[rds-devel] IB/iWARP code separation

[Or Gerlitz]

> All the evidences show that the iWARP patch doesn't fall into that category.  At a minimum it had to be broken into set of logical changes patches that built one on top of the other and can be easily reviewed at least by someone that knows the code being changed.
OK, to be more precise, doing some checks over the various branches 
where the iwarp changes were merged, I understand the set of changes is 
made of the three patches listed below, where the 1st and 2nd ones are 
of modest / relative small size and the 3rd being quite large, but not 
something which totally doesn't make sense (below is also the diffstat 
for that patch). Was it really unavoidable to get this level of 
instability, I don't know. Also, I understood that part of the story was 
other changes which are not specific to iwarp?

Or.



From: Jon Mason <jon at opengridcomputing.com>
Date: Tue, 15 Jul 2008 07:02:00 +0000 (-0700)
Subject: RDS: add iWARP bcopy support

From: Jon Mason <jon at opengridcomputing.com>
Date: Thu, 31 Jul 2008 18:55:06 +0000 (-0500)
Subject: rds: support for IB_DEVICE_LOCAL_DMA_LKEY

From: Jon Mason <jon at opengridcomputing.com>
Date: Fri, 10 Oct 2008 21:33:21 +0000 (-0700)
Subject: RDS: iWARP RDMA enablement
> $ diffstat rds_iwarp_rdma.patch
>  ib.c      |   24 +
>  ib.h      |   41 ++
>  ib_cm.c   |  145 +++++----
>  ib_rdma.c |  978 ++++++++++++++++++++++++++++++++++++++++++++++----------------
>  ib_recv.c |    4
>  ib_send.c |  152 ++++++++-
>  message.c |    2
>  rdma.c    |   22 -
>  rdma.h    |    9
>  rds.h     |    6
>  send.c    |    3
>  11 files changed, 1045 insertions(+), 341 deletions(-)
>
>