Hold slack

Also called: hold margin

How much later new data arrives than a flip-flop's hold time requires: tcq(min) + tcd − th. Negative means a hold violation.

The hold constraint asks whether the fastest new data arrives after the capturing flip-flop's hold window has closed. Hold slack measures the margin:

hold slack = tcq(min) + tcd − th

where tcd is the contamination delay of the short path through the logic.

Positive or zero: the new data arrives after the window closes. Safe.
Negative: a hold violation. The new data arrives while the flip-flop is still locking in the old value, and may corrupt it.

With clock skew, the requirement grows by the skew: hold slack = tcq(min) + tcd − (th + tskew).

Unlike setup slack, hold slack has no clock period in it, so it's the same at every clock frequency. A negative hold slack needs a hold fix: more delay on the short path.

Worked example

Example

Hold slack on two paths

Flip-flops have tcq(min) = 30 ps and th = 45 ps. Path P has a logic tcd of 25 ps; path Q is a direct wire.

1.
Path P: 30 + 25 − 45 = +10 ps. It passes.
2.
Path Q: 30 + 0 − 45 = −15 ps. A hold violation.
3.
Path Q needs at least 15 ps of extra contamination delay, for example a buffer.

Common mistakes

Subtracting the wrong way. Hold slack is arrival time minus th; positive is good.
Using tpd or tcq in place of tcd and tcq(min).
Expecting a slower clock to improve hold slack. The period isn't in the formula.

Practice Hold slack

Interactive questions with instant feedback and a worked solution for every wrong answer.

Timing and Sequential Logic lesson full course

Practice exams

Learn it step by step

Hold slack is taught in Timing and Sequential Logic.

Timing and Sequential Logic →

All terms