aboutsummaryrefslogtreecommitdiff
path: root/fs/smb/client
AgeCommit message (Collapse)AuthorFilesLines
2024-11-03smb: client: Fix use-after-free of network namespace.Kuniyuki Iwashima1-3/+11
Recently, we got a customer report that CIFS triggers oops while reconnecting to a server. [0] The workload runs on Kubernetes, and some pods mount CIFS servers in non-root network namespaces. The problem rarely happened, but it was always while the pod was dying. The root cause is wrong reference counting for network namespace. CIFS uses kernel sockets, which do not hold refcnt of the netns that the socket belongs to. That means CIFS must ensure the socket is always freed before its netns; otherwise, use-after-free happens. The repro steps are roughly: 1. mount CIFS in a non-root netns 2. drop packets from the netns 3. destroy the netns 4. unmount CIFS We can reproduce the issue quickly with the script [1] below and see the splat [2] if CONFIG_NET_NS_REFCNT_TRACKER is enabled. When the socket is TCP, it is hard to guarantee the netns lifetime without holding refcnt due to async timers. Let's hold netns refcnt for each socket as done for SMC in commit 9744d2bf1976 ("smc: Fix use-after-free in tcp_write_timer_handler()."). Note that we need to move put_net() from cifs_put_tcp_session() to clean_demultiplex_info(); otherwise, __sock_create() still could touch a freed netns while cifsd tries to reconnect from cifs_demultiplex_thread(). Also, maybe_get_net() cannot be put just before __sock_create() because the code is not under RCU and there is a small chance that the same address happened to be reallocated to another netns. [0]: CIFS: VFS: \\XXXXXXXXXXX has not responded in 15 seconds. Reconnecting... CIFS: Serverclose failed 4 times, giving up Unable to handle kernel paging request at virtual address 14de99e461f84a07 Mem abort info: ESR = 0x0000000096000004 EC = 0x25: DABT (current EL), IL = 32 bits SET = 0, FnV = 0 EA = 0, S1PTW = 0 FSC = 0x04: level 0 translation fault Data abort info: ISV = 0, ISS = 0x00000004 CM = 0, WnR = 0 [14de99e461f84a07] address between user and kernel address ranges Internal error: Oops: 0000000096000004 [#1] SMP Modules linked in: cls_bpf sch_ingress nls_utf8 cifs cifs_arc4 cifs_md4 dns_resolver tcp_diag inet_diag veth xt_state xt_connmark nf_conntrack_netlink xt_nat xt_statistic xt_MASQUERADE xt_mark xt_addrtype ipt_REJECT nf_reject_ipv4 nft_chain_nat nf_nat xt_conntrack nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 xt_comment nft_compat nf_tables nfnetlink overlay nls_ascii nls_cp437 sunrpc vfat fat aes_ce_blk aes_ce_cipher ghash_ce sm4_ce_cipher sm4 sm3_ce sm3 sha3_ce sha512_ce sha512_arm64 sha1_ce ena button sch_fq_codel loop fuse configfs dmi_sysfs sha2_ce sha256_arm64 dm_mirror dm_region_hash dm_log dm_mod dax efivarfs CPU: 5 PID: 2690970 Comm: cifsd Not tainted 6.1.103-109.184.amzn2023.aarch64 #1 Hardware name: Amazon EC2 r7g.4xlarge/, BIOS 1.0 11/1/2018 pstate: 00400005 (nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : fib_rules_lookup+0x44/0x238 lr : __fib_lookup+0x64/0xbc sp : ffff8000265db790 x29: ffff8000265db790 x28: 0000000000000000 x27: 000000000000bd01 x26: 0000000000000000 x25: ffff000b4baf8000 x24: ffff00047b5e4580 x23: ffff8000265db7e0 x22: 0000000000000000 x21: ffff00047b5e4500 x20: ffff0010e3f694f8 x19: 14de99e461f849f7 x18: 0000000000000000 x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000 x14: 0000000000000000 x13: 0000000000000000 x12: 3f92800abd010002 x11: 0000000000000001 x10: ffff0010e3f69420 x9 : ffff800008a6f294 x8 : 0000000000000000 x7 : 0000000000000006 x6 : 0000000000000000 x5 : 0000000000000001 x4 : ffff001924354280 x3 : ffff8000265db7e0 x2 : 0000000000000000 x1 : ffff0010e3f694f8 x0 : ffff00047b5e4500 Call trace: fib_rules_lookup+0x44/0x238 __fib_lookup+0x64/0xbc ip_route_output_key_hash_rcu+0x2c4/0x398 ip_route_output_key_hash+0x60/0x8c tcp_v4_connect+0x290/0x488 __inet_stream_connect+0x108/0x3d0 inet_stream_connect+0x50/0x78 kernel_connect+0x6c/0xac generic_ip_connect+0x10c/0x6c8 [cifs] __reconnect_target_unlocked+0xa0/0x214 [cifs] reconnect_dfs_server+0x144/0x460 [cifs] cifs_reconnect+0x88/0x148 [cifs] cifs_readv_from_socket+0x230/0x430 [cifs] cifs_read_from_socket+0x74/0xa8 [cifs] cifs_demultiplex_thread+0xf8/0x704 [cifs] kthread+0xd0/0xd4 Code: aa0003f8 f8480f13 eb18027f 540006c0 (b9401264) [1]: CIFS_CRED="/root/cred.cifs" CIFS_USER="Administrator" CIFS_PASS="Password" CIFS_IP="X.X.X.X" CIFS_PATH="//${CIFS_IP}/Users/Administrator/Desktop/CIFS_TEST" CIFS_MNT="/mnt/smb" DEV="enp0s3" cat <<EOF > ${CIFS_CRED} username=${CIFS_USER} password=${CIFS_PASS} domain=EXAMPLE.COM EOF unshare -n bash -c " mkdir -p ${CIFS_MNT} ip netns attach root 1 ip link add eth0 type veth peer veth0 netns root ip link set eth0 up ip -n root link set veth0 up ip addr add 192.168.0.2/24 dev eth0 ip -n root addr add 192.168.0.1/24 dev veth0 ip route add default via 192.168.0.1 dev eth0 ip netns exec root sysctl net.ipv4.ip_forward=1 ip netns exec root iptables -t nat -A POSTROUTING -s 192.168.0.2 -o ${DEV} -j MASQUERADE mount -t cifs ${CIFS_PATH} ${CIFS_MNT} -o vers=3.0,sec=ntlmssp,credentials=${CIFS_CRED},rsize=65536,wsize=65536,cache=none,echo_interval=1 touch ${CIFS_MNT}/a.txt ip netns exec root iptables -t nat -D POSTROUTING -s 192.168.0.2 -o ${DEV} -j MASQUERADE " umount ${CIFS_MNT} [2]: ref_tracker: net notrefcnt@000000004bbc008d has 1/1 users at sk_alloc (./include/net/net_namespace.h:339 net/core/sock.c:2227) inet_create (net/ipv4/af_inet.c:326 net/ipv4/af_inet.c:252) __sock_create (net/socket.c:1576) generic_ip_connect (fs/smb/client/connect.c:3075) cifs_get_tcp_session.part.0 (fs/smb/client/connect.c:3160 fs/smb/client/connect.c:1798) cifs_mount_get_session (fs/smb/client/trace.h:959 fs/smb/client/connect.c:3366) dfs_mount_share (fs/smb/client/dfs.c:63 fs/smb/client/dfs.c:285) cifs_mount (fs/smb/client/connect.c:3622) cifs_smb3_do_mount (fs/smb/client/cifsfs.c:949) smb3_get_tree (fs/smb/client/fs_context.c:784 fs/smb/client/fs_context.c:802 fs/smb/client/fs_context.c:794) vfs_get_tree (fs/super.c:1800) path_mount (fs/namespace.c:3508 fs/namespace.c:3834) __x64_sys_mount (fs/namespace.c:3848 fs/namespace.c:4057 fs/namespace.c:4034 fs/namespace.c:4034) do_syscall_64 (arch/x86/entry/common.c:52 arch/x86/entry/common.c:83) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130) Fixes: 26abe14379f8 ("net: Modify sk_alloc to not reference count the netns of kernel sockets.") Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Acked-by: Tom Talpey <tom@talpey.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-23cifs: fix warning when destroy 'cifs_io_request_pool'Ye Bin1-1/+1
