RE: [Patch v5 19/21] CIFS: SMBD: Read correct returned data length for RDMA write (SMB read) I/O
From: Long Li
Date: Wed Nov 01 2017 - 14:48:25 EST
> -----Original Message-----
> From: Pavel Shilovsky [mailto:piastryyy@xxxxxxxxx]
> Sent: Wednesday, November 1, 2017 9:50 AM
> To: Long Li <longli@xxxxxxxxxxxxx>
> Cc: Steve French <sfrench@xxxxxxxxx>; linux-cifs <linux-
> cifs@xxxxxxxxxxxxxxx>; samba-technical <samba-technical@xxxxxxxxxxxxxxx>;
> Kernel Mailing List <linux-kernel@xxxxxxxxxxxxxxx>; linux-
> rdma@xxxxxxxxxxxxxxx; Tom Talpey <ttalpey@xxxxxxxxxxxxx>; Matthew
> Wilcox <mawilcox@xxxxxxxxxxxxx>; Stephen Hemminger
> <sthemmin@xxxxxxxxxxxxx>; Long Li <longli@xxxxxxxxxxxxx>
> Subject: Re: [Patch v5 19/21] CIFS: SMBD: Read correct returned data length
> for RDMA write (SMB read) I/O
>
> 2017-10-18 16:09 GMT-07:00 Long Li <longli@xxxxxxxxxxxxxxxxxxxxxx>:
> > From: Long Li <longli@xxxxxxxxxxxxx>
> >
> > This patch is for preparing upper layer doing SMB read via RDMA write.
> >
> > When RDMA write is used for SMB read, the returned data length is in
> > DataRemaining in the response packet. Reading it properly by adding a
> > parameter to specifiy where the returned data length is.
> >
> > Add the defition for memory registration to wdata and return the
> > correct length based on if RDMA write is used.
> >
> > Signed-off-by: Long Li <longli@xxxxxxxxxxxxx>
> > ---
> > fs/cifs/cifsglob.h | 13 +++++++++++-- fs/cifs/cifssmb.c | 7
> > ++++++- fs/cifs/smb1ops.c | 6 +++++- fs/cifs/smb2ops.c | 12
> > ++++++++++--
> > 4 files changed, 32 insertions(+), 6 deletions(-)
> >
> > diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index
> > 2ae7d02..ddf83d8 100644
> > --- a/fs/cifs/cifsglob.h
> > +++ b/fs/cifs/cifsglob.h
> > @@ -228,8 +228,14 @@ struct smb_version_operations {
> > __u64 (*get_next_mid)(struct TCP_Server_Info *);
> > /* data offset from read response message */
> > unsigned int (*read_data_offset)(char *);
> > - /* data length from read response message */
> > - unsigned int (*read_data_length)(char *);
> > + /*
> > + * Data length from read response message
> > + * When in_remaining is true, the returned data length is in
> > + * message field DataRemaining for out-of-band data read (e.g
> through
> > + * Memory Registration RDMA write in SMBD).
> > + * Otherwise, the returned data length is in message field DataLength.
> > + */
> > + unsigned int (*read_data_length)(char *, bool in_remaining);
> > /* map smb to linux error */
> > int (*map_error)(char *, bool);
> > /* find mid corresponding to the response message */ @@
> > -1148,6 +1154,9 @@ struct cifs_readdata {
> > struct cifs_readdata *rdata,
> > struct iov_iter *iter);
> > struct kvec iov[2];
> > +#ifdef CONFIG_CIFS_SMB_DIRECT
> > + struct smbd_mr *mr;
> > +#endif
> > unsigned int pagesz;
> > unsigned int tailsz;
> > unsigned int credits;
> > diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index
> > 08ff56a..a343db4 100644
> > --- a/fs/cifs/cifssmb.c
> > +++ b/fs/cifs/cifssmb.c
> > @@ -1533,8 +1533,13 @@ cifs_readv_receive(struct TCP_Server_Info
> *server, struct mid_q_entry *mid)
> > rdata->iov[0].iov_base, server->total_read);
> >
> > /* how much data is in the response? */
> > - data_len = server->ops->read_data_length(buf);
> > +#ifdef CONFIG_CIFS_SMB_DIRECT
> > + data_len = server->ops->read_data_length(buf, rdata->mr);
> > + if (!rdata->mr && (data_offset + data_len > buflen)) { #else
> > + data_len = server->ops->read_data_length(buf, false);
> > if (data_offset + data_len > buflen) {
> > +#endif
> > /* data_len is corrupt -- discard frame */
> > rdata->result = -EIO;
> > return cifs_readv_discard(server, mid); diff --git
> > a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c index a723df3..b718bb8 100644
> > --- a/fs/cifs/smb1ops.c
> > +++ b/fs/cifs/smb1ops.c
> > @@ -87,9 +87,13 @@ cifs_read_data_offset(char *buf) }
> >
> > static unsigned int
> > -cifs_read_data_length(char *buf)
> > +cifs_read_data_length(char *buf, bool in_remaining)
> > {
> > READ_RSP *rsp = (READ_RSP *)buf;
> > + if (in_remaining) {
> > + cifs_dbg(VFS, "Invalid SMB1 calling path, check
> > + code\n");
>
> I would suggest to do WARN_ON(1) instead.
I will make the change.
>
> > + return 0;
> > + }
> > return (le16_to_cpu(rsp->DataLengthHigh) << 16) +
> > le16_to_cpu(rsp->DataLength); }
>
>
> --
> Best regards,
> Pavel Shilovsky