IV98971: POTENTIAL SYSTEM CRASH IF ASO IS ENABLED APPLIES TO AIX 7100-05

APAR status

• Closed as program error.

Error description

• System can crash with Machine Check error, due to
  duplicate SLB entries.
  This can only happen if Active System Optimizer (ASO)
  is enabled.
  This can happen when U-block of the process is in
  corrupted state:
  Crash symptom can look like this:
  Machine Check - RTAS log Version 6 Details:
  Severity:        3 (Error Sync)
  Disposition:     2 (Not Recovered)
  Initiator:       1 (Cpu)
  Target:          0 (Unknown)
  Type:            0 (Unknown)
  - Unrecoverable error
  FRU ID:      1           Processor ID:     12
  Machine Check Type:   1 - SLB Error
  - Multiple hit error. There are two or more entries in
  the
   SLB that translate the same effective address
        Eaddr:   0xF000000020000010
        Duplicate/Overlapping entries in SLB:
          01  F000000028000000  000000298E618590
          15  F000000028000000  00000000054B0310
  

Local fix

• Following Tunable can work as work around
  # vmo -r -o vmm_mpsize_support=2
  

Problem summary

• System can crash with Machine Check error, due to
  duplicate SLB entries.
  This can only happen if Active System Optimizer (ASO)
  is enabled.
  This can happen when U-block of the process is in
  corrupted state:
  Crash symptom can look like this:
  Machine Check - RTAS log Version 6 Details:
  Severity:        3 (Error Sync)
  Disposition:     2 (Not Recovered)
  Initiator:       1 (Cpu)
  Target:          0 (Unknown)
  Type:            0 (Unknown)
  - Unrecoverable error
  FRU ID:      1           Processor ID:     12
  Machine Check Type:   1 - SLB Error
  - Multiple hit error. There are two or more entries in
  the
   SLB that translate the same effective address
        Eaddr:   0xF000000020000010
        Duplicate/Overlapping entries in SLB:
          01  F000000028000000  000000298E618590
          15  F000000028000000  00000000054B0310
  

Problem conclusion

• Correclty reset internal state before attempting to use a new
  16MB region for the 16MB MPSS promotion request.
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

  IV99023 IV99101 IJ00747

Fix information

• Fixed component name

  AIX V7.1

• Fixed component ID

  5765H4000

Applicable component levels

• R710 PSY U878659

     UP17/10/30 I 1000