[openib-general] RE: [PATCH 1 of 3] mad: large RMPP support, Round 2

[Sean Hefty]

Thanks for updating this.  I know that Hal already committed these changes, but
I have a couple of additional comments that I want to capture.

>+struct ib_mad_multipacket_seg {
>+	struct list_head list;
>+	u32 num;
>+	u16 size;
>+	u8 data[0];
>+};

Does size change between segments?  It doesn't seem like it should need to for
segments that belong to the same MAD.  We can remove size from here and add a
seg_size to ib_mad_send_buf.

> /**
>+ * *ib_mad_get_multipacket_seg - returns a given RMPP segment.
>+ * @send_buf: Previously allocated send data buffer.
>+ * @seg_num: number of segment to return
>+ *
>+ * This routine returns a pointer to a segment of a multipacket RMPP message.
>+ */
>+struct ib_mad_multipacket_seg
>+*ib_mad_get_multipacket_seg(struct ib_mad_send_buf *send_buf, int seg_num);

I think we're serialized everywhere this call is made, but this call is not
thread safe.  We may want to make a note of this in the comment.

- Sean