[openfabrics-ewg] Cisco SQA results so far for OFED 1.1 rc3
Tziporet Koren
tziporet at mellanox.co.il
Thu Sep 7 04:24:05 PDT 2006
Hi Scott,
Thanks for the details report.
This is the status of bugs that are not Cisco specific (e.g. tvflash)
for RC4:
219 OFED 1.1rc3 contains prerelease unstable libibverbs code -
fixed
221 SRP on V40Z and Sun T4 gets Kernel BUG at spinlock:118 - open
and Roland should work on this
222 ib_uverbs fails to load on ia64, OFED 1.1 - rc3 (opened by Bob
Woodruff, but we saw it too) - fixed
Tziporet
-----Original Message-----
From: openfabrics-ewg-bounces at openib.org
[mailto:openfabrics-ewg-bounces at openib.org] On Behalf Of Scott
Weitzenkamp (sweitzen)
Sent: Thursday, September 07, 2006 9:23 AM
To: EWG
Cc: OPENIB
Subject: [openfabrics-ewg] Cisco SQA results so far for OFED 1.1 rc3
Testing is still continuing, we have not started testing RHEL4 U4, SLES
10, IPoIB HA, or SRP HA yet.
Some high points of rc3 (all testing done with Mellanox HCAs and Cisco
switches):
* We have migrated from Intel 9.0 compilers to 9.1.
* Are seeing 2 million msg/sec with MVAPICH.
* Sinai 1.1.000 firmware fixes SDP scalability.
* Open MPI 1.1.1 is working better than 1.1.
* We see up to 3.5 Gb/sec max throughput with IPoIB on latest
Intel Xeon and AMD Opteron processors.
Many bug fixes have been tested:
* 193 OFED 1.1 rc1: openib-diags should not be linked with
opensm libs
* 197 OFED 1.1rc1: Open MPI fails on RHEL4 64-bit
* 74 OFED 1.0 rc4: Open MPI Pallas test hangs
* 109 OFED 1.0 rc5: SDP can't sustain 100+ concurrent SDP
connections (mem leak?)
* 101 OFED 1.0: need documentation on openib-diags
* 80 OFED 1.0 rc4: Open MPI fails on RHEL4 U3 ppc64
* 135 OFED 1.0: MVAPICH doesn't work on RHEL4 U3 ppc64
* 176 OFED 1.0: mpicc fails with Intel C on RHEL4 IA64
* 179 move /usr/local/ofed/sbin binaries to
/usr/local/ofed/bin
* 103 OFED 1.0: change ibutils to not depend on opensm
Scott Weitzenkamp
SQA and Release Manager
Server Virtualization Business Unit
Cisco Systems
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/ewg/attachments/20060907/46c63bbb/attachment.html>
More information about the ewg
mailing list