2_set_related(&rqst[num_rqst]);
}
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
}
cifs: fix crash in smb2_compound_op()/smb2_set_next_command() RHBZ: 1722704 In low memory situations the various SMB2_*_init() functions can fail to allocate a request PDU and thus leave the request iovector as NULL. If we don't check the return code for failure we end up calling smb2_set_next_command() with a NULL iovector causing a crash when it tries to dereference it. CC: Stable <stable@vger.kernel.org> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-07-16 08:41:46 +08:00
if (rc)
goto finished;
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
num_rqst++;
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
trace_smb3_rename_enter(xid, ses->Suid, tcon->tid, full_path);
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
break;
case SMB2_OP_HARDLINK:
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
rqst[num_rqst].rq_iov = &vars->si_iov[0];
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
rqst[num_rqst].rq_nvec = 2;
len = (2 * UniStrnlen((wchar_t *)ptr, PATH_MAX));
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
vars->link_info.ReplaceIfExists = 0;
vars->link_info.RootDirectory = 0;
vars->link_info.FileNameLength = cpu_to_le32(len);
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
size[0] = sizeof(struct smb2_file_link_info);
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
data[0] = &vars->link_info;
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
size[1] = len + 2 /* null */;
data[1] = (__le16 *)ptr;
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_set_info_init(tcon, server,
&rqst[num_rqst], COMPOUND_FID,
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
COMPOUND_FID, current->tgid,
FILE_LINK_INFORMATION,
SMB2_O_INFO_FILE, 0, data, size);
cifs: fix crash in smb2_compound_op()/smb2_set_next_command() RHBZ: 1722704 In low memory situations the various SMB2_*_init() functions can fail to allocate a request PDU and thus leave the request iovector as NULL. If we don't check the return code for failure we end up calling smb2_set_next_command() with a NULL iovector causing a crash when it tries to dereference it. CC: Stable <stable@vger.kernel.org> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-07-16 08:41:46 +08:00
if (rc)
goto finished;
cifs: we can not use small padding iovs together with encryption We can not append small padding buffers as separate iovs when encryption is used. For this case we must flatten the request into a single buffer containing both the data from all the iovs as well as the padding bytes. This is at least needed for 4.20 as well due to compounding changes. CC: Stable <stable@vger.kernel.org> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-12-31 11:43:40 +08:00
smb2_set_next_command(tcon, &rqst[num_rqst]);
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
smb2_set_related(&rqst[num_rqst++]);
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
trace_smb3_hardlink_enter(xid, ses->Suid, tcon->tid, full_path);
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
break;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
default:
cifs_dbg(VFS, "Invalid command\n");
rc = -EINVAL;
}
if (rc)
goto finished;
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
/* We already have a handle so we can skip the close */
if (cfile)
goto after_close;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
/* Close */
cifs: do not send close in compound create+close requests In case of interrupted syscalls, prevent sending CLOSE commands for compound CREATE+CLOSE requests by introducing an CIFS_CP_CREATE_CLOSE_OP flag to indicate lower layers that it should not send a CLOSE command to the MIDs corresponding the compound CREATE+CLOSE request. A simple reproducer: #!/bin/bash mount //server/share /mnt -o username=foo,password=*** tc qdisc add dev eth0 root netem delay 450ms stat -f /mnt &>/dev/null & pid=$! sleep 0.01 kill $pid tc qdisc del dev eth0 root umount /mnt Before patch: ... 6 0.256893470 192.168.122.2 → 192.168.122.15 SMB2 402 Create Request File: ;GetInfo Request FS_INFO/FileFsFullSizeInformation;Close Request 7 0.257144491 192.168.122.15 → 192.168.122.2 SMB2 498 Create Response File: ;GetInfo Response;Close Response 9 0.260798209 192.168.122.2 → 192.168.122.15 SMB2 146 Close Request File: 10 0.260841089 192.168.122.15 → 192.168.122.2 SMB2 130 Close Response, Error: STATUS_FILE_CLOSED Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com> CC: <stable@vger.kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-03-08 23:00:50 +08:00
flags |= CIFS_CP_CREATE_CLOSE_OP;
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
rqst[num_rqst].rq_iov = &vars->close_iov[0];
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
rqst[num_rqst].rq_nvec = 1;
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = SMB2_close_init(tcon, server,
&rqst[num_rqst], COMPOUND_FID,
smb3: query attributes on file close Since timestamps on files on most servers can be updated at close, and since timestamps on our dentries default to one second we can have stale timestamps in some common cases (e.g. open, write, close, stat, wait one second, stat - will show different mtime for the first and second stat). The SMB2/SMB3 protocol allows querying timestamps at close so add the code to request timestamp and attr information (which is cheap for the server to provide) to be returned when a file is closed (it is not needed for the many paths that call SMB2_close that are from compounded query infos and close nor is it needed for some of the cases where a directory close immediately follows a directory open. Signed-off-by: Steve French <stfrench@microsoft.com> Acked-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com> Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com>
2019-12-03 11:46:54 +08:00
COMPOUND_FID, false);
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
smb2_set_related(&rqst[num_rqst]);
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
if (rc)
goto finished;
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
after_close:
num_rqst++;
if (cfile) {
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = compound_send_recv(xid, ses, server,
flags, num_rqst - 2,
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
&rqst[1], &resp_buftype[1],
&rsp_iov[1]);
} else
cifs: multichannel: move channel selection above transport layer Move the channel (TCP_Server_Info*) selection from the tranport layer to higher in the call stack so that: - credit handling is done with the server that will actually be used to send. * ->wait_mtu_credit * ->set_credits / set_credits * ->add_credits / add_credits * add_credits_and_wake_if - potential reconnection (smb2_reconnect) done when initializing a request is checked and done with the server that will actually be used to send. To do this: - remove the cifs_pick_channel() call out of compound_send_recv() - select channel and pass it down by adding a cifs_pick_channel(ses) call in: - smb311_posix_mkdir - SMB2_open - SMB2_ioctl - __SMB2_close - query_info - SMB2_change_notify - SMB2_flush - smb2_async_readv (if none provided in context param) - SMB2_read (if none provided in context param) - smb2_async_writev (if none provided in context param) - SMB2_write (if none provided in context param) - SMB2_query_directory - send_set_info - SMB2_oplock_break - SMB311_posix_qfs_info - SMB2_QFS_info - SMB2_QFS_attr - smb2_lockv - SMB2_lease_break - smb2_compound_op - smb2_set_ea - smb2_ioctl_query_info - smb2_query_dir_first - smb2_query_info_comound - smb2_query_symlink - cifs_writepages - cifs_write_from_iter - cifs_send_async_read - cifs_read - cifs_readpages - add TCP_Server_Info *server param argument to: - cifs_send_recv - compound_send_recv - SMB2_open_init - SMB2_query_info_init - SMB2_set_info_init - SMB2_close_init - SMB2_ioctl_init - smb2_iotcl_req_init - SMB2_query_directory_init - SMB2_notify_init - SMB2_flush_init - build_qfs_info_req - smb2_hdr_assemble - smb2_reconnect - fill_small_buf - smb2_plain_req_init - __smb2_plain_req_init The read/write codepath is different than the rest as it is using pages, io iterators and async calls. To deal with those we add a server pointer in the cifs_writedata/cifs_readdata/cifs_io_parms context struct and set it in: - cifs_writepages (wdata) - cifs_write_from_iter (wdata) - cifs_readpages (rdata) - cifs_send_async_read (rdata) The [rw]data->server pointer is eventually copied to cifs_io_parms->server to pass it down to SMB2_read/SMB2_write. If SMB2_read/SMB2_write is called from a different place that doesn't set the server field it will pick a channel. Some places do not pick a channel and just use ses->server or cifs_ses_server(ses). All cifs_ses_server(ses) calls are in codepaths involving negprot/sess.setup. - SMB2_negotiate (binding channel) - SMB2_sess_alloc_buffer (binding channel) - SMB2_echo (uses provided one) - SMB2_logoff (uses master) - SMB2_tdis (uses master) (list not exhaustive) Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-06-01 01:38:22 +08:00
rc = compound_send_recv(xid, ses, server,
flags, num_rqst,
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
rqst, resp_buftype,
rsp_iov);
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
finished:
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
if (cfile)
cifsFileInfo_put(cfile);
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
SMB2_open_free(&rqst[0]);
smb3: improve handling of share deleted (and share recreated) When a share is deleted, returning EIO is confusing and no useful information is logged. Improve the handling of this case by at least logging a better error for this (and also mapping the error differently to EREMCHG). See e.g. the new messages that would be logged: [55243.639530] server share \\192.168.1.219\scratch deleted [55243.642568] CIFS VFS: \\192.168.1.219\scratch BAD_NETWORK_NAME: \\192.168.1.219\scratch In addition for the case where a share is deleted and then recreated with the same name, have now fixed that so it works. This is sometimes done for example, because the admin had to move a share to a different, bigger local drive when a share is running low on space. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-09-11 13:07:36 +08:00
if (rc == -EREMCHG) {
smb3: add dynamic trace points for tree disconnect Needed this for debugging a failing xfstest. Also change camel case for "treeName" to "tree_name" in tcon struct. Example trace output (from "trace-cmd record -e smb3_tdis*"): umount-9718 [006] ..... 5909.780244: smb3_tdis_enter: xid=206 sid=0xcf38894e tid=0x3d0b8cf8 path=\\localhost\test umount-9718 [007] ..... 5909.780878: smb3_tdis_done: xid=206 sid=0xcf38894e tid=0x3d0b8cf8 Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-09-22 03:05:53 +08:00
pr_warn_once("server share %s deleted\n", tcon->tree_name);
smb3: improve handling of share deleted (and share recreated) When a share is deleted, returning EIO is confusing and no useful information is logged. Improve the handling of this case by at least logging a better error for this (and also mapping the error differently to EREMCHG). See e.g. the new messages that would be logged: [55243.639530] server share \\192.168.1.219\scratch deleted [55243.642568] CIFS VFS: \\192.168.1.219\scratch BAD_NETWORK_NAME: \\192.168.1.219\scratch In addition for the case where a share is deleted and then recreated with the same name, have now fixed that so it works. This is sometimes done for example, because the admin had to move a share to a different, bigger local drive when a share is running low on space. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-09-11 13:07:36 +08:00
tcon->need_reconnect = true;
}
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
switch (command) {
case SMB2_OP_QUERY_INFO:
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
idata = ptr;
if (rc == 0 && cfile && cfile->symlink_target) {
idata->symlink_target = kstrdup(cfile->symlink_target, GFP_KERNEL);
if (!idata->symlink_target)
rc = -ENOMEM;
}
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
if (rc == 0) {
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
qi_rsp = (struct smb2_query_info_rsp *)
rsp_iov[1].iov_base;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
rc = smb2_validate_and_copy_iov(
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
le16_to_cpu(qi_rsp->OutputBufferOffset),
le32_to_cpu(qi_rsp->OutputBufferLength),
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
&rsp_iov[1], sizeof(idata->fi), (char *)&idata->fi);
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
}
if (rqst[1].rq_iov)
SMB2_query_info_free(&rqst[1]);
if (rqst[2].rq_iov)
SMB2_close_free(&rqst[2]);
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
if (rc)
trace_smb3_query_info_compound_err(xid, ses->Suid,
tcon->tid, rc);
else
trace_smb3_query_info_compound_done(xid, ses->Suid,
tcon->tid);
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
break;
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
case SMB2_OP_POSIX_QUERY_INFO:
cifs: fix uninitialised var in smb2_compound_op() Fix uninitialised variable @idata when calling smb2_compound_op() with SMB2_OP_POSIX_QUERY_INFO. Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-05 02:10:09 +08:00
idata = ptr;
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
if (rc == 0 && cfile && cfile->symlink_target) {
idata->symlink_target = kstrdup(cfile->symlink_target, GFP_KERNEL);
if (!idata->symlink_target)
rc = -ENOMEM;
}
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
if (rc == 0) {
qi_rsp = (struct smb2_query_info_rsp *)
rsp_iov[1].iov_base;
rc = smb2_validate_and_copy_iov(
le16_to_cpu(qi_rsp->OutputBufferOffset),
le32_to_cpu(qi_rsp->OutputBufferLength),
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
&rsp_iov[1], sizeof(idata->posix_fi) /* add SIDs */,
(char *)&idata->posix_fi);
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
}
cifs: Parse owner/group for stat in smb311 posix extensions stat was returning default owner and group (unlike readdir) for SMB3.1.1 POSIX extensions Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:37:44 +08:00
if (rc == 0) {
unsigned int length = le32_to_cpu(qi_rsp->OutputBufferLength);
if (length > sizeof(idata->posix_fi)) {
char *base = (char *)rsp_iov[1].iov_base +
le16_to_cpu(qi_rsp->OutputBufferOffset) +
sizeof(idata->posix_fi);
*extbuflen = length - sizeof(idata->posix_fi);
*extbuf = kmemdup(base, *extbuflen, GFP_KERNEL);
if (!*extbuf)
rc = -ENOMEM;
} else {
rc = -EINVAL;
}
}
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
if (rqst[1].rq_iov)
SMB2_query_info_free(&rqst[1]);
if (rqst[2].rq_iov)
SMB2_close_free(&rqst[2]);
if (rc)
smb311: Add tracepoints for new compound posix query info Add dynamic tracepoints for new SMB3.1.1. posix extensions query info level (100) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 12:02:33 +08:00
trace_smb3_posix_query_info_compound_err(xid, ses->Suid, tcon->tid, rc);
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
else
smb311: Add tracepoints for new compound posix query info Add dynamic tracepoints for new SMB3.1.1. posix extensions query info level (100) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 12:02:33 +08:00
trace_smb3_posix_query_info_compound_done(xid, ses->Suid, tcon->tid);
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
break;
cifs: change unlink to use a compound This,and previous patches, drops the number of roundtrips from five to two for unlink() Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:43 +08:00
case SMB2_OP_DELETE:
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
if (rc)
trace_smb3_delete_err(xid, ses->Suid, tcon->tid, rc);
else
trace_smb3_delete_done(xid, ses->Suid, tcon->tid);
if (rqst[1].rq_iov)
SMB2_close_free(&rqst[1]);
break;
cifs: change mkdir to use a compound This with the previous patch changes mkdir() from needing 6 roundtrips to just 3. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:42 +08:00
case SMB2_OP_MKDIR:
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
if (rc)
trace_smb3_mkdir_err(xid, ses->Suid, tcon->tid, rc);
else
trace_smb3_mkdir_done(xid, ses->Suid, tcon->tid);
cifs: change mkdir to use a compound This with the previous patch changes mkdir() from needing 6 roundtrips to just 3. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:42 +08:00
if (rqst[1].rq_iov)
SMB2_close_free(&rqst[1]);
break;
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
case SMB2_OP_HARDLINK:
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
if (rc)
trace_smb3_hardlink_err(xid, ses->Suid, tcon->tid, rc);
else
trace_smb3_hardlink_done(xid, ses->Suid, tcon->tid);
free_set_inf_compound(rqst);
break;
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
case SMB2_OP_RENAME:
