What's the actual tradeoff between fishbone, comb, and sub_strap topologies when routing to clock straps?

Short Answer

Fishbone (the default) connects each driver pin individually to the nearest stripe with comb routing for loads onto a single finger, controlled by -fishbone_fanout, -fishbone_span/-fishbone_sub_span, -fishbone_layers. Comb routes driver and load pins directly to the nearest stripe, falling back to Steiner topology beyond the comb distance (default 2 global routing cells) -- good for many pins directly under stripes, but creates many stacked vias, a real physical DRC risk. Sub_strap adds parallel straps on intermediate layers specifically to reduce those stacked vias versus comb.

Technical Explanation

• Fishbone (the default -topology) connects each driver pin individually to the nearest stripe, with loads comb-routed onto a single finger -- controlled by -fishbone_fanout (max loads per finger), -fishbone_span/-fishbone_sub_span (max span between loads), and -fishbone_layers.
• Comb routes driver and load pins directly to the nearest stripe; if the Manhattan distance exceeds the comb distance (default 2 global routing cells, set via route.common.comb_distance), it falls back to Steiner topology instead.
• Comb's real risk: it's suitable when many pins sit directly under mesh stripes, but it creates many stacked vias in that case -- a genuine physical DRC concern, not just a routing-quality nitpick.
• Sub_strap adds parallel straps on intermediate layers specifically to reduce stacked vias versus comb -- it requires -sub_strap_layers and has an optional -sub_strap_max_delay (default 2ps RC delay budget).

Formula Or Decision Rule

fishbone (default): individual driver connections + comb-routed loads on fingers. comb: direct-to-stripe routing, risks stacked vias. sub_strap: reduces comb's stacked-via risk via intermediate parallel straps.

What To Check

• Warning sign: physical DRC after clock strap routing shows a cluster of stacked-via violations.
• Inspect: check which -topology was used -- comb topology is the documented source of this specific risk when many pins sit directly under stripes.
• Correct: switch to sub_strap (with -sub_strap_layers configured) or fishbone if comb's stacked-via risk is the actual cause.

Healthy, Suspicious & Hard-stop Results

• Expected: strap routing completes without a cluster of stacked-via DRC violations, regardless of which topology was chosen.
• Investigate: a stacked-via DRC cluster appears specifically under mesh stripes after routing -- check whether comb topology was used in that region.
• Stop: comb topology is being used across a design with many pins directly under stripes despite repeated stacked-via violations -- switch topology rather than repeatedly re-routing the same way.

What The Interviewer Is Testing

Whether you know the real physical tradeoff (stacked vias) that distinguishes comb from the alternatives, not just that "there are three topology options."