[ewg] New name for IBV_QPT_RAW_ETH

[Alex Rosenbaum]

It is true that on the send path you can send any sort of packet. 
The user of the QP is required to build the MAC header so he could
potentially build any other sort of header instead.
AFAIK the NIC will wrap the post_sent data in an Ethernet type II frame
which includes the tailing CRC.

But on the receive path this QP assume an MAC header. The qp attach GID
value gets the multicast MAC destination address to catch the correct
ingress packets.

Bottom line is that I think RAW_ETH is a good name.
On the other hand, I never fully understand what does RAW_ETY stand for?
Maybe we should change it instead to better represent what it does.

PS: I agree with you that all ibverbs enums should start with a
standard, in this case using IBV_QPT_xxxx.

_________________
 Alex Rosenbaum 

 


-----Original Message-----
From: ewg-bounces at lists.openfabrics.org
[mailto:ewg-bounces at lists.openfabrics.org] On Behalf Of Moni Shoua
Sent: Tuesday, June 22, 2010 11:35 AM
To: Aleksey Senin; Walukiewicz, Miroslaw; Liran Liss; EliCohen
Cc: ewg
Subject: [ewg] New name for IBV_QPT_RAW_ETH

Hi,
In yesterday's EWG meeting we agreed that a better name is required
instead of IBV_QPT_RAW_ETH.

The qp type IBV_QPT_RAW_ETH was introduced lately and it represents a QP
that doesn't add headers to the data in post_send operation. Today, this
QP is used in applications that want to send Ethernet frames to the
wire. However, it is not limited to the Ethernet protocol and one can
write an application that sends any type of packet (am I wrong here?)

Therefore I can think of the following names

IBV_QPT_FRAME
IBV_QPT_PACKET
IBV_QPT_NOHDR

Also, the prefix IBV may be misleading (it's not an IB QP) so this it
can be replaced in the above suggestions to RAW.
I don't tend to agree to this because I think that all QP types should
start with the same prefix. 

You are all welcome to state your opinion so please do so.


thanks 

Moni S.
_______________________________________________
ewg mailing list
ewg at lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg