summaryrefslogtreecommitdiff
path: root/fs
AgeCommit message (Collapse)Author
2025-01-16xfs: check for dead buffers in xfs_buf_find_insertChristoph Hellwig
Commit 32dd4f9c506b ("xfs: remove a superflous hash lookup when inserting new buffers") converted xfs_buf_find_insert to use rhashtable_lookup_get_insert_fast and thus an operation that returns the existing buffer when an insert would duplicate the hash key. But this code path misses the check for a buffer with a reference count of zero, which could lead to reusing an about to be freed buffer. Fix this by using the same atomic_inc_not_zero pattern as xfs_buf_insert. Fixes: 32dd4f9c506b ("xfs: remove a superflous hash lookup when inserting new buffers") Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Cc: stable@vger.kernel.org # v6.0 Signed-off-by: Carlos Maiolino <cem@kernel.org>
2025-01-15ksmbd: fix integer overflows on 32 bit systemsDan Carpenter
On 32bit systems the addition operations in ipc_msg_alloc() can potentially overflow leading to memory corruption. Add bounds checking using KSMBD_IPC_MAX_PAYLOAD to avoid overflow. Fixes: 0626e6641f6b ("cifsd: add server handler for central processing and tranport layers") Cc: stable@vger.kernel.org Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Namjae Jeon <linkinjeon@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2025-01-15ksmbd: browse interfaces list on FSCTL_QUERY_INTERFACE_INFO IOCTLNamjae Jeon
ksmbd.mount will give each interfaces list and bind_interfaces_only flags to ksmbd server. Previously, the interfaces list was sent only when bind_interfaces_only was enabled. ksmbd server browse only interfaces list given from ksmbd.conf on FSCTL_QUERY_INTERFACE_INFO IOCTL. Signed-off-by: Namjae Jeon <linkinjeon@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2025-01-15ksmbd: Remove unused functionsDr. David Alan Gilbert
ksmbd_rpc_rap() was added in 2021 as part of commit 0626e6641f6b ("cifsd: add server handler for central processing and tranport layers") ksmbd_vfs_posix_lock_wait_timeout() was added in 2021 as part of commit f44158485826 ("cifsd: add file operations") both have remained unused. Remove them. Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org> Acked-by: Namjae Jeon <linkinjeon@kernel.org> Signed-off-by: Steve French <stfrench@microsoft.com>
2025-01-15ocfs2: check dir i_size in ocfs2_find_entrySu Yue
syz reports an out of bounds read: ================================================================== BUG: KASAN: slab-out-of-bounds in ocfs2_match fs/ocfs2/dir.c:334 [inline] BUG: KASAN: slab-out-of-bounds in ocfs2_search_dirblock+0x283/0x6e0 fs/ocfs2/dir.c:367 Read of size 1 at addr ffff88804d8b9982 by task syz-executor.2/14802 CPU: 0 UID: 0 PID: 14802 Comm: syz-executor.2 Not tainted 6.13.0-rc4 #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 Sched_ext: serialise (enabled+all), task: runnable_at=-10ms Call Trace: <TASK> __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x229/0x350 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:378 [inline] print_report+0x164/0x530 mm/kasan/report.c:489 kasan_report+0x147/0x180 mm/kasan/report.c:602 ocfs2_match fs/ocfs2/dir.c:334 [inline] ocfs2_search_dirblock+0x283/0x6e0 fs/ocfs2/dir.c:367 ocfs2_find_entry_id fs/ocfs2/dir.c:414 [inline] ocfs2_find_entry+0x1143/0x2db0 fs/ocfs2/dir.c:1078 ocfs2_find_files_on_disk+0x18e/0x530 fs/ocfs2/dir.c:1981 ocfs2_lookup_ino_from_name+0xb6/0x110 fs/ocfs2/dir.c:2003 ocfs2_lookup+0x30a/0xd40 fs/ocfs2/namei.c:122 lookup_open fs/namei.c:3627 [inline] open_last_lookups fs/namei.c:3748 [inline] path_openat+0x145a/0x3870 fs/namei.c:3984 do_filp_open+0xe9/0x1c0 fs/namei.c:4014 do_sys_openat2+0x135/0x1d0 fs/open.c:1402 do_sys_open fs/open.c:1417 [inline] __do_sys_openat fs/open.c:1433 [inline] __se_sys_openat fs/open.c:1428 [inline] __x64_sys_openat+0x15d/0x1c0 fs/open.c:1428 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf6/0x210 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f01076903ad Code: c3 e8 a7 2b 00 00 0f 1f 80 00 00 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b0 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f01084acfc8 EFLAGS: 00000246 ORIG_RAX: 0000000000000101 RAX: ffffffffffffffda RBX: 00007f01077cbf80 RCX: 00007f01076903ad RDX: 0000000000105042 RSI: 0000000020000080 RDI: ffffffffffffff9c RBP: 00007f01077cbf80 R08: 0000000000000000 R09: 0000000000000000 R10: 00000000000001ff R11: 0000000000000246 R12: 0000000000000000 R13: 00007f01077cbf80 R14: 00007f010764fc90 R15: 00007f010848d000 </TASK> ================================================================== And a general protection fault in