diff --git a/go.mod b/go.mod index 11c1f58..f0db622 100644 --- a/go.mod +++ b/go.mod @@ -2,4 +2,4 @@ module github.com/go-ndarray/go-ndarray.github.io go 1.27.4 -require github.com/go-fleettools/landingkit v0.21.0 +require github.com/go-fleettools/landingkit v0.29.0 diff --git a/go.sum b/go.sum index d7a658b..5a9c481 100644 --- a/go.sum +++ b/go.sum @@ -1,2 +1,2 @@ -github.com/go-fleettools/landingkit v0.21.0 h1:fcg3aS5pTvzzR+ht/LcKY7SXeFW+6PMYvB7TODmzL1c= -github.com/go-fleettools/landingkit v0.21.0/go.mod h1:jSXTO7FVKNXCK1aJjLq7H7x8d8wxDoXe3ncXWGcAKRE= +github.com/go-fleettools/landingkit v0.29.0 h1:DLjwM6jHnfymwN3j3gyuJW3v+dR16YlM05zMPV0QWCY= +github.com/go-fleettools/landingkit v0.29.0/go.mod h1:jSXTO7FVKNXCK1aJjLq7H7x8d8wxDoXe3ncXWGcAKRE= diff --git a/hugo.toml b/hugo.toml index 02b03d5..c510817 100644 --- a/hugo.toml +++ b/hugo.toml @@ -106,3 +106,7 @@ disableKinds = ['taxonomy', 'term', 'rss', 'sitemap', '404'] [[params.families]] name = "Workspace" what = "An arena for loops: `Use`, `Reset`, `Detach` — results stop costing the garbage collector from one pass to the next." + +# The architectures the tests RUN on, read from the library's CI. +[params.mantra] + goarch = ["amd64", "arm64", "riscv64", "loong64", "ppc64le", "s390x", "arm"] diff --git a/layouts/index.html b/layouts/index.html index 70c7d54..d59b66f 100644 --- a/layouts/index.html +++ b/layouts/index.html @@ -79,15 +79,15 @@

The example above, actually run.

The n-dimensional array, without cgo.

NumPy is the lingua franca of numerical computing, but in Go the choices are gonum (matrix-centric, amd64-only assembly) or a C extension. go-ndarray brings the n-dimensional array vocabulary — shapes, strided views, broadcasting, ufuncs, reductions and linear algebra — to Go with no cgo, cross-compiling to a static binary everywhere. The hot paths are multicore, and SIMD on amd64, arm64, ppc64le, loong64 and riscv64.

-
🐹

Pure Go

CGO_ENABLED=0 across the org. No cgo, no BLAS shim.

+ {{ if not (partial "mantra-on.html" $) }}
🐹

Pure Go

CGO_ENABLED=0 across the org. No cgo, no BLAS shim.

{{ end }}
✅

100% test coverage

Gated in CI, differentially checked against numpy.

🔢

NumPy semantics

Row-major strided arrays, NumPy broadcasting, and NumPy basic-indexing Slice returning strided views that share data.

♻️

Workspace arena for loops

Bind the inputs with ws.Use, compute as usual, ws.Reset() at the end of each pass: results stop costing the garbage collector anything. On a Zen 3, x + y on 1 024 elements drops from 2.6 to 0.38 µs, Go against Go.

-
⚡

SIMD on every 64-bit target

go-asmgen SIMD kernels on amd64 (SSE2, AVX2/FMA picked at run time), arm64 (NEON), ppc64le (VSX), loong64 (LASX, when the CPU has it), riscv64 (RVV, when the CPU has it) and s390x (vector facility, measured on a real IBM z17); the 32-bit targets run the same pure-Go code those kernels are tested against.

-
🎯

Accurate Exp/Log

Ports of Arm's optimized-routines: worst 0.504 / 0.508 ULP against a 300-bit reference, and correct where Go's amd64 math.Exp returns +Inf (golang/go#81995) and math.Log is wrong on subnormals (golang/go#56600).

-
🧪

Multi-arch CI

The suite runs on amd64, arm64 and 386 natively and on riscv64/loong64/ppc64le/s390x/arm under qemu, on Linux, macOS and Windows, and compiles for every GOOS/GOARCH pair Go supports; v0.2.0 was also run on real amd64, arm64, ppc64le, riscv64 and loong64 hardware.

+
⚡

SIMD on every 64-bit target

{{ if not (and $.Site.Params.mantra $.Site.Params.mantra.goarch) }}go-asmgen SIMD kernels on amd64 (SSE2, AVX2/FMA picked at run time), arm64 (NEON), ppc64le (VSX), loong64 (LASX, when the CPU has it), riscv64 (RVV, when the CPU has it) and s390x (vector facility, measured on a real IBM z17); the 32-bit targets run the same pure-Go code those kernels are tested against.{{ end }}

+
🎯

Accurate Exp/Log

{{ if not (and $.Site.Params.mantra $.Site.Params.mantra.goarch) }}Ports of Arm's optimized-routines: worst 0.504 / 0.508 ULP against a 300-bit reference, and correct where Go's amd64 math.Exp returns +Inf (golang/go#81995) and math.Log is wrong on subnormals (golang/go#56600).{{ end }}

+
🧪

Multi-arch CI

{{ if not (and $.Site.Params.mantra $.Site.Params.mantra.goarch) }}The suite runs on amd64, arm64 and 386 natively and on riscv64/loong64/ppc64le/s390x/arm under qemu, on Linux, macOS and Windows, and compiles for every GOOS/GOARCH pair Go supports; v0.2.0 was also run on real amd64, arm64, ppc64le, riscv64 and loong64 hardware.{{ end }}

💎

The cgo-free ndarray that was missing

Ruby has no cgo-free ndarray (Numo::NArray and NMatrix are C extensions) and gonum's optimized assembly is amd64-only. go-ndarray pairs a portable pure-Go core with multicore fan-out and SIMD kernels.

-
⚖️

BSD-3-Clause

Exp and Log are MIT-licensed ports of Arm's optimized-routines.

+ {{ if not (partial "mantra-on.html" $) }}
⚖️

BSD-3-Clause

Exp and Log are MIT-licensed ports of Arm's optimized-routines.

{{ end }}
@@ -140,6 +140,6 @@

The cgo-free ndarray the Go and Ruby ecosystems lacked.

- +