There's a issue as follows: WARNING: CPU: 1 PID: 27826 at mm/slub.c:4698 free_large_kmalloc+0xac/0xe0 RIP: 0010:free_large_kmalloc+0xac/0xe0 Call Trace: <TASK> ? __warn+0xea/0x330 mempool_destroy+0x13f/0x1d0 init_cifs+0xa50/0xff0 [cifs] do_one_initcall+0xdc/0x550 do_init_module+0x22d/0x6b0 load_module+0x4e96/0x5ff0 init_module_from_file+0xcd/0x130 idempotent_init_module+0x330/0x620 __x64_sys_finit_module+0xb3/0x110 do_syscall_64+0xc1/0x1d0 entry_SYSCALL_64_after_hwframe+0x77/0x7f Obviously, 'cifs_io_request_pool' is not created by mempool_create(). So just use mempool_exit() to revert 'cifs_io_request_pool'. Fixes: edea94a69730 ("cifs: Add mempools for cifs_io_request and cifs_io_subrequest structs") Signed-off-by: Ye Bin <yebin10@huawei.com> Acked-by: David Howells <dhowells@redhat.com Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-23smb: client: Handle kstrdup failures for passwordsHenrique Carvalho1-0/+7
In smb3_reconfigure(), after duplicating ctx->password and ctx->password2 with kstrdup(), we need to check for allocation failures. If ses->password allocation fails, return -ENOMEM. If ses->password2 allocation fails, free ses->password, set it to NULL, and return -ENOMEM. Fixes: c1eb537bf456 ("cifs: allow changing password during remount") Reviewed-by: David Howells <dhowells@redhat.com Signed-off-by: Haoxiang Li <make24@iscas.ac.cn> Signed-off-by: Henrique Carvalho <henrique.carvalho@suse.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-16cifs: Remove unused functionsDr. David Alan Gilbert3-53/+0
cifs_ses_find_chan() has been unused since commit f486ef8e2003 ("cifs: use the chans_need_reconnect bitmap for reconnect status") cifs_read_page_from_socket() has been unused since commit d08089f649a0 ("cifs: Change the I/O paths to use an iterator rather than a page list") cifs_chan_in_reconnect() has been unused since commit bc962159e8e3 ("cifs: avoid race conditions with parallel reconnects") Remove them. Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-16smb/client: Fix logically dead codeAdvait Dhamorikar1-4/+0
The if condition in collect_sample: can never be satisfied because of a logical contradiction. The indicated dead code may have performed some action; that action will never occur. Fixes: 94ae8c3fee94 ("smb: client: compress: LZ77 code improvements cleanup") Signed-off-by: Advait Dhamorikar <advaitdhamorikar@gmail.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-16smb: client: fix OOBs when building SMB2_IOCTL requestPaulo Alcantara1-0/+9
When using encryption, either enforced by the server or when using 'seal' mount option, the client will squash all compound request buffers down for encryption into a single iov in smb2_set_next_command(). SMB2_ioctl_init() allocates a small buffer (448 bytes) to hold the SMB2_IOCTL request in the first iov, and if the user passes an input buffer that is greater than 328 bytes, smb2_set_next_command() will end up writing off the end of @rqst->iov[0].iov_base as shown below: mount.cifs //srv/share /mnt -o ...,seal ln -s $(perl -e "print('a')for 1..1024") /mnt/link BUG: KASAN: slab-out-of-bounds in smb2_set_next_command.cold+0x1d6/0x24c [cifs] Write of size 4116 at addr ffff8881148fcab8 by task ln/859 CPU: 1 UID: 0 PID: 859 Comm: ln Not tainted 6.12.0-rc3 #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-2.fc40 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x5d/0x80 ? smb2_set_next_command.cold+0x1d6/0x24c [cifs] print_report+0x156/0x4d9 ? smb2_set_next_command.cold+0x1d6/0x24c [cifs] ? __virt_addr_valid+0x145/0x310 ? __phys_addr+0x46/0x90 ? smb2_set_next_command.cold+0x1d6/0x24c [cifs] kasan_report+0xda/0x110 ? smb2_set_next_command.cold+0x1d6/0x24c [cifs] kasan_check_range+0x10f/0x1f0 __asan_memcpy+0x3c/0x60 smb2_set_next_command.cold+0x1d6/0x24c [cifs] smb2_compound_op+0x238c/0x3840 [cifs] ? kasan_save_track+0x14/0x30 ? kasan_save_free_info+0x3b/0x70 ? vfs_symlink+0x1a1/0x2c0 ? do_symlinkat+0x108/0x1c0 ? __pfx_smb2_compound_op+0x10/0x10 [cifs] ? kmem_cache_free+0x118/0x3e0 ? cifs_get_writable_path+0xeb/0x1a0 [cifs] smb2_get_reparse_inode+0x423/0x540 [cifs] ? __pfx_smb2_get_reparse_inode+0x10/0x10 [cifs] ? rcu_is_watching+0x20/0x50 ? __kmalloc_noprof+0x37c/0x480 ? smb2_create_reparse_symlink+0x257/0x490 [cifs] ? smb2_create_reparse_symlink+0x38f/0x490 [cifs] smb2_create_reparse_symlink+0x38f/0x490 [cifs] ? __pfx_smb2_create_reparse_symlink+0x10/0x10 [cifs] ? find_held_lock+0x8a/0xa0 ? hlock_class+0x32/0xb0 ? __build_path_from_dentry_optional_prefix+0x19d/0x2e0 [cifs] cifs_symlink+0x24f/0x960 [cifs] ? __pfx_make_vfsuid+0x10/0x10 ? __pfx_cifs_symlink+0x10/0x10 [cifs] ? make_vfsgid+0x6b/0xc0 ? generic_permission+0x96/0x2d0 vfs_symlink+0x1a1/0x2c0 do_symlinkat+0x108/0x1c0 ? __pfx_do_symlinkat+0x10/0x10 ? strncpy_from_user+0xaa/0x160 __x64_sys_symlinkat+0xb9/0xf0 do_syscall_64+0xbb/0x1d0 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f08d75c13bb Reported-by: David Howells <dhowells@redhat.com> Fixes: e77fe73c7e38 ("cifs: we can not use small padding iovs together with encryption") Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-16smb: client: fix possible double free in smb2_set_ea()Su Hui1-1/+2