ocfs2_prepare_dir_for_insert: ================================================================== loop0: detected capacity change from 0 to 32768 JBD2: Ignoring recovery information on journal ocfs2: Mounting device (7,0) on (node local, slot 0) with ordered data mode. Oops: general protection fault, probably for non-canonical address 0xdffffc0000000001: 0000 [#1] PREEMPT SMP KASAN NOPTI KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f] CPU: 0 UID: 0 PID: 5096 Comm: syz-executor792 Not tainted 6.11.0-rc4-syzkaller-00002-gb0da640826ba #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014 RIP: 0010:ocfs2_find_dir_space_id fs/ocfs2/dir.c:3406 [inline] RIP: 0010:ocfs2_prepare_dir_for_insert+0x3309/0x5c70 fs/ocfs2/dir.c:4280 Code: 00 00 e8 2a 25 13 fe e9 ba 06 00 00 e8 20 25 13 fe e9 4f 01 00 00 e8 16 25 13 fe 49 8d 7f 08 49 8d 5f 09 48 89 f8 48 c1 e8 03 <42> 0f b6 04 20 84 c0 0f 85 bd 23 00 00 48 89 d8 48 c1 e8 03 42 0f RSP: 0018:ffffc9000af9f020 EFLAGS: 00010202 RAX: 0000000000000001 RBX: 0000000000000009 RCX: ffff88801e27a440 RDX: 0000000000000000 RSI: 0000000000000400 RDI: 0000000000000008 RBP: ffffc9000af9f830 R08: ffffffff8380395b R09: ffffffff838090a7 R10: 0000000000000002 R11: ffff88801e27a440 R12: dffffc0000000000 R13: ffff88803c660878 R14: f700000000000088 R15: 0000000000000000 FS: 000055555a677380(0000) GS:ffff888020800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000560bce569178 CR3: 000000001de5a000 CR4: 0000000000350ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> ocfs2_mknod+0xcaf/0x2b40 fs/ocfs2/namei.c:292 vfs_mknod+0x36d/0x3b0 fs/namei.c:4088 do_mknodat+0x3ec/0x5b0 __do_sys_mknodat fs/namei.c:4166 [inline] __se_sys_mknodat fs/namei.c:4163 [inline] __x64_sys_mknodat+0xa7/0xc0 fs/namei.c:4163 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f2dafda3a99 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 f1 17 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffe336a6658 EFLAGS: 00000246 ORIG_RAX: 0000000000000103 RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f2dafda3a99 RDX: 00000000000021c0 RSI: 0000000020000040 RDI: 00000000ffffff9c RBP: 00007f2dafe1b5f0 R08: 0000000000004480 R09: 000055555a6784c0 R10: 0000000000000103 R11: 0000000000000246 R12: 00007ffe336a6680 R13: 00007ffe336a68a8 R14: 431bde82d7b634db R15: 00007f2dafdec03b </TASK> ================================================================== The two reports are all caused invalid negative i_size of dir inode. For ocfs2, dir_inode can't be negative or zero. Here add a check in which is called by ocfs2_check_dir_for_entry(). It fixes the second report as ocfs2_check_dir_for_entry() must be called before ocfs2_prepare_dir_for_insert(). Also set a up limit for dir with OCFS2_INLINE_DATA_FL. The i_size can't be great than blocksize. Link: https://lkml.kernel.org/r/20250106140640.92260-1-glass.su@suse.com Reported-by: Jiacheng Xu <stitch@zju.edu.cn> Link: https://lore.kernel.org/ocfs2-devel/17a04f01.1ae74.19436d003fc.Coremail.stitch@zju.edu.cn/T/#u Reported-by: syzbot+5a64828fcc4c2ad9b04f@syzkaller.appspotmail.com Link: https://lore.kernel.org/all/0000000000005894f3062018caf1@google.com/T/ Signed-off-by: Su Yue <glass.su@suse.com> Reviewed-by: Heming Zhao <heming.zhao@suse.com> Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Changwei Ge <gechangwei@live.cn> Cc: Jun Piao <piaojun@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
2025-01-15smb: client: fix double free of TCP_Server_Info::hostnamePaulo Alcantara
When shutting down the server in cifs_put_tcp_session(), cifsd thread might be reconnecting to multiple DFS targets before it realizes it should exit the loop, so @server->hostname can't be freed as long as cifsd thread isn't done. Otherwise the following can happen: RIP: 0010:__slab_free+0x223/0x3c0 Code: 5e 41 5f c3 cc cc cc cc 4c 89 de 4c 89 cf 44 89 44 24 08 4c 89 1c 24 e8 fb cf 8e 00 44 8b 44 24 08 4c 8b 1c 24 e9 5f fe ff ff <0f> 0b 41 f7 45 08 00 0d 21 00 0f 85 2d ff ff ff e9 1f ff ff ff 80 RSP: 0018:ffffb26180dbfd08 EFLAGS: 00010246 RAX: ffff8ea34728e510 RBX: ffff8ea34728e500 RCX: 0000000000800068 RDX: 0000000000800068 RSI: 0000000000000000 RDI: ffff8ea340042400 RBP: ffffe112041ca380 R08: 0000000000000001 R09: 0000000000000000 R10: 6170732e31303000 R11: 70726f632e786563 R12: ffff8ea34728e500 R13: ffff8ea340042400 R14: ffff8ea34728e500 R15: 0000000000800068 FS: 0000000000000000(0000) GS:ffff8ea66fd80000(0000) 000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007ffc25376080 CR3: 000000012a2ba001 CR4: PKRU: 55555554 Call Trace: <TASK> ? show_trace_log_lvl+0x1c4/0x2df ? show_trace_log_lvl+0x1c4/0x2df ? __reconnect_target_unlocked+0x3e/0x160 [cifs] ? __die_body.cold+0x8/0xd ? die+0x2b/0x50 ? do_trap+0xce/0x120 ? __slab_free+0x223/0x3c0 ? do_error_trap+0x65/0x80 ? __slab_free+0x223/0x3c0 ? exc_invalid_op+0x4e/0x70 ? __slab_free+0x223/0x3c0 ? asm_exc_invalid_op+0x16/0x20 ? __slab_free+0x223/0x3c0 ? extract_hostname+0x5c/0xa0 [cifs] ? extract_hostname+0x5c/0xa0 [cifs] ? __kmalloc+0x4b/0x140 __reconnect_target_unlocked+0x3e/0x160 [cifs] reconnect_dfs_server+0x145/0x430 [cifs] cifs_handle_standard+0x1ad/0x1d0 [cifs] cifs_demultiplex_thread+0x592/0x730 [cifs] ? __pfx_cifs_demultiplex_thread+0x10/0x10 [cifs] kthread+0xdd/0x100 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x29/0x50 </TASK> Fixes: 7be3248f3139 ("cifs: To match file servers, make sure the server hostname matches") Reported-by: Jay Shin <jaeshin@redhat.com> Signed-off-by: Paulo Alcantara (Red Hat) <pc@manguebit.com> Signed-off-by: Steve French <stfrench@microsoft.com>
2025-01-15bcachefs: Fix check_inode_hash_info_matches_root()Kent Overstreet
Can't use memcmp() when the struct contains padding. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-15ceph: fix memory leak in ceph_mds_auth_match()Antoine Viallon
We now free the temporary target path substring allocation on every possible branch, instead of omitting the default branch. In some cases, a memory leak occured, which could rapidly crash the system (depending on how many file accesses were attempted). This was detected in production because it caused a continuous memory growth, eventually triggering kernel OOM and completely hard-locking the kernel. Relevant kmemleak stacktrace: unreferenced object 0xffff888131e69900 (size 128): comm "git", pid 66104, jiffies 4295435999 hex dump (first 32 bytes): 76 6f 6c 75 6d 65 73 2f 63 6f 6e 74 61 69 6e 65 volumes/containe 72 73 2f 67 69 74 65 61 2f 67 69 74 65 61 2f 67 rs/gitea/gitea/g backtrace (crc 2f3bb450): [<ffffffffaa68fb49>] __kmalloc_noprof+0x359/0x510 [<ffffffffc32bf1df>] ceph_mds_check_access+0x5bf/0x14e0 [ceph] [<ffffffffc3235722>] ceph_open+0x312/0xd80 [ceph] [<ffffffffaa7dd786>] do_dentry_open+0x456/0x1120 [<ffffffffaa7e3729>] vfs_open+0x79/0x360 [<ffffffffaa832875>] path_openat+0x1de5/0x4390 [<ffffffffaa834fcc>] do_filp_open+0x19c/0x3c0 [<ffffffffaa7e44a1>] do_sys_openat2+0x141/0x180 [<ffffffffaa7e4945>] __x64_sys_open+0xe5/0x1a0 [<ffffffffac2cc2f7>] do_syscall_64+0xb7/0x210 [<ffffffffac400130>] entry_SYSCALL_64_after_hwframe+0x77/0x7f It can be triggered by mouting a subdirectory of a CephFS filesystem, and then trying to access files on this subdirectory with an auth token using a path-scoped capability: $ ceph auth get client.services [client.services] key = REDACTED caps mds = "allow rw fsname=cephfs path=/volumes/" caps mon = "allow r fsname=cephfs" caps osd = "allow rw tag cephfs data=cephfs" $ cat /proc/self/mounts services@[REDACTED].cephfs=/volumes/containers /ceph/containers ceph rw,noatime,name=services,secret=<hidden>,ms_mode=prefer-crc,mount_timeout=300,acl,mon_addr=[REDACTED]:3300,recover_session=clean 0 0 $ seq 1 1000000 | xargs -P32 --replace={} touch /ceph/containers/file-{} && \ seq 1 1000000 | xargs -P32 --replace={} cat /ceph/containers/file-{} [ idryomov: combine if statements, rename rc to path_matched and make it a bool, formatting ] Cc: stable@vger.kernel.org Fixes: 596afb0b8933 ("ceph: add ceph_mds_check_access() helper") Signed-off-by: Antoine Viallon <antoine@lesviallon.fr> Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko@ibm.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
2025-01-15saner replacement for debugfs_rename()Al Viro
Existing primitive has several problems: 1) calling conventions are clumsy - it returns a dentry reference that is either identical to its second argument or is an ERR_PTR(-E...); in both cases no refcount changes happen. Inconvenient for users and bug-prone; it would be better to have it return 0 on success and -E... on failure. 2) it allows cross-directory moves; however, no such caller have ever materialized and considering the way debugfs is used, it's unlikely to happen in the future. What's more, any such caller would have fun issues to deal with wrt interplay with recursive removal. It also makes the calling conventions clumsier... 3) tautological rename fails; the callers have no race-free way to deal with that. 4) new name must have been formed by the caller; quite a few callers have it done by sprintf/kasprintf/etc., ending up with considerable boilerplate. Proposed replacement: int debugfs_change_name(dentry, fmt, ...). All callers convert to that easily, and it's simpler internally. IMO debugfs_rename() should go; if we ever get a real-world use case for cross-directory moves in debugfs, we can always look into the right way to handle that. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Link: https://lore.kernel.org/r/20250112080705.141166-21-viro@zeniv.linux.org.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-01-15orangefs-debugfs: don't mess with ->d_nameAl Viro
