Creează un cont pentru a te abona la acest blog și a-i urmări articolele pe pagina Lectură.
-
Hot path optimization: unsafe pointer arithmetic to eliminate bound checks the Go compiler can't remove, given you can prove they are truly unnecessary.
-
An alignment quirk on Intel and AMD CPUs: keeping a Go array 8-byte aligned (vs offset 4) makes clearing it up to 49% faster, thanks to how REP STOSQ works.
-
A counterintuitive hot-path optimization from my Go work: swapping integer division (IDIVQ) for floating-point division (DIVSD) to divide integers faster.
-
A practical tour of compression in Go web servers: static assets, dynamic content, shared dictionaries, training, and streaming.
-
How a change to one Go function slowed unrelated code by 3%.
-
Lessons from porting Brotli to pure Go.
-
Notes on setting up an LVM dm-cache volume on Linux.
-
Performance notes on reusing the body of an outgoing http.Request without redundant allocations and copies in Go middleware.