On Sep 19, 2014, at 6:39 AM, Samuel Ghinet
wrote:
> Hi Nithin,
>
> Thanks for the clarifications on the Tcp segmentation and NBs sharing info. I
> had missed that detail :)
> I will use the flow from the first NBL for all the NBLs that resulted from
> tcp segmentation.
>
> Yes you're right:
Hi Nithin,
Thanks for the clarifications on the Tcp segmentation and NBs sharing info. I
had missed that detail :)
I will use the flow from the first NBL for all the NBLs that resulted from tcp
segmentation.
Yes you're right: since OvsProcessOneNb is wrapped within a rw lock and unlock,
we are
hi Sam,
Pls. find my replies inline.
On Sep 14, 2014, at 9:12 PM, Samuel Ghinet
wrote:
> Hello Nithin,
>
> Thanks a lot again for the review!
Welcome. As you said, the code is intricate. It is fine though. Thanks for
working on this one.
> -
>> More than that I was a
Eitan: yes, I agree. That's why we need to keep the atDispatch variable (in the
patch, which holds the result of the flag check).
From: Eitan Eliahu [elia...@vmware.com]
Sent: Monday, September 15, 2014 7:06 PM
To: Samuel Ghinet; Nithin Raju
Cc: dev@openvsw
"Actually, we should be at DPC level, because OvsProcessOneNb is called by the
NDIS Send callback, which is always called (from observation) at DPC level."
This is usually true but we could not count on that as it can be called in a
passive IRQL as well.
Eitan
-Original Message-
From: Sa
Hello Nithin,
Thanks a lot again for the review!
-
> More than that I was a little concerned about the changes in OvsTunnelTx to
> split a TSOed NBL (with multiple NBs) into multiple smaller NBLs. I believe
> this would have a performance impact (when we measure it).
> T
hi Sam,
Thanks for the updating the patch to make it specific to "handle NBLs with
multiple NBs".
In general it looks good. There are a few bugs w.r.t NBL completion that I have
highlighted.
More than that I was a little concerned about the changes in OvsTunnelTx to
split a TSOed NBL (with mul
All NET_BUFFERs of a NET_BUFFER_LIST must go through the pipeline: extract,
find flow, execute. Previously, only the first NET_BUFFER of a NET_BUFFER_LIST
was going through this pipeline, which was erroneous.
OvsPartialCopyToMultipleNBLs is used to make each NET_BUFFER have its own
NET_BUFFER_LIST
