This seemingly irrelevant change shaves several gate delays off the I$
tl.a.valid path.

Fix D$ ready-valid signaling bug

tilelink: use the Monitor to enforce Probe sourcing

The SystemBus already has a flow buffer on outputs.

This reverts commit b64b87ad.

The crossings already have buffering in those places where it was
appropriate. Adding more does not help flow through paths.

The order should be:
  master => buffer|xing => fifofixer => splitter => xbar

Perf improvements

diplomacy: More verbose require

This avoids pipeline flushes due to subword WAW hazards, as with
consecutive byte stores.

The 4-cycle hit time is 1 cycle too long to avoid a second
pipeline replay, so it was effectively 9 cycles instead of 4.

That effectively increased the miss latency by 5 cycles when there was
a store hit followed by a load miss.  Since pending stores are drained
when releaseInFlight, the check I removed was redundant.

Stabilize tile wrappers for downstream tools

tilelink: it is now legal to support Acquire for UNCACHED regions

- Acquire.BtoT succeeds with toT instantly
- AccessAckData.error causes Grant.toN.error

We need this if we want to divert traffic to it from a TL-C slave.

These cases exist:
  GET_EFFECTS, PUT_EFFECTS, UNCACHEABLE && !supportsAcquire: MMIO
  UNCACHED && !supportsAcquire: speculation ok and may be cached
  UNCACHED && supportsAcquire: LLC/CacheCork applied (slave never probes)
  CACHED, TRACKED && supportsAcquire: slave might probe

When we first implemented TL, we thought this was helpful, because
it made WidthWidgets stateless in all cases. However, it put too
much burden on all other masters and slaves, none of which benefitted
from this signal. Furthermore, even with addr_lo, WidthWidgets were
information lossy because when they widen, they have no information
about what to fill in the new high bits of addr_lo.

Fuzzer rage

The new fuzzer already found it's first victim.

Width widget enhancements