Intel C2518 FH8065501516710 Data Sheet

Intel

®

 Atom™ Processor C2000 Product Family for Microserver

September 2014

Datasheet, Vol. 2 of 3

Order Number: 330061-002US

67

Reliability refers to how often errors occur in the system, and whether the system 

recovers from an error condition.

Availability refers to how flexible the system resources are allocated or redistributed for 

system utilizations and system recovery from errors.

Serviceability refers to how well the system reports and handles events related to 

errors.

The RAS features aim to achieve the following:

• Hardware-based error recovery on PCI Express* links.

— Packet re-transmission on detecting CRC errors.

• Clearly identify non-fatal errors and minimize fatal errors.

— Error reporting of the affected transactions by the appropriate completion 

responses or data poisoning.

— Correctable, non-fatal, and fatal errors are forwarded to the CPU via 

Non-Maskable Interrupt (NMI) or System Management Interrupt (SMI), or to 

an external device via the ERROR2_B, ERROR1_B, and ERROR0_B pins (see 

).

— Error logging/reporting to assist error containment and recovery.

 shows the high-level SoC error handling scheme. The SoC receives the 

PCIe* error messages from downstream devices. The SoC logs these errors and other 

internal device errors.