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Link: https://lore.kernel.org/r/20250112080705.141166-20-viro@zeniv.linux.org.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-01-15debugfs: allow to store an additional opaque pointer at file creationAl Viro
Set by debugfs_create_file_aux(name, mode, parent, data, aux, fops). Plain debugfs_create_file() has it set to NULL. Accessed by debugfs_get_aux(file). Convenience macros for numeric opaque data - debugfs_create_file_aux_num and debugfs_get_aux_num, resp. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Reviewed-by: Christian Brauner <brauner@kernel.org> Link: https://lore.kernel.org/r/20250112080705.141166-5-viro@zeniv.linux.org.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-01-15debugfs: don't mess with bits in ->d_fsdataAl Viro
The reason we need that crap is the dual use ->d_fsdata has there - it's both holding a debugfs_fsdata reference after the first debugfs_file_get() (actually, after the call of proxy ->open()) *and* it serves as a place to stash a reference to real file_operations from object creation to the first open. Oh, and it's triple use, actually - that stashed reference might be to debugfs_short_fops. Bugger that for a game of solidiers - just put the operations reference into debugfs-private augmentation of inode. And split debugfs_full_file_operations into full and short cases, so that debugfs_get_file() could tell one from another. Voila - ->d_fsdata holds NULL until the first (successful) debugfs_get_file() and a reference to struct debugfs_fsdata afterwards. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Reviewed-by: Christian Brauner <brauner@kernel.org> Link: https://lore.kernel.org/r/20250112080705.141166-4-viro@zeniv.linux.org.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-01-15debugfs: get rid of dynamically allocation proxy_opsAl Viro
All it takes is having full_proxy_open() collect the information about available methods and store it in debugfs_fsdata. Wrappers are called only after full_proxy_open() has succeeded calling debugfs_get_file(), so they are guaranteed to have ->d_fsdata already pointing to debugfs_fsdata. As the result, they can check if method is absent and bugger off early, without any atomic operations, etc. - same effect as what we'd have from NULL method. Which makes the entire proxy_fops contents unconditional, making it completely pointless - we can just put those methods (unconditionally) into debugfs_full_proxy_file_operations and forget about dynamic allocation, replace_fops, etc. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Reviewed-by: Christian Brauner <brauner@kernel.org> Link: https://lore.kernel.org/r/20250112080705.141166-3-viro@zeniv.linux.org.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-01-15debugfs: move ->automount into debugfs_inode_infoAl Viro
... and don't bother with debugfs_fsdata for those. Life's simpler that way... Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Reviewed-by: Christian Brauner <brauner@kernel.org> Link: https://lore.kernel.org/r/20250112080705.141166-2-viro@zeniv.linux.org.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-01-15debugfs: separate cache for debugfs inodesAl Viro
Embed them into container (struct debugfs_inode_info, with nothing else in it at the moment), set the cache up, etc. Just the infrastructure changes letting us augment debugfs inodes here; adding stuff will come at the next step. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Reviewed-by: Christian Brauner <brauner@kernel.org> Link: https://lore.kernel.org/r/20250112080705.141166-1-viro@zeniv.linux.org.uk Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2025-01-15afs: Fix the fallback handling for the YFS.RemoveFile2 RPC callDavid Howells
Fix a pair of bugs in the fallback handling for the YFS.RemoveFile2 RPC call: (1) Fix the abort code check to also look for RXGEN_OPCODE. The lack of this masks the second bug. (2) call->server is now not used for ordinary filesystem RPC calls that have an operation descriptor. Fix to use call->op->server instead. Fixes: e49c7b2f6de7 ("afs: Build an abstraction around an "operation" concept") Signed-off-by: David Howells <dhowells@redhat.com> Link: https://lore.kernel.org/r/109541.1736865963@warthog.procyon.org.uk cc: Marc Dionne <marc.dionne@auristor.com> cc: linux-afs@lists.infradead.org Signed-off-by: Christian Brauner <brauner@kernel.org>
