mise-pin zig 0.15.2; rebuild libghostty-vt ReleaseFast — 27-32x pipeline speedup
Added .mise.toml pinning zig = "0.15.2" (the minimum the vendored
Ghostty commit requires) and taught the Makefile to resolve zig
through mise when available, falling back to PATH. Contributors run
`mise install` once and `make deps` just works.
Re-ran the pipeline benchmarks after rebuilding libghostty-vt with
ReleaseFast (same hardware, AMD Ryzen 7 7800X3D):
Debug ReleaseFast speedup
Pipeline 8-colour @120fps 63 fps 2030 fps 32x
Pipeline truecolor @120fps 34 fps 931 fps 27x
Emulator-only truecolor 34 fps 2051 fps 60x
7-16x headroom over 120 fps for the heaviest workload (truecolor
full-screen redraws). Static library size 33 MiB -> 13 MiB.
TODO.md baseline numbers updated to reflect post-fix throughput;
the "Debug-mode lib" finding is folded into the result it produced
rather than left as an open item.
This commit is contained in:
37
TODO.md
37
TODO.md
@@ -3,8 +3,8 @@ Findings from a codebase sweep — not user-reported, needs review before
|
||||
action. Each item names the anchor and a sketched fix.
|
||||
|
||||
Baseline benchmark numbers (`go test -bench=. ./internal/app/`, AMD
|
||||
Ryzen 7 7800X3D, libghostty-vt **Debug-mode** — see the first item
|
||||
below):
|
||||
Ryzen 7 7800X3D, libghostty-vt **ReleaseFast** after the Makefile
|
||||
fix landed):
|
||||
|
||||
```
|
||||
# Renderer alone
|
||||
@@ -19,35 +19,18 @@ ASCIIVideo_Stream_8Color_120fps 260 µs/frame 3845 fps_ceiling 3.1% budge
|
||||
ASCIIVideo_Stream_TrueColor_120fps 576 µs/frame 1735 fps_ceiling 6.9% budget
|
||||
|
||||
# Full pipeline (em.Write + renderer + io.Discard write)
|
||||
Pipeline_ASCIIVideo_8Color_120fps 15838 µs/frame 63 fps_ceiling 190% budget
|
||||
Pipeline_ASCIIVideo_TrueColor_120fps 29224 µs/frame 34 fps_ceiling 350% budget
|
||||
Pipeline_ASCIIVideo_8Color_120fps 493 µs/frame 2030 fps_ceiling 5.9% budget
|
||||
Pipeline_ASCIIVideo_TrueColor_120fps 1075 µs/frame 931 fps_ceiling 12.9% budget
|
||||
|
||||
# Emulator alone (libghostty-vt CSI/SGR parser)
|
||||
Emulator_Write_Stream_8Color_120fps 15930 µs/frame 63 fps_ceiling
|
||||
Emulator_Write_Stream_TrueColor_120fps 29241 µs/frame 34 fps_ceiling
|
||||
Emulator_Write_Stream_8Color_120fps 257 µs/frame 3890 fps_ceiling
|
||||
Emulator_Write_Stream_TrueColor_120fps 488 µs/frame 2051 fps_ceiling
|
||||
```
|
||||
|
||||
The renderer alone hits 1700-3800 fps with margin. The full pipeline
|
||||
caps at 34-63 fps. **The whole gap is libghostty-vt's em.Write — its
|
||||
parser is shipping in Debug mode, which is also a 33 MiB static
|
||||
library file (release builds are a fraction of that).**
|
||||
|
||||
- [ ] **libghostty-vt was being built in Debug mode.** [HIGH — partially fixed]
|
||||
- `Makefile` used `zig build -Demit-lib-vt` with no
|
||||
`-Doptimize`. Zig's `standardOptimizeOption` defaults to
|
||||
`.Debug`, so the shipped static lib was unoptimised. Effect:
|
||||
the SGR/CSI parser eats 16-29 ms per 30-70 KiB full-screen
|
||||
frame, capping the entire patterm pipeline at 34-63 fps. The
|
||||
Makefile now defaults to `ReleaseFast` (override via
|
||||
`make deps GHOSTTY_VT_OPTIMIZE=Debug` if you ever need a
|
||||
debug build of the upstream lib for diagnosing a bug in it).
|
||||
- To apply: `make clean-deps && make deps`, then re-run
|
||||
`go test -bench=BenchmarkPipeline -benchmem ./internal/app/`
|
||||
and confirm the truecolor 120fps stream drops well under 100%
|
||||
budget. Update the numbers in this section after rebuilding.
|
||||
- Severity HIGH because it's the single biggest perf win on the
|
||||
table; the renderer optimisations below are second-order until
|
||||
this lands.
|
||||
Result of the fix below: 27-32× pipeline speedup, 60× emulator
|
||||
speedup. Pipeline hits 930-2030 fps end-to-end — 7-16× headroom
|
||||
over the 120 fps target on the heaviest workload (truecolor
|
||||
full-screen redraws).
|
||||
|
||||
|
||||
- [ ] **viewport renderer allocates ~1 alloc per 4 input bytes on SGR/CSI-heavy chunks.** [MEDIUM]
|
||||
|
||||
Reference in New Issue
Block a user