mirror of
https://github.com/bdbrown4/peltier.io.git
synced 2026-11-07 10:58:28 +00:00
phase0: dry run — comrak & tokei vendored, 3 attempts in ledger
Target selection (SPEC §6): comrak 0.53.0 (45c1995, BSD-2-Clause) and tokei 14.0.0 (fa44e51, MIT/Apache-2.0); both CPU-bound with pinned corpora (progit book 11MB; 8x replicated pinned trees, 221MB) and committed MANIFEST.sha256 + golden-output pins + regen scripts. Calibration: A/A null verdicts on both real workloads; taskset -c 2 pinning tightens the A/A CI from [0.938, 1.008] to [0.993, 1.008], making the 2% accept threshold meaningful in this container. Profiles (callgrind; perf unavailable): comrak allocator symbols 26.7% of Ir; tokei perform_multi_line_analysis 31.7%. Attempts (ledger + results/phase0/): - comrak class-1 target-cpu=native: null [0.983, 1.011], rejected-bench - comrak class-2 mimalloc swap: +4.6% [1.033, 1.058], 848 upstream tests green, golden byte-identical, ASan-proper clean; LSan flags 484B in onig (reachability false-positive analysis in case study) -> needs-human-review per SPEC §8 - tokei class-1 fat LTO+cgu=1: +1.6% [1.003, 1.026], real but under the 2% bar, rejected-bench Also: ledger-cli for manual rows, targets/ excluded from trust-layer workspace, corpora/build artifacts gitignored, case study written. Phase 0 exit criteria (>=10% verified win) not yet met; next hypotheses logged in results/phase0/attempts.md. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01EcKPWQ3MQD4NdEHwar96u2
This commit is contained in:
Claude
@@ -0,0 +1,4 @@
|
||||
env fingerprint: {"kernel":"6.18.5","cpu_model":"Intel(R) Xeon(R) Processor @ 2.80GHz","governor":"unknown","smt":"0","turbo":"unknown","aslr":"mode=system-default randomize_va_space=2","cpu_pinning":"none"}
|
||||
plan: 30 measured + 3 warm-up runs per side, interleaved
|
||||
speedup (baseline/candidate): median 1.0006, 95% CI [0.9932, 1.0080]
|
||||
A/A self-test passed (null verdict)
|
||||
@@ -0,0 +1,4 @@
|
||||
env fingerprint: {"kernel":"6.18.5","cpu_model":"Intel(R) Xeon(R) Processor @ 2.80GHz","governor":"unknown","smt":"0","turbo":"unknown","aslr":"mode=system-default randomize_va_space=2","cpu_pinning":"none"}
|
||||
plan: 30 measured + 3 warm-up runs per side, interleaved
|
||||
speedup (baseline/candidate): median 0.9773, 95% CI [0.9376, 1.0080]
|
||||
A/A self-test passed (null verdict)
|
||||
@@ -0,0 +1,4 @@
|
||||
env fingerprint: {"kernel":"6.18.5","cpu_model":"Intel(R) Xeon(R) Processor @ 2.80GHz","governor":"unknown","smt":"0","turbo":"unknown","aslr":"mode=system-default randomize_va_space=2","cpu_pinning":"none"}
|
||||
plan: 30 measured + 3 warm-up runs per side, interleaved
|
||||
speedup (baseline/candidate): median 0.9984, 95% CI [0.9891, 1.0118]
|
||||
A/A self-test passed (null verdict)
|
||||
@@ -0,0 +1,40 @@
|
||||
--------------------------------------------------------------------------------
|
||||
Profile data file 'results/comrak/callgrind.out' (creator: callgrind-3.22.0)
|
||||
--------------------------------------------------------------------------------
|
||||
I1 cache:
|
||||
D1 cache:
|
||||
LL cache:
|
||||
Timerange: Basic block 0 - 196578070
|
||||
Trigger: Program termination
|
||||
Profiled target: targets/comrak/workspace/target/release/comrak (PID 17601, part 1)
|
||||
Events recorded: Ir
|
||||
Events shown: Ir
|
||||
Event sort order: Ir
|
||||
Thresholds: 99
|
||||
Include dirs:
|
||||
User annotated:
|
||||
Auto-annotation: on
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
Ir
|
||||
--------------------------------------------------------------------------------
|
||||
883,766,827 (100.0%) PROGRAM TOTALS
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
Ir file:function
|
||||
--------------------------------------------------------------------------------
|
||||
88,282,747 ( 9.99%) ./malloc/./malloc/malloc.c:_int_malloc [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
60,568,128 ( 6.85%) ./malloc/./malloc/malloc.c:_int_free [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
58,911,428 ( 6.67%) ???:comrak::parser::inlines::Subject::parse_inline [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
57,026,217 ( 6.45%) ???:comrak::parser::Parser::process_line [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
45,107,073 ( 5.10%) ???:core::str::converts::from_utf8 [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
41,962,238 ( 4.75%) ???:<comrak::plugins::syntect::SyntectAdapter as comrak::adapters::SyntaxHighlighterAdapter>::write_highlighted [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
34,662,512 ( 3.92%) ./malloc/./malloc/malloc.c:malloc [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
33,819,139 ( 3.83%) ???:comrak::parser::Parser::parse [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
30,762,786 ( 3.48%) ./malloc/./malloc/malloc.c:malloc_consolidate [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
27,843,363 ( 3.15%) ???:syntect::parsing::parser::ParseState::parse_line [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
27,272,869 ( 3.09%) ./string/../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S:__memcpy_avx_unaligned_erms [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
26,634,752 ( 3.01%) ???:comrak::parser::Parser::open_new_blocks [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
22,049,414 ( 2.49%) ./malloc/./malloc/malloc.c:free [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
21,200,547 ( 2.40%) ???:jetscii::simd::Bytes::find [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
20,549,182 ( 2.33%) ???:alloc::raw_vec::RawVecInner<A>::finish_grow [/home/user/peltier.io/targets/comrak/workspace/target/release/comrak]
|
||||
Binary file not shown.
