BMC/Intel-BMC/linux.git/fs/nfs, branch dev-4.17

NFS: Fix up nfs_post_op_update_inode() to force ctime updates

2018-08-03T05:47:38+00:00

[ Upstream commit d554168f87a55b35b7c59921a0dc45b6ba17d08d ] We do not want to ignore ctime updates that originate from functions such as link(). Signed-off-by: Trond Myklebust Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

pnfs: Don't release the sequence slot until we've processed layoutget on open

2018-08-03T05:47:38+00:00

[ Upstream commit ae55e59da0e401893b3c52b575fc18a00623d0a1 ] If the server recalls the layout that was just handed out, we risk hitting a race as described in RFC5661 Section 2.10.6.3 unless we ensure that we release the sequence slot after processing the LAYOUTGET operation that was sent as part of the OPEN compound. Signed-off-by: Trond Myklebust Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

NFSv4.1: Fix the client behaviour on NFS4ERR_SEQ_FALSE_RETRY

2018-08-03T05:47:34+00:00

[ Upstream commit f9312a541050007ec59eb0106273a0a10718cd83 ] If the server returns NFS4ERR_SEQ_FALSE_RETRY or NFS4ERR_RETRY_UNCACHED_REP, then it thinks we're trying to replay an existing request. If so, then let's just bump the sequence ID and retry the operation. Signed-off-by: Trond Myklebust Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

skip LAYOUTRETURN if layout is invalid

2018-08-03T05:47:33+00:00

[ Upstream commit 93b7f7ad2018d2037559b1d0892417864c78b371 ] Currently, when IO to DS fails, client returns the layout and retries against the MDS. However, then on umounting (inode eviction) it returns the layout again. This is because pnfs_return_layout() was changed in commit d78471d32bb6 ("pnfs/blocklayout: set PNFS_LAYOUTRETURN_ON_ERROR") to always set NFS_LAYOUT_RETURN_REQUESTED so even if we returned the layout, it will be returned again. Instead, let's also check if we have already marked the layout invalid. Signed-off-by: Olga Kornievskaia Signed-off-by: Trond Myklebust Signed-off-by: Sasha Levin Signed-off-by: Greg Kroah-Hartman

NFSv4: Fix a typo in nfs41_sequence_process

2018-07-03T09:27:08+00:00

commit 995891006ccbb73c0c9c3923cf9d25c4d07ec16b upstream. We want to compare the slot_id to the highest slot number advertised by the server. Fixes: 3be0f80b5fe9c ("NFSv4.1: Fix up replays of interrupted requests") Cc: stable@vger.kernel.org # 4.15+ Signed-off-by: Trond Myklebust Signed-off-by: Greg Kroah-Hartman

NFSv4: Revert commit 5f83d86cf531d ("NFSv4.x: Fix wraparound issues..")

2018-07-03T09:27:08+00:00

commit fc40724fc6731d90cc7fb6d62d66135f85a33dd2 upstream. The correct behaviour for NFSv4 sequence IDs is to wrap around to the value 0 after 0xffffffff. See https://tools.ietf.org/html/rfc5661#section-2.10.6.1 Fixes: 5f83d86cf531d ("NFSv4.x: Fix wraparound issues when validing...") Cc: stable@vger.kernel.org # 4.6+ Signed-off-by: Trond Myklebust Signed-off-by: Greg Kroah-Hartman

NFSv4: Fix possible 1-byte stack overflow in nfs_idmap_read_and_verify_message

2018-07-03T09:27:08+00:00

commit d68894800ec5712d7ddf042356f11e36f87d7f78 upstream. In nfs_idmap_read_and_verify_message there is an incorrect sprintf '%d' that converts the __u32 'im_id' from struct idmap_msg to 'id_str', which is a stack char array variable of length NFS_UINT_MAXLEN == 11. If a uid or gid value is > 2147483647 = 0x7fffffff, the conversion overflows into a negative value, for example: crash> p (unsigned) (0x80000000) $1 = 2147483648 crash> p (signed) (0x80000000) $2 = -2147483648 The '-' sign is written to the buffer and this causes a 1 byte overflow when the NULL byte is written, which corrupts kernel stack memory. If CONFIG_CC_STACKPROTECTOR_STRONG is set we see a stack-protector panic: [11558053.616565] Kernel panic - not syncing: stack-protector: Kernel stack is corrupted in: ffffffffa05b8a8c [11558053.639063] CPU: 6 PID: 9423 Comm: rpc.idmapd Tainted: G W ------------ T 3.10.0-514.el7.x86_64 #1 [11558053.641990] Hardware name: Red Hat OpenStack Compute, BIOS 1.10.2-3.el7_4.1 04/01/2014 [11558053.644462] ffffffff818c7bc0 00000000b1f3aec1 ffff880de0f9bd48 ffffffff81685eac [11558053.646430] ffff880de0f9bdc8 ffffffff8167f2b3 ffffffff00000010 ffff880de0f9bdd8 [11558053.648313] ffff880de0f9bd78 00000000b1f3aec1 ffffffff811dcb03 ffffffffa05b8a8c [11558053.650107] Call Trace: [11558053.651347] [] dump_stack+0x19/0x1b [11558053.653013] [] panic+0xe3/0x1f2 [11558053.666240] [] ? kfree+0x103/0x140 [11558053.682589] [] ? idmap_pipe_downcall+0x1cc/0x1e0 [nfsv4] [11558053.689710] [] __stack_chk_fail+0x1b/0x30 [11558053.691619] [] idmap_pipe_downcall+0x1cc/0x1e0 [nfsv4] [11558053.693867] [] rpc_pipe_write+0x56/0x70 [sunrpc] [11558053.695763] [] vfs_write+0xbd/0x1e0 [11558053.702236] [] ? task_work_run+0xac/0xe0 [11558053.704215] [] SyS_write+0x7f/0xe0 [11558053.709674] [] system_call_fastpath+0x16/0x1b Fix this by calling the internally defined nfs_map_numeric_to_string() function which properly uses '%u' to convert this __u32. For consistency, also replace the one other place where snprintf is called. Signed-off-by: Dave Wysochanski Reported-by: Stephen Johnston Fixes: cf4ab538f1516 ("NFSv4: Fix the string length returned by the idmapper") Cc: stable@vger.kernel.org # v3.4+ Signed-off-by: Trond Myklebust Signed-off-by: Greg Kroah-Hartman

