Add support for Emulex UCNA iscsi be2iscsi driver for both BE2 and BE3 asics
*** Bug 569539 has been marked as a duplicate of this bug. ***
*** Bug 572880 has been marked as a duplicate of this bug. ***
Created attachment 405438 [details] The be2iscso patch for RHEL5.6 This contains the be2iscsi changes intended for inclusion in RHEL5.6. These have been approved in linux-scsi list
Created attachment 417760 [details] Some more fixes for be2iscsi intended for RHEL5.6 Some more fixes for be2iscsi
Jay, When you send the batch of fixes you were telling me about to the list, then attach a patch with all the fixes and changes you want here. I will then get the ball rolling here on getting it all merged up in RHEL 5.6 and 5.5.z. Thanks.
Laurie - just wanted to confirm that *only* what's in Comment #4 is what's required for the DUP. Correct?
Yes...
(In reply to comment #8) > Yes... In that case, Mike - can you proceed in POSTing what's in this BZ now and then waiting for any other patches later?
Jay made this bz: https://bugzilla.redhat.com/show_bug.cgi?id=608801 and set it urgent and high priority. Emulex, did you guys want that in too?
(In reply to comment #10) > Jay made this bz: > https://bugzilla.redhat.com/show_bug.cgi?id=608801 > and set it urgent and high priority. Emulex, did you guys want that in too? Mike, I believe not, since that request just came in, and was past the cutoff for the DUP.
We can do it then. Jay, could you give me a list of upstream (linus's or James's tree) git commits for the code in comment 3 and 4?
in for 5.6 but not for dup.
Laurie - waiting on Jay's answer to Comment #12. Mike cannot proceed without knowing what commits to include.
Mike, This is a crucial patch that needs to go in. http://www.spinics.net/lists/linux-scsi/msg44908.html There was a confusion for me reg the approval as a whole lot of time was spent on the list on BSG, some of the checkin's did not go through as unfortunately they were tied to BSG.I have resubmitted many of the patches. Request to include only whatever has got thru
Just to confirm that http://marc.info/?a=124876522100002&r=1&w=2 (the 64K) patch is minimum of what I need into the RHEL5.6
Andrius, Emulex/SE are tied out. Let's proceed with the this set of RHEL5.6 iSCSI patches for the time critical RHEL5.5 DUP OEM programs. thank you laurie
Mike - are these patches ready for internal POSTing? HP/Emulex are waiting for this for their DUP. Thanks!
Jay at Emulex/SE just sent a patch for review, but is having trouble testing it.
in kernel-2.6.18-214.el5 You can download this test kernel from http://people.redhat.com/jwilson/el5 Detailed testing feedback is always welcomed.
We've verified the following DUPs but were unable to get the RHEL5.5 (IA32 or X86_64) to detect the iSCSI target configured in the Emulex option ROM setup utility on the BL460cG7 server. The MSA2324i target was tested. dd-be2iscsi-2.0.549.0-i686.iso dd-be2iscsi-2.0.549.0-x86_64.iso Thanks, Sandy
This bug must remain assigned to someone that can commit patches internally. Feel free to add anyone in the CC-list.
I saw a mention of option ROM in the bug report. That method of automatically detecting the target once it is configured in option ROM is only valid for our proprietary driver. For the open-iscsi driver , the configuring the the option ROM is unnecessary. Also, open-iscsi (i.e. the Inbox driver) has to be setup using the standard open-iscsi process utilizing the iscsiadm
Jay, We are using the be2iscsi driver and not open-iscsi.
Working communications between ELX and HP re: inbox open-iscsi configuration model.
This driver seems to be not included in snapshot -2 Thanks Narayanan
Can you confirm all the patches for be2iscsi driver included in snapshot 2 ?
(In reply to comment #29) > This driver seems to be not included in snapshot -2 > Do you mean the be2iscsi driver is not there or support for a specific chipset?
(In reply to comment #32) be2sicsi driver is available in snap-2. Wanted to confirm all the patches got into snap-2 to support this feature ?
Putting back to ON_QA until concrete negative failing results are provided.
Jay, can you confirm that all the patches you requested for 5.6 got in? Kernel is here http://people.redhat.com/jwilson/el5
Per Ludek, this feature was fully implemented by Beta, so Beta or any of the subsequent snapshots, should have the necessary patches. Ashokan -- Can you please proceed with the testing and let us know if you uncover any problems? Thanks!
We have verified that the driver initializes the CNA controller. Thanks Narayanan
Thanks, Narayanan!!
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2011-0017.html