diff options
| author | Chuck Lever <chuck.lever@oracle.com> | 2023-01-06 20:43:37 +0300 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2023-01-14 12:33:42 +0300 |
| commit | 8cc0e63ba86c9f4e732de9f42642231c8e3a31d2 (patch) | |
| tree | b7a96b8336cff433e64f0597462365d4d91ca193 /fs/nfsd/nfs4proc.c | |
| parent | 29fbaa434c223f29917e8de054cc48a96e6c0ed4 (diff) | |
| download | linux-8cc0e63ba86c9f4e732de9f42642231c8e3a31d2.tar.xz | |
Revert "SUNRPC: Use RMW bitops in single-threaded hot paths"
commit 7827c81f0248e3c2f40d438b020f3d222f002171 upstream.
The premise that "Once an svc thread is scheduled and executing an
RPC, no other processes will touch svc_rqst::rq_flags" is false.
svc_xprt_enqueue() examines the RQ_BUSY flag in scheduled nfsd
threads when determining which thread to wake up next.
Found via KCSAN.
Fixes: 28df0988815f ("SUNRPC: Use RMW bitops in single-threaded hot paths")
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'fs/nfsd/nfs4proc.c')
| -rw-r--r-- | fs/nfsd/nfs4proc.c | 7 |
1 files changed, 3 insertions, 4 deletions
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 34d1cd5883fb..c7329523a10f 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -928,7 +928,7 @@ nfsd4_read(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, * the client wants us to do more in this compound: */ if (!nfsd4_last_compound_op(rqstp)) - __clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags); + clear_bit(RQ_SPLICE_OK, &rqstp->rq_flags); /* check stateid */ status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh, @@ -2615,12 +2615,11 @@ nfsd4_proc_compound(struct svc_rqst *rqstp) cstate->minorversion = args->minorversion; fh_init(current_fh, NFS4_FHSIZE); fh_init(save_fh, NFS4_FHSIZE); - /* * Don't use the deferral mechanism for NFSv4; compounds make it * too hard to avoid non-idempotency problems. */ - __clear_bit(RQ_USEDEFERRAL, &rqstp->rq_flags); + clear_bit(RQ_USEDEFERRAL, &rqstp->rq_flags); /* * According to RFC3010, this takes precedence over all other errors. @@ -2742,7 +2741,7 @@ encode_op: out: cstate->status = status; /* Reset deferral mechanism for RPC deferrals */ - __set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags); + set_bit(RQ_USEDEFERRAL, &rqstp->rq_flags); return rpc_success; } |