Hello all.
This patch was updated according new API of RDMAoE (It looks that
community agreed with it).
Thank you for your comments.
Jeff Squyres wrote:
Sorry for the delay -- SC and the Forum meeting made a disaster of my
INBOX (again).
Pasha and I just talked about this patch quite
Sorry for the delay -- SC and the Forum meeting made a disaster of my
INBOX (again).
Pasha and I just talked about this patch quite a bit in IM. He's got
my feedback (pretty minor, actually). I'm ok with the changes Pasha
and I discussed (i.e., no need for another review).
Does anyone e
Hello Jeff.
The most code of the patch was removed.
Please see below the new patch.(file lle.patch is attached)
Thank you for your time.
Jeff Squyres wrote:
Perhaps since we now have 2 levels of m4 checking that is necessary
(1) check to see if we have transport_type, and 2) check to see if
h
Jeff Squyres wrote:
Perhaps since we now have 2 levels of m4 checking that is necessary
(1) check to see if we have transport_type, and 2) check to see if
have RDMA_TRANSPORT_RDMAOE), perhaps it would be better to put all
these checks into a single place somewhere to avoid proliferating the
#i
Perhaps since we now have 2 levels of m4 checking that is necessary
(1) check to see if we have transport_type, and 2) check to see if
have RDMA_TRANSPORT_RDMAOE), perhaps it would be better to put all
these checks into a single place somewhere to avoid proliferating the
#if's for these 2 t
I see you remove support for #if
defined(HAVE_STRUCT_IBV_DEVICE_TRANSPORT_TYPE) -- that doesn't seem
like a good idea. We still have users on older OFED's without that
field.
Can you create a 1.5 ticket for this item?
On Nov 1, 2009, at 6:44 AM, Vasily Philipov wrote:
The attached patch
Jeff,
Can you please check that we do not brake iwarp devices functionality ?
Pasha
Vasily Philipov wrote:
The attached patch adds support for RDMAoE (RDMA over Ethernet)
devices to Openib BTL. The code changes are very minimal, actually we
only modified the RDMACM code to provide better suppo
