Chuck Lever d3d885507b NFSD: Fix verifier returned in stable WRITEs ... commit f11ad7aa65 upstream. RFC 8881 explains the purpose of the write verifier this way: > The final portion of the result is the field writeverf. This field > is the write verifier and is a cookie that the client can use to > determine whether a server has changed instance state (e.g., server > restart) between a call to WRITE and a subsequent call to either > WRITE or COMMIT. But then it says: > This cookie MUST be unchanged during a single instance of the > NFSv4.1 server and MUST be unique between instances of the NFSv4.1 > server. If the cookie changes, then the client MUST assume that > any data written with an UNSTABLE4 value for committed and an old > writeverf in the reply has been lost and will need to be > recovered. RFC 1813 has similar language for NFSv3. NFSv2 does not have a write verifier since it doesn't implement the COMMIT procedure. Since commit 19e0663ff9 ("nfsd: Ensure sampling of the write verifier is atomic with the write"), the Linux NFS server has returned a boot-time-based verifier for UNSTABLE WRITEs, but a zero verifier for FILE_SYNC and DATA_SYNC WRITEs. FILE_SYNC and DATA_SYNC WRITEs are not followed up with a COMMIT, so there's no need for clients to compare verifiers for stable writes. However, by returning a different verifier for stable and unstable writes, the above commit puts the Linux NFS server a step farther out of compliance with the first MUST above. At least one NFS client (FreeBSD) noticed the difference, making this a potential regression. [Removed down_write to fix the conflict in the cherry-pick. The down_write functionality was no longer needed there. Upstream commit 555dbf1a9a titled nfsd: Replace use of rwsem with errseq_t removed those and replace it with new functionality that was more scalable. This commit is already backported onto 5.10 and so removing down_write ensures consistency with that change. Tested by compiling and booting successfully. - kochera] Reported-by: Rick Macklem <rmacklem@uoguelph.ca> Link: https://lore.kernel.org/linux-nfs/YQXPR0101MB096857EEACF04A6DF1FC6D9BDD749@YQXPR0101MB0968.CANPRD01.PROD.OUTLOOK.COM/T/ Fixes: 19e0663ff9 ("nfsd: Ensure sampling of the write verifier is atomic with the write") Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Michael Kochera <kochera@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>		2022-09-15 11:32:02 +02:00
..
acl.h
auth.c
auth.h
blocklayout.c
blocklayoutxdr.c
blocklayoutxdr.h
cache.h
current_stateid.h
export.c
export.h
fault_inject.c
filecache.c	nfsd: Replace use of rwsem with errseq_t	2022-06-22 14:13:12 +02:00
filecache.h	nfsd: Replace use of rwsem with errseq_t	2022-06-22 14:13:12 +02:00
flexfilelayout.c
flexfilelayoutxdr.c
flexfilelayoutxdr.h
idmap.h
Kconfig
lockd.c
Makefile
netns.h
nfs2acl.c
nfs3acl.c
nfs3proc.c	NFSD: Clamp WRITE offsets	2022-02-16 12:54:17 +01:00
nfs3xdr.c
nfs4acl.c
nfs4callback.c	NFSD: fix error handling in NFSv4.0 callbacks	2021-04-07 15:00:05 +02:00
nfs4idmap.c
nfs4layouts.c
nfs4proc.c	nfsd: Replace use of rwsem with errseq_t	2022-06-22 14:13:12 +02:00
nfs4recover.c	nfsd: Fix nsfd startup race (again)	2021-12-14 11:32:39 +01:00
nfs4state.c	NFSD: Fix possible sleep during nfsd4_release_lockowner()	2022-06-06 08:42:45 +02:00
nfs4xdr.c	nfsd4: Handle the NFSv4 READDIR 'dircount' hint being zero	2021-10-13 10:04:24 +02:00
nfscache.c
nfsctl.c	fsnotify: fix fsnotify hooks in pseudo filesystems	2022-02-01 17:25:39 +01:00
nfsd.h
nfsfh.c
nfsfh.h
nfsproc.c	NFSD: prevent underflow in nfssvc_decode_writeargs()	2022-04-08 14:39:51 +02:00
nfssvc.c
nfsxdr.c
pnfs.h
state.h
stats.c
stats.h
trace.c
trace.h	NFSD: Fix offset type in I/O trace points	2022-02-16 12:54:17 +01:00
vfs.c	NFSD: Fix verifier returned in stable WRITEs	2022-09-15 11:32:02 +02:00
vfs.h
xdr.h	NFSD: prevent underflow in nfssvc_decode_writeargs()	2022-04-08 14:39:51 +02:00
xdr3.h
xdr4.h
xdr4cb.h