Clang static checker(scan-build) warning： fs/smb/client/smb2ops.c:1304:2: Attempt to free released memory. 1304 | kfree(ea); | ^~~~~~~~~ There is a double free in such case: 'ea is initialized to NULL' -> 'first successful memory allocation for ea' -> 'something failed, goto sea_exit' -> 'first memory release for ea' -> 'goto replay_again' -> 'second goto sea_exit before allocate memory for ea' -> 'second memory release for ea resulted in double free'. Re-initialie 'ea' to NULL near to the replay_again label, it can fix this double free problem. Fixes: 4f1fffa23769 ("cifs: commands that are retried should have replay flag set") Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Su Hui <suhui@nfschina.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-06cifs: Fix creating native symlinks pointing to current or parent directoryPali Rohár1-3/+14
Calling 'ln -s . symlink' or 'ln -s .. symlink' creates symlink pointing to some object name which ends with U+F029 unicode codepoint. This is because trailing dot in the object name is replaced by non-ASCII unicode codepoint. So Linux SMB client currently is not able to create native symlink pointing to current or parent directory on Windows SMB server which can be read by either on local Windows server or by any other SMB client which does not implement compatible-reverse character replacement. Fix this problem in cifsConvertToUTF16() function which is doing that character replacement. Function comment already says that it does not need to handle special cases '.' and '..', but after introduction of native symlinks in reparse point form, this handling is needed. Note that this change depends on the previous change "cifs: Improve creating native symlinks pointing to directory". Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-06cifs: Improve creating native symlinks pointing to directoryPali Rohár3-4/+164
SMB protocol for native symlinks distinguish between symlink to directory and symlink to file. These two symlink types cannot be exchanged, which means that symlink of file type pointing to directory cannot be resolved at all (and vice-versa). Windows follows this rule for local filesystems (NTFS) and also for SMB. Linux SMB client currenly creates all native symlinks of file type. Which means that Windows (and some other SMB clients) cannot resolve symlinks pointing to directory created by Linux SMB client. As Linux system does not distinguish between directory and file symlinks, its API does not provide enough information for Linux SMB client during creating of native symlinks. Add some heuristic into the Linux SMB client for choosing the correct symlink type during symlink creation. Check if the symlink target location ends with slash, or last path component is dot or dot-dot, and check if the target location on SMB share exists and is a directory. If at least one condition is truth then create a new SMB symlink of directory type. Otherwise create it as file type symlink. This change improves interoperability with Windows systems. Windows systems would be able to resolve more SMB symlinks created by Linux SMB client which points to existing directory. Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-04Merge tag 'v6.12-rc1-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6Linus Torvalds23-56/+163
Pull smb client fixes from Steve French: - statfs fix (e.g. when limited access to root directory of share) - special file handling fixes: fix packet validation to avoid buffer overflow for reparse points, fixes for symlink path parsing (one for reparse points, and one for SFU use case), and fix for cleanup after failed SET_REPARSE operation. - fix for SMB2.1 signing bug introduced by recent patch to NFS symlink path, and NFS reparse point validation - comment cleanup * tag 'v6.12-rc1-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6: cifs: Do not convert delimiter when parsing NFS-style symlinks cifs: Validate content of NFS reparse point buffer cifs: Fix buffer overflow when parsing NFS reparse points smb: client: Correct typos in multiple comments across various files smb: client: use actual path when queryfs cifs: Remove intermediate object of failed create reparse call Revert "smb: client: make SHA-512 TFM ephemeral" smb: Update comments about some reparse point tags cifs: Check for UTF-16 null codepoint in SFU symlink target location
2024-10-03cifs: Do not convert delimiter when parsing NFS-style symlinksPali Rohár1-1/+0
NFS-style symlinks have target location always stored in NFS/UNIX form where backslash means the real UNIX backslash and not the SMB path separator. So do not mangle slash and backslash content of NFS-style symlink during readlink() syscall as it is already in the correct Linux form. This fixes interoperability of NFS-style symlinks with backslashes created by Linux NFS3 client throw Windows NFS server and retrieved by Linux SMB client throw Windows SMB server, where both Windows servers exports the same directory. Fixes: d5ecebc4900d ("smb3: Allow query of symlinks stored as reparse points") Acked-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-03cifs: Validate content of NFS reparse point bufferPali Rohár1-0/+23
Symlink target location stored in DataBuffer is encoded in UTF-16. So check that symlink DataBuffer length is non-zero and even number. And check that DataBuffer does not contain UTF-16 null codepoint because Linux cannot process symlink with null byte. DataBuffer for char and block devices is 8 bytes long as it contains two 32-bit numbers (major and minor). Add check for this. DataBuffer buffer for sockets and fifos zero-length. Add checks for this. Signed-off-by: Pali Rohár <pali@kernel.org> Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-03cifs: Fix buffer overflow when parsing NFS reparse pointsPali Rohár1-1/+14
