> -----Original Message-----
> From: Tom Talpey
> Sent: Monday, August 14, 2017 1:57 PM
> To: Long Li <[email protected]>; Steve French <[email protected]>;
> [email protected]; [email protected]; linux-
> [email protected]
> Subject: RE: [[PATCH v1] 21/37] [CIFS] SMBD: Implement API for upper layer
> to receive data
>
> > -----Original Message-----
> > From: [email protected] [mailto:linux-cifs-
> > [email protected]] On Behalf Of Long Li
> > Sent: Wednesday, August 2, 2017 4:11 PM
> > To: Steve French <[email protected]>; [email protected];
> > samba- [email protected]; [email protected]
> > Cc: Long Li <[email protected]>
> > Subject: [[PATCH v1] 21/37] [CIFS] SMBD: Implement API for upper layer
> > to receive data
> >
> > /*
> > + * Read data from receive reassembly queue
> > + * All the incoming data packets are placed in reassembly queue
> > + * buf: the buffer to read data into
> > + * size: the length of data to read
> > + * return value: actual data read
> > + */
> > +int cifs_rdma_read(struct cifs_rdma_info *info, char *buf, unsigned
> > +int size) {
> >...
> > + spin_lock_irqsave(&info->reassembly_queue_lock, flags);
> > + log_cifs_read("size=%d info->reassembly_data_length=%d\n", size,
> > + atomic_read(&info->reassembly_data_length));
> > + if (atomic_read(&info->reassembly_data_length) >= size) {
>
> If the reassembly queue is protected by a lock, why is an atomic_read() of its
> length needed?
Will change this to non-atomic.
>
> > + // this is for reading rfc1002 length
> > + if (response->first_segment && size==4) {
> > + unsigned int rfc1002_len =
> > + data_length + remaining_data_length;
> > + *((__be32*)buf) = cpu_to_be32(rfc1002_len);
> > + data_read = 4;
> > + response->first_segment = false;
> > + log_cifs_read("returning rfc1002 length
> > %d\n",
> > + rfc1002_len);
> > + goto read_rfc1002_done;
> > + }
>
> I am totally confused. What does RFC1002 framing have to do with receiving
> an SMB Direct packet???
The upper layer expects RFC1002 length at the beginning of the payload. A lot
of protocol processing logic check and act on this value. Returning this value
will avoid changes to lots of other upper layer code.
This will be eventually fixed when a transport layer is added to upper layer
code. I recommend we do it in another patch.
>
> > +
> > + to_copy = min_t(int, data_length - offset, to_read);
> > + memcpy(
> > + buf + data_read,
> > + (char*)data_transfer + data_offset + offset,
> > + to_copy);
>
> Is it really necessary to perform all these data copies, especially under the
> reassembly_queue spinlock? This seems quite inefficient. Can the receive
> buffers not be loaned out and chained logically?
This will require upper layer code changes to move to use new buffers
allocated/loaned this way, and also deal with packet boundaries.
This code is not used to actually carry file data, which are normally done
through RDMA read/write.
If we want to do it, I suggest do another patch since more changes other than
transport are involved.
>
> Tom.
