PCIe/CXL Deep Dive · All levels

Host-Device Coherency Contracts: Interview Drills

Interview Drills for Host-Device Coherency Contracts.

Interview drills

Interview Drills for Host-Device Coherency Contracts focuses on Coherency miss rate, snoop stall cycles, and fence-to-visible latency. The purpose is to turn memory observations into mechanism-backed actions with explicit owners and release-safe validation.

diagram
PROMPT
You observe Coherency miss rate, snoop stall cycles, and fence-to-visible latency on Host-Device Coherency Contracts. Explain root cause and release decision.

STRONG ANSWER
1. Defines failing traffic context and first transition loss.
2. Explains mechanism: Host CPUs and CXL devices share a coherency domain defined by snoop filters, directories, and invalidation paths. Contracts specify which agent owns lines, how evictions propagate, and when stores become globally visible.
3. Requests proving artifact: Coherency contract doc, snoop latency histogram, and fence benchmark
4. Proposes bounded fix + owner + rollback-safe validation.

WEAK ANSWER
Gives generic PCIe tuning ideas without command evidence, owner accountability, or risk controls.

Interview evidence matrix

diagram
PCIe/CXL EVIDENCE MATRIX - Host-Device Coherency Contracts

+-------------------------------+--------------------------------+--------------------------------+---------------------------+
| Evidence                      | Tells you                      | Does not prove                 | Next action               |
+-------------------------------+--------------------------------+--------------------------------+---------------------------+
| TLP type mix + credit stall counters    | protocol-layer stall cost    | link integrity and replay behavior   | inspect training margins  |
| queue age + class breakdown   | fairness and starvation risk   | command legality details       | parse command timeline    |
| LTSSM timeline + ordered set progression | timing-window pressure         | root cause by itself           | correlate with topology map|
| eye / Vref / skew snapshots   | PHY margin and drift behavior  | controller policy quality      | pair with schedule logs   |
| CE/UE + scrub telemetry       | reliability trajectory         | immediate perf bottleneck only | map to hotspot apcieesses  |
+-------------------------------+--------------------------------+--------------------------------+---------------------------+

PCIe/CXL deep dive

Memory expansion and coherency require HDM windows, ownership discipline, and NUMA-aware software policies.

Concept diagram

diagram
COHERENCY + HDM

CPU caches <-> CXL.cache <-> device memory (CXL.mem/HDM)

Metric graph

diagram
EXPANSION BOTTLENECK SHARE

remote latency      ██████
ownership retry     ████
interleave skew     ███

Reports and artifacts

  • HDM decode table

  • ownership transition trace

  • NUMA distance profile

  • RAS region policy

Mini case study

Fabric-attached memory increased capacity but p99 regressed until page placement respected NUMA distance.

Debug branches

  • Map HDM windows and interleave groups

  • Run ownership litmus under contention

  • Correlate RAS events with region offline policy

Senior review question

Ask: which latency, bandwidth, and reliability evidence proves this PCIe/CXL topic is closed under real traffic?

Key takeaways

  • Always tie controller and PHY counter shifts to application latency and throughput outcomes.

  • Lock firmware timing profile, thermal condition, and DIMM state before comparing PCIe/CXL captures.

Common pitfalls

  • Chasing peak bandwidth while ignoring p99 latency and fairness tails.

  • Changing timing guardbands without separating SI noise from scheduling issues.

  • Declaring closure without reliability gates, fault injection, and regression replay.

Interview answer expansion

Strong interview answers for Host-Device Coherency Contracts start with workload framing and metric framing, then explain mechanism plainly: Host CPUs and CXL devices share a coherency domain defined by snoop filters, directories, and invalidation paths. Contracts specify which agent owns lines, how evictions propagate, and when stores become globally visible.

Then propose a measurement plan: TLP routing, credit dynamics, turnaround cost, RAS interference, and PHY margin where relevant.

Finally, present one bounded fix plus regression risk. PCIe/CXL interviews reward explicit tradeoff ownership, not generic tuning slogans.