ReparseDataLength is sum of the InodeType size and DataBuffer size. So to get DataBuffer size it is needed to subtract InodeType's size from ReparseDataLength. Function cifs_strndup_from_utf16() is currentlly accessing buf->DataBuffer at position after the end of the buffer because it does not subtract InodeType size from the length. Fix this problem and correctly subtract variable len. Member InodeType is present only when reparse buffer is large enough. Check for ReparseDataLength before accessing InodeType to prevent another invalid memory access. Major and minor rdev values are present also only when reparse buffer is large enough. Check for reparse buffer size before calling reparse_mkdev(). Fixes: d5ecebc4900d ("smb3: Allow query of symlinks stored as reparse points") Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-02smb: client: Correct typos in multiple comments across various filesShen Lichuan16-24/+24
Fixed some confusing typos that were currently identified witch codespell, the details are as follows: -in the code comments: fs/smb/client/cifsacl.h:58: inheritence ==> inheritance fs/smb/client/cifsencrypt.c:242: origiginal ==> original fs/smb/client/cifsfs.c:164: referece ==> reference fs/smb/client/cifsfs.c:292: ned ==> need fs/smb/client/cifsglob.h:779: initital ==> initial fs/smb/client/cifspdu.h:784: altetnative ==> alternative fs/smb/client/cifspdu.h:2409: conrol ==> control fs/smb/client/cifssmb.c:1218: Expirement ==> Experiment fs/smb/client/cifssmb.c:3021: conver ==> convert fs/smb/client/cifssmb.c:3998: asterik ==> asterisk fs/smb/client/file.c:2505: useable ==> usable fs/smb/client/fs_context.h:263: timemout ==> timeout fs/smb/client/misc.c:257: responsbility ==> responsibility fs/smb/client/netmisc.c:1006: divisable ==> divisible fs/smb/client/readdir.c:556: endianess ==> endianness fs/smb/client/readdir.c:818: bu ==> by fs/smb/client/smb2ops.c:2180: snaphots ==> snapshots fs/smb/client/smb2ops.c:3586: otions ==> options fs/smb/client/smb2pdu.c:2979: timestaps ==> timestamps fs/smb/client/smb2pdu.c:4574: memmory ==> memory fs/smb/client/smb2transport.c:699: origiginal ==> original fs/smb/client/smbdirect.c:222: happenes ==> happens fs/smb/client/smbdirect.c:1347: registartions ==> registrations fs/smb/client/smbdirect.h:114: accoutning ==> accounting Signed-off-by: Shen Lichuan <shenlichuan@vivo.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-10-02move asm/unaligned.h to linux/unaligned.hAl Viro2-2/+2
asm/unaligned.h is always an include of asm-generic/unaligned.h; might as well move that thing to linux/unaligned.h and include that - there's nothing arch-specific in that header. auto-generated by the following: for i in `git grep -l -w asm/unaligned.h`; do sed -i -e "s/asm\/unaligned.h/linux\/unaligned.h/" $i done for i in `git grep -l -w asm-generic/unaligned.h`; do sed -i -e "s/asm-generic\/unaligned.h/linux\/unaligned.h/" $i done git mv include/asm-generic/unaligned.h include/linux/unaligned.h git mv tools/include/asm-generic/unaligned.h tools/include/linux/unaligned.h sed -i -e "/unaligned.h/d" include/asm-generic/Kbuild sed -i -e "s/__ASM_GENERIC/__LINUX/" include/linux/unaligned.h tools/include/linux/unaligned.h
2024-10-01smb: client: use actual path when queryfswangrong4-10/+26
Due to server permission control, the client does not have access to the shared root directory, but can access subdirectories normally, so users usually mount the shared subdirectories directly. In this case, queryfs should use the actual path instead of the root directory to avoid the call returning an error (EACCES). Signed-off-by: wangrong <wangrong@uniontech.com> Reviewed-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-30cifs: Remove intermediate object of failed create reparse callPali Rohár1-2/+22
If CREATE was successful but SMB2_OP_SET_REPARSE failed then remove the intermediate object created by CREATE. Otherwise empty object stay on the server when reparse call failed. This ensures that if the creating of special files is unsupported by the server then no empty file stay on the server as a result of unsupported operation. Fixes: 102466f303ff ("smb: client: allow creating special files via reparse points") Signed-off-by: Pali Rohár <pali@kernel.org> Acked-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-30Revert "smb: client: make SHA-512 TFM ephemeral"Steve French6-17/+47
The original patch causes a crash with signed mounts when using the SMB2.1 dialect RIP: 0010:smb2_calc_signature+0x10e/0x460 [cifs] Code: 46 30 00 00 00 00 49 c7 46 38 00 00 00 00 0f 85 3e 01 00 00 48 8b 83 a8 02 00 00 48 89 85 68 ff ff ff 49 8b b4 24 58 01 00 00 <48> 8b 38 ba 10 00 00 00 e8 55 0f 0c e0 41 89 c7 85 c0 0f 85 44 01 RSP: 0018:ffffb349422fb5c8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff98028765b800 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffff980200f2b100 RDI: 0000000000000000 RBP: ffffb349422fb680 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000000 R12: ffff980235e37800 R13: ffffb349422fb900 R14: ffff98027c160700 R15: ffff98028765b820 FS: 000074139b98f780(0000) GS:ffff98097b980000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 000000011cb78006 CR4: 00000000003726f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> ? show_regs+0x6c/0x80 ? __die+0x24/0x80 ? page_fault_oops+0x175/0x5c0 ? hrtimer_try_to_cancel.part.0+0x55/0xf0 ? do_user_addr_fault+0x4b2/0x870 ? exc_page_fault+0x85/0x1c0 ? asm_exc_page_fault+0x27/0x30 ? smb2_calc_signature+0x10e/0x460 [cifs] ? smb2_calc_signature+0xa7/0x460 [cifs] ? kmem_cache_alloc_noprof+0x101/0x300 smb2_sign_rqst+0xa2/0xe0 [cifs] smb2_setup_request+0x12d/0x240 [cifs] compound_send_recv+0x304/0x1220 [cifs] cifs_send_recv+0x22/0x40 [cifs] SMB2_tcon+0x2d9/0x8c0 [cifs] cifs_get_smb_ses+0x910/0xef0 [cifs] ? cifs_get_smb_ses+0x910/0xef0 [cifs] cifs_mount_get_session+0x6a/0x250 [cifs] Reported-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Suggested-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com> This reverts commit 220d83b52c7d16ec3c168b82f4e6ce59c645f7ab.
