[ofw] [Fwd: Connection teardown in WinSock Direct and CM related questions]

Fab Tillier ftillier at windows.microsoft.com
Wed Mar 12 12:51:51 PDT 2008


>From memory, thing might have changed, but:

The CM will move the QP to the error state when the DREQ receives a response or times out.  The application can move it early though, which is what the WSD provider does, probably just to help expedite things and not strictly necessary.

-Fab

-----Original Message-----
From: ofw-bounces at lists.openfabrics.org [mailto:ofw-bounces at lists.openfabrics.org] On Behalf Of Hal Rosenstock
Sent: Monday, March 10, 2008 8:50 AM
To: ofw at lists.openfabrics.org
Subject: [ofw] [Fwd: Connection teardown in WinSock Direct and CM related questions]

Hi,

I'm posting this again hoping someone can shed some light on the below.
Thanks in advance.

-- Hal

-------- Forwarded Message --------
From: Hal Rosenstock <hrosenstock at xsigo.com>
To: ofw at lists.openfabrics.org
Subject: Connection teardown in WinSock Direct and CM related questions
Date: Wed, 27 Feb 2008 15:39:42 -0800
Hi,

In the the Windows user mode code for Windows Sockets Direct, it sends
a CM DREQ (ib_cm_dreq) and then just modifies the QP state to error,
forcing a flush of outstanding work items. Can someone explain this ?

Is this a workaround for a CM issue or the normal way to handle
connection termination ? Isn't the DREQ guaranteed to either have a DREP
reply event, or a timeout or error ? Who's responsibility is it to
assure that the QP is moved to an appropriate state (CM or "user"/ULP) ?

Thanks.

-- Hal
_______________________________________________
ofw mailing list
ofw at lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ofw



More information about the ofw mailing list