From: Anna Schumaker <Anna.Schumaker@xxxxxxxxxx>
When we left off with READ_PLUS, Chuck had suggested reverting the
server to reply with a single NFS4_CONTENT_DATA segment essentially
mimicing how the READ operation behaves. Then, a future sparse read
function can be added and the server modified to support it without
needing to rip out the old READ_PLUS code at the same time.
This patch takes that first step. I was even able to re-use the
nfsd4_encode_readv() function to remove some duplicate code.
Chuck, I tried to add in sparse read support by adding this extra
change. Unfortunately it leads to a bunch of new failing xfstests. Do
you have any thoughts about what might be going on? Is the patch okay
without the splice support?
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index adbff7737c14..e21e6cfd1c6d 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -4733,6 +4733,7 @@ static __be32
nfsd4_encode_read_plus_data(struct nfsd4_compoundres *resp,
struct nfsd4_read *read)
{
+ bool splice_ok = test_bit(RQ_SPLICE_OK, &resp->rqstp->rq_flags);
unsigned long maxcount;
struct xdr_stream *xdr = resp->xdr;
struct file *file = read->rd_nf->nf_file;
@@ -4747,7 +4748,10 @@ nfsd4_encode_read_plus_data(struct nfsd4_compoundres *resp,
maxcount = min_t(unsigned long, read->rd_length,
(xdr->buf->buflen - xdr->buf->len));
- nfserr = nfsd4_encode_readv(resp, read, file, maxcount);
+ if (file->f_op->splice_read && splice_ok)
+ nfserr = nfsd4_encode_splice_read(resp, read, file, maxcount);
+ else
+ nfserr = nfsd4_encode_readv(resp, read, file, maxcount)
if (nfserr)
return nfserr;
Thanks,
Anna
Anna Schumaker (1):
NFSD: Simplify READ_PLUS
fs/nfsd/nfs4xdr.c | 122 ++++++++--------------------------------------
1 file changed, 20 insertions(+), 102 deletions(-)
--
2.37.2
