REQ#/GNT# behavior at RST# deassertion

[kevin.normoyle@Eng.Sun.COM (Kevin Normoyle)]

The spec says
	"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." (GNT# is also undriven during RST#)

The problem:

IF REQ# inputs only become valid the cycle after RST# deasserts,
then the GNT# won't be valid until the cycle after that.
(assuming a fast reacting arbiter)

So I would think a master should also ignore GNT# for 1 or 2 more
cycles after the RST# deassertion, otherwise there's the potential
for multiple people seeing GNT# in the cycle after the RST# deassertion.

This makes me want to put pullups on GNT#, since that isn't a spec
requirement, and I don't want contention on the RST# deassertion.


Any comments?

-kevin
��