v0.7.12 - FFI Hardening & Library Loading

Changes

Dependencies Updated

Package	Old	New
go-webgpu/webgpu	v0.3.2	v0.4.0
go-webgpu/goffi	v0.4.0	v0.4.0 (unchanged)

Upstream Improvements

• Null handle guards on 27 public FFI methods — prevents SIGSEGV on nil/released objects
• ptrFromUintptr helper eliminates all go vet unsafe.Pointer warnings
• WGPU_NATIVE_PATH env var for custom wgpu-native library path
• loadLibrary returns (Library, error) with proper error propagation
• Windows DLL eager loading — errors surface at init, not at first use
• Enhanced Init() error messages with library path and remediation suggestions
• 85 new null guard test cases upstream

Impact

Significantly improved safety and debuggability of GPU backend initialization.

Full Changelog: v0.7.11...v0.7.12
Upstream: https://github.com/go-webgpu/webgpu/releases/tag/v0.4.0

v0.7.11 - Crosscall2 Callback Integration

Changes

Dependencies Updated

Package	Old	New
go-webgpu/webgpu	v0.3.1	v0.3.2
go-webgpu/goffi	v0.3.9	v0.4.0 (indirect)

Upstream Improvements

• crosscall2 integration — callbacks now work from C-library-created threads (Metal, wgpu-native)
• fakecgo trampoline register fixes synced with purego v0.10.0

Bug Fixes

• Xavier init test: added float32 rounding tolerance to prevent false failures

Impact

Improved callback reliability on macOS Metal and native WebGPU implementations.

Full Changelog: v0.7.10...v0.7.11
Upstream: https://github.com/go-webgpu/webgpu/releases/tag/v0.3.2

v0.7.10

🔧 Dependencies Update + Lint Cleanup

Update WebGPU backend to v0.3.1 with critical ARM64 callback fix. Clean up 101 stale //nolint:gosec directives.

Updated Dependencies:

• go-webgpu/webgpu v0.3.0 → v0.3.1
• go-webgpu/goffi v0.3.8 → v0.3.9 (indirect)

Upstream Fixes:

• ARM64 callback trampoline rewrite — fixes LR corruption for callbacks at index > 0
• Symbol rename to prevent linker collision with purego

Code Quality:

• Removed 101 unused //nolint:gosec directives (gosec linter updated, no longer flags these)
• Standardized remaining nolint comments to short format
• 0 lint issues across all platforms

Impact: Critical fix for macOS Apple Silicon and Linux ARM64 users.

Links:

• Upstream release: go-webgpu v0.3.1
• PRs: #29, #30

v0.7.9

🔧 Dependencies Update

Update WebGPU backend to v0.3.0 with new capability-querying API and typed errors.

Updated Dependencies:

• go-webgpu/webgpu v0.2.1 → v0.3.0

New Upstream Features Available:

• Surface.GetCapabilities() — query supported formats, present modes, alpha modes
• Device.GetFeatures() / Device.HasFeature() — feature enumeration
• Device.GetLimits() — device limits (experimental)
• Typed errors with errors.Is() / errors.As() support (ErrValidation, ErrOutOfMemory, ErrInternal, ErrDeviceLost)
• Resource leak detection via SetDebugMode(true) / ReportLeaks()

Links:

• Upstream release: go-webgpu v0.3.0
• PR: #28

v0.7.8 - GoGPU Ecosystem Integration (Phase 1)

🔧 GoGPU Ecosystem Integration (Phase 1)

Migrate WebGPU backend to unified gputypes for future dual-backend support.