2025-01-14nfs: probe for LOCALIO when v3 client reconnects to serverMike Snitzer
Re-enabling NFSv3 LOCALIO is made more complex (than NFSv4) because v3 is stateless. As such, the hueristic used to identify a LOCALIO probe point is more adhoc by nature: if/when NFSv3 client IO begins to complete again in terms of normal RPC-based NFSv3 server IO, attempt nfs_local_probe_async(). Care is taken to throttle the frequency of nfs_local_probe_async(), otherwise there could be a flood of repeat calls to nfs_local_probe_async(). The throttle is admin controlled using a new module parameter for nfsv3, e.g.: echo 512 > /sys/module/nfsv3/parameters/nfs3_localio_probe_throttle Probe for NFSv3 LOCALIO every N IO requests (512 in this case). Must be power-of-2, defaults to 0 (probing disabled). On systems that expect to use LOCALIO with NFSv3 the admin should configure the 'nfs3_localio_probe_throttle' module parameter. This commit backfills module parameter documentation in localio.rst Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs: probe for LOCALIO when v4 client reconnects to serverMike Snitzer
Introduce nfs_local_probe_async() for the NFS client to initiate if/when it reconnects with server. For NFSv4 it is a simple matter to call nfs_local_probe_async() from nfs4_do_reclaim (during NFSv4 grace). Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs/localio: remove redundant code and simplify LOCALIO enablementMike Snitzer
Remove nfs_local_enable and nfs_local_disable, instead use nfs_localio_enable_client and nfs_localio_disable_client. Discontinue use of the NFS_CS_LOCAL_IO bit in the nfs_client struct's cl_flags to reflect that LOCALIO is enabled; instead just test if the net member of the nfs_uuid_t struct is set. Also remove NFS_CS_LOCAL_IO. Lastly, remove trace_nfs_local_enable and trace_nfs_local_disable because comparable traces are available from nfs_localio.ko. Suggested-by: NeilBrown <neilb@suse.de> Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs_common: add nfs_localio trace eventsMike Snitzer
The nfs_localio.ko now exposes /sys/kernel/tracing/events/nfs_localio with nfs_localio_enable_client and nfs_localio_disable_client events. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs_common: track all open nfsd_files per LOCALIO nfs_clientMike Snitzer
This tracking enables __nfsd_file_cache_purge() to call nfs_localio_invalidate_clients(), upon shutdown or export change, to nfs_close_local_fh() all open nfsd_files that are still cached by the LOCALIO nfs clients associated with nfsd_net that is being shutdown. Now that the client must track all open nfsd_files there was more work than necessary being done with the global nfs_uuids_lock contended. This manifested in various RCU issues, e.g.: hrtimer: interrupt took 47969440 ns rcu: INFO: rcu_sched detected stalls on CPUs/tasks: Use nfs_uuid->lock to protect all nfs_uuid_t members, instead of nfs_uuids_lock, once nfs_uuid_is_local() adds the client to nn->local_clients. Also add 'local_clients_lock' to 'struct nfsd_net' to protect nn->local_clients. And store a pointer to spinlock in the 'list_lock' member of nfs_uuid_t so nfs_localio_disable_client() can use it to avoid taking the global nfs_uuids_lock. In combination, these split out locks eliminate the use of the single nfslocalio.c global nfs_uuids_lock in the IO paths (open and close). Also refactored associated fs/nfs_common/nfslocalio.c methods' locking to reduce work performed with spinlocks held in general. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs_common: rename nfslocalio nfs_uuid_lock to nfs_uuids_lockMike Snitzer
This global spinlock protects all nfs_uuid_t relative to the global nfs_uuids list. A later commit will split this global spinlock so prepare by renaming this lock to reflect its intended narrow scope. Signed-off-by: Mike Snitzer <snitzer@redhat.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfsd: nfsd_file_acquire_local no longer returns GC'd nfsd_fileMike Snitzer
Now that LOCALIO no longer leans on NFSD's filecache for caching open files (and instead uses NFS client-side open nfsd_file caching) there is no need to use NFSD filecache's GC feature. Avoiding GC will speed up nfsd_file initial opens. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Acked-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfsd: rename nfsd_serv_ prefixed methods and variables with nfsd_net_Mike Snitzer
Also update Documentation/filesystems/nfs/localio.rst accordingly and reduce the technical documentation debt that was previously captured in that document. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Acked-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfsd: update percpu_ref to manage references on nfsd_netMike Snitzer
Holding a reference on nfsd_net is what is required, it was never actually about ensuring nn->nfsd_serv available. Move waiting for outstanding percpu references from nfsd_destroy_serv() to nfsd_shutdown_net(). By moving it later it will be possible to invalidate localio clients during nfsd_file_cache_shutdown_net() via __nfsd_file_cache_purge(). Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Acked-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs: cache all open LOCALIO nfsd_file(s) in clientMike Snitzer
This commit switches from leaning heavily on NFSD's filecache (in terms of GC'd nfsd_files) back to caching nfsd_files in the client. A later commit will add the callback mechanism needed to allow NFSD to force the NFS client to cleanup all cached nfsd_files. Add nfs_fh_localio_init() and 'struct nfs_fh_localio' to cache opened nfsd_file(s) (both a RO and RW nfsd_file is able to be opened and cached for a given nfs_fh). Update nfs_local_open_fh() to cache the nfsd_file once it is opened using __nfs_local_open_fh(). Introduce nfs_close_local_fh() to clear the cached open nfsd_files and call nfs_to_nfsd_file_put_local(). Refcounting is such that: - nfs_local_open_fh() is paired with nfs_close_local_fh(). - __nfs_local_open_fh() is paired with nfs_to_nfsd_file_put_local(). - nfs_local_file_get() is paired with nfs_local_file_put(). Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs_common: move localio_lock to new lock member of nfs_uuid_tMike Snitzer
Remove cl_localio_lock from 'struct nfs_client' in favor of adding a lock to the nfs_uuid_t struct (which is embedded in each nfs_client). Push nfs_local_{enable,disable} implementation down to nfs_common. Those methods now call nfs_localio_{enable,disable}_client. This allows implementing nfs_localio_invalidate_clients in terms of nfs_localio_disable_client. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs_common: rename functions that invalidate LOCALIO nfs_clientsMike Snitzer
Rename nfs_uuid_invalidate_one_client to nfs_localio_disable_client. Rename nfs_uuid_invalidate_clients to nfs_localio_invalidate_clients. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfsd: add nfsd_file_{get,put} to 'nfs_to' nfsd_localio_operationsMike Snitzer
In later a commit LOCALIO must call both nfsd_file_get and nfsd_file_put to manage extra nfsd_file references. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Acked-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14nfs/localio: add direct IO enablement with sync and async IO supportMike Snitzer
This commit simply adds the required O_DIRECT plumbing. It doesn't address the fact that NFS doesn't ensure all writes are page aligned (nor device logical block size aligned as required by O_DIRECT). Because NFS will read-modify-write for IO that isn't aligned, LOCALIO will not use O_DIRECT semantics by default if/when an application requests the use of O_DIRECT. Allow the use of O_DIRECT semantics by: 1: Adding a flag to the nfs_pgio_header struct to allow the NFS O_DIRECT layer to signal that O_DIRECT was used by the application 2: Adding a 'localio_O_DIRECT_semantics' NFS module parameter that when enabled will cause LOCALIO to use O_DIRECT semantics (this may cause IO to fail if applications do not properly align their IO). This commit is derived from code developed by Weston Andros Adamson. Signed-off-by: Mike Snitzer <snitzer@kernel.org> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Anna Schumaker <anna.schumaker@oracle.com>
2025-01-14gfs2: Truncate address space when flipping GFS2_DIF_JDATA flagAndreas Gruenbacher
Truncate an inode's address space when flipping the GFS2_DIF_JDATA flag: depending on that flag, the pages in the address space will either use buffer heads or iomap_folio_state structs, and we cannot mix the two. Reported-by: Kun Hu <huk23@m.fudan.edu.cn>, Jiaji Qin <jjtan24@m.fudan.edu.cn> Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
2025-01-14bcachefs: Document issue with bch_stripe layoutKent Overstreet
We've got a problem with bch_stripe that is going to take an on disk format rev to fix - we can't access the block sector counts if the checksum type is unknown. Document it for now, there are a few other things to fix as well. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14bcachefs: Fix self healing on read errorKent Overstreet
