Chapter 9. System Management

The Previous Chapter

HT uses an interrupt signaling scheme very similar to PCI's Message Signaled Interrupts. The previous chapter defines how HT delivers interrupts to the Host Bridge via posted memory writes . The chapter also defined an End of Interrupt Message and details the mechanism that HT uses for configuring and setting up interrupt transactions (which is different from the PCI-defined mechanisms).

This Chapter

Rather than requiring HT devices to incorporate additional pins for signaling system- related items such as power management events, System Management messages are defined that permit signaling via transactions, thereby creating virtual wires. This chapter defines the mechanism used to send these messages, by detailing the System Management packets and protocol. Some System Management functions such as changing the operating frequency of an HT link require that transactions be stopped . This chapter also introduces the ability to temporarily disconnect the links to allow these types of state change and to save power.

The Next Chapter

There are two aspects in dealing with link or internal errors in HyperTransport: detection and handling. The next chapter describes the error types defined by the specification and what devices may do about them. Some devices may choose to detect and handle some errors but not others. The PCI configuration space registers, used to program the error strategy and log errors, are described here ” as are the reporting mechanisms: error response, fatal and non-fatal interrupts, and Sync flood.