ta
TanStack
GitHub
router
Workspace
GitHub
CI Pipeline Executions
Current execution
Succeeded
perf/frame-decoder-zero-copy-extract
Runs and Agents
Configuration
Self-Healing CI
Project Graph
Circular Dependencies
Enterprise
Resource Usage
Analysis
Conformance
Enterprise
Runs
Current run
Succeeded
Tasks
Resource Usage
Insights
Compare tasks
Analytics
Sign in
Toggle sidebar
perf(start-client-core): zero-copy frame payload extraction extractFlattened() always allocated a new Uint8Array and copied `count` bytes, even when the requested bytes were fully contained in the first buffered chunk (the common case, since most frames arrive within a single network read). Add a fast path that returns a subarray view of the first chunk when the bytes are contiguous, avoiding the allocation and the byte copy. The multi-chunk path is unchanged. The returned view shares the chunk's backing ArrayBuffer, which is safe because buffered chunks are never mutated in place after being read. This is on the hot path for decoding every streamed server-function response and RawStream binary payload on the client. A micro-benchmark shows ~3x faster extraction for 1KB frames and ~27x for 64KB frames (the win scales with payload size).
nx run-many --target=build --exclude=examples/** --exclude=e2e/**
⌘K
Succeeded
nx run-many --target=build --exclude=examples/** --exclude=e2e/**
Click to copy
Linux
4 CPU cores
6f696aaf
7662
Get faster results
Learn how to enable distribution.
Atomizer enabled
43 groups of tasks optimized.
GitHub
router
Workspace