From sacadmin Tue May 11 12:51:57 2010 Received: from sac.sfbay.sun.com (localhost [127.0.0.1]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id o4BJpved024822; Tue, 11 May 2010 12:51:57 -0700 (PDT) Received: (from ehring@localhost) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8/Submit) id o4BJpun1024818; Tue, 11 May 2010 12:51:56 -0700 (PDT) Date: Tue, 11 May 2010 12:51:56 -0700 (PDT) From: Stephen Ehring Message-Id: <201005111951.o4BJpun1024818@sac.sfbay.sun.com> To: FWARC-record@sac.sfbay.sun.com Subject: SAS Phy Identifier Binary Representation Update [FWARC/2010/163 FastTrack timeout 05/17/2010] Status: RO Content-Length: 627 Template Version: @(#)sac_nextcase 1.70 03/30/10 SMI This information is Copyright (c) 2010, Oracle and/or its affiliates. All rights reserved. 1. Introduction 1.1. Project/Component Working Name: SAS Phy Identifier Binary Representation Update 1.2. Name of Document Author/Supplier: Author: Tarl Neustaedter 1.3 Date of This Document: 11 May, 2010 4. Technical Description See the case directory for more detail 6. Resources and Schedule 6.4. Steering Committee requested information 6.4.1. Consolidation C-team Name: unknown 6.5. ARC review type: FastTrack 6.6. ARC Exposure: open From sacadmin Tue May 11 13:22:51 2010 Received: from sunmail6brm.central.sun.com (sunmail6brm.Central.Sun.COM [129.147.4.169]) by sac.sfbay.sun.com (8.13.8+Sun/8.13.8) with ESMTP id o4BKMpOu026238 for ; Tue, 11 May 2010 13:22:51 -0700 (PDT) Received: from nwk-avmta-1.SFBay.Sun.COM (nwk-avmta-1.SFBay.Sun.COM [129.146.11.74]) by sunmail6brm.central.sun.com (8.13.8+Sun/8.13.8/ENSMAIL,v2.4) with ESMTP id o4BKMnK0009409 for <@sunmail2sca.sfbay.sun.com:fwarc@sun.com>; Tue, 11 May 2010 15:22:51 -0500 (CDT) Received: from pmxchannel-daemon.nwk-avmta-1.sfbay.Sun.COM by nwk-avmta-1.sfbay.Sun.COM (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) id <0L2900L2BVA2JC00@nwk-avmta-1.sfbay.Sun.COM> for fwarc@sun.com (ORCPT fwarc@sun.com); Tue, 11 May 2010 13:22:50 -0700 (PDT) Received: from brmea-mail-4.sun.com ([192.18.98.36]) by nwk-avmta-1.sfbay.Sun.COM (Sun Java System Messaging Server 6.2-3.04 (built Jul 15 2005)) with ESMTP id <0L2900KA4VA0PWB0@nwk-avmta-1.sfbay.Sun.COM> for fwarc@sun.com (ORCPT fwarc@sun.com); Tue, 11 May 2010 13:22:49 -0700 (PDT) Received: from rcsinet15.oracle.com (rcsinet15.oracle.com [148.87.113.117]) by brmea-mail-4.sun.com (8.13.6+Sun/8.12.9) with ESMTP id o4BKMmhX010070 for ; Tue, 11 May 2010 20:22:48 +0000 (GMT) Received: from acsmt354.oracle.com (acsmt354.oracle.com [141.146.40.154]) by rcsinet15.oracle.com (Switch-3.4.2/Switch-3.4.1) with ESMTP id o4BA0Mkq001199 for ; Tue, 11 May 2010 20:22:47 +0000 (GMT) Received: from abhmt017.oracle.com by acsmt355.oracle.com with ESMTP id 232757581273609259; Tue, 11 May 2010 13:20:59 -0700 Received: from dhcp-ubur-189-205.East.Sun.COM (/129.148.189.205) by default (Oracle Beehive Gateway v4.0) with ESMTP ; Tue, 11 May 2010 13:20:59 -0700 Date: Tue, 11 May 2010 16:20:55 -0400 From: Stephen Ehring Subject: FWARC 2010/163 SAS Phy Identifier Binary Representation Update In-reply-to: <4BE886B1.7060301@oracle.com> To: sunit.jain@oracle.com Cc: fwarc@sun.com, minn.ayetut@oracle.com, Dan Mahoney Message-id: <121D55CA-E5F3-4063-B5F1-A970A048DDF5@oracle.com> MIME-version: 1.0 X-Mailer: Apple Mail (2.936) Content-type: multipart/alternative; boundary="Boundary_(ID_r5zl+vUHouBktViABhO0+Q)" X-PMX-Version: 5.4.1.325704 X-Source-IP: acsmt354.oracle.com [141.146.40.154] X-Auth-Type: Internal IP X-CT-RefId: str=0001.0A090204.4BE9BC97.022A:SCFMA4539814,ss=1,fgs=0 References: <4BE886B1.7060301@oracle.com> Status: RO Content-Length: 11175 --Boundary_(ID_r5zl+vUHouBktViABhO0+Q) Content-type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-transfer-encoding: 7BIT On May 10, 2010, at 6:20 PM, Sunit Jain wrote: > Hi Steve, > > FWARC 2010/150 Platform Specific NIU SERDES and PHY values > which just got approved. Seems like you are reusing the same case > number for SAS Phy changes too. There was some confusion yesterday since sac.eng was down and no new case number was checked out. It was rebooted and is working now. Resending original notification with correct case number. I'm sponsoring this fast-track case for Tarl Neustaedter and the Serpa platform team. The case timer is set to expire 5/17/2010. The case updates the SAS binding to allow for a new phy identifier binary representation to allow for the case where some of the disks are behind a SAS expander. The driving requirement for the update is the Serpa platform's team need to have static disk devaliases. Requested release binding: - Firmware - Any Steve 1. Introduction 1.1. Project/Component Working Name: SAS Phy Identifier Binary Representation Update 1.2. Name of Document Author/Supplier: Tarl Neustaedter Stephen Ehring 1.3. Date of This Document: 05/10/2010 2. Project Summary This project modifies the SAS phy identifier binary representation in order to handle phys on SAS expanders. 3. Business Summary 3.1. Problem Area: Some SPARC platforms (Serpa in this case) support different disk backplanes, some of which contain a SAS expander. With the current SAS binding implementation, there is currently no way to uniquely identify a disk in a static manner since the phy identifier binary representation only allows for one phy number. What is needed is some way to represent this nested phy identifier configuration. 4. Technical Description: 4.1. Details: To solve this problem we are proposing an update the the SAS binding [1], specifically the section 3.1.1.1.2 "Phy Identifier Binary Representation". The new description will read: 3.1.1.1.2 Phy Identifier Binary Representation This representation identifies a phy on the SAS adapter or a directly connected expander. Bit # 33222222 22221111 11111100 00000000 10987654 32109876 54321098 76543210 sas.hi cell: 00000000 00000000 00000000 00000000 sas.lo cell: 00000000 00000001 jjjjjjjj iiiiiiii lun.hi cell: uuuuuuuu uuuuuuuu uuuuuuuu uuuuuuuu lun.lo cell: uuuuuuuu uuuuuuuu uuuuuuuu uuuuuuuu where: 00..00 Bits with the value zero ii..ii 8-bit unsigned number phy identifier in the range of 0..FE . jj..jj Expander identifier. Either zero (indicating the PHY number iiiiiiii is on the SAS adapter itself) or identifies the PHY connecting to the expander, in which case iiiiiiii identifies a PHY on a SAS expander. In the non-zero case, jjjjjjjj is an 8-bit unsigned number of the PHY plus one, in the range 1..FF uu..uu 64-bit unsigned number logical unit number Note: The high order 47 bits sas.hi ... sas.lo are required to be zero in this form, and bit 16 of sas.lo is required to be one. 4.2 Exported Interfaces: Interface Classification Comments ------------------------ --------------- ---------------------------- SAS Phy Identifier Binary Uncommitted New binding is located in materials Representation as described directory. in this document (4.1) 5. Reference Documents: [1] SAS Binding http://sac.eng/Archives/CaseLog/arc/FWARC/2008/013/commitment2.materials/sas-bindings-r1.9.txt --Boundary_(ID_r5zl+vUHouBktViABhO0+Q) Content-type: text/html; charset=US-ASCII Content-transfer-encoding: quoted-printable
On May 10, 2010, = at 6:20 PM, Sunit Jain wrote:

Hi Steve,

FWARC 2010/150 Platform Specific NIU = SERDES and PHY values
which just got approved. Seems like you are = reusing the same case
number for SAS Phy changes = too.


There = was some confusion yesterday since sac.eng was down and no new case = number was checked out. It was rebooted and is working now. Resending = original notification with correct case = number.

I'm sponsoring this fast-track case for = Tarl Neustaedter and the Serpa platform team. The case timer is set to = expire 5/17/2010. The case updates the SAS binding to allow for a new = phy identifier binary representation to allow for the case where some of = the disks are behind a SAS expander. The driving requirement for the = update is the Serpa platform's team need to have static disk = devaliases.

Requested release binding:

- = Firmware - Any

Steve

1. Introduction
  1.1. = Project/Component Working Name:
SAS Phy Identifier Binary = Representation Update

  1.2. Name of Document = Author/Supplier:
Tarl = Neustaedter
       Stephen = Ehring

  1.3. Date of This Document:
= 05/10/2010

2. Project Summary

  This project = modifies the SAS phy identifier binary representation in = order
  to handle phys on SAS expanders.

3. Business = Summary

  3.1. Problem = Area:

       Some SPARC = platforms (Serpa in this case) support different = disk
       backplanes, some of = which contain a SAS expander. With the = current
       SAS binding = implementation, there is currently no way to = uniquely
       identify a disk in = a static manner since the phy identifier = binary
       representation only = allows for one phy number. What is needed is = some
       way to represent this = nested phy identifier configuration.


4. Technical = Description:

   4.1. = Details:

       To solve this = problem we are proposing an update the the = SAS
       binding [1], = specifically the section 3.1.1.1.2 "Phy = Identifier
       Binary = Representation". The new description will read:

3.1.1.1.2 =  Phy Identifier Binary Representation

This representation = identifies a phy on the SAS adapter or a directly connected = expander.

         &nb= sp;    Bit #   33222222 22221111 11111100 = 00000000
          &n= bsp;           1098= 7654 32109876 54321098 = 76543210

      sas.hi cell: =    00000000 00000000 00000000 = 00000000
      sas.lo cell: =    00000000 00000001 jjjjjjjj = iiiiiiii
      lun.hi cell: =    uuuuuuuu uuuuuuuu uuuuuuuu = uuuuuuuu
      lun.lo cell: =    uuuuuuuu uuuuuuuu uuuuuuuu = uuuuuuuu

where:
  00..00 =        Bits with the value = zero
      ii..ii =          8-bit unsigned = number phy identifier in the = range
           = ;           of = 0..FE .
      jj..jj =          Expander = identifier. Either zero (indicating the PHY = number
          &nbs= p;           iiiiii= ii is on the SAS adapter itself) or identifies the = PHY
           &= nbsp;          connectin= g to the expander, in which case iiiiiiii = identifies
          =             a = PHY on a SAS expander. In the non-zero case, jjjjjjjj is = an
           &n= bsp;          8-bit = unsigned number of the PHY plus one, in the range = 1..FF
      uu..uu =          64-bit unsigned = number logical unit number

Note: The high order 47 bits sas.hi = ... sas.lo are required to be zero in
this form, and bit 16 of sas.lo = is required to be one.

   4.2 Exported = Interfaces:

   Interface = Classification Comments
   ------------------------ =    --------------- = ----------------------------
   SAS Phy Identifier = Binary   Uncommitted     New binding is = located in materials
   Representation as described =             &n= bsp;   directory.
   in this document = (4.1)

5. Reference Documents:

  [1] SAS = Binding

  http://sac.eng/Archives/CaseLog/arc/FWARC/20= 08/013/commitment2.materials/sas-bindings-r1.9.txt
= --Boundary_(ID_r5zl+vUHouBktViABhO0+Q)--