2024-09-29cifs: Check for UTF-16 null codepoint in SFU symlink target locationPali Rohár1-1/+7
Check that read buffer of SFU symlink target location does not contain UTF-16 null codepoint (via UniStrnlen() call) because Linux cannot process symlink with null byte, it truncates everything in buffer after null byte. Fixes: cf2ce67345d6 ("cifs: Add support for reading SFU symlink location") Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-26smb: client: make SHA-512 TFM ephemeralEnzo Matsumiya6-47/+17
The SHA-512 shash TFM is used only briefly during Session Setup stage, when computing SMB 3.1.1 preauth hash. There's no need to keep it allocated in servers' secmech the whole time, so keep its lifetime inside smb311_update_preauth_hash(). This also makes smb311_crypto_shash_allocate() redundant, so expose smb3_crypto_shash_allocate() and use that. Signed-off-by: Enzo Matsumiya <ematsumiya@suse.de> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-26smb: client: make HMAC-MD5 TFM ephemeralEnzo Matsumiya2-84/+50
The HMAC-MD5 shash TFM is used only briefly during Session Setup stage, when computing NTLMv2 hashes. There's no need to keep it allocated in servers' secmech the whole time, so keep its lifetime inside setup_ntlmv2_rsp(). Signed-off-by: Enzo Matsumiya <ematsumiya@suse.de> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-26smb: client: stop flooding dmesg in smb2_calc_signature()Paulo Alcantara1-1/+1
When having several mounts that share same credential and the client couldn't re-establish an SMB session due to an expired kerberos ticket or rotated password, smb2_calc_signature() will end up flooding dmesg when not finding SMB sessions to calculate signatures. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-26smb: client: allocate crypto only for primary serverEnzo Matsumiya2-9/+18
For extra channels, point ->secmech.{enc,dec} to the primary server ones. Signed-off-by: Enzo Matsumiya <ematsumiya@suse.de> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-26smb: client: fix UAF in async decryptionEnzo Matsumiya2-19/+34
Doing an async decryption (large read) crashes with a slab-use-after-free way down in the crypto API. Reproducer: # mount.cifs -o ...,seal,esize=1 //srv/share /mnt # dd if=/mnt/largefile of=/dev/null ... [ 194.196391] ================================================================== [ 194.196844] BUG: KASAN: slab-use-after-free in gf128mul_4k_lle+0xc1/0x110 [ 194.197269] Read of size 8 at addr ffff888112bd0448 by task kworker/u77:2/899 [ 194.197707] [ 194.197818] CPU: 12 UID: 0 PID: 899 Comm: kworker/u77:2 Not tainted 6.11.0-lku-00028-gfca3ca14a17a-dirty #43 [ 194.198400] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.2-3-gd478f380-prebuilt.qemu.org 04/01/2014 [ 194.199046] Workqueue: smb3decryptd smb2_decrypt_offload [cifs] [ 194.200032] Call Trace: [ 194.200191] <TASK> [ 194.200327] dump_stack_lvl+0x4e/0x70 [ 194.200558] ? gf128mul_4k_lle+0xc1/0x110 [ 194.200809] print_report+0x174/0x505 [ 194.201040] ? __pfx__raw_spin_lock_irqsave+0x10/0x10 [ 194.201352] ? srso_return_thunk+0x5/0x5f [ 194.201604] ? __virt_addr_valid+0xdf/0x1c0 [ 194.201868] ? gf128mul_4k_lle+0xc1/0x110 [ 194.202128] kasan_report+0xc8/0x150 [ 194.202361] ? gf128mul_4k_lle+0xc1/0x110 [ 194.202616] gf128mul_4k_lle+0xc1/0x110 [ 194.202863] ghash_update+0x184/0x210 [ 194.203103] shash_ahash_update+0x184/0x2a0 [ 194.203377] ? __pfx_shash_ahash_update+0x10/0x10 [ 194.203651] ? srso_return_thunk+0x5/0x5f [ 194.203877] ? crypto_gcm_init_common+0x1ba/0x340 [ 194.204142] gcm_hash_assoc_remain_continue+0x10a/0x140 [ 194.204434] crypt_message+0xec1/0x10a0 [cifs] [ 194.206489] ? __pfx_crypt_message+0x10/0x10 [cifs] [ 194.208507] ? srso_return_thunk+0x5/0x5f [ 194.209205] ? srso_return_thunk+0x5/0x5f [ 194.209925] ? srso_return_thunk+0x5/0x5f [ 194.210443] ? srso_return_thunk+0x5/0x5f [ 194.211037] decrypt_raw_data+0x15f/0x250 [cifs] [ 194.212906] ? __pfx_decrypt_raw_data+0x10/0x10 [cifs] [ 194.214670] ? srso_return_thunk+0x5/0x5f [ 194.215193] smb2_decrypt_offload+0x12a/0x6c0 [cifs] This is because TFM is being used in parallel. Fix this by allocating a new AEAD TFM for async decryption, but keep the existing one for synchronous READ cases (similar to what is done in smb3_calc_signature()). Also remove the calls to aead_request_set_callback() and crypto_wait_req() since it's always going to be a synchronous operation. Signed-off-by: Enzo Matsumiya <ematsumiya@suse.de> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-26Merge tag 'v6.12-rc-smb3-client-fixes-part2' of ↵Linus Torvalds17-224/+281
git://git.samba.org/sfrench/cifs-2.6 Pull smb client fixes from Steve French: "Most are from the recent SMB3.1.1 test event, and also an important netfs fix for a cifs mtime write regression - fix mode reported by stat of readonly directories and files - DFS (global namespace) related fixes - fixes for special file support via reparse points - mount improvement and reconnect fix - fix for noisy log message on umount - two netfs related fixes, one fixing a recent regression, and add new write tracepoint" * tag 'v6.12-rc-smb3-client-fixes-part2' of git://git.samba.org/sfrench/cifs-2.6: netfs, cifs: Fix mtime/ctime update for mmapped writes cifs: update internal version number smb: client: print failed session logoffs with FYI cifs: Fix reversion of the iter in cifs_readv_receive(). smb3: fix incorrect mode displayed for read-only files smb: client: fix parsing of device numbers smb: client: set correct device number on nfs reparse points smb: client: propagate error from cifs_construct_tcon() smb: client: fix DFS failover in multiuser mounts cifs: Make the write_{enter,done,err} tracepoints display netfs info smb: client: fix DFS interlink failover smb: client: improve purging of cached referrals smb: client: avoid unnecessary reconnects when refreshing referrals
