Explain the transparent-latch two-CRPR mechanism's effect on time borrowing, with the sign of each effect.
From PDVerse PrimeTime STA Interview Guide · pdVerse Mentor Guide
Ten-second Interview Answer
PrimeTime computes a separate Rise CRPR and Fall CRPR and shifts each latch edge later. Opening-edge pessimism shifts the opening later and reduces borrow capacity; closing-edge pessimism shifts the closing later and increases the maximum borrowing allowed.
Complete Technical Explanation
A transparent latch endpoint has two edges that matter: an opening edge, after which data can start borrowing time, and a closing edge, by which borrowing must end. Because the two edges are different polarities on the clock, PrimeTime computes a Rise CRPR and a Fall CRPR separately and shifts each edge by its own pessimism. The two shifts then act in opposite directions on borrowing. Pessimism on the opening edge shifts the opening later, which reduces the amount of time available to borrow - and it also worsens the slack of any non-borrowing path referenced to that edge. Call that a tightening on borrow capacity. Pessimism on the closing edge shifts the closing later, which increases the maximum borrowing allowed and reduces the violation of a path that fails setup at the latch. That is a loosening. So one credit tightens the borrow window from the front while the other opens it at the back, and their effects do not cancel or combine into a single number. That is precisely why a latch needs both CRPRs computed separately rather than one lumped value: a single average would misstate both the borrow capacity and the setup slack at the latch, and it would get the sign of at least one of them wrong for any path that actually borrows.
In Pd / Signoff
You need this when reading a latch-based path report where borrowed time and slack both look off. Knowing the opening and closing edges carry independent CRPR shifts, in opposite directions, tells you whether a reported borrow limit is a real timing constraint or an artifact of clock pessimism before you go asking for a latch-side ECO.
Common Trap
Assuming one lumped CRPR value applies to the whole latch endpoint. Because opening-edge pessimism reduces borrow capacity while closing-edge pessimism increases the borrowing allowed, a single average misstates both, and can predict the wrong sign for the change in either borrowed time or the setup slack at the latch.
Expect Next
If the Rise and Fall CRPR values at a latch differ substantially, what does that tell you about the clock path feeding it?
Topic And Primetime Commands
Operating Conditions / General. No specific PrimeTime command is required for this conceptual answer.
Keep building interview depth
Continue practising STA
Continue learning free
Get a practical low-power chapter
Receive the existing “Low Power and Multivoltage Fundamentals” PDF chapter and its download link by email.
Continue practising