Roland Dreier [EMAIL PROTECTED] wrote on 17.07.2007 19:52:55:
At a higher level, I'm left wondering why nobody talked about multiple
EQs during the last months of the 2.6.22 process and now all of a
sudden it becomes urgent in the last few days of the 2.6.23 merge
window. That's not really
Here's some anecdotal evidence :)
http://lists.openfabrics.org/pipermail/general/2007-May/035758.html
Right, but then we went on to say that we probably want to use
multiple vectors to separate out multiple HCA ports rather than
send/sreceive on the same port. And the current IPoIB
Quoting Roland Dreier [EMAIL PROTECTED]:
Subject: Re: [PATCH 01/10] IB/ehca: Support for multiple event queues
Here's some anecdotal evidence :)
http://lists.openfabrics.org/pipermail/general/2007-May/035758.html
Right, but then we went on to say that we probably want to use
multiple
The eHCA driver can now handle multiple event queues (read: interrupt
sources) instead of one. The number of available EQs is selected via the
nr_eqs module parameter.
CQs are either assigned to the EQs based on the comp_vector index or, if the
dist_eqs module parameter is supplied,
Roland Dreier [EMAIL PROTECTED] wrote on 16.07.2007 18:04:26:
It seems not quite right to me for the driver to advertise nr_eqs
completion vectors, but then if round-robin is turned on to ignore the
consumer's decision about which vector to use.
The round-robin feature was primarily meant as
Roland Dreier [EMAIL PROTECTED] wrote on 16.07.2007 18:04:26:
Do you have any data on how well this round-robin assignment works?
It seems not quite right to me for the driver to advertise nr_eqs
completion vectors, but then if round-robin is turned on to ignore the
consumer's decision about
No, I've no figures to provide here. The background of this dist_eqs
option is actually to allow us testing across all event queues
without to change the testcases resp consumers to use certain
event queue number. Thus, I should comment it as EXPERIMENTAL?
Seems like it's just
I still haven't seen much code using the feature or
even any anecdotal information about the performance impact.
Here's some anecdotal evidence :)
http://lists.openfabrics.org/pipermail/general/2007-May/035758.html
--
MST
___
general mailing list
Hello Roland,
I still haven't seen much code using the feature or
even any anecdotal information about the performance impact.
The multiple links performance has been significant improved according to
the prototype IPoIB-UD mode test for eHCA driver esepcially for two links
on the same