Hi,
Some time ago, I thought there were some thoughts about making the
outstanding RDMA count to be configurable via a patch to the mthca
driver (the 'rdb_per_qp' parameter in mthca_main.c).
Just curious if the patch was going to go in at some point.
Thanks,
Manpreet.
We have seen this happen over an IB analyzer. Recompiling the mthca
driver with a high value like 64 or 128 works around this problem.
When the condition hits, the HCA receiving the 4+ RDMAs generates an invalid request error.
Any ideas as to when this patch might enter the mainline sources?
Hi,
It seems that the number of outstanding RDMAs that a Mellanox HCA can
handle has been configured at 4 (mthca_main.c: default_profile:
rdb_per_qp). And the HCAs can support a much higher value (128 I think).
Could we move this value higher or atleast make it configurable?
Thanks,
Manpreet.
, Manpreet Singh wrote: Hi, I am observing the following with redhat kernel rpm at: http://people.redhat.com/dledford/Infiniband
, which uses openib version 3965. This is on an RHEL4 install. When the system is rebooted, ib_core, ib_mad and ib_mthca modules get loaded automatically. When I load
Hi,
I was wondering if anyone has seen this or knows the cause. I am using
the redhat RPM from:
https://openib.org/svn/gen2/branches/backport-to-2.6.9/RPMS/i686/kernel-smp-2.6.9-11.OpenIB.4016.EL.root.i686.rpm
.
I did a 'ib_register_mad_agent' which was successful, then tried to send a MAD out
Hi Roland,
The following is the output with the card after adding
the lines you mentioned:
ib_mthca :04:00.0: Missing UAR, aborting.
ib_mthca :04:00.0: flags: 0, len: 0
Thanks,
Manpreet.
__
Yahoo! Mail - PC Magazine Editors' Choice
Hi,
I was wondering if the Mellanox PCI-X cards (with
memory) are still supported in the openib stack.
Although the code seems to include it in the PCI
device id list, but I get the following error when I
load ib_mthca:
ib_mthca :04:00.0: Missing UAR, aborting
Details of the configuration: