At 10:41 PM 10/18/2005, Mohit Katiyar, Noida wrote:
Content-class:
urn:content-classes:message
Content-Type: multipart/alternative;
boundary=_=_NextPart_001_01C5D46F.B45AF930
Hi all,
Can anyone tell me are there any specific I/O controller for the
connection between the TCA and SCSI devices
I put out a initial set of files (as
discussed) on openib.org svn.
The Kconfig option will follow when
I've verified that it really compiles as is in svn.
The same set with makefiles and install
scripts is available from sourceforge as ehca2_0033
Christoph
Roland Dreier wrote on 04.10.2005
Hi there,
I was wondering whether the CM API currently (I'm currently using svn 3470) supports establishing UC connections? I have the RC transport type working fine using the CM.
I've found that somewhere between me sending and receiving the REQ message, the qp_type variable changes from UC to
Arkady,
Intel MPI (real consumer of uDAPL) has no
problem with this change.
-arlin
From:
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Kanevsky, Arkady
Sent: Wednesday, October 19, 2005
6:40 AM
To: Grant Grundler; Caitlin
Bestler
Cc: Roland Dreier;
[EMAIL
On Tue, 2005-10-18 at 17:11, Eitan Zahavi wrote:
Hal Rosenstock wrote:
Currently, OpenSM does not support an interactive console. There has
been a desire to introduce the ability to change certain parameters (as
well as display things) once OpenSM has started. This patch introduces
the
Title: Message
Arlin,
just
to clarify, Intel MPI will not have problems with useing less than 64
bytes
of
private data.
Ifa solution will provide you with 48 bytes of
private data will it be sufficient?
Arkady
Arkady Kanevsky
email: [EMAIL PROTECTED]
Network
Appliance
phone:
Title: Message
Yes, 48 bytes would be sufficient.
From: Kanevsky, Arkady
[mailto:[EMAIL PROTECTED]
Sent: Wednesday, October 19, 2005
8:32 AM
To: Davis, Arlin R;
[EMAIL PROTECTED]; Grant Grundler
Cc: [EMAIL PROTECTED];
openib-general@openib.org
Subject: RE: [dat-discussions]
Title: Message
Oracle currently depends on 64 bytes of private
data for connect and accept.
- Original Message -
From:
Kanevsky,
Arkady
To: Davis, Arlin R ; [EMAIL PROTECTED]
; Grant Grundler
Cc: [EMAIL PROTECTED] ; openib-general@openib.org
Sent: Wednesday,
Hal Rosenstock wrote:
On Tue, 2005-10-18 at 17:11, Eitan Zahavi wrote:
Hal Rosenstock wrote:
Currently, OpenSM does not support an interactive console. There has
been a desire to introduce the ability to change certain parameters
(as
well as display things) once OpenSM has started. This
Steven Wooding wrote:
I was wondering whether the CM API currently (I'm currently using svn
3470) supports establishing UC connections? I have the RC transport type
working fine using the CM.
Unless there's a bug, nothing in the CM should prevent UC from working.
I had a quick look in the
On Wed, 2005-10-19 at 12:28, Eitan Zahavi wrote:
Hal Rosenstock wrote:
On Tue, 2005-10-18 at 17:11, Eitan Zahavi wrote:
Hal Rosenstock wrote:
Currently, OpenSM does not support an interactive console. There has
been a desire to introduce the ability to change certain parameters
(as
Richard MessageOracle currently depends on 64 bytes of private
Richard data for connect and accept.
How do you work with RDS, which is not connection oriented and hence
does not even have connect or accept?
- R.
___
openib-general mailing list
Richard Frank wrote:
Oracle currently depends on 64 bytes of private data for connect and
accept.
Is any of that data used to exchange address information?
It's impossible to provide both the source and destination address in the CM REQ
private data and still give the user 64 bytes. The
Sean,
if look at the proposal it shows 2 ways to address this.
1. Have 2 protocols.
One just send SRC IP address and port, and provdie 64 bytes to ULP.
Another one send both SRC and DEST info and leaves 48(+-) bytes of
private data for ULP.
2. Have 2 protocols.
Split IPv4 and IPv6 methods.