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
if (rc)
trace_smb3_rename_err(xid, ses->Suid, tcon->tid, rc);
else
trace_smb3_rename_done(xid, ses->Suid, tcon->tid);
free_set_inf_compound(rqst);
break;
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
case SMB2_OP_RMDIR:
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
if (rc)
trace_smb3_rmdir_err(xid, ses->Suid, tcon->tid, rc);
else
trace_smb3_rmdir_done(xid, ses->Suid, tcon->tid);
free_set_inf_compound(rqst);
break;
cifs: change SMB2_OP_SET_INFO to use compounding Cuts number of network roundtrips significantly for some common syscalls Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:48 +08:00
case SMB2_OP_SET_EOF:
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
if (rc)
trace_smb3_set_eof_err(xid, ses->Suid, tcon->tid, rc);
else
trace_smb3_set_eof_done(xid, ses->Suid, tcon->tid);
free_set_inf_compound(rqst);
break;
cifs: change SMB2_OP_SET_INFO to use compounding Cuts number of network roundtrips significantly for some common syscalls Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:48 +08:00
case SMB2_OP_SET_INFO:
smb3: Add dynamic trace points for various compounded smb3 ops Adds trace points for enter and exit (done vs. error) for: compounded query and setinfo, hardlink, rename, mkdir, rmdir, set_eof, delete (unlink) Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2019-03-13 13:02:47 +08:00
if (rc)
trace_smb3_set_info_compound_err(xid, ses->Suid,
tcon->tid, rc);
else
trace_smb3_set_info_compound_done(xid, ses->Suid,
tcon->tid);
free_set_inf_compound(rqst);
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
break;
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
}
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
if (rc && err_iov && err_buftype) {
memcpy(err_iov, rsp_iov, 3 * sizeof(*err_iov));
memcpy(err_buftype, resp_buftype, 3 * sizeof(*err_buftype));
} else {
free_rsp_buf(resp_buftype[0], rsp_iov[0].iov_base);
free_rsp_buf(resp_buftype[1], rsp_iov[1].iov_base);
free_rsp_buf(resp_buftype[2], rsp_iov[2].iov_base);
}
cifs: reduce stack use in smb2_compound_op Move a lot of structures and arrays off the stack and into a dynamically allocated structure instead. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-05-20 10:19:59 +08:00
kfree(vars);
cifs: add a smb2_compound_op and change QUERY_INFO to use it This turns most open/query-info/close patterns in cifs.ko to become compounds. This changes stat from using 3 roundtrips to just a single one. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:41 +08:00
return rc;
}
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
int smb2_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
struct cifs_sb_info *cifs_sb, const char *full_path,
struct cifs_open_info_data *data, bool *adjust_tz, bool *reparse)
CIFS: Query SMB2 inode info Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com>
2011-12-29 21:06:33 +08:00
{
smb3: send backup intent on compounded query info When mounting with backupuid set, we should be setting CREATE_OPEN_BACKUP_INTENT flag on compounded opens as well, especially the case of compounded smb2_query_path_info. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2018-10-19 13:32:41 +08:00
__u32 create_options = 0;
cifs: add a helper to find an existing readable handle to a file and convert smb2_query_path_info() to use it. This will eliminate the need for a SMB2_Create when we already have an open handle that can be used. This will also prevent a oplock break in case the other handle holds a lease. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-09-09 13:30:00 +08:00
struct cifsFileInfo *cfile;
cifs: return cached_fid from open_shroot Cleanup patch for followon to cache additional information for the root directory when directory lease held. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2020-10-05 10:37:52 +08:00
struct cached_fid *cfid = NULL;
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
struct kvec err_iov[3] = {};
int err_buftype[3] = {};
cifs: improve checking of DFS links over STATUS_OBJECT_NAME_INVALID Do not map STATUS_OBJECT_NAME_INVALID to -EREMOTE under non-DFS shares, or 'nodfs' mounts or CONFIG_CIFS_DFS_UPCALL=n builds. Otherwise, in the slow path, get a referral to figure out whether it is an actual DFS link. This could be simply reproduced under a non-DFS share by running the following $ mount.cifs //srv/share /mnt -o ... $ cat /mnt/$(printf '\U110000') cat: '/mnt/'$'\364\220\200\200': Object is remote Fixes: c877ce47e137 ("cifs: reduce roundtrips on create/qinfo requests") CC: stable@vger.kernel.org # 6.2 Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2023-03-01 06:01:54 +08:00
bool islink;
int rc, rc2;
CIFS: Query SMB2 inode info Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com>
2011-12-29 21:06:33 +08:00
*adjust_tz = false;
smb3: add support for stat of WSL reparse points for special file types This is needed so when mounting to Windows we do not misinterpret various special files created by Linux (WSL) as symlinks. An earlier patch addressed readdir. This patch fixes stat (getattr). With this patch:   File: /mnt1/char   Size: 0          Blocks: 0          IO Block: 16384  character special file Device: 34h/52d Inode: 844424930132069  Links: 1     Device type: 0,0 Access: (0755/crwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:46:51.839458900 -0500 Modify: 2020-10-21 17:46:51.839458900 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/fifo   Size: 0          Blocks: 0          IO Block: 16384  fifo Device: 34h/52d Inode: 1125899906842722  Links: 1 Access: (0755/prwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 16:21:37.259249700 -0500 Modify: 2020-10-21 16:21:37.259249700 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/block   Size: 0          Blocks: 0          IO Block: 16384  block special file Device: 34h/52d Inode: 844424930132068  Links: 1     Device type: 0,0 Access: (0755/brwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:10:47.913103200 -0500 Modify: 2020-10-21 17:10:47.913103200 -0500 Change: 2020-10-21 18:30:39.796725500 -0500  Birth: - without the patch all show up incorrectly as symlinks with annoying "operation not supported error also returned"   File: /mnt1/charstat: cannot read symbolic link '/mnt1/char': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 844424930132069  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:46:51.839458900 -0500 Modify: 2020-10-21 17:46:51.839458900 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/fifostat: cannot read symbolic link '/mnt1/fifo': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 1125899906842722  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 16:21:37.259249700 -0500 Modify: 2020-10-21 16:21:37.259249700 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/blockstat: cannot read symbolic link '/mnt1/block': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 844424930132068  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:10:47.913103200 -0500 Modify: 2020-10-21 17:10:47.913103200 -0500 Change: 2020-10-21 18:30:39.796725500 -0500 Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2020-10-23 11:03:14 +08:00
*reparse = false;
CIFS: Query SMB2 inode info Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com>
2011-12-29 21:06:33 +08:00
cifs: set the CREATE_NOT_FILE when opening the directory in use_cached_dir() This enforces that we can only do this for directories and not normal files or else the server will return an error. This means that we will have conditionally check IF the path refers to a directory or not in all the call-sites where we are unsure. Right now this check is for "" i.e. root. Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Enzo Matsumiya <ematsumiya@suse.de> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-05-10 07:42:06 +08:00
if (strcmp(full_path, ""))
rc = -ENOENT;
else
cifs: Do not access tcon->cfids->cfid directly from is_path_accessible cfids will soon keep a list of cached fids so we should not access this directly from outside of cached_dir.c Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-08-12 08:51:18 +08:00
rc = open_cached_dir(xid, tcon, full_path, cifs_sb, false, &cfid);
