diff options
Diffstat (limited to 'gui/bench/RESULTS.md')
| -rw-r--r-- | gui/bench/RESULTS.md | 102 |
1 files changed, 102 insertions, 0 deletions
diff --git a/gui/bench/RESULTS.md b/gui/bench/RESULTS.md new file mode 100644 index 0000000..d24659b --- /dev/null +++ b/gui/bench/RESULTS.md @@ -0,0 +1,102 @@ +# Debug V8 versus a small C target + +Measured locally on 2026-11-12, Linux 7.2.11-arch1-1, Intel i7-11700, +Zig 0.17.0, ReleaseFast controller. The targets were a `cc -g -O0` C loop +and **unoptimized debug V8**, not release d8. Both used the same controller: +512 stack bytes, 24 disassembled instructions and telescope depth 2, with a +recording flush at every stop. No syscall tracing and no continuous stepping +while the user has selected Continue. + +Final controller run: + +```sh +python3 gui/bench/steps.py \ + --d8 /tmp/computeranim-v8-debug-remote/d8 \ + --ztrace zig-out/bin/ztrace \ + --out /tmp/ztrace-step-bench-cleared +``` + +| 1,000 instruction steps | C | Debug V8 / generated JS | +| --- | ---: | ---: | +| Capture + flush, median | 35.73 µs | 38.47 µs | +| Capture + flush, p95 | 45.13 µs | 59.18 µs | +| Controller round trip, median | 46.94 µs | 49.53 µs | +| Controller round trip, p95 | 59.23 µs | 73.00 µs | +| Controller round trip, maximum | 310.72 µs | 381.48 µs | +| Launch through first requested breakpoint | 9.50 ms | 901.84 ms | + +This supports comparable steady instruction-step cost for these workloads. +It does not claim that V8 starts as quickly as a small program, or that all C++ +source/DWARF operations have this cost. The earlier 1,000-step run measured +39.09 / 44.42 µs capture medians and 53.69 / 60.57 µs round-trip medians for +C / V8. Scheduling and cache state affect these microsecond measurements; no +CPU isolation or filesystem-cache dropping was used. GUI presentation is not +included in the controller timings. + +The V8 fixture compiles `jitWork` with TurboFan, traps in the native runtime, +then installs an actual software breakpoint at `addl r8,r11` inside the +optimized loop. The harness asserts the returned RIP equals that address and +lies in an anonymous executable mapping. It then tests source-step fallback +without JIT source lines, records the instruction-step sample, and checks that +the recording no longer reports the previous native source line. The loop +address is read from V8's own output; the harness fails rather than guesses if +a different V8 compiler allocates different registers. + +The raw records, logs and JSON results are in `/tmp/ztrace-step-bench-cleared`. +The script is the durable reproducer; `/tmp` artifacts are not permanent. + +## V8 provenance + +The user's remote Arch build host (`chaincli-box`) has the existing checkout at +`/home/goblin/v8-research/v8`. Ninja verified the compatible debug build up to date +(`ninja: no work to do`). No V8 compilation was performed on the workstation. +The complete runtime artifacts were copied to `/tmp/computeranim-v8-debug-remote` +and smoke-tested locally. + +- Revision: `84703f702dca525edd469f3a98d784faa46a02f8`. +- d8 version: `15.2.0 (candidate)`. +- Output: `out/x64.debug-vuln` (historical directory name). +- Existing unrelated local `BUILD.gn` change to the `v8_fuzzer` source list was + preserved; the checkout was not reset or updated. +- `is_debug=true`, `v8_optimized_debug=false`, `symbol_level=2`, + `v8_symbol_level=2`, `use_debug_fission=true`, `target_cpu="x64"`. +- Disassembler, object printing, heap verification, slow DCHECKs and backtraces + enabled; `use_remoteexec=false`, `use_siso=false`. +- Component build: d8 plus its shared libraries (including the large debug + libv8.so), snapshots and ICU data. Split `.dwo` files were not copied, so this + run is not a validation of complete C++ local-variable/source debugging. + +## Checks + +The GUI's real-target script exercises launch, a named breakpoint, continue, +keyboard and mouse instruction stepping, source stepping, pause during execution, +and browsing history back to live. The core tests additionally check pause +signal suppression and source stepping without line information. + +The root suite had two failures before this work; they remain in the existing +loader/syscall formatting and watch/string assertions. The new debugger tests +pass (86/88 in the full root suite; 25/27 debugger tests, with those same two +failures). No assertion was +weakened to hide them. + +## Rich GUI verification + +A separate UI agent compared rendered C and V8 frames against the existing +animation design, checked strike/highlight/settled frames, and approved the +final native window. The checked-in GUI test asserts that the stack highlight +changes between strike and settled screenshots, as well as exercising controls. + +At 1920×1080, an approximate headless redraw measurement was **1.23 ms/frame +for C and 2.41 ms/frame for V8**. This subtracts the median elapsed time of +30-frame runs from 330-frame runs (three repeats each), then divides by 300. +Both opened the corresponding 1,000-step recording. V8 displayed 24 instruction +rows while C displayed 10; this is a whole-view cost, not an equal-row renderer +comparison. It excludes live trace reopening and display presentation latency. +Raw timings are in `/tmp/ztrace-gui-redraw-results.json`. + +The final native session was verified with d8 PID 2502490 in Linux tracing-stop +state and recorded RIP `0x7f09921c0196`: breakpoint 2 on `add r8d, r9d`, offset +`0x156` inside TurboFan-generated `jitWork`. It is left open in the GUI. The +recording is `/tmp/ztrace-gui-2502452-357545622054931.ztrace`; the screenshot is +`/tmp/ztrace-v8-live-final.png`. These addresses and PIDs describe this session +only. The complete animation gallery was not re-rendered. |