We were incorrectly checking if there'd been an io error. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14bcachefs: Pop all the transactions from the abort oneAlan Huang
The transaction is going to abort, so there will be no cycle involving this transaction anymore. Signed-off-by: Alan Huang <mmpgouride@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14bcachefs: Only abort the transactions in the cycleAlan Huang
When the cycle doesn't involve the initiator of the cycle detection, we might choose a transaction that is not involved in the cycle to abort. It shouldn't be that since it won't break the cycle, this patch therefore chooses the transaction in the cycle to abort. Signed-off-by: Alan Huang <mmpgouride@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14bcachefs: Introduce lock_graph_pop_fromAlan Huang
This patch introduces a helper function called lock_graph_pop_from, it pops the graph from i. Signed-off-by: Alan Huang <mmpgouride@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14bcachefs: Convert open-coded lock_graph_pop_all to helperAlan Huang
Signed-off-by: Alan Huang <mmpgouride@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14bcachefs: Do not allow no fail lock request to failAlan Huang
If the transaction chose itself as a victim before and restarted, it might request a no fail lock request this time. But it might be added to others' lock graph and be chose as the victim again, it's no longer safe without additional check. We can also convert the cycle detector to be fully RCU-based to solve that unsoundness, but the latency added to trans_put and additional memory required may not worth it. Signed-off-by: Alan Huang <mmpgouride@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14bcachefs: Merge the condition to avoid additional invocationAlan Huang
If the lock has been acquired and unlocked, we don't have to do clear and wakeup again, though harmless since we hold the intent lock. Merge the condition might be clearer. Signed-off-by: Alan Huang <mmpgouride@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14Revert "bcachefs: Fix bch2_btree_node_upgrade()"Alan Huang
This reverts commit 62448afee714354a26db8a0f3c644f58628f0792. six_lock_tryupgrade fails only if there is an intent lock held, it won't fail no matter how many read locks are held. Signed-off-by: Alan Huang <mmpgouride@gmail.com> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-01-14btrfs: selftests: add a selftest for deleting two out of three extentsJohannes Thumshirn
Add a selftest creating three extents and then deleting two out of the three extents. Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: selftests: add test for punching a hole into 3 RAID stripe-extentsJohannes Thumshirn
Test creating a range of three RAID stripe-extents and then punch a hole in the middle, deleting all of the middle extents and partially deleting the "book ends". Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: selftests: add selftest for punching holes into the RAID stripe extentsJohannes Thumshirn
Add a selftest for punching a hole into a RAID stripe extent. The test create an 1M extent and punches a 64k bytes long hole at offset of 32k from the start of the extent. Afterwards it verifies the start and length of both resulting new extents "left" and "right" as well as the absence of the hole. Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: selftests: test RAID stripe-tree deletion spanning two itemsJohannes Thumshirn
Add a selftest for RAID stripe-tree deletion with a delete range spanning two items, so that we're punching a hole into two adjacent RAID stripe extents truncating the first and "moving" the second to the right. The following diagram illustrates the operation: |--- RAID Stripe Extent ---||--- RAID Stripe Extent ---| |----- keep -----|--- drop ---|----- keep ----| Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: selftests: don't split RAID extents in halfJohannes Thumshirn
The selftests for partially deleting the start or tail of RAID stripe-extents split these extents in half. This can hide errors in the calculation, so don't split the RAID stripe-extents in half but delete the first or last 16K of the 64K extents. Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: selftests: check for correct return value of failed lookupJohannes Thumshirn
Commit 5e72aabc1fff ("btrfs: return ENODATA in case RST lookup fails") changed btrfs_get_raid_extent_offset()'s return value to ENODATA in case the RAID stripe-tree lookup failed. Adjust the test cases which check for absence of a given range to check for ENODATA as return value in this case. Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: don't use btrfs_set_item_key_safe on RAID stripe-extentsJohannes Thumshirn