CIFS: Fix mounts if the client is low on credits If the server doesn't grant us at least 3 credits during the mount we won't be able to complete it because query path info operation requires 3 credits. Use the cached file handle if possible to allow the mount to succeed. Signed-off-by: Pavel Shilovsky <pshilov@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-01-17 03:48:42 +08:00
/* If it is a root and its handle is cached then use it */
cifs: pass a path to open_shroot and check if it is the root or not Move the check for the directory path into the open_shroot() function but still fail for any non-root directories. This is preparation for later when we will start using the cache also for other directories than the root. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-03-09 07:07:28 +08:00
if (!rc) {
cifs: Do not use tcon->cfid directly, use the cfid we get from open_cached_dir They are the same right now but tcon-> will later point to a different type of struct containing a list of cfids. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-08-09 10:11:49 +08:00
if (cfid->file_all_info_is_valid) {
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
memcpy(&data->fi, &cfid->file_all_info, sizeof(data->fi));
cifs: pass a path to open_shroot and check if it is the root or not Move the check for the directory path into the open_shroot() function but still fail for any non-root directories. This is preparation for later when we will start using the cache also for other directories than the root. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-03-09 07:07:28 +08:00
} else {
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
rc = SMB2_query_info(xid, tcon, cfid->fid.persistent_fid,
cfid->fid.volatile_fid, &data->fi);
cifs: cache FILE_ALL_INFO for the shared root handle When we open the shared root handle also ask for FILE_ALL_INFORMATION since we can do this at zero cost as part of a compound. Cache this information as long as the lease is held and return and serve any future requests from cache. This allows us to serve "stat /<mountpoint>" directly from cache and avoid a network roundtrip. Since clients often want to do this quite a lot this improve performance slightly. As an example: xfstest generic/533 performs 43 stat operations on the root of the share while it is run. Which are eliminated with this patch. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com>
2019-03-12 11:58:31 +08:00
}
cifs: rename the *_shroot* functions to *_cached_dir* These functions will eventually be used to cache any directory, not just the root so change the names. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-03-09 07:07:29 +08:00
close_cached_dir(cfid);
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
return rc;
CIFS: Fix mounts if the client is low on credits If the server doesn't grant us at least 3 credits during the mount we won't be able to complete it because query path info operation requires 3 credits. Use the cached file handle if possible to allow the mount to succeed. Signed-off-by: Pavel Shilovsky <pshilov@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-01-17 03:48:42 +08:00
}
cifs: add a helper to find an existing readable handle to a file and convert smb2_query_path_info() to use it. This will eliminate the need for a SMB2_Create when we already have an open handle that can be used. This will also prevent a oplock break in case the other handle holds a lease. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-09-09 13:30:00 +08:00
cifs_get_readable_path(tcon, full_path, &cfile);
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, FILE_READ_ATTRIBUTES, FILE_OPEN,
create_options, ACL_NO_MODE, data, SMB2_OP_QUERY_INFO, cfile,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
NULL, NULL, err_iov, err_buftype);
cifs: reduce roundtrips on create/qinfo requests To work around some Window servers that return STATUS_OBJECT_NAME_INVALID on query infos under DFS namespaces that contain non-ASCII characters, we started checking for -ENOENT on every file open, and if so, then send additional requests to figure out whether it is a DFS link or not. It means that all those requests will be sent to every non-existing file. So, in order to reduce the number of roundtrips, check earlier whether status code is STATUS_OBJECT_NAME_INVALID and tcon supports dfs, and if so, then map -ENOENT to -EREMOTE so mount or automount will take care of chasing the DFS link -- if it isn't an DFS link, then -ENOENT will be returned appropriately. Before patch SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 228 Ioctl Request FSCTL_DFS_GET_REFERRALS, File: \ada.test\dfs\foo SMB2 143 Ioctl Response, Error: STATUS_OBJECT_PATH_NOT_FOUND SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 228 Ioctl Request FSCTL_DFS_GET_REFERRALS, File: \ada.test\dfs\foo SMB2 143 Ioctl Response, Error: STATUS_OBJECT_PATH_NOT_FOUND After patch SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-12-13 10:39:37 +08:00
if (rc) {
struct smb2_hdr *hdr = err_iov[0].iov_base;
if (unlikely(!hdr || err_buftype[0] == CIFS_NO_BUFFER))
goto out;
if (rc == -EOPNOTSUPP && hdr->Command == SMB2_CREATE &&
hdr->Status == STATUS_STOPPED_ON_SYMLINK) {
rc = smb2_parse_symlink_response(cifs_sb, err_iov,
&data->symlink_target);
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
if (rc)
goto out;
cifs: reduce roundtrips on create/qinfo requests To work around some Window servers that return STATUS_OBJECT_NAME_INVALID on query infos under DFS namespaces that contain non-ASCII characters, we started checking for -ENOENT on every file open, and if so, then send additional requests to figure out whether it is a DFS link or not. It means that all those requests will be sent to every non-existing file. So, in order to reduce the number of roundtrips, check earlier whether status code is STATUS_OBJECT_NAME_INVALID and tcon supports dfs, and if so, then map -ENOENT to -EREMOTE so mount or automount will take care of chasing the DFS link -- if it isn't an DFS link, then -ENOENT will be returned appropriately. Before patch SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 228 Ioctl Request FSCTL_DFS_GET_REFERRALS, File: \ada.test\dfs\foo SMB2 143 Ioctl Response, Error: STATUS_OBJECT_PATH_NOT_FOUND SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 228 Ioctl Request FSCTL_DFS_GET_REFERRALS, File: \ada.test\dfs\foo SMB2 143 Ioctl Response, Error: STATUS_OBJECT_PATH_NOT_FOUND After patch SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-12-13 10:39:37 +08:00
*reparse = true;
create_options |= OPEN_REPARSE_POINT;
/* Failed on a symbolic link - query a reparse point info */
cifs_get_readable_path(tcon, full_path, &cfile);
rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
FILE_READ_ATTRIBUTES, FILE_OPEN,
create_options, ACL_NO_MODE, data,
SMB2_OP_QUERY_INFO, cfile, NULL, NULL,
NULL, NULL);
goto out;
cifs: improve checking of DFS links over STATUS_OBJECT_NAME_INVALID Do not map STATUS_OBJECT_NAME_INVALID to -EREMOTE under non-DFS shares, or 'nodfs' mounts or CONFIG_CIFS_DFS_UPCALL=n builds. Otherwise, in the slow path, get a referral to figure out whether it is an actual DFS link. This could be simply reproduced under a non-DFS share by running the following $ mount.cifs //srv/share /mnt -o ... $ cat /mnt/$(printf '\U110000') cat: '/mnt/'$'\364\220\200\200': Object is remote Fixes: c877ce47e137 ("cifs: reduce roundtrips on create/qinfo requests") CC: stable@vger.kernel.org # 6.2 Signed-off-by: Paulo Alcantara (SUSE) <pc@manguebit.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2023-03-01 06:01:54 +08:00
} else if (rc != -EREMOTE && hdr->Status == STATUS_OBJECT_NAME_INVALID) {
rc2 = cifs_inval_name_dfs_link_error(xid, tcon, cifs_sb,
full_path, &islink);
if (rc2) {
rc = rc2;
goto out;
}
if (islink)
rc = -EREMOTE;
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
}
cifs: reduce roundtrips on create/qinfo requests To work around some Window servers that return STATUS_OBJECT_NAME_INVALID on query infos under DFS namespaces that contain non-ASCII characters, we started checking for -ENOENT on every file open, and if so, then send additional requests to figure out whether it is a DFS link or not. It means that all those requests will be sent to every non-existing file. So, in order to reduce the number of roundtrips, check earlier whether status code is STATUS_OBJECT_NAME_INVALID and tcon supports dfs, and if so, then map -ENOENT to -EREMOTE so mount or automount will take care of chasing the DFS link -- if it isn't an DFS link, then -ENOENT will be returned appropriately. Before patch SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 228 Ioctl Request FSCTL_DFS_GET_REFERRALS, File: \ada.test\dfs\foo SMB2 143 Ioctl Response, Error: STATUS_OBJECT_PATH_NOT_FOUND SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 228 Ioctl Request FSCTL_DFS_GET_REFERRALS, File: \ada.test\dfs\foo SMB2 143 Ioctl Response, Error: STATUS_OBJECT_PATH_NOT_FOUND After patch SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... SMB2 438 Create Request File: ada.test\dfs\foo;GetInfo Request... SMB2 310 Create Response, Error: STATUS_OBJECT_NAME_NOT_FOUND;... Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-12-13 10:39:37 +08:00
