Home /Modules/cpu/TIMING.md

TIMING.md

lib/cpu/powerpc/TIMING.md · 54 line(s) · markdown

PowerPC Timing

Timing model

PowerPC implementations differ substantially in pipeline depth, dispatch width, execution latency, cache latency, and completion behavior.

SLeeLa therefore records timing as implementation data rather than asserting a universal cycle count.

The timing record can include:

Fixed instruction size

Instructions are 32-bit words, giving a stable fetch/decode granularity. citeturn0search9

Branch timing

Branch processing is modeled independently because CR/CTR/LR state can affect control flow. Conditional branch forms may also update LR when LK is set. citeturn0search11

Accuracy boundary

A PowerPC 601-style timing model must not be presented as a 604/750/970 timing model. Each concrete implementation gets its own timing profile.

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.