Why cgo is slow @ CapitalGo 2018

https://capitalgolang.com/program#filippo_valsorda

This talk uses cgo and its below-average performance as an excuse to look into Go internals and what makes Go different from C.

We learn about calling conventions and the code-generated cgo trampolines; about the small goroutines stacks and how C doesn't know how to grow them; about the Go scheduler and how C doesn't yield to it; and about the garbage collector and how pointers in C memory can't be tracked.

Transcript

1. Why cgo is slow Filippo Valsorda
3. cgo is a FFI (Foreign Function Interface)
4. I like FFIs. • From cgo back to Go @ GopherCon 2016
 https://speakerdeck.com/filosottile/from-cgo-back-to-go-gophercon-2016 • rustgo: Building your own FFI @ GothamGo 2017
 https://speakerdeck.com/filosottile/calling-rust-from-go-without-cgo-at-gothamgo-2017 • Why cgo is slow @ CapitalGo 2018
 Hi!
5. C function call Java FFI Rust FFI LuaJIT FFI Node.js FFI cgo 2.364 ns 9.01 ns 2.386 ns 1.81 ns (!) https://nullprogram.com/blog/2018/05/27/ 18.33 ns 75.95 ns https://github.com/dyu/ffi-overhead
6. name old time/op new time/op delta CgoCall-4 63.1ns ± 3% 57.1ns ± 0% -9.43%
7. C function call Java FFI Rust FFI LuaJIT FFI Node.js FFI cgo 2.364 ns 9.01 ns 2.386 ns 1.81 ns (!) https://nullprogram.com/blog/2018/05/27/ 18.33 ns 68.77 ns https://github.com/dyu/ffi-overhead
8. C function call Java FFI Rust FFI LuaJIT FFI Node.js FFI cgo 2.364 ns 9.01 ns 2.386 ns 1.81 ns (!) https://nullprogram.com/blog/2018/05/27/ 18.33 ns 68.77 ns (29x) https://github.com/dyu/ffi-overhead
9. cgo: • cmd/cgo • runtime/cgo • a sprinkle of cmd/link/internal/ld support
10. cgo: • cmd/cgo • runtime/cgo • a sprinkle of cmd/link/internal/ld support • not a compiler feature!
11. cgo: • cmd/cgo — a code generator • runtime/cgo • a sprinkle of cmd/link/internal/ld support • not a compiler feature!
12. Reason 1: calling conventions
13. C compiler Go compiler
14. go build -x -work
19. src/runtime/cgocall.go
23. rewritten Go function calling convention trampoline arg unpacking real C function Go ASM C
24. Learn more: • src/runtime/cgocall.go • rustgo: Building your own FFI @ GothamGo 2017
 https://speakerdeck.com/filosottile/calling-rust-from-go-without-cgo-at-gothamgo-2017
25. Reason 2: small stacks
26. Initial goroutine stack size: 2048 bytes System stack size: 1–8 megabytes
27. stack
28. function frame
34. Function preamble
35. C doesn't call morestack C code needs to run on a system stack cgocall / asmcgocall
36. Learn more: • src/runtime/stack.go • src/runtime/cgocall.go • How stacks are handled in Go by Daniel Morsing
 https://blog.cloudflare.com/how-stacks-are-handled-in-go/
37. Reason 3: the scheduler
38. From https://morsmachine.dk/go-scheduler
39. From https://morsmachine.dk/go-scheduler
40. The Go scheduler is collaborative. It can't preempt running code. (ProTip: for {} is never what you want. Use select {}.)
42. From https://morsmachine.dk/go-scheduler
43. Learn more: • src/runtime/proc.go → reentersyscall • The Go scheduler by Daniel Morsing
 https://morsmachine.dk/go-scheduler • Performance without the event loop by Dave Cheney
 https://dave.cheney.net/2015/08/08/performance-without-the-event-loop
44. Reason 4: the garbage collector
45. Go memory C memory []byte
46. Go memory C memory GC []byte
47. Go memory C memory GC []byte
48. Go memory C memory GC []byte *uint8_t C.some_func()
49. The cgo rules You may pass a Go pointer … if it doesn’t point to other pointers … and C can’t keep a reference to it
50. The GC must see all the Go pointers.
51. panic: runtime error: cgo argument has Go pointer to Go pointer GODEBUG=cgocheck=2
52. Learn more: • From cgo back to Go @ GopherCon 2016
 https://speakerdeck.com/filosottile/from-cgo-back-to-go-gophercon-2016
53. Thank you! filippo@golang.org @FiloSottile Olga Shalakhina artwork under CC 3.0 license based on Renee French under Creative Commons 3.0 Attributions.