Hi Byran,
Can you please include this patch submitted by Shirley Ma that went
into the main openIB trunk in revision 6255 into the
release 1.0 branch.
Shirley Ma xma at us.ibm.com
Wed Apr 5 10:46:39 PDT 2006
[openib-general] Re: [PATCH] repost: IPoIB queue size tune patch
---
t;[EMAIL PROTECTED] To
il> Kevin Reilly/Poughkeepsie/[EMAIL
PROTECTED]
cc
03/25/2006 12:04 openib-general@openib.org
We wanted to let others in the OpenIB community know that IBM is using
and testing openIB as it's base support for it's IBM HCA
as part of linux on Power offerings. We are in the process of testing
much of our clustering software and high performance subsystems
layered over some of the openIB
Was the intent that the openIB verbs be "thread safe" or was the intend
that if thread safety was required it be implemented in the layers above
the
verbs. If there was a discussion on verb thread safety can somebody please
point me to it? Since the verbs are a pretty low level access to IB
I
Thanks Sean,
I think the rdma_resolve_addr() does what we want. Translate a local IP
to a ib_device structure that i can use in the ibverbs.
What we want to do is pretty simple and we won't need to create a
connection.
Can we have a discussion on the timeframe for this?
Kevin J. Reilly
STSM,
2005 12:47 Satyanarayana/Beaverton/[EMAIL
PROTECTED]
PM cc
Hal Rosenstock <[EMAIL PROTECTED]>,
On Mon, 2005-10-18 at 10:07, Kevin Reilly wrote:
>On Mon, 2005-10-17 at 10:07, Hal Rosenstock wrote:
>> > Should this code work, because it seems that out_dev is a kernel
>> > address (platform: PPC64) which cannot accessed by a userspace
>> > program. Via
Thanks Sean and Roland for your response. You right it's not a primitive in
the IB verbs or iWARP unless you think the verbs define a
non-consuming poll(). Which is a stretch for sure. We fell into the trap
of structuring with it because it was in VAPI and maybe other
that coded over VAPI fell
AIL PROTECTED]
m> To
Kevin Reilly/Poughkeepsie/[EMAIL
PROTECTED]
07/08/2005 05:38 cc
reier
<[EMAIL PROTECTED]
m> To
Kevin Reilly/Poughkeepsie/[EMAIL
PROTECTED]
07/08
struct ibv_device_attr {
..
uint32_t vendor_id;
uint32_t
vendor_part_id;
uint32_t hw_ver;
Does gen2 user access lib have a way to peek at the CQ to see if it has
entries on it like VAPI did? All i see is ways to pop an entry
off the CQ.
Kevin J. Reilly
STSM, HPC Architecture
-Federation/HPS Chief Engineer
-HPC interconnect architect
(office) 845-433-7976 (tieline) 8-293-7976
___
People need to excuse all my annoying startup questions. I just notices
that the user space ib verbs lib doesn't have any functions
to register completion or event handlers. Is the intent that the API will
only support polling or are the asynchronous event handling stuff
in somebodies branch
ROTECTED]
m> To
Kevin Reilly/Poughkeepsie/[EMAIL
PROTECTED]
06/21/2005 09:05 cc
AM
Maybe somebody could help me understand the proper way to map between an IP
address assigned to a port to the
"device name" and "port number" in the gen2 architecture. If I have an IP
address can I map it to a name that i get back
from ibv_get_device_name() or pass to ibv_open_device().
Roland Dreier
<[EMAIL PROTECTED]
om>To
Kevin Reilly/Poughkeepsie/[EMAIL
PROT
ROTECTED]
om>To
Kevin Reilly/Poughkeepsie/[EMAIL
PROTECTED]
06/06/2005 11:18 cc
AM
I'm been hunting around for the proper method for determining the number of
ports a device/hca has by using the gen2
user access library. I can't figure out how except by loop through port
numbers calling ibv_query_port(). I figure maybe
i'm missing something? Can somebody steer me in the right
om>To
Kevin Reilly/Poughkeepsie/[EMAIL
PROTECTED]
04/14/2005 10:55 cc
PMopenib-general@open
I was wonder where i could find information on the openIB gen2 user space
verbs API? One of my key questions
is how much different then VAPI. Is it a superset of VAPI and if so what
function was added?
Kevin J. Reilly
STSM, HPC Architecture
-Federation/HPS Chief Engineer
-HPC interconnect a
20 matches
Mail list logo