if (rc == -EREMOTE && IS_ENABLED(CONFIG_CIFS_DFS_UPCALL) && cifs_sb &&
(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS))
rc = -EOPNOTSUPP;
CIFS: Fix symbolic links usage Now we treat any reparse point as a symbolic link and map it to a Unix one that is not true in a common case due to many reparse point types supported by SMB servers. Distinguish reparse point types into two groups: 1) that can be accessed directly through a reparse point (junctions, deduplicated files, NFS symlinks); 2) that need to be processed manually (Windows symbolic links, DFS); and map only Windows symbolic links to Unix ones. Cc: <stable@vger.kernel.org> Acked-by: Jeff Layton <jlayton@redhat.com> Reported-and-tested-by: Joao Correia <joaomiguelcorreia@gmail.com> Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com>
2013-10-23 21:49:47 +08:00
}
CIFS: Query SMB2 inode info Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com>
2011-12-29 21:06:33 +08:00
out:
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
free_rsp_buf(err_buftype[0], err_iov[0].iov_base);
free_rsp_buf(err_buftype[1], err_iov[1].iov_base);
free_rsp_buf(err_buftype[2], err_iov[2].iov_base);
CIFS: Query SMB2 inode info Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by: Steve French <smfrench@gmail.com>
2011-12-29 21:06:33 +08:00
return rc;
}
CIFS: Add SMB2 support for mkdir operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2011-07-19 16:56:37 +08:00
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
int smb311_posix_query_path_info(const unsigned int xid, struct cifs_tcon *tcon,
struct cifs_sb_info *cifs_sb, const char *full_path,
cifs: Parse owner/group for stat in smb311 posix extensions stat was returning default owner and group (unlike readdir) for SMB3.1.1 POSIX extensions Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:37:44 +08:00
struct cifs_open_info_data *data,
struct cifs_sid *owner,
struct cifs_sid *group,
bool *adjust_tz, bool *reparse)
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
{
int rc;
__u32 create_options = 0;
struct cifsFileInfo *cfile;
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
struct kvec err_iov[3] = {};
int err_buftype[3] = {};
cifs: Parse owner/group for stat in smb311 posix extensions stat was returning default owner and group (unlike readdir) for SMB3.1.1 POSIX extensions Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:37:44 +08:00
__u8 *sidsbuf = NULL;
__u8 *sidsbuf_end = NULL;
size_t sidsbuflen = 0;
size_t owner_len, group_len;
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
*adjust_tz = false;
smb3: add support for stat of WSL reparse points for special file types This is needed so when mounting to Windows we do not misinterpret various special files created by Linux (WSL) as symlinks. An earlier patch addressed readdir. This patch fixes stat (getattr). With this patch:   File: /mnt1/char   Size: 0          Blocks: 0          IO Block: 16384  character special file Device: 34h/52d Inode: 844424930132069  Links: 1     Device type: 0,0 Access: (0755/crwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:46:51.839458900 -0500 Modify: 2020-10-21 17:46:51.839458900 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/fifo   Size: 0          Blocks: 0          IO Block: 16384  fifo Device: 34h/52d Inode: 1125899906842722  Links: 1 Access: (0755/prwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 16:21:37.259249700 -0500 Modify: 2020-10-21 16:21:37.259249700 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/block   Size: 0          Blocks: 0          IO Block: 16384  block special file Device: 34h/52d Inode: 844424930132068  Links: 1     Device type: 0,0 Access: (0755/brwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:10:47.913103200 -0500 Modify: 2020-10-21 17:10:47.913103200 -0500 Change: 2020-10-21 18:30:39.796725500 -0500  Birth: - without the patch all show up incorrectly as symlinks with annoying "operation not supported error also returned"   File: /mnt1/charstat: cannot read symbolic link '/mnt1/char': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 844424930132069  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:46:51.839458900 -0500 Modify: 2020-10-21 17:46:51.839458900 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/fifostat: cannot read symbolic link '/mnt1/fifo': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 1125899906842722  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 16:21:37.259249700 -0500 Modify: 2020-10-21 16:21:37.259249700 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/blockstat: cannot read symbolic link '/mnt1/block': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 844424930132068  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:10:47.913103200 -0500 Modify: 2020-10-21 17:10:47.913103200 -0500 Change: 2020-10-21 18:30:39.796725500 -0500 Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2020-10-23 11:03:14 +08:00
*reparse = false;
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
/*
* BB TODO: Add support for using the cached root handle.
* Create SMB2_query_posix_info worker function to do non-compounded query
* when we already have an open file handle for this. For now this is fast enough
* (always using the compounded version).
*/
cifs_get_readable_path(tcon, full_path, &cfile);
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, FILE_READ_ATTRIBUTES, FILE_OPEN,
create_options, ACL_NO_MODE, data, SMB2_OP_POSIX_QUERY_INFO, cfile,
cifs: Parse owner/group for stat in smb311 posix extensions stat was returning default owner and group (unlike readdir) for SMB3.1.1 POSIX extensions Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:37:44 +08:00
&sidsbuf, &sidsbuflen, err_iov, err_buftype);
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
if (rc == -EOPNOTSUPP) {
/* BB TODO: When support for special files added to Samba re-verify this path */
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
if (err_iov[0].iov_base && err_buftype[0] != CIFS_NO_BUFFER &&
((struct smb2_hdr *)err_iov[0].iov_base)->Command == SMB2_CREATE &&
((struct smb2_hdr *)err_iov[0].iov_base)->Status == STATUS_STOPPED_ON_SYMLINK) {
rc = smb2_parse_symlink_response(cifs_sb, err_iov, &data->symlink_target);
if (rc)
goto out;
}
smb3: add support for stat of WSL reparse points for special file types This is needed so when mounting to Windows we do not misinterpret various special files created by Linux (WSL) as symlinks. An earlier patch addressed readdir. This patch fixes stat (getattr). With this patch:   File: /mnt1/char   Size: 0          Blocks: 0          IO Block: 16384  character special file Device: 34h/52d Inode: 844424930132069  Links: 1     Device type: 0,0 Access: (0755/crwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:46:51.839458900 -0500 Modify: 2020-10-21 17:46:51.839458900 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/fifo   Size: 0          Blocks: 0          IO Block: 16384  fifo Device: 34h/52d Inode: 1125899906842722  Links: 1 Access: (0755/prwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 16:21:37.259249700 -0500 Modify: 2020-10-21 16:21:37.259249700 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/block   Size: 0          Blocks: 0          IO Block: 16384  block special file Device: 34h/52d Inode: 844424930132068  Links: 1     Device type: 0,0 Access: (0755/brwxr-xr-x)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:10:47.913103200 -0500 Modify: 2020-10-21 17:10:47.913103200 -0500 Change: 2020-10-21 18:30:39.796725500 -0500  Birth: - without the patch all show up incorrectly as symlinks with annoying "operation not supported error also returned"   File: /mnt1/charstat: cannot read symbolic link '/mnt1/char': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 844424930132069  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:46:51.839458900 -0500 Modify: 2020-10-21 17:46:51.839458900 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/fifostat: cannot read symbolic link '/mnt1/fifo': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 1125899906842722  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 16:21:37.259249700 -0500 Modify: 2020-10-21 16:21:37.259249700 -0500 Change: 2020-10-21 18:30:39.797358800 -0500  Birth: -   File: /mnt1/blockstat: cannot read symbolic link '/mnt1/block': Operation not supported   Size: 0          Blocks: 0          IO Block: 16384  symbolic link Device: 34h/52d Inode: 844424930132068  Links: 1 Access: (0000/l---------)  Uid: (    0/    root)   Gid: (    0/    root) Access: 2020-10-21 17:10:47.913103200 -0500 Modify: 2020-10-21 17:10:47.913103200 -0500 Change: 2020-10-21 18:30:39.796725500 -0500 Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com>
