Ilp

2026-05-21 19:09:24 +02:00
parent 3f26921177
commit ef67c7e5c0
6 changed files with 1598 additions and 2 deletions
@@ -0,0 +1,85 @@
+# Optimization: Inlining
+
+- **Author**: Smail
+- **Predecessor**: variant_inlining_7_symmetry_2.cpp
+- **File(s)**: variants/variant_inlining_7_symmetry_2_ilp_1.cpp
+- **Status**: TBA
+
+## What this changes
+
+Implemented ILP with loop unrolling and accumulators, each factor 4.
+
+
+## Code (relevant snippet)
+
+TBD - see `variants/variant_inlining_7_symmetry_1.cpp` for the final version, and the
+individual file headers for the incremental steps.
+
+## Benchmark
+
+Conditions: single run (TODO: switch to min-of-N), N_ITER=1000,
+-O3 -march=native -flto. Predecessor reference for the same config is the
+baseline row.
+
+### x86
+
+**Machine**:
+
+```
+Processors: 16 x AMD Ryzen AI 7 350 w/ Radeon 860M
+Memory: 32 GiB of RAM (30.6 GiB usable)
+Graphics Processor: AMD Radeon 860M Graphics
+Manufacturer: LENOVO
+Product Name: 83HX
+System Version: IdeaPad Slim 5 14AKP10
+```
+
+**Frequency**: 2 GHz
+
+Existing single-point measurement (N unspecified, kept from the previous
+OPTIMIZATIONS.md; W matches roughly N=120 at N_ITER=1000):
+
+| Variant  | W(n) [Flops]     | T(n) [Cycles]  | Performance [Flops/Cycle] |
+| -------- | ---------------- | -------------- | ------------------------- |
+| Baseline | 1'058'400'000.00 | 850'352'900.00 | 1.24                      |
+| Inlining | 1'058'400'000.00 | 756'088'260.00 | 1.40                      |
+| Symmetry |  |  |                       |
+
+Inlining is about 12.9% faster on this point. A full N sweep with min-of-3 is TODO.
+
+Full sweep (TODO):
+
+| N    | W(n) [Flops] | T(n) [Cycles] | Performance [Flops/Cycle] | Diff vs Predecessor |
+| ---- | ------------ | ------------- | ------------------------- | ------------------- |
+| 50   |              |               |                           |                     |
+| 100  |              |               |                           |                     |
+| 200  |              |               |                           |                     |
+| 500  |              |               |                           |                     |
+| 1000 |              |               |                           |                     |
+
+### ARM (Apple)
+
+**Machine**:
+
+```
+TBD
+```
+
+**Frequency**: TBD
+
+| N    | W(n) [Flops] | T(n) [Cycles] | Performance [Flops/Cycle] | Diff vs Predecessor |
+| ---- | ------------ | ------------- | ------------------------- | ------------------- |
+| 50   |              |               |                           |                     |
+| 100  |              |               |                           |                     |
+| 200  |              |               |                           |                     |
+| 500  |              |               |                           |                     |
+| 1000 |              |               |                           |                     |
+
+## Analysis
+
+Speedup likely comes mainly from halving the number of `sqrt` flops.
+
+
+## Takeaway
+
+TBD