> Are there any examples out there of IB applications that use
> RDMA_OPTION_ID_TOS? I'm looking to make this available for iwarp
> transports, but I'm still trying to understand how its currently used,
> and what values are used. Since this option is level RDMA_OPTION_ID, I
> guess it is transpo
Changes from V1:
Fix spacing
Changes from V2:
Add option to disable query.
Signed-off-by: Ira Weiny
---
include/opensm/osm_perfmgr.h | 15 +++
include/opensm/osm_subnet.h |1 +
opensm/osm_console.c | 13 ++-
opensm/osm_perfmgr.c | 230 ++
Subject: [PATCH 02/13] opensm/perfmgr: clean up: break out redirect processing
from pc_recv_process
Changes from V1:
Cleaned up processing when CPI redirect is invalid
Changes from V2:
If CPI is a redirection only set cpi_valid if the redirection was valid
(Also
Hi Ira,
We (Oracle IB development) are submitting six patches against
libibmad-1.3.9 (as attached).
Please review our submission of changes.
Many thanks,
Boris
libibmad: To reserve upper 8 bits of tid used by solaris SRIOV driver
Signed-off-by: Brendan Doyle
diff -uprN ./SOURCES/libibmad-1.3.
On Wed, Feb 27, 2013 at 11:11 AM, Or Gerlitz wrote:
> You can also take a look on the upstream trees for libibverbs and
> libmlx4 maintained by Roland which have a debian dir, which you
> probably can re-use for the managment libraries and opensm, here are
> the gitweb pointers,
>
> http://git.ker
On Wed, Feb 27, 2013 at 5:47 PM, Narayan Desai wrote:
> Hi Ilya. The easiest way to see packaging for a library (from an
> upstream apt source) is to run
[...]
Ilya,
You can also take a look on the upstream trees for libibverbs and
libmlx4 maintained by Roland which have a debian dir, which you
Hi Ilya. The easiest way to see packaging for a library (from an
upstream apt source) is to run
# This downloads the original source, debian specific patches, and a
package description file
$ apt-get source
# this unpacks all of the files
$ dpkg-source -x *.dsc
This produces a source directory wi
- add transport type for making transport-specific decisions
- rdma read sink mrs need REMOTE_WRITE access for iwarp transports
- add chelsio devices
Changes since V1:
Restored the use of RDMA_MAX_RESP_RES/RDMA_MAX_INIT_DEPTH for rdmacm
connection setup.
Signed-off-by: Steve Wise
---
src/perf
Hey Sean,
Are there any examples out there of IB applications that use
RDMA_OPTION_ID_TOS? I'm looking to make this available for iwarp
transports, but I'm still trying to understand how its currently used,
and what values are used. Since this option is level RDMA_OPTION_ID, I
guess it is t
On 2/27/2013 9:27 AM, Roland Dreier wrote:
> On Tue, Feb 26, 2013 at 11:22 PM, Ilya Nelkenbaum
> wrote:
>> Does the libibumad that is installed in Debian synchronized with the
>> upstream version?
>
> I'm not sure what you're asking. At appropriate times, new upstream
> versions of software are
1) if use -G or -D option : we obtain sl before doing perf query.
2) if no destination is given : we obtain sl for every pair source-destination.
3) if no destination is given and use --skip-sl option : we don't obtain
sl to all nodes in the fabric,in this case sl=0 for all node pairs.
Signed-off-
On 26.02.2013 17:55, Roland Dreier wrote:
[...]
> In fact I bet this is why the bug has been there as long as it has
> been: almost no one is using IPv6 on IPoIB seriously, and IPv4 should
> work OK as you point out.
Thanks a lot, Unfortunately, we are using IPoIB with IPv6 in
production for t
On 2/21/2013 8:34 PM, Steve Wise wrote:
On 2/21/2013 12:27 PM, Hefty, Sean wrote:
ib_read_lat displays RTT latency, while the other lat programs display
RTT/2. Is this intended? Or is read_lat.c/print_report() missing a
divide-by-two?
The tests should measure how long it takes the data to reac
13 matches
Mail list logo