Updated Dependencies

Package	Old	New
go-webgpu/webgpu	v0.1.4	v0.2.1
go-webgpu/goffi	v0.3.7	v0.3.8
gogpu/gputypes	-	v0.2.0 (new)
dlclark/regexp2	v1.10.0	v1.11.5
google/uuid	v1.3.0	v1.6.0

Changes

• Migrated all WebGPU types from wgpu.* to gputypes.*:
  • BufferUsage, BufferUsageStorage, BufferUsageCopySrc, BufferUsageCopyDst
  • PowerPreferenceHighPerformance
• Updated 10 files in internal/backend/webgpu/
• Fixed 3 prealloc warnings in linter

Why This Matters

Prepares codebase for Pure Go WebGPU backend (gogpu/wgpu):

• Unified type system enables future dual-backend architecture
• Build tags will allow: go build (Rust FFI) vs go build -tags purego (Pure Go)

Links

• Upstream: go-webgpu v0.2.1
• GoGPU ecosystem: github.com/gogpu
• PR: #27

Installation

```bash
go get github.com/born-ml/born@v0.7.8
```

v0.7.7: Public API Improvements

🔧 Public API Improvements

Refactored public API packages to use proper Go interfaces instead of type aliases where possible.

Improvements

• tensor/: Added Backend interface with 40+ methods (was type alias)
• nn/: Added Module interface with full method definitions
• onnx/: Added Model interface for ONNX model operations
• optim/: Now uses public nn.Parameter in function signatures
• autodiff/: Now uses public tensor types
• backend/cpu, backend/webgpu: Added compile-time interface checks

Technical Details

• Improves pkg.go.dev documentation by hiding internal paths
• External packages can now properly import and use the public API
• Some interfaces (Optimizer, ModelReader) remain as type aliases due to Go's type system constraints

Fixed Issues

• #25 — ONNX package not accessible from external packages

Full Changelog: v0.7.6...v0.7.7

v0.7.6: ARM64 Darwin Enhancement

ARM64 Darwin Enhancement

Comprehensive ARM64 Darwin support with enhanced struct handling, tested on M3 Pro hardware.

Updated Dependencies

• go-webgpu/webgpu v0.1.3 → v0.1.4
• go-webgpu/goffi v0.3.6 → v0.3.7

Improvements

• Struct Layout: Proper layout for nested and complex struct types
• Mixed Data Types: Automatic struct layout computation for integer/float combinations
• Register Optimization: Enhanced struct return handling (9-16 bytes) utilizing X0 and X1 registers

Fixed Issues

• Resolved segmentation fault in string output benchmarks on Darwin systems

Links

• PR: #24
• Upstream: go-webgpu v0.1.4

Contributors

• @ppoage — ARM64 Darwin implementation, Objective-C test suite, assembly verification

v0.7.5: ARM64 Hotfix

ARM64 Hotfix

Update GPU backend dependencies with critical ARM64 fixes for Apple Silicon.

Updated Dependencies

• go-webgpu/webgpu v0.1.2 → v0.1.3
• go-webgpu/goffi v0.3.5 → v0.3.6

Fixed Issues

• ARM64 HFA returns: NSRect with 4×float64 now correctly returns all values on Apple Silicon
• Large struct returns: Structs exceeding 16 bytes now properly use X8 register for implicit pointer passing
• macOS ARM64 display: Resolved blank window issue where GPU dimensions were returning 0×0

Links

• PR: #23
• Upstream: go-webgpu v0.1.3

v0.7.4: Linear Layer Without Bias

✨ New Feature: Linear Layer Without Bias

Add WithBias option to nn.NewLinear for creating Linear layers without bias term.

New API

// With bias (default, backwards compatible)
layer := nn.NewLinear(784, 128, backend)

// Without bias (for LLaMA-style models, LM head, etc.)
lmHead := nn.NewLinear(hiddenSize, vocabSize, backend, nn.WithBias(false))

Changes

• Add LinearOption type and WithBias(bool) functional option
• Add HasBias() method for introspection
• Update SwiGLUFFN to use public API
• Export WithBias in public nn package

Use Cases

• LM Head in language models (GPT, LLaMA, HRM)
• Attention projections (some architectures)
• SwiGLU FFN layers

Full Changelog: v0.7.3...v0.7.4

v0.7.3 - Dependencies Update

Dependencies Update

Hotfix release updating GPU backend dependencies to latest versions.

Updated Dependencies

• go-webgpu/webgpu v0.1.1 → v0.1.2
• go-webgpu/goffi v0.3.3 → v0.3.5

Links

• PR: #21
• Changelog: https://github.com/born-ml/born/blob/main/CHANGELOG.md