2024-09-24cifs: update internal version numberSteve French1-2/+2
To 2.51 Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb: client: print failed session logoffs with FYIPaulo Alcantara1-2/+1
Do not flood dmesg with failed session logoffs as kerberos tickets getting expired or passwords being rotated is a very common scenario. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24cifs: Fix reversion of the iter in cifs_readv_receive().David Howells3-11/+7
cifs_read_iter_from_socket() copies the iterator that's passed in for the socket to modify as and if it will, and then advances the original iterator by the amount sent. However, both callers revert the advancement (although receive_encrypted_read() zeros beyond the iterator first). The problem is, though, that cifs_readv_receive() reverts by the original length, not the amount transmitted which can cause an oops in iov_iter_revert(). Fix this by: (1) Remove the iov_iter_advance() from cifs_read_iter_from_socket(). (2) Remove the iov_iter_revert() from both callers. This fixes the bug in cifs_readv_receive(). (3) In receive_encrypted_read(), if we didn't get back as much data as the buffer will hold, copy the iterator, advance the copy and use the copy to drive iov_iter_zero(). As a bonus, this gets rid of some unnecessary work. This was triggered by generic/074 with the "-o sign" mount option. Fixes: 3ee1a1fc3981 ("cifs: Cut over to using netfslib") Signed-off-by: David Howells <dhowells@redhat.com> cc: Steve French <sfrench@samba.org> cc: Paulo Alcantara <pc@manguebit.com> cc: Shyam Prasad N <nspmangalore@gmail.com> cc: Rohith Surabattula <rohiths.msft@gmail.com> cc: Jeff Layton <jlayton@kernel.org> cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb3: fix incorrect mode displayed for read-only filesSteve French1-8/+11
Commands like "chmod 0444" mark a file readonly via the attribute flag (when mapping of mode bits into the ACL are not set, or POSIX extensions are not negotiated), but they were not reported correctly for stat of directories (they were reported ok for files and for "ls"). See example below: root:~# ls /mnt2 -l total 12 drwxr-xr-x 2 root root 0 Sep 21 18:03 normaldir -rwxr-xr-x 1 root root 0 Sep 21 23:24 normalfile dr-xr-xr-x 2 root root 0 Sep 21 17:55 readonly-dir -r-xr-xr-x 1 root root 209716224 Sep 21 18:15 readonly-file root:~# stat -c %a /mnt2/readonly-dir 755 root:~# stat -c %a /mnt2/readonly-file 555 This fixes the stat of directories when ATTR_READONLY is set (in cases where the mode can not be obtained other ways). root:~# stat -c %a /mnt2/readonly-dir 555 Cc: stable@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb: client: fix parsing of device numbersPaulo Alcantara2-11/+4
Report correct major and minor numbers from special files created with NFS reparse points. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb: client: set correct device number on nfs reparse pointsPaulo Alcantara1-2/+2
Fix major and minor numbers set on special files created with NFS reparse points. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb: client: propagate error from cifs_construct_tcon()Paulo Alcantara1-6/+10
Propagate error from cifs_construct_tcon() in cifs_sb_tlink() instead of always returning -EACCES. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb: client: fix DFS failover in multiuser mountsPaulo Alcantara1-1/+2
For sessions and tcons created on behalf of new users accessing a multiuser mount, matching their sessions in tcon_super_cb() with master tcon will always lead to false as every new user will have its own session and tcon. All multiuser sessions, however, will inherit ->dfs_root_ses from master tcon, so match it instead. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24cifs: Make the write_{enter,done,err} tracepoints display netfs infoDavid Howells2-10/+18
Make the write RPC tracepoints use the same trace macro complexes as the read tracepoints and display the netfs request and subrequest IDs where available (see commit 519be989717c "cifs: Add a tracepoint to track credits involved in R/W requests"). Signed-off-by: David Howells <dhowells@redhat.com> cc: Steve French <stfrench@microsoft.com> cc: Paulo Alcantara (Red Hat) <pc@manguebit.com> cc: Jeff Layton <jlayton@kernel.org> cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb: client: fix DFS interlink failoverPaulo Alcantara9-86/+94
The DFS interlinks point to different DFS namespaces so make sure to use the correct DFS root server to chase any DFS links under it by storing the SMB session in dfs_ref_walk structure and then using it on every referral walk. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb: client: improve purging of cached referralsPaulo Alcantara1-17/+15
Purge cached referrals that have a single target when reaching maximum of cache size as the client won't need them to failover. Otherwise remove oldest cache entry. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-24smb: client: avoid unnecessary reconnects when refreshing referralsPaulo Alcantara1-70/+117
Do not mark tcons for reconnect when current connection matches any of the targets returned by new referral even when there is no cached entry. Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-23Merge tag 'pull-stable-struct_fd' of ↵Linus Torvalds1-4/+4
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs Pull 'struct fd' updates from Al Viro: "Just the 'struct fd' layout change, with conversion to accessor helpers" * tag 'pull-stable-struct_fd' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: add struct fd constructors, get rid of __to_fd() struct fd: representation change introduce fd_file(), convert all accessors to it.