Don't use btrfs_set_item_key_safe() to modify the keys in the RAID stripe-tree, as this can lead to corruption of the tree, which is caught by the checks in btrfs_set_item_key_safe(): BTRFS info (device nvme1n1): leaf 49168384 gen 15 total ptrs 194 free space 8329 owner 12 BTRFS info (device nvme1n1): refs 2 lock_owner 1030 current 1030 [ snip ] item 105 key (354549760 230 20480) itemoff 14587 itemsize 16 stride 0 devid 5 physical 67502080 item 106 key (354631680 230 4096) itemoff 14571 itemsize 16 stride 0 devid 1 physical 88559616 item 107 key (354631680 230 32768) itemoff 14555 itemsize 16 stride 0 devid 1 physical 88555520 item 108 key (354717696 230 28672) itemoff 14539 itemsize 16 stride 0 devid 2 physical 67604480 [ snip ] BTRFS critical (device nvme1n1): slot 106 key (354631680 230 32768) new key (354635776 230 4096) ------------[ cut here ]------------ kernel BUG at fs/btrfs/ctree.c:2602! Oops: invalid opcode: 0000 [#1] PREEMPT SMP PTI CPU: 1 UID: 0 PID: 1055 Comm: fsstress Not tainted 6.13.0-rc1+ #1464 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.2-3-gd478f380-rebuilt.opensuse.org 04/01/2014 RIP: 0010:btrfs_set_item_key_safe+0xf7/0x270 Code: <snip> RSP: 0018:ffffc90001337ab0 EFLAGS: 00010287 RAX: 0000000000000000 RBX: ffff8881115fd000 RCX: 0000000000000000 RDX: 0000000000000001 RSI: 0000000000000001 RDI: 00000000ffffffff RBP: ffff888110ed6f50 R08: 00000000ffffefff R09: ffffffff8244c500 R10: 00000000ffffefff R11: 00000000ffffffff R12: ffff888100586000 R13: 00000000000000c9 R14: ffffc90001337b1f R15: ffff888110f23b58 FS: 00007f7d75c72740(0000) GS:ffff88813bd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fa811652c60 CR3: 0000000111398001 CR4: 0000000000370eb0 Call Trace: <TASK> ? __die_body.cold+0x14/0x1a ? die+0x2e/0x50 ? do_trap+0xca/0x110 ? do_error_trap+0x65/0x80 ? btrfs_set_item_key_safe+0xf7/0x270 ? exc_invalid_op+0x50/0x70 ? btrfs_set_item_key_safe+0xf7/0x270 ? asm_exc_invalid_op+0x1a/0x20 ? btrfs_set_item_key_safe+0xf7/0x270 btrfs_partially_delete_raid_extent+0xc4/0xe0 btrfs_delete_raid_extent+0x227/0x240 __btrfs_free_extent.isra.0+0x57f/0x9c0 ? exc_coproc_segment_overrun+0x40/0x40 __btrfs_run_delayed_refs+0x2fa/0xe80 btrfs_run_delayed_refs+0x81/0xe0 btrfs_commit_transaction+0x2dd/0xbe0 ? preempt_count_add+0x52/0xb0 btrfs_sync_file+0x375/0x4c0 do_fsync+0x39/0x70 __x64_sys_fsync+0x13/0x20 do_syscall_64+0x54/0x110 entry_SYSCALL_64_after_hwframe+0x76/0x7e RIP: 0033:0x7f7d7550ef90 Code: <snip> RSP: 002b:00007ffd70237248 EFLAGS: 00000202 ORIG_RAX: 000000000000004a RAX: ffffffffffffffda RBX: 0000000000000004 RCX: 00007f7d7550ef90 RDX: 000000000000013a RSI: 000000000040eb28 RDI: 0000000000000004 RBP: 000000000000001b R08: 0000000000000078 R09: 00007ffd7023725c R10: 00007f7d75400390 R11: 0000000000000202 R12: 028f5c28f5c28f5c R13: 8f5c28f5c28f5c29 R14: 000000000040b520 R15: 00007f7d75c726c8 </TASK> While the root cause of the tree order corruption isn't clear, using btrfs_duplicate_item() to copy the item and then adjusting both the key and the per-device physical addresses is a safe way to counter this problem. Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: implement hole punching for RAID stripe extentsJohannes Thumshirn
If the stripe extent we want to delete starts before the range we want to delete and ends after the range we want to delete we're punching a hole in the stripe extent: |--- RAID Stripe Extent ---| | keep |--- drop ---| keep | This means we need to a) truncate the existing item and b) create a second item for the remaining range. Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: fix deletion of a range spanning parts two RAID stripe extentsJohannes Thumshirn
When a user requests the deletion of a range that spans multiple stripe extents and btrfs_search_slot() returns us the second RAID stripe extent, we need to pick the previous item and truncate it, if there's still a range to delete left, move on to the next item. The following diagram illustrates the operation: |--- RAID Stripe Extent ---||--- RAID Stripe Extent ---| |--- keep ---|--- drop ---| While at it, comment the trivial case of a whole item delete as well. Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
2025-01-14btrfs: fix tail delete of RAID stripe-extentsJohannes Thumshirn
Fix tail delete of RAID stripe-extents, if there is a range to be deleted as well after the tail delete of the extent. Reviewed-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Signed-off-by: David Sterba <dsterba@suse.com>
generated by cgit (git 2.34.1) at 2025-07-18 16:40:10 +0000