
IDT Upstream Port Failover
PES34H16 User Manual
2 - 3
October 30, 2008
Notes
At a system level, a dynamic upstream port failover appears as a full link retrain of the upstream port,
i.e., the Link State Sequence State Machine (LTSSM) transitions to the Detect state, and the data link layer
transitions to a DL_Down state. This typically results in a hot-reset of the PES34H16 and devices below the
PES34H16 in the PCIe hierarchy. A hot-reset due to a DL_Down state may be disabled by setting the
Disable Link Down Hot Reset (DLDHRST) bit in the Switch Control (SWCTL) register. With this bit set, an
upstream port failover does not reset the PES34H16 or the PCIe hierarchy and system state is preserved.
When a dynamic upstream port failover occurs, upstream port data queued in the switch, data being
transmitted, and data in the replay buffer may be lost. Thus, some interruption of PCIe traffic should be
expected with an upstream port failover. When the switch is configured to operate in a mode that supports
upstream port failover, the Upstream Port Failover Enabled (USPFEN) bit is set in the Upstream Port
Failover Status (USPFSTS) register.
In all switch modes, the current external PES34H16 switch port associated with the upstream port may
be determined by reading the Current Upstream Port (CUSP) field in the USPFSTS register. Whenever a
dynamic upstream port failover occurs, the Upstream Port Change (USPC) bit is set in the USPFSTS
register. This bit is sticky and thus its status is preserved across a hot-reset.
The operation of the upstream port failover mechanism is unaffected by a hot-reset. Fields in the
USPFEN register have no effect on system operation when an upstream port failover switch mode has not
been selected during a fundamental reset.
Static Upstream Port Failover
A static upstream port failover requires a fundamental reset to be initiated whenever the upstream port
is changed. Since the initial upstream port is selected by the switch mode in the boot configuration vector,
the static upstream port failover feature may be viewed as nothing more than the ability to select the
upstream port during a fundamental reset.
A static upstream port failover consists of the following steps:
–
Assert the PCIe fundamental reset signal (PERSTN)
–
Modify the switch mode (SWMODE) signals to select the external PES34H16 port associated with
the upstream port.
–
Negate the PCIe fundamental reset signal (PERSTN).
The following switch modes select port 0 as the upstream port during a fundamental reset:
–
Normal switch mode
–
Normal switch mode with Serial EEPROM initialization
–
Normal switch mode with upstream port failover (port 0 selected as the upstream port)
–
Normal switch mode with Serial EEPROM initialization and upstream port failover (port 0 selected
as the upstream port).
The following switch modes select port 2 as the upstream port during a fundamental reset:
–
Normal switch mode with upstream port failover (port 2 selected as the upstream port)
–
Normal switch mode with Serial EEPROM initialization and upstream port failover (port 2 selected
as the upstream port)
Since initiation of an upstream port failover requires a fundamental reset of the PCIe hierarchy, many
systems may require the use of dynamic upstream port failover.
Dynamic Upstream Port Failover
Dynamic upstream port failover allows the external PES34H16 port associated with the upstream port to
be modified while the system is live and in a manner which preserves the system state.
Summary of Contents for 89HPES34H16
Page 10: ...IDT Table of Contents PES34H16 User Manual iv October 30 2008 Notes...
Page 12: ...IDT List of Tables PES34H16 User Manual vi October 30 2008 Notes...
Page 18: ...IDT Register List PES34H16 User Manual xii October 30 2008 Notes...
Page 40: ...IDT Upstream Port Failover PES34H16 User Manual 2 6 October 30 2008 Notes...
Page 86: ...IDT Power Management PES34H16 User Manual 7 4 October 30 2008 Notes...
Page 172: ...IDT Configuration Registers PES34H16 User Manual 9 80 October 30 2008 Notes...