2024-09-19Merge tag 'v6.12-rc-smb3-client-fixes-part1' of ↵Linus Torvalds34-2098/+1072
git://git.samba.org/sfrench/cifs-2.6 Pull smb client updates from Steve French: - cleanups (moving duplicated code, removing unused code etc) - fixes relating to "sfu" mount options (for better handling special file types) - SMB3.1.1 compression fixes/improvements * tag 'v6.12-rc-smb3-client-fixes-part1' of git://git.samba.org/sfrench/cifs-2.6: (24 commits) smb: client: fix compression heuristic functions cifs: Update SFU comments about fifos and sockets cifs: Add support for creating SFU symlinks smb: use LIST_HEAD() to simplify code cifs: Recognize SFU socket type cifs: Show debug message when SFU Fifo type was detected cifs: Put explicit zero byte into SFU block/char types cifs: Add support for reading SFU symlink location cifs: Fix recognizing SFU symlinks smb: client: compress: fix an "illegal accesses" issue smb: client: compress: fix a potential issue of freeing an invalid pointer smb: client: compress: LZ77 code improvements cleanup smb: client: insert compression check/call on write requests smb3: mark compression as CONFIG_EXPERIMENTAL and fix missing compression operation cifs: Remove obsoleted declaration for cifs_dir_open smb: client: Use min() macro cifs: convert to use ERR_CAST() smb: add comment to STATUS_MCA_OCCURED smb: move SMB2 Status code to common header file smb: move some duplicate definitions to common/smbacl.h ...
2024-09-16smb: client: fix compression heuristic functionsEnzo Matsumiya1-50/+55
Change is_compressible() return type to bool, use WARN_ON_ONCE(1) for internal errors and return false for those. Renames: check_repeated_data -> has_repeated_data check_ascii_bytes -> is_mostly_ascii (also refactor into a single loop) calc_shannon_entropy -> has_low_entropy Also wraps "wreq->Length" in le32_to_cpu() in should_compress() (caught by sparse). Signed-off-by: Enzo Matsumiya <ematsumiya@suse.de> Suggested-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-16cifs: Update SFU comments about fifos and socketsPali Rohár3-6/+6
In SFU mode, activated by -o sfu mount option is now also support for creating new fifos and sockets. Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-16cifs: Add support for creating SFU symlinksPali Rohár5-29/+77
Linux cifs client can already detect SFU symlinks and reads it content (target location). But currently is not able to create new symlink. So implement this missing support. When 'sfu' mount option is specified and 'mfsymlinks' is not specified then create new symlinks in SFU-style. This will provide full SFU compatibility of symlinks when mounting cifs share with 'sfu' option. 'mfsymlinks' option override SFU for better Apple compatibility as explained in fs_context.c file in smb3_update_mnt_flags() function. Extend __cifs_sfu_make_node() function, which now can handle also S_IFLNK type and refactor structures passed to sync_write() in this function, by splitting SFU type and SFU data from original combined struct win_dev as combined fixed-length struct cannot be used for variable-length symlinks. Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-16cifs: Remove redundant setting of NETFS_SREQ_HIT_EOFDavid Howells1-2/+0
Fix an upstream merge resolution issue[1]. The NETFS_SREQ_HIT_EOF flag, and code to set it, got added via two different paths. The original path saw it added in the netfslib read improvements[2], but it was also added, and slightly differently, in a fix that was committed before v6.11: 1da29f2c39b67b846b74205c81bf0ccd96d34727 netfs, cifs: Fix handling of short DIO read However, the code added to smb2_readv_callback() to set the flag in didn't get removed when the netfs read improvements series was rebased to take account of the cifs fixes. The proposed merge resolution[2] deleted it rather than rebase the patches. Fix this by removing the redundant lines. Code to set the bit that derives from the fix patch is still there, a few lines above in the source. Fixes: 35219bc5c71f ("Merge tag 'vfs-6.12.netfs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs") Signed-off-by: David Howells <dhowells@redhat.com> cc: Steve French <stfrench@microsoft.com> cc: Paulo Alcantara <pc@manguebit.com> cc: Christian Brauner <brauner@kernel.org> cc: Jeff Layton <jlayton@kernel.org> cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org Link: https://lore.kernel.org/r/CAHk-=wjr8fxk20-wx=63mZruW1LTvBvAKya1GQ1EhyzXb-okMA@mail.gmail.com/ [1] Link: https://lore.kernel.org/linux-fsdevel/20240913-vfs-netfs-39ef6f974061@brauner/ [2] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-09-16cifs: Fix cifs readv callback merge resolution issueDavid Howells1-1/+1
Fix an upstream merge resolution issue[1]. Prior to the netfs read healpers, the SMB1 asynchronous read callback, cifs_readv_worker() performed the cleanup for the operation in the network message processing loop, potentially slowing down the processing of incoming SMB messages. With commit a68c74865f51 ("cifs: Fix SMB1 readv/writev callback in the same way as SMB2/3"), this was moved to a worker thread (as is done in the SMB2/3 transport variant). However, the "was_async" argument to netfs_subreq_terminated (which was originally incorrectly "false" got flipped to "true" - which was then incorrect because, being in a kernel thread, it's not in an async context). This got corrected in the sample merge[2], but Linus, not unreasonably, switched it back to its previous value. Note that this value tells netfslib whether or not it can run sleepable stuff or stuff that takes a long time, such as retries and cleanups, in the calling thread, or whether it should offload to a worker thread. Fix this so that it is "false". The callback to netfslib in both SMB1 and SMB2/3 now gets offloaded from the network message thread to a separate worker thread and thus it's fine to do the slow work in this thread. Fixes: 35219bc5c71f ("Merge tag 'vfs-6.12.netfs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs") Signed-off-by: David Howells <dhowells@redhat.com> cc: Steve French <stfrench@microsoft.com> cc: Paulo Alcantara <pc@manguebit.com> cc: Christian Brauner <brauner@kernel.org> cc: Jeff Layton <jlayton@kernel.org> cc: linux-cifs@vger.kernel.org cc: netfs@lists.linux.dev cc: linux-fsdevel@vger.kernel.org Link: https://lore.kernel.org/r/CAHk-=wjr8fxk20-wx=63mZruW1LTvBvAKya1GQ1EhyzXb-okMA@mail.gmail.com/ [1] Link: https://lore.kernel.org/linux-fsdevel/20240913-vfs-netfs-39ef6f974061@brauner/ [2] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-09-16Merge tag 'vfs-6.12.netfs' of ↵Linus Torvalds7-344/+239
