possible hang condition in GHES code on sandy bridge
Issue
- The error is exhibited error with EDAC_DEBUG enabled:
EDAC DEBUG: in drivers/edac/sb_edac.c, line at 911: SAD interleave #0: 0
EDAC DEBUG: in drivers/edac/sb_edac.c, line at 919: mc#0: Error detected on SAD#0: address 0x0000000000000000 <
0x000000043fffffff, Interleave [8:6]XOR[18:16]
EDAC DEBUG: in drivers/edac/sb_edac.c, line at 943: SAD interleave index: 0 (wayness 1) = CPU socket 0
EDAC DEBUG: in drivers/edac/sb_edac.c, line at 1043: TAD#0: address 0x0000000000000000 < 0x00000000bfffffff, socket
interleave 0, channel interleave 4 (offset 0x00000000), index 0, base ch: 0, ch mask: 0x01
EDAC MC0: CE - no information available: Can't discover the memory rank for ch addr 0x00000000
not that one appears to get 0x0 as the address
EDAC DEBUG: in drivers/edac/sb_edac.c, line at 911: SAD interleave #0: 1
EDAC DEBUG: in drivers/edac/sb_edac.c, line at 919: mc#1: Error detected on SAD#1: address 0x00000006b5dc3700 <
0x000000083fffffff, Interleave [8:6]XOR[18:16]
EDAC DEBUG: in drivers/edac/sb_edac.c, line at 943: SAD interleave index: 0 (wayness 1) = CPU socket 1
EDAC DEBUG: in drivers/edac/sb_edac.c, line at 1043: TAD#0: address 0x00000006b5dc3700 < 0x000000083fffffff, socket
interleave 0, channel interleave 4 (offset 0x440000000), index 2, base ch: 2, ch mask: 0x04
EDAC MC1: CE - no information available: Can't discover the memory rank for ch addr 0x4ebb86c0
- This one gets a more valid address but still cant decode rank.
- You never see the error decoded back to the dimm.
Environment
- Red Hat Enterprise Linux 6
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