2020-10-23 11:03:14 +08:00
*reparse = true;
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
create_options |= OPEN_REPARSE_POINT;
/* Failed on a symbolic link - query a reparse point info */
cifs: for compound requests, use open handle if possible For smb2_compound_op, it is possible to pass a ref to an already open file. We should be passing it whenever possible. i.e. if a matching handle is already kept open. If we don't do that, we will end up breaking leases for files kept open on the same client. Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-07-30 14:43:09 +08:00
cifs_get_readable_path(tcon, full_path, &cfile);
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
rc = smb2_compound_op(xid, tcon, cifs_sb, full_path, FILE_READ_ATTRIBUTES,
FILE_OPEN, create_options, ACL_NO_MODE, data,
cifs: Parse owner/group for stat in smb311 posix extensions stat was returning default owner and group (unlike readdir) for SMB3.1.1 POSIX extensions Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:37:44 +08:00
SMB2_OP_POSIX_QUERY_INFO, cfile,
&sidsbuf, &sidsbuflen, NULL, NULL);
}
if (rc == 0) {
sidsbuf_end = sidsbuf + sidsbuflen;
owner_len = posix_info_sid_size(sidsbuf, sidsbuf_end);
if (owner_len == -1) {
rc = -EINVAL;
goto out;
}
memcpy(owner, sidsbuf, owner_len);
group_len = posix_info_sid_size(
sidsbuf + owner_len, sidsbuf_end);
if (group_len == -1) {
rc = -EINVAL;
goto out;
}
memcpy(group, sidsbuf + owner_len, group_len);
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
}
out:
cifs: Parse owner/group for stat in smb311 posix extensions stat was returning default owner and group (unlike readdir) for SMB3.1.1 POSIX extensions Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:37:44 +08:00
kfree(sidsbuf);
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
free_rsp_buf(err_buftype[0], err_iov[0].iov_base);
free_rsp_buf(err_buftype[1], err_iov[1].iov_base);
free_rsp_buf(err_buftype[2], err_iov[2].iov_base);
SMB311: Add support for query info using posix extensions (level 100) Adds support for better query info on dentry revalidation (using the SMB3.1.1 POSIX extensions level 100). Followon patch will add support for translating the UID/GID from the SID and also will add support for using the posix query info on lookup. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2020-06-12 08:25:47 +08:00
return rc;
}
CIFS: Add SMB2 support for mkdir operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2011-07-19 16:56:37 +08:00
int
smb3: pass mode bits into create calls We need to populate an ACL (security descriptor open context) on file and directory correct. This patch passes in the mode. Followon patch will build the open context and the security descriptor (from the mode) that goes in the open context. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2019-09-25 13:32:13 +08:00
smb2_mkdir(const unsigned int xid, struct inode *parent_inode, umode_t mode,
struct cifs_tcon *tcon, const char *name,
CIFS: Add SMB2 support for mkdir operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2011-07-19 16:56:37 +08:00
struct cifs_sb_info *cifs_sb)
{
cifs: change mkdir to use a compound This with the previous patch changes mkdir() from needing 6 roundtrips to just 3. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:42 +08:00
return smb2_compound_op(xid, tcon, cifs_sb, name,
FILE_WRITE_ATTRIBUTES, FILE_CREATE,
smb3: pass mode bits into create calls We need to populate an ACL (security descriptor open context) on file and directory correct. This patch passes in the mode. Followon patch will build the open context and the security descriptor (from the mode) that goes in the open context. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2019-09-25 13:32:13 +08:00
CREATE_NOT_FILE, mode, NULL, SMB2_OP_MKDIR,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
NULL, NULL, NULL, NULL, NULL);
CIFS: Add SMB2 support for mkdir operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2011-07-19 16:56:37 +08:00
}
void
smb2_mkdir_setinfo(struct inode *inode, const char *name,
struct cifs_sb_info *cifs_sb, struct cifs_tcon *tcon,
const unsigned int xid)
{
FILE_BASIC_INFO data;
struct cifsInodeInfo *cifs_i;
cifs: use existing handle for compound_op(OP_SET_INFO) when possible If we already have a writable handle for a path we want to set the attributes for then use that instead of a create/set-info/close compound. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 07:53:56 +08:00
struct cifsFileInfo *cfile;
CIFS: Add SMB2 support for mkdir operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2011-07-19 16:56:37 +08:00
u32 dosattrs;
int tmprc;
memset(&data, 0, sizeof(data));
cifs_i = CIFS_I(inode);
dosattrs = cifs_i->cifsAttrs | ATTR_READONLY;
data.Attributes = cpu_to_le32(dosattrs);
cifs: fix rename() by ensuring source handle opened with DELETE bit To rename a file in SMB2 we open it with the DELETE access and do a special SetInfo on it. If the handle is missing the DELETE bit the server will fail the SetInfo with STATUS_ACCESS_DENIED. We currently try to reuse any existing opened handle we have with cifs_get_writable_path(). That function looks for handles with WRITE access but doesn't check for DELETE, making rename() fail if it finds a handle to reuse. Simple reproducer below. To select handles with the DELETE bit, this patch adds a flag argument to cifs_get_writable_path() and find_writable_file() and the existing 'bool fsuid_only' argument is converted to a flag. The cifsFileInfo struct only stores the UNIX open mode but not the original SMB access flags. Since the DELETE bit is not mapped in that mode, this patch stores the access mask in cifs_fid on file open, which is accessible from cifsFileInfo. Simple reproducer: #include <stdio.h> #include <stdlib.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #define E(s) perror(s), exit(1) int main(int argc, char *argv[]) { int fd, ret; if (argc != 3) { fprintf(stderr, "Usage: %s A B\n" "create&open A in write mode, " "rename A to B, close A\n", argv[0]); return 0; } fd = openat(AT_FDCWD, argv[1], O_WRONLY|O_CREAT|O_SYNC, 0666); if (fd == -1) E("openat()"); ret = rename(argv[1], argv[2]); if (ret) E("rename()"); ret = close(fd); if (ret) E("close()"); return ret; } $ gcc -o bugrename bugrename.c $ ./bugrename /mnt/a /mnt/b rename(): Permission denied Fixes: 8de9e86c67ba ("cifs: create a helper to find a writeable handle by path name") CC: Stable <stable@vger.kernel.org> Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
2020-02-21 18:19:06 +08:00
cifs_get_writable_path(tcon, name, FIND_WR_ANY, &cfile);
cifs: change SMB2_OP_SET_INFO to use compounding Cuts number of network roundtrips significantly for some common syscalls Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:48 +08:00
tmprc = smb2_compound_op(xid, tcon, cifs_sb, name,
FILE_WRITE_ATTRIBUTES, FILE_CREATE,
smb3: pass mode bits into create calls We need to populate an ACL (security descriptor open context) on file and directory correct. This patch passes in the mode. Followon patch will build the open context and the security descriptor (from the mode) that goes in the open context. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2019-09-25 13:32:13 +08:00
CREATE_NOT_FILE, ACL_NO_MODE,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
&data, SMB2_OP_SET_INFO, cfile, NULL, NULL, NULL, NULL);
