Contents

• 1. Introduction
• 2. Overview
• 3. Implementation
  • 3.1. Use Case 1: DBE Detected
  • 3.2. Use Case 2: Two SBEs Detected at the Same Location
• 4. Blacklisting and ECC Error Recovery
  • 4.1. Verifying Retired Pages are Pending
  • 4.2. Stopping GPU Clients
  • 4.3. Reattaching the GPU
• 5. Availability
• 6. Visibility
  • 6.1. XIDs
  • 6.2. NVML
  • 6.3. nvidia-smi
• 7. Caveats
• 8. FAQ
  • 8.1. RMA Eligibility
  • 8.2. Memory Impact
  • 8.3. Configuration
  • 8.4. App Behavior
  • 8.5. App Performance
  • 8.6. Driver Behavior
• 9. Notices
  • 9.1. Trademarks