Home /Modules/cpu/TIMING.md

TIMING.md

lib/cpu/z-architecture/TIMING.md · 42 line(s) · markdown

z/Architecture Timing and Performance Model

Timing policy

There is no single reliable cycles-per-instruction figure for the entire z/Architecture family. Performance depends on processor generation, instruction facilities, operand alignment, cache/storage hierarchy, translation, contention, and system configuration.

Evidence levels

Performance dimensions

Record instruction mix, facility use, branch behavior, storage translation, cache assumptions, synchronization, I/O waits, and multiprocessor contention separately where measurable. CPU instruction timing must not include unmodeled operating-system or channel-subsystem time as if it were intrinsic core latency.

Reports

Every timing report identifies the processor profile, enabled facilities, system configuration and evidence class. Do not claim exact or guaranteed execution time from the generic family model.

Expert timing, voltages, and variable clocks (shared across all CPU models)

The per-CPU timing above is the architectural/behavioral view for this part. The adjustable, quality-first timing system is shared by every CPU model in lib/cpu and is documented once, in the package guide:

Every CPU model (this one included) inherits the clock/overclock surface from SLCPURuntime and can be assembled with those shared expert-timing, voltage, and component options; they are a build-layer concern, not re-declared per CPU.