CIFS: Add SMB2 support for mkdir operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2011-07-19 16:56:37 +08:00
if (tmprc == 0)
cifs_i->cifsAttrs = dosattrs;
}
CIFS: Add SMB2 support for rmdir Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-07-10 20:14:38 +08:00
int
smb2_rmdir(const unsigned int xid, struct cifs_tcon *tcon, const char *name,
struct cifs_sb_info *cifs_sb)
{
cifs: drop the lease for cached directories on rmdir or rename When we delete or rename a directory we must also drop any cached lease we have on the directory. Fixes: a350d6e73f5e ("cifs: enable caching of directories for which a lease is held") Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-18 15:39:10 +08:00
drop_cached_dir_by_name(xid, tcon, name, cifs_sb);
cifs: make rmdir() use compounding This and previous patches drop the number of roundtrips we need for rmdir() from 6 to 2. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:45 +08:00
return smb2_compound_op(xid, tcon, cifs_sb, name, DELETE, FILE_OPEN,
smb3: pass mode bits into create calls We need to populate an ACL (security descriptor open context) on file and directory correct. This patch passes in the mode. Followon patch will build the open context and the security descriptor (from the mode) that goes in the open context. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2019-09-25 13:32:13 +08:00
CREATE_NOT_FILE, ACL_NO_MODE,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
NULL, SMB2_OP_RMDIR, NULL, NULL, NULL, NULL, NULL);
CIFS: Add SMB2 support for rmdir Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-07-10 20:14:38 +08:00
}
CIFS: Add SMB2 support for unlink Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:25 +08:00
int
smb2_unlink(const unsigned int xid, struct cifs_tcon *tcon, const char *name,
struct cifs_sb_info *cifs_sb)
{
cifs: change unlink to use a compound This,and previous patches, drops the number of roundtrips from five to two for unlink() Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:43 +08:00
return smb2_compound_op(xid, tcon, cifs_sb, name, DELETE, FILE_OPEN,
CREATE_DELETE_ON_CLOSE | OPEN_REPARSE_POINT,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
ACL_NO_MODE, NULL, SMB2_OP_DELETE, NULL, NULL, NULL, NULL, NULL);
CIFS: Add SMB2 support for unlink Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:25 +08:00
}
CIFS: Add SMB2 support for rename operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:31 +08:00
CIFS: Add SMB2 support for hardlink operation Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:31 +08:00
static int
smb2_set_path_attr(const unsigned int xid, struct cifs_tcon *tcon,
const char *from_name, const char *to_name,
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
struct cifs_sb_info *cifs_sb, __u32 access, int command,
struct cifsFileInfo *cfile)
CIFS: Add SMB2 support for rename operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:31 +08:00
{
__le16 *smb2_to_name = NULL;
int rc;
smb2_to_name = cifs_convert_path_to_utf16(to_name, cifs_sb);
if (smb2_to_name == NULL) {
rc = -ENOMEM;
goto smb2_rename_path;
}
cifs: change SMB2_OP_RENAME and SMB2_OP_HARDLINK to use compounding Get rid of smb2_open_op_close() as all operations are now migrated to smb2_compound_op(). Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:49 +08:00
rc = smb2_compound_op(xid, tcon, cifs_sb, from_name, access,
smb3: pass mode bits into create calls We need to populate an ACL (security descriptor open context) on file and directory correct. This patch passes in the mode. Followon patch will build the open context and the security descriptor (from the mode) that goes in the open context. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2019-09-25 13:32:13 +08:00
FILE_OPEN, 0, ACL_NO_MODE, smb2_to_name,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
command, cfile, NULL, NULL, NULL, NULL);
CIFS: Add SMB2 support for rename operation Signed-off-by: Pavel Shilovsky <piastryyy@gmail.com> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:31 +08:00
smb2_rename_path:
kfree(smb2_to_name);
return rc;
}
CIFS: Add SMB2 support for hardlink operation Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:31 +08:00
int
smb2_rename_path(const unsigned int xid, struct cifs_tcon *tcon,
const char *from_name, const char *to_name,
struct cifs_sb_info *cifs_sb)
{
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
struct cifsFileInfo *cfile;
cifs: drop the lease for cached directories on rmdir or rename When we delete or rename a directory we must also drop any cached lease we have on the directory. Fixes: a350d6e73f5e ("cifs: enable caching of directories for which a lease is held") Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-18 15:39:10 +08:00
drop_cached_dir_by_name(xid, tcon, from_name, cifs_sb);
cifs: fix rename() by ensuring source handle opened with DELETE bit To rename a file in SMB2 we open it with the DELETE access and do a special SetInfo on it. If the handle is missing the DELETE bit the server will fail the SetInfo with STATUS_ACCESS_DENIED. We currently try to reuse any existing opened handle we have with cifs_get_writable_path(). That function looks for handles with WRITE access but doesn't check for DELETE, making rename() fail if it finds a handle to reuse. Simple reproducer below. To select handles with the DELETE bit, this patch adds a flag argument to cifs_get_writable_path() and find_writable_file() and the existing 'bool fsuid_only' argument is converted to a flag. The cifsFileInfo struct only stores the UNIX open mode but not the original SMB access flags. Since the DELETE bit is not mapped in that mode, this patch stores the access mask in cifs_fid on file open, which is accessible from cifsFileInfo. Simple reproducer: #include <stdio.h> #include <stdlib.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #define E(s) perror(s), exit(1) int main(int argc, char *argv[]) { int fd, ret; if (argc != 3) { fprintf(stderr, "Usage: %s A B\n" "create&open A in write mode, " "rename A to B, close A\n", argv[0]); return 0; } fd = openat(AT_FDCWD, argv[1], O_WRONLY|O_CREAT|O_SYNC, 0666); if (fd == -1) E("openat()"); ret = rename(argv[1], argv[2]); if (ret) E("rename()"); ret = close(fd); if (ret) E("close()"); return ret; } $ gcc -o bugrename bugrename.c $ ./bugrename /mnt/a /mnt/b rename(): Permission denied Fixes: 8de9e86c67ba ("cifs: create a helper to find a writeable handle by path name") CC: Stable <stable@vger.kernel.org> Signed-off-by: Aurelien Aptel <aaptel@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
2020-02-21 18:19:06 +08:00
cifs_get_writable_path(tcon, from_name, FIND_WR_WITH_DELETE, &cfile);
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
return smb2_set_path_attr(xid, tcon, from_name, to_name,
cifs_sb, DELETE, SMB2_OP_RENAME, cfile);
CIFS: Add SMB2 support for hardlink operation Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:31 +08:00
}
int
smb2_create_hardlink(const unsigned int xid, struct cifs_tcon *tcon,
const char *from_name, const char *to_name,
struct cifs_sb_info *cifs_sb)
{
return smb2_set_path_attr(xid, tcon, from_name, to_name, cifs_sb,
cifs: create a helper to find a writeable handle by path name rename() takes a path for old_file and in SMB2 we used to just create a compound for create(old_path)/rename/close(). If we already have a writable handle we can avoid the create() and close() altogether and just use the existing handle. For this situation, as we avoid doing the create() we also avoid triggering an oplock break for the existing handle. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2019-08-30 06:25:46 +08:00
FILE_READ_ATTRIBUTES, SMB2_OP_HARDLINK,
NULL);
CIFS: Add SMB2 support for hardlink operation Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:31 +08:00
}
CIFS: Add SMB2 support for set_file_size Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:32 +08:00
int
smb2_set_path_size(const unsigned int xid, struct cifs_tcon *tcon,
const char *full_path, __u64 size,
struct cifs_sb_info *cifs_sb, bool set_alloc)
{
__le64 eof = cpu_to_le64(size);
cifs: for compound requests, use open handle if possible For smb2_compound_op, it is possible to pass a ref to an already open file. We should be passing it whenever possible. i.e. if a matching handle is already kept open. If we don't do that, we will end up breaking leases for files kept open on the same client. Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-07-30 14:43:09 +08:00
