A single shader that ramps *several* realistic dimensions together as n grows, modelling how a real shader's compile cost behaves going from simple to highly complex — as opposed to the single-axis stressors, which isolate one pass. Each step adds: a branchy helper (control flow / SSA + phi), a generic call (sema + specialization), a small bounded inner loop, resource reads, and a dynamic-dispatch site. Helpers are chained in bounded-depth groups (call graph depth) so doubling n roughly doubles total work across all of front-end, IR opt, and codegen at once. Sweep this to get the holistic complexity->compile-time curve and a floor+slope fit. Scaling reference (not a null): several axes co-scale, so the sweep's linear line means 'cost tracks code size' — a user-meaningful reference. Standing super-linearity here is expected; the actionable signal is its release-over-release movement.
bucket: realistic_scaling · mode: target · flags: -target spirv -emit-spirv-directly
compileInner split into phase buckets (named leaves + (self) residuals) stacked across the sweep sizes — the top edge is compileInner, so you can see which phase drives the scaling.
floor-subtracted power-law fit (t − floor) = a·Nk; floor = the minimal workload (fixed per-compile cost), k the global exponent, top-2× the local high-end doubling ratio.
| N range | floor (ms) | k (work) | fit R² | t(Nmin) | t(Nmax) | top-2× |
|---|---|---|---|---|---|---|
| 160–1280 | 11 | 1.22 | 0.998 | 210 | 2592 | 2.58× |
compileInner grows by 2382 ms across the sweep; the mutually-exclusive phase buckets below partition that growth exactly (no nested-timer double counting). × lin is the same metric as the top-level panels, per bucket: the end point vs a linear expectation anchored to the bucket's share of the minimal floor and fitted on the low-N half — 1.0 = grew exactly linearly, >1 bends up. The super-linearity lives where × lin (and k) are red.
| bucket | t@N=160 | t@N=1280 | Δ ms | share | × lin | ∝Nk |
|---|---|---|---|---|---|---|
| generateOutput (self) | 66 | 862 | +796 | 33% | 1.40× | 1.24 |
| linkAndOptimizeIR (self) | 24 | 460 | +436 | 18% | 1.91× | 1.42 |
| specializeModule | 20 | 258 | +238 | 10% | 1.52× | 1.23 |
| legalizeExistentialTypeLayout | 4 | 218 | +213 | 9% | 3.79× | 1.89 |
| legalizeResourceTypes | 4 | 217 | +213 | 9% | 3.86× | 1.91 |
Also growing (below top-5): simplifyIR (+170 ms, 7%), SemanticChecking (+148 ms, 6%), generateIR (+130 ms, 5%).
Near-constant (≤2% of growth each): parseTranslationUnit (2→13 ms), linkIR (2→12 ms), performMandatoryEarlyInlining (1→11 ms), performForceInlining (1→4 ms), unrollLoopsInModule (0→3 ms), frontEndExecute (self) (0→2 ms), compileInner (self) (0→0 ms).
| N | compileInner | frontEndExecute | linkAndOptimizeIR | simplifyIR |
|---|---|---|---|---|
| 160 | 210 | 64 | 81 | 25 |
| 320 | 460 | 100 | 197 | 50 |
| 640 | 1006 | 178 | 478 | 97 |
| 1280 | 2592 | 354 | 1377 | 195 |