For
IBMEHCA I put out a initial set of files (as discussed) on
IBMEHCA openib.org svn. The Kconfig option will follow when I've
IBMEHCA verified that it really compiles as is in svn.
Great, it's good to have the code in svn at last.
From a first look I don't see any ebus files checked
Here is the Kconfig/Makefile patch I
used to build ehca as a part of linux build.
diff -urN infiniband/Kconfig infiniband-patch/Kconfig
--- infiniband/Kconfig2005-10-17 09:28:41.0
-0700
+++ infiniband-patch/Kconfig2005-10-17
09:51:08.0 -0700
@@ -30,6 +30,8 @@
source
On Wed, 2005-10-19 at 16:09, Troy Benjegerdes wrote:
I'm trying to rebuild opensm, and the libibumad configure is failing because
IB_WARN is apparently not defined anyplace I can find it.
It should be IBWARN. Where is IB_WARN or do you mean IBWARN ?
It should be found in
On Tue, 2005-10-18 at 18:40, Kevin Reilly wrote:
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
Hi Sean,
I've modified cmpost to try to UC and get similar
results as my app. The changes I made to cmpost.c was
to change RC to UC (two places) and remove the
following req and rep parameters as I beleive these
are not required for UC:
req.retry_count = 5;
rep.rnr_retry_count =
Does the patch below help? It looks like there is a missing break
after the UC case in the kernel CM, so the code falls through and
overwrites the field with the value for RC.
- R.
diff --git a/drivers/infiniband/core/cm_msgs.h
b/drivers/infiniband/core/cm_msgs.h
---
Roland Dreier wrote:
Does the patch below help? It looks like there is a missing break
after the UC case in the kernel CM, so the code falls through and
overwrites the field with the value for RC.
Good catch. I overlooked this about 10 times now... The break should be there.
- Sean
On Wed, 19 Oct 2005, Roland Dreier wrote:
roland Richard MessageOracle currently depends on 64 bytes of private
roland Richard data for connect and accept.
roland
roland How do you work with RDS, which is not connection oriented and hence
roland does not even have connect or accept?
Sorry Roland,
My fault. I had the wrong access flags set when I
registered the memory region.
Thanks for reply though.
Cheers,
Steve.
--- Roland Dreier [EMAIL PROTECTED] wrote:
However, reading the completion contents, I see that
it is a receive
completion with status local protection
James The D is somewhat misleading. It refers to the
James functionality provider to the consumer application.
Right, that's what we're talking about. The RDS implementation only
needs a few bytes of private data on top of the IP address info. So
the RDS implementation itself is clearly
On 10/16/05 08:26, Steve Wooding wrote:
Hi there,
I'm trying to make a QP connection using the CM, but the active side
cannot get to the RTR state. ibv_modify_qp returns errorno -22, invalid
argument.
What would the best way to find out exactly what the error is and help
me fix my app
On Wed, Oct 19, 2005 at 04:30:06PM -0400, Hal Rosenstock wrote:
On Wed, 2005-10-19 at 16:09, Troy Benjegerdes wrote:
I'm trying to rebuild opensm, and the libibumad configure is failing
because
IB_WARN is apparently not defined anyplace I can find it.
It should be IBWARN. Where is
On Wed, 2005-10-19 at 21:02, Troy Benjegerdes wrote:
On Wed, Oct 19, 2005 at 04:30:06PM -0400, Hal Rosenstock wrote:
On Wed, 2005-10-19 at 16:09, Troy Benjegerdes wrote:
I'm trying to rebuild opensm, and the libibumad configure is failing
because
IB_WARN is apparently not defined
make[1]: Entering directory `/usr/src/openib-src/userspace/libehca'
if /bin/sh ./libtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I.
-I.-O3 -g -Wall -D_GNU_SOURCE -DP_SERIES -I../libibverbs/include -Isrc
-g -O2 -MTsrc_libehca_la-ehca_umain.lo -MD -MP -MF
Hrrm.. it looks like for some reason the top level makefile didn't
rebuild libibcommon.
Not sure why that would be.
In my top level generated Makefile,
LIBS:=libibcommon libibumad libibmad
@for i in $(LIBS); do\
if [ -x $$i/autogen.sh ]; then\
29 matches
Mail list logo
