Hi Woody,
I have heard back from Marty Schlining who is the engineer involved with all the testing at UNH-IOL and his response is below. It appears that the original concerns have been resolved but Bug 2395 presents a serious problem for us if we are going to base OFED 3.5 on the older kernel. Thanks Rupert From: Marty Schlining [mailto:[email protected]] Sent: Monday, November 19, 2012 5:38 PM To: Rupert Dance Cc: Edward Mossman; [email protected] Subject: RE: [ewg] EWG/OFED meeting agenda for today Nov-13 2012 Rupert, Bart has made all of the changes required for OFED and ib_srp. The delay for unloading ib_srp is expected at this time according to Bart. There are no more error messages in the kernel log. The main issue is this bug: <http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2395> Bug 2395 - Kernel crash using SRP extended sg descriptors, increased max_sectors_kb, when running direct IO. http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2395 According to Bart, this is a kernel bug that was fixed in 2.6.38 or greater. So, the new features in OFED 3.5 regarding ib_srp will crash the present test setup. You only ran ½ of the ib_srp capabilities. Im not really sure how to proceed here. If OFED 3.5 is released and it supports a kernel that has this bug, then it is possible to crash the system. And, since DDN supports this feature, we would not like a version of OFED that does not work properly with RHEL 6.3, SL 6.3, etc. We can talk more about this. Getting ready to leave for the day. -Marty From: Rupert Dance [mailto:[email protected]] Sent: Monday, November 19, 2012 4:18 PM To: Marty Schlining Cc: Edward Mossman; [email protected] Subject: FW: [ewg] EWG/OFED meeting agenda for today Nov-13 2012 Hi Marty, The EWG is anxious to release RC3 and I know there have been several SRP patches. I was able to successfully run the OFA SRP Test Plan while at SC12. Can you provide me with a status update regarding your testing an whether you think we are ready for RC3. I only see one open bug for SRP and there has been no update since 10/26/2012. I remember seeing a response saying that 20-20 seconds was expected behavior. If you feel there are other issues, can you create a new but or provide me with a concise list of the problems and the solutions you would like to see? http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2396 Thanks Rupert From: Woodruff, Robert J [mailto:[email protected]] Sent: Tuesday, November 13, 2012 12:16 PM To: Steve Wise; Tziporet Koren; Rupert Dance Cc: Vipul Pandya; [email protected]; Divy Le Ray Subject: RE: [ewg] EWG/OFED meeting agenda for today Nov-13 2012 Well the meeting was short today with only attendance from the Intel QIB team and SGI. Perhaps we can get a status update from the Intel iWarp folks via email on this thread ? I also wanted to know from Rupert if he was able to test the final SRP fix that was open last meeting, I know there was some delay in that testing due to Hurricane Sandy. Rupert ? So it looks like we are still a couple weeks away from getting the last critical bugs resolved from Chelsio, as per Steves email below. If anyone knows of any other outstanding Critical bugs still being investigated, please respond on this email thread. woody From: [email protected] [mailto:[email protected]] On Behalf Of Steve Wise Sent: Tuesday, November 13, 2012 8:30 AM To: Tziporet Koren Cc: Vipul Pandya; [email protected]; Divy Le Ray Subject: Re: [ewg] EWG/OFED meeting agenda for today Nov-13 2012 On 11/13/2012 10:27 AM, Tziporet Koren wrote: Hi All Unfortunately I cannot attend the meeting today Woody - I hope you can replace me All we need to close today is whether Chelsio and Intel drivers are ready so that Vlad can build RC3 I cannot attend today either. Vipul is out for Diwali. We have made great progress, but I see 2 blocker bugs still. Our goal will be to close them out by the end of next week. * Severity: blocker,critical * Status: NEW, ASSIGNED, REOPENED * Version: 3.5,3.5-rc1,3.5-rc2 * Component: cxgb3 driver,cxgb4 driver,iWarp Core _____ ID <http://bugs.openfabrics.org/bugzilla/buglist.cgi?bug_severity=blocker&bug_s everity=critical&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&comp onent=cxgb3%20driver&component=cxgb4%20driver&component=iWarp%20Core&query_f ormat=advanced&version=3.5&version=3.5-rc1&version=3.5-rc2&order=bug_id&quer y_based_on=> Sev <http://bugs.openfabrics.org/bugzilla/buglist.cgi?bug_severity=blocker&bug_s everity=critical&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&comp onent=cxgb3%20driver&component=cxgb4%20driver&component=iWarp%20Core&query_f ormat=advanced&version=3.5&version=3.5-rc1&version=3.5-rc2&order=bug_severit y%20DESC%2Cpriority%2C&query_based_on=> Pri <http://bugs.openfabrics.org/bugzilla/buglist.cgi?bug_severity=blocker&bug_s everity=critical&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&comp onent=cxgb3%20driver&component=cxgb4%20driver&component=iWarp%20Core&query_f ormat=advanced&version=3.5&version=3.5-rc1&version=3.5-rc2&order=priority%20 DESC%2Cbug_severity&query_based_on=> OS <http://bugs.openfabrics.org/bugzilla/buglist.cgi?bug_severity=blocker&bug_s everity=critical&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&comp onent=cxgb3%20driver&component=cxgb4%20driver&component=iWarp%20Core&query_f ormat=advanced&version=3.5&version=3.5-rc1&version=3.5-rc2&order=op_sys%2Cpr iority%2Cbug_severity&query_based_on=> Assignee <http://bugs.openfabrics.org/bugzilla/buglist.cgi?bug_severity=blocker&bug_s everity=critical&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&comp onent=cxgb3%20driver&component=cxgb4%20driver&component=iWarp%20Core&query_f ormat=advanced&version=3.5&version=3.5-rc1&version=3.5-rc2&order=assigned_to %2Cpriority%2Cbug_severity&query_based_on=> Status <http://bugs.openfabrics.org/bugzilla/buglist.cgi?bug_severity=blocker&bug_s everity=critical&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&comp onent=cxgb3%20driver&component=cxgb4%20driver&component=iWarp%20Core&query_f ormat=advanced&version=3.5&version=3.5-rc1&version=3.5-rc2&order=bug_status% 2Cpriority%2Cbug_severity&query_based_on=> Resolution <http://bugs.openfabrics.org/bugzilla/buglist.cgi?bug_severity=blocker&bug_s everity=critical&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&comp onent=cxgb3%20driver&component=cxgb4%20driver&component=iWarp%20Core&query_f ormat=advanced&version=3.5&version=3.5-rc1&version=3.5-rc2&order=resolution% 2Cpriority%2Cbug_severity&query_based_on=> Summary <http://bugs.openfabrics.org/bugzilla/buglist.cgi?bug_severity=blocker&bug_s everity=critical&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&comp onent=cxgb3%20driver&component=cxgb4%20driver&component=iWarp%20Core&query_f ormat=advanced&version=3.5&version=3.5-rc1&version=3.5-rc2&order=short_desc% 2Cpriority%2Cbug_severity&query_based_on=> 2380 <http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2380> blo P1 SLES [email protected] ASSI Kernel panic observed on client side while running iozone using NFSoRDMA on multiple shares. 2404 <http://bugs.openfabrics.org/bugzilla/show_bug.cgi?id=2404> cri P1 Othe [email protected] NEW rping fails to run with mismatched mpa_rev _____ No virus found in this message. Checked by AVG - www.avg.com Version: 2013.0.2793 / Virus Database: 2624/5890 - Release Date: 11/12/12 Internal Virus Database is out of date.
_______________________________________________ ewg mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