git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs Pull netfs updates from Christian Brauner: "This contains the work to improve read/write performance for the new netfs library. The main performance enhancing changes are: - Define a structure, struct folio_queue, and a new iterator type, ITER_FOLIOQ, to hold a buffer as a replacement for ITER_XARRAY. See that patch for questions about naming and form. ITER_FOLIOQ is provided as a replacement for ITER_XARRAY. The problem with an xarray is that accessing it requires the use of a lock (typically the RCU read lock) - and this means that we can't supply iterate_and_advance() with a step function that might sleep (crypto for example) without having to drop the lock between pages. ITER_FOLIOQ is the iterator for a chain of folio_queue structs, where each folio_queue holds a small list of folios. A folio_queue struct is a simpler structure than xarray and is not subject to concurrent manipulation by the VM. folio_queue is used rather than a bvec[] as it can form lists of indefinite size, adding to one end and removing from the other on the fly. - Provide a copy_folio_from_iter() wrapper. - Make cifs RDMA support ITER_FOLIOQ. - Use folio queues in the write-side helpers instead of xarrays. - Add a function to reset the iterator in a subrequest. - Simplify the write-side helpers to use sheaves to skip gaps rather than trying to work out where gaps are. - In afs, make the read subrequests asynchronous, putting them into work items to allow the next patch to do progressive unlocking/reading. - Overhaul the read-side helpers to improve performance. - Fix the caching of a partial block at the end of a file. - Allow a store to be cancelled. Then some changes for cifs to make it use folio queues instead of xarrays for crypto bufferage: - Use raw iteration functions rather than manually coding iteration when hashing data. - Switch to using folio_queue for crypto buffers. - Remove the xarray bits. Make some adjustments to the /proc/fs/netfs/stats file such that: - All the netfs stats lines begin 'Netfs:' but change this to something a bit more useful. - Add a couple of stats counters to track the numbers of skips and waits on the per-inode writeback serialisation lock to make it easier to check for this as a source of performance loss. Miscellaneous work: - Ensure that the sb_writers lock is taken around vfs_{set,remove}xattr() in the cachefiles code. - Reduce the number of conditional branches in netfs_perform_write(). - Move the CIFS_INO_MODIFIED_ATTR flag to the netfs_inode struct and remove cifs_post_modify(). - Move the max_len/max_nr_segs members from netfs_io_subrequest to netfs_io_request as they're only needed for one subreq at a time. - Add an 'unknown' source value for tracing purposes. - Remove NETFS_COPY_TO_CACHE as it's no longer used. - Set the request work function up front at allocation time. - Use bh-disabling spinlocks for rreq->lock as cachefiles completion may be run from block-filesystem DIO completion in softirq context. - Remove fs/netfs/io.c" * tag 'vfs-6.12.netfs' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (25 commits) docs: filesystems: corrected grammar of netfs page cifs: Don't support ITER_XARRAY cifs: Switch crypto buffer to use a folio_queue rather than an xarray cifs: Use iterate_and_advance*() routines directly for hashing netfs: Cancel dirty folios that have no storage destination cachefiles, netfs: Fix write to partial block at EOF netfs: Remove fs/netfs/io.c netfs: Speed up buffered reading afs: Make read subreqs async netfs: Simplify the writeback code netfs: Provide an iterator-reset function netfs: Use new folio_queue data type and iterator instead of xarray iter cifs: Provide the capability to extract from ITER_FOLIOQ to RDMA SGEs iov_iter: Provide copy_folio_from_iter() mm: Define struct folio_queue and ITER_FOLIOQ to handle a sequence of folios netfs: Use bh-disabling spinlocks for rreq->lock netfs: Set the request work function upon allocation netfs: Remove NETFS_COPY_TO_CACHE netfs: Reserve netfs_sreq_source 0 as unset/unknown netfs: Move max_len/max_nr_segs from netfs_io_subrequest to netfs_io_stream ...
2024-09-15smb: use LIST_HEAD() to simplify codeHongbo Li4-16/+7
list_head can be initialized automatically with LIST_HEAD() instead of calling INIT_LIST_HEAD(). No functional impact. Signed-off-by: Hongbo Li <lihongbo22@huawei.com> Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-15cifs: Recognize SFU socket typePali Rohár1-1/+5
SFU since its (first) version 3.0 supports AF_LOCAL sockets and stores them on filesytem as system file with one zero byte. Add support for detecting this SFU socket type into cifs_sfu_type() function. With this change cifs_sfu_type() would correctly detect all special file types created by SFU: fifo, socket, symlink, block and char. Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-15cifs: Show debug message when SFU Fifo type was detectedPali Rohár1-0/+1
For debugging purposes it is a good idea to show detected SFU type also for Fifo. Debug message is already print for all other special types. Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-15cifs: Put explicit zero byte into SFU block/char typesPali Rohár2-4/+4
SFU types IntxCHR and IntxBLK are 8 bytes with zero as last byte. Make it explicit in memcpy and memset calls, so the zero byte is visible in the code (and not hidden as string trailing nul byte). It is important for reader to show the last byte for block and char types because it differs from the last byte of symlink type (which has it 0x01). Also it is important to show that the type is not nul-term string, but rather 8 bytes (with some printable bytes). Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2024-09-15cifs: Add support for reading SFU symlink locationPali Rohár1-0/+29
Currently when sfu mount option is specified then CIFS can recognize SFU symlink, but is not able to read symlink target location. readlink() syscall just returns that operation is not supported. Implement this missing functionality in cifs_sfu_type() function. Read target location of SFU-style symlink, parse it and fill into fattr's cf_symlink_target member. SFU-style symlink is file which has system attribute set and file content is buffer "IntxLNK\1" (8th byte is 0x01) followed by the target location encoded in little endian UCS-2/UTF-16. This format was introduced in Interix 3.0 subsystem, as part of the Microsoft SFU 3.0 and is used also by all later versions. Previous versions had no symlink support. Signed-off-by: Pali Rohár <pali@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
Powered by cgit, Gruvboxed by Blaster4385.