Grant> FMR is a red herring. I tried SVN r2080 and it has roughly
Grant> the same performance as r2082 (when FMR was committed) and
Grant> later r210x. "packed" attribute is a red herring too.
Grant> Performance stunk with r2050 and I will do a binary search
Grant> this week until I sort out which changes doubled the
Grant> perf. ISTR there was one change related to a "double
Grant> mapping" issue and I will be tracking that down in a few
Grant> days.
A binary search to find the changeset that makes the difference would
be really useful. I read through the svn log from r2046 through r2082
and I don't see anything that should make a difference to IPoIB.
The only changes that seem remotely plausible are
r2059 "Set skb->mac.raw on receive"
r2068 "Make address handle verbs usable from interrupt context"
but I don't see how either one could really have an effect.
So I wonder what obvious thing I'm missing...
- R.
_______________________________________________
openib-general mailing list
[email protected]
http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