struct cifsFileInfo *cfile;
cifs: change SMB2_OP_SET_EOF to use compounding This changes SMB2_OP_SET_EOF to use compounding in some situations. This is part of the path based API to truncate a file. Most of the time this will however not be invoked for SMB2 since cifs_set_file_size() will as far as I can tell almost always just open the file synchronously and switch to the handle based truncate code path, thus bypassing the compounding we add here. Rewriting cifs_set_file_size() and make that whole pile of code more compounding friendly, and also easier to read and understand, is a different project though and not for this patch. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:46 +08:00
cifs: for compound requests, use open handle if possible For smb2_compound_op, it is possible to pass a ref to an already open file. We should be passing it whenever possible. i.e. if a matching handle is already kept open. If we don't do that, we will end up breaking leases for files kept open on the same client. Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-07-30 14:43:09 +08:00
cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile);
cifs: change SMB2_OP_SET_EOF to use compounding This changes SMB2_OP_SET_EOF to use compounding in some situations. This is part of the path based API to truncate a file. Most of the time this will however not be invoked for SMB2 since cifs_set_file_size() will as far as I can tell almost always just open the file synchronously and switch to the handle based truncate code path, thus bypassing the compounding we add here. Rewriting cifs_set_file_size() and make that whole pile of code more compounding friendly, and also easier to read and understand, is a different project though and not for this patch. Signed-off-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-09-03 11:33:46 +08:00
return smb2_compound_op(xid, tcon, cifs_sb, full_path,
smb3: pass mode bits into create calls We need to populate an ACL (security descriptor open context) on file and directory correct. This patch passes in the mode. Followon patch will build the open context and the security descriptor (from the mode) that goes in the open context. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2019-09-25 13:32:13 +08:00
FILE_WRITE_DATA, FILE_OPEN, 0, ACL_NO_MODE,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
&eof, SMB2_OP_SET_EOF, cfile, NULL, NULL, NULL, NULL);
CIFS: Add SMB2 support for set_file_size Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:32 +08:00
}
CIFS: Add set_file_info support for SMB2 Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:32 +08:00
int
smb2_set_file_info(struct inode *inode, const char *full_path,
FILE_BASIC_INFO *buf, const unsigned int xid)
{
struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
struct tcon_link *tlink;
cifs: for compound requests, use open handle if possible For smb2_compound_op, it is possible to pass a ref to an already open file. We should be passing it whenever possible. i.e. if a matching handle is already kept open. If we don't do that, we will end up breaking leases for files kept open on the same client. Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-07-30 14:43:09 +08:00
struct cifs_tcon *tcon;
struct cifsFileInfo *cfile;
CIFS: Add set_file_info support for SMB2 Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:32 +08:00
int rc;
Do not send SMB3 SET_INFO request if nothing is changing [CIFS] We had cases where we sent a SMB2/SMB3 setinfo request with all timestamp (and DOS attribute) fields marked as 0 (ie do not change) e.g. on chmod or chown. Signed-off-by: Steve French <steve.french@primarydata.com> CC: Stable <stable@vger.kernel.org>
2016-09-27 03:23:08 +08:00
if ((buf->CreationTime == 0) && (buf->LastAccessTime == 0) &&
smb3: Do not send SMB3 SET_INFO if nothing changed An earlier commit had a typo which prevented the optimization from working: commit 18dd8e1a65dd ("Do not send SMB3 SET_INFO request if nothing is changing") Thank you to Metze for noticing this. Also clear a reserved field in the FILE_BASIC_INFO struct we send that should be zero (all the other fields in that struct were set or cleared explicitly already in cifs_set_file_info). Reviewed-by: Pavel Shilovsky <pshilov@microsoft.com> CC: Stable <stable@vger.kernel.org> # 4.9.x+ Reported-by: Stefan Metzmacher <metze@samba.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2018-08-03 09:28:18 +08:00
(buf->LastWriteTime == 0) && (buf->ChangeTime == 0) &&
Do not send SMB3 SET_INFO request if nothing is changing [CIFS] We had cases where we sent a SMB2/SMB3 setinfo request with all timestamp (and DOS attribute) fields marked as 0 (ie do not change) e.g. on chmod or chown. Signed-off-by: Steve French <steve.french@primarydata.com> CC: Stable <stable@vger.kernel.org>
2016-09-27 03:23:08 +08:00
(buf->Attributes == 0))
return 0; /* would be a no op, no sense sending this */
CIFS: Add set_file_info support for SMB2 Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:32 +08:00
tlink = cifs_sb_tlink(cifs_sb);
if (IS_ERR(tlink))
return PTR_ERR(tlink);
cifs: for compound requests, use open handle if possible For smb2_compound_op, it is possible to pass a ref to an already open file. We should be passing it whenever possible. i.e. if a matching handle is already kept open. If we don't do that, we will end up breaking leases for files kept open on the same client. Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-07-30 14:43:09 +08:00
tcon = tlink_tcon(tlink);
Do not send SMB3 SET_INFO request if nothing is changing [CIFS] We had cases where we sent a SMB2/SMB3 setinfo request with all timestamp (and DOS attribute) fields marked as 0 (ie do not change) e.g. on chmod or chown. Signed-off-by: Steve French <steve.french@primarydata.com> CC: Stable <stable@vger.kernel.org>
2016-09-27 03:23:08 +08:00
cifs: for compound requests, use open handle if possible For smb2_compound_op, it is possible to pass a ref to an already open file. We should be passing it whenever possible. i.e. if a matching handle is already kept open. If we don't do that, we will end up breaking leases for files kept open on the same client. Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2021-07-30 14:43:09 +08:00
cifs_get_writable_path(tcon, full_path, FIND_WR_ANY, &cfile);
rc = smb2_compound_op(xid, tcon, cifs_sb, full_path,
smb3: pass mode bits into create calls We need to populate an ACL (security descriptor open context) on file and directory correct. This patch passes in the mode. Followon patch will build the open context and the security descriptor (from the mode) that goes in the open context. Signed-off-by: Steve French <stfrench@microsoft.com> Reviewed-by: Aurelien Aptel <aaptel@suse.com>
2019-09-25 13:32:13 +08:00
FILE_WRITE_ATTRIBUTES, FILE_OPEN,
cifs: improve symlink handling for smb2+ When creating inode for symlink, the client used to send below requests to fill it in: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK) * create(+reparse_flag)+query_info+close (set file attrs) * create+ioctl(get_reparse)+close (query reparse tag) and then for every access to the symlink dentry, the ->link() method would send another: * create+ioctl(get_reparse)+close (parse symlink) So, in order to improve: (i) Get rid of unnecessary roundtrips and then resolve symlinks as follows: * create+query_info+close (STATUS_STOPPED_ON_SYMLINK + parse symlink + get reparse tag) * create(+reparse_flag)+query_info+close (set file attrs) (ii) Set the resolved symlink target directly in inode->i_link and use simple_get_link() for ->link() to simply return it. Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Reviewed-by: Ronnie Sahlberg <lsahlber@redhat.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-10-04 05:43:50 +08:00
0, ACL_NO_MODE, buf, SMB2_OP_SET_INFO, cfile,
cifs: Add "extbuf" and "extbuflen" args to smb2_compound_op() Will carry the variable-sized reply from SMB_FIND_FILE_POSIX_INFO Signed-off-by: Volker Lendecke <vl@samba.org> Reviewed-by: Paulo Alcantara (SUSE) <pc@cjr.nz> Signed-off-by: Steve French <stfrench@microsoft.com>
2022-11-25 19:26:00 +08:00
NULL, NULL, NULL, NULL);
CIFS: Add set_file_info support for SMB2 Signed-off-by: Pavel Shilovsky <pshilovsky@samba.org> Signed-off-by: Steve French <smfrench@gmail.com>
2012-09-19 07:20:32 +08:00
cifs_put_tlink(tlink);
return rc;
}