It would have its own 4360-7 IC. The one on the scope can technically be used and reprogrammed but that would require doing some damage to the PCB traces to hijack the SPI bus. We can do a less intrusive by just unsoldering the existing two decoupling capacitors and injecting our own clock. That is what i am thinking.
Alternatively, we could remove the decoupling cap on the 25 Mhz reference line and connect an external signal gen and vary frequency a little and see if the PLL locks. If Nonorthogonal has a suitable signal gen he could try this now.
ADDED: And the indication of a lock condition would be the voltage on the 4360-7 CP pin going from sawtooth to a straight line at some level of DC. Once that is achieved, go and recheck if the 5us jitter gone.
If the on-board 4360-7 is to be replaced with something else, it should be verified that Rigol is not programming any other frequency into it when other sweep speeds are selected. Since the part is only spec'd for 350MHz to 1800MHz, the only other option would be 500MHz to fit nicely into the ADC and sweep selection.
I think it's highly unlikely they're changing the frequency anyway, but it's not beyond possibility.
For me, I'm going to wait to see what Rigol comes up with. No word yet.