Tuesday Nov 03, 2009

Putback: x86gentopo Project

A few months ago, I talked about the x86gentopo project - a reusable baseboard enumeration approach for x86 systems. It's the same spirit that drove the platform independent sun4v FMA work.

Yesterday evening, the project integrated into OpenSolaris. If you missed the flag day message, here it is again.

If you don't use x86 systems, or don't care about x86 FMA, you can hit delete now. There is a new x86 generic FMA topology enumerator available with the putback of: PSARC/2009/490 x86 Generic FMA Topology Enumerator 6785310 Implement SMBIOS contained elements/handles 6841286 Need x86 generic FMA topo enumerator 6853537 x86gentopo needs OEM-Specific SMBIOS structures 6865771 Topology relationships should be derived from contained handles & elements of SMBIOS 6865814 Chip enumerator should derive serials & labels using libsmbios, if SMBIOS is FM aware 6865845 /dev/fm should export the Initial APICID, SMBIOS based ID/instance to the chip enumerator 6866456 Generic Topology FMRI ereport The new x86 generic enumerator creates physical topology, as well as identity information (serial number, part number, etc...), for i86pc class systems which contain a compatible SMBIOS. The X64 Platform Resource Management Specification (PRMS-1) describes what a compliant SMBIOS is (currently in uncirculated draft form). To correctly diagnose faults the cpu and memory ereport generators have been modified to report x86 generic topology, when a compliant SMBIOS is found. If a compliant platform SMBIOS is not found, the x86 generic enumerator and x86 generic ereport generators will revert to existing (legacy) enumeration and ereport generation. If a platform does contain a compliant SMBIOS and wishes to force legacy enumeration, the kernel tunable variable x86gentopo_legacy can be set in /etc/system: set x86gentopo_legacy = 1 To report bugs against the x86 generic topology enumerator and/or the cpu/mem ereport generators please use the following product/cat/subcat: solaris/fma/other : x86 generic enumerator solaris/fma/mem : memory ereport generator solaris/fma/cpu : cpu ereport generator To report bugs against the SMBIOS structure use please use the following product/cat/subcat: solaris/library/libsmbios For more information please visit the OpenSolaris x86gentopo project page: http://hub.opensolaris.org/bin/view/Project+x86gentopo/WebHome Thank you, The x86gentopo team

Now to get that PRMS finalized and made public...

:wq

Friday Sep 25, 2009

Improvements in Nehalelm Page Retire

6734814 Intel address translation Phase II

This change hit build 125 today. Solaris has supported memory page retire since the initial launch of Nehalem EP. Today's putback improves that support in the area of fault replay.

FMA persists page retires (and all other faults) across reboots via the on-disk fault cache. When FMD starts, the fault cache is consulted and (provided the affected resource is still in the configuration) replays the cached faults.

For page retires, the faults are associated with a physical address (PA). Between OS reboots, it's possible the memory topology can change - DIMMs can be added/removed, interleaves changed, etc. In such cases, the physical/virtual mappings change, and the PA in the on-disk fault cache could point at a healthy page. FMD would then retire a page that had experienced no errors.

This putback adds code to recalculate the PA (if necessary) after reboots to ensure the correct, faulty page is re-retired.

:wq

Friday Sep 18, 2009

Disabling Fast Reboot Upon Failure

Build 100 of OpenSolaris introduced fast reboot, quite a nifty feature. However, in certain fault scenarios, having fast reboot on by default is not desired. Fast reboot bypasses BIOS involvement in the boot process. But after certain classes of failure, BIOS engagement is desired. Examples include BIOS deconfiguring faulty components or BIOS collecting error information that may be lost or discarded after OS reboot.

Earlier this week 6880616 putback interfaces that allow Solaris FMA to disable fast reboot on terminal errors. The FMA changes to take advantage of these interfaces is close behind, planned to make build 125 (6883623.

:wq

About

user9148476

Search

Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today