Merge tag 'nfs-for-4.17-1' of git://git.linux-nfs.org/projects/anna/linux-nfs

2018-04-12T19:55:50+00:00

Pull NFS client updates from Anna Schumaker: "Stable bugfixes: - xprtrdma: Fix corner cases when handling device removal # v4.12+ - xprtrdma: Fix latency regression on NUMA NFS/RDMA clients # v4.15+ Features: - New sunrpc tracepoint for RPC pings - Finer grained NFSv4 attribute checking - Don't unnecessarily return NFS v4 delegations Other bugfixes and cleanups: - Several other small NFSoRDMA cleanups - Improvements to the sunrpc RTT measurements - A few sunrpc tracepoint cleanups - Various fixes for NFS v4 lock notifications - Various sunrpc and NFS v4 XDR encoding cleanups - Switch to the ida_simple API - Fix NFSv4.1 exclusive create - Forget acl cache after setattr operation - Don't advance the nfs_entry readdir cookie if xdr decoding fails" * tag 'nfs-for-4.17-1' of git://git.linux-nfs.org/projects/anna/linux-nfs: (47 commits) NFS: advance nfs_entry cookie only after decoding completes successfully NFSv3/acl: forget acl cache after setattr NFSv4.1: Fix exclusive create NFSv4: Declare the size up to date after it was set. nfs: Use ida_simple API NFSv4: Fix the nfs_inode_set_delegation() arguments NFSv4: Clean up CB_GETATTR encoding NFSv4: Don't ask for attributes when ACCESS is protected by a delegation NFSv4: Add a helper to encode/decode struct timespec NFSv4: Clean up encode_attrs NFSv4; Clean up XDR encoding of type bitmap4 NFSv4: Allow GFP_NOIO sleeps in decode_attr_owner/decode_attr_group SUNRPC: Add a helper for encoding opaque data inline SUNRPC: Add helpers for decoding opaque and string types NFSv4: Ignore change attribute invalidations if we hold a delegation NFS: More fine grained attribute tracking NFS: Don't force unnecessary cache invalidation in nfs_update_inode() NFS: Don't redirty the attribute cache in nfs_wcc_update_inode() NFS: Don't force a revalidation of all attributes if change is missing NFS: Convert NFS_INO_INVALID flags to unsigned long ...

NFS: advance nfs_entry cookie only after decoding completes successfully

2018-04-10T20:06:22+00:00

In nfs[34]_decode_dirent, the cookie is advanced as soon as it is read, but decoding may still fail later in the function, returning an error. Because the cookie has been advanced, the failing entry is not re-requested from the server, resulting in a missing directory entry. In addition, nfs v3 and v4 read the cookie at different locations in the xdr_stream, so the behavior of the two can be inconsistent. Fix these by reading the cookie into a temporary variable, and only advancing the cookie once the entire entry has been decoded from the xdr_stream successfully. Signed-off-by: Frank Sorenson Signed-off-by: Anna Schumaker

NFSv3/acl: forget acl cache after setattr

2018-04-10T20:06:22+00:00

Sync of ACL with std permissions fail，We need to forget the ACL cache after setattr. Reproduction: #!/bin/bash touch testfile cat <testfile #!/bin/bash echo "Test was executed" EOF chmod u=rwx testfile chmod g=rw- testfile chmod o=r-- testfile chacl u::r--,g::rwx,o:rw- testfile chmod u+w testfile ls -l testfile chacl -l testfile Output： -rw-rwxrw- 1 root root 0 Mar 28 05:29 testfile testfile [u::r--,g::rwx,o::rw-] Signed-off-by: chendt.fnst Reviewed-by: Benjamin Coddington Reviewed-by: Kinglong Mee Signed-off-by: Anna Schumaker