kernel fusion
From Dontopedia, the open, paraconsistent wiki. (Last updated 2026-06-06.)
kernel fusion has 9 facts recorded in Dontopedia across 4 references.
Mostly:needs to start(1), is what we want(1), rdf:type(1)
Maturity scale
raw canonical shape-checked rule-derived certifiedInbound mentions (4)
Other subjects in dontopedia point AT this entity as a value. These are inverse relationships — e.g. "X motherOf this subject" — and answer questions the forward facts can't. Grouped by predicate.
allowsAllows(1)
- Doubling Scan Eval Cadence
ex:doubling-scan-eval-cadence
benefitsPythonBenefits Python(1)
- Torch Compile
ex:torch-compile
doesKernelFusionAutomaticallyDoes Kernel Fusion Automatically(1)
- Apples Graph Compiler
ex:apples-graph-compiler
optimizedByOptimized by(1)
- Strang Step
ex:strang-step
Other facts (8)
The long tail: predicates that appear too rarely to warrant their own section. Filter or scroll to find a specific one. Each row links to its source.
| Predicate | Value | Ref |
|---|---|---|
| Needs to Start | true | [1] |
| Is What We Want | Desired | [2] |
| Rdf:type | Optimization Technique | [3] |
| Reduces Dispatch Count | 5 | [4] |
| Involves | Double Buffering | [4] |
| Buffer Read Count | 1 | [4] |
| Buffer Write Count | 1 | [4] |
| Eliminates | Cpu Gpu Round Trips | [4] |
Timeline
Timeline axis is valid_time — when each source says the fact was true in the world, not when Dontopedia learned about it. Retracted rows are kept for provenance; coloured stripes indicate the context kind.
References (4)
ctx:discord/blah/watt-activation/part-15ctx:discord/blah/watt-activation/part-642ctx:discord/blah/watt-activation/15- full textwatt-activation-15text/plain3 KB
doc:agent/watt-activation-15/13ad2519-f6c2-47c1-afd8-14c1f26821f5Show excerpt
[2026-03-03 03:48] xenonfun: some nocopy memory things still same speed, needs to start fusing kernels. [2026-03-03 04:30] xenonfun: Now I have a complete picture. Let me write the documentation first, then plan and implement the kernel fus…
ctx:discord/blah/watt-activation/525- full textwatt-activation-525text/plain3 KB
doc:agent/watt-activation-525/133fb518-fad6-4e16-a093-f95c1d94dda4Show excerpt
[2026-03-23 01:57] xenonfun: ⏺ 16.7 million grid points, 384 MB, 104M point-updates/s, 13.1 GFLOP/s. That's 10x the original single-core throughput at scale. Rayon is fully saturating all M4 performance cores. The scaling curve shows we'…
See also
Keep researching
Missing something or suspicious of what's here? Kick off a research session — a Claude agent will investigate, cite its sources, and file new facts into a dedicated context you can review before accepting into the shared view.