Kontron KT965 Series - Manual (Page 54)

KT965 Family
KTD-00699-S Public User Manual Date: 2010-04-15 Page 54 of 90
ARBITRATION PINS (BUS MASTERS ONLY)
REQ# Request indicates to the arbiter that this agent desires use of the bus. This is a point to point signal. Every
master has its own REQ# which must be tri-stated while RST# is asserted.
GNT# Grant indicates to the agent that access to the bus has been granted. This is a point to point signal. Every
master has its own GNT# which must be ignored while RST# is asserted.
While RST# is asserted, the arbiter must ignore all REQ# lines since they are tri-stated and do not contain
a valid request. The arbiter can only perform arbitration after RST# is deasserted. A master must ignore its
GNT# while RST# is asserted. REQ# and GNT# are tri-state signals due to power sequencing
requirements when 3.3V or 5.0V only add-in boards are used with add-in boards that use a universal I/O
buffer.
ERROR REPORTING PINS.
The error reporting pins are required by all devices and maybe asserted when enabled
PERR#
Parity Error is only for the reporting of data parity errors during all PCI transactions except a Special Cycle.
The PERR# pin is sustained tri-state and must be driven active by the agent receiving data two clocks
following the data when a data parity error is detected. The minimum duration of PERR# is one clock for
each data phase that a data parity error is detected. (If sequential data phases each have a data parity
error, the PERR# signal will be asserted for more than a single clock.) PERR# must be driven high for one
clock before being tri-stated as with all sustained tri-state signals. There are no special conditions when a
data parity error may be lost or when reporting of an error may be delayed. An agent cannot report a
PERR# until it has claimed the access by asserting DEVSEL# (for a target) and completed a data phase
or is the master of the current transaction.
SERR#
System Error is for reporting address parity errors, data parity errors on the Special Cycle command, or
any other system error where the result will be catastrophic. If an agent does not want a non-maskable
interrupt (NMI) to be generated, a different reporting mechanism is required. SERR# is pure open drain
and is actively driven for a single PCI clock by the agent reporting the error. The assertion of SERR# is
synchronous to the clock and meets the setup and hold times of all bused signals. However, the restoring
of SERR# to the deasserted state is accomplished by a weak pullup (same value as used for s/t/s) which
is provided by the system designer and not by the signaling agent or central resource. This pull-up may
take two to three clock periods to fully restore SERR#. The agent that reports SERR#s to the operating
system does so anytime SERR# is sampled asserted.
INTERRUPT PINS (OPTIONAL).
Interrupts on PCI are optional and defined as “level sensitive,” asserted low (negative true), using open drain output
drivers. The assertion and deassertion of INTx# is asynchronous to CLK. A device asserts its INTx# line when requesting
attention from its device driver. Once the INTx# signal is asserted, it remains asserted until the device driver clears the
pending request. When the request is cleared, the device deasserts its INTx# signal. PCI defines one interrupt line for a
single function device and up to four interrupt lines for a multi-function device or connector. For a single function device,
only INTA# may be used while the other three interrupt lines have no meaning.
INTA#
Interrupt A is used to request an interrupt.
INTB# Interrupt B is used to request an interrupt and only has meaning on a multi-function device.
INTC# Interrupt C is used to request an interrupt and only has meaning on a multi-function device.
INTD# Interrupt D is used to request an interrupt and only has meaning on a multi-function device.