@@ -0,0 +1,4 @@
|
||||
env fingerprint: {"kernel":"6.18.5","cpu_model":"Intel(R) Xeon(R) Processor @ 2.80GHz","governor":"unknown","smt":"0","turbo":"unknown","aslr":"mode=system-default randomize_va_space=2","cpu_pinning":"none"}
|
||||
plan: 30 measured + 3 warm-up runs per side, interleaved
|
||||
speedup (baseline/candidate): median 0.9965, 95% CI [0.9826, 1.0110]
|
||||
verdict: rejected-bench
|
||||
@@ -0,0 +1,4 @@
|
||||
env fingerprint: {"kernel":"6.18.5","cpu_model":"Intel(R) Xeon(R) Processor @ 2.80GHz","governor":"unknown","smt":"0","turbo":"unknown","aslr":"mode=system-default randomize_va_space=2","cpu_pinning":"none"}
|
||||
plan: 30 measured + 3 warm-up runs per side, interleaved
|
||||
speedup (baseline/candidate): median 1.0459, 95% CI [1.0333, 1.0577]
|
||||
verdict: accepted
|
||||
@@ -0,0 +1,4 @@
|
||||
env fingerprint: {"kernel":"6.18.5","cpu_model":"Intel(R) Xeon(R) Processor @ 2.80GHz","governor":"unknown","smt":"0","turbo":"unknown","aslr":"mode=system-default randomize_va_space=2","cpu_pinning":"none"}
|
||||
plan: 30 measured + 3 warm-up runs per side, interleaved
|
||||
speedup (baseline/candidate): median 1.0161, 95% CI [1.0026, 1.0262]
|
||||
verdict: rejected-bench
|
||||
@@ -0,0 +1,84 @@
|
||||
# Phase 0 dry-run attempt log
|
||||
|
||||
Hypotheses are recorded here *before* any patch is applied; final
|
||||
evidence and verdicts go to the SQLite ledger and are mirrored here.
|
||||
|
||||
Environment for the whole session: remote 4-core container
|
||||
(Intel Xeon @ 2.10GHz, kernel 6.18.5), governor/turbo not controllable,
|
||||
benches pinned with `taskset -c 2`. A/A calibration on both target
|
||||
workloads passed with pinning (see results/calibration/).
|
||||
|
||||
## Deviations from SPEC (Phase 0 manual dry run)
|
||||
|
||||
- No nested no-network container available (no docker daemon): target
|
||||
code runs in this session's own isolated ephemeral container, which
|
||||
serves as the sandbox. Phase 1 must restore the dedicated container.
|
||||
- `perf` unavailable (container): profiling via valgrind/callgrind
|
||||
(deterministic instruction counts, Ir), not cycle sampling. Ir% is a
|
||||
proxy for cycles%; it under-weights cache/branch effects.
|
||||
- Corpus files >10 MB are pinned by committed MANIFEST.sha256 +
|
||||
regeneration scripts rather than committed raw (repo bloat).
|
||||
|
||||
## Attempt 1 — comrak, class 1 (build configuration)
|
||||
|
||||
- **Logged**: 2026-07-06, before patching.
|
||||
- **Hotspot**: whole-program; upstream already ships lto=true +
|
||||
codegen-units=1, so classic class-1 flags are exhausted upstream.
|
||||
- **Hypothesis**: `RUSTFLAGS=-C target-cpu=native` (baseline ISA: the
|
||||
bench machine's Xeon) enables wider SIMD paths in string scanning
|
||||
(jetscii, memchr paths visible in profile) and better scheduling;
|
||||
expect a small win, 2-6% median speedup. No FP semantics change
|
||||
(integer/string workload; not a -ffast-math-class flag), so
|
||||
auto-acceptable if gates pass. Deploy caveat: win only valid for
|
||||
fleets whose ISA baseline matches; recorded in workload statement.
|
||||
- **Patch**: none (env-level build config); candidate built to a
|
||||
separate CARGO_TARGET_DIR from the same pinned commit.
|
||||
|
||||
## Attempt 2 — comrak, class 2 (allocator swap)
|
||||
|
||||
- **Logged**: 2026-07-06, before patching, after Attempt 1's hypothesis.
|
||||
- **Hotspot**: allocator symbols sum to ~26.7% of Ir (_int_malloc 9.99%,
|
||||
_int_free 6.85%, malloc 3.92%, malloc_consolidate 3.48%, free 2.49%)
|
||||
— the class-2 precondition (≥5%) is exceeded 5x.
|
||||
- **Hypothesis**: glibc malloc is the allocator; comrak's AST/inline
|
||||
parsing churns small allocations. Swapping the CLI's global allocator
|
||||
to mimalloc cuts allocator instructions substantially; expect 8-20%
|
||||
median speedup on the progit workload.
|
||||
- **Patch plan**: 2-line change in comrak CLI main + Cargo.toml dep,
|
||||
inside targets/comrak/workspace only.
|
||||
|
||||
## Attempt 3 — tokei, class 1 (build configuration)
|
||||
|
||||
- **Logged**: 2026-07-06, before patching.
|
||||
- **Hotspot**: whole-program; upstream ships lto="thin", panic="abort",
|
||||
default codegen-units (16).
|
||||
- **Hypothesis**: lto=true (fat) + codegen-units=1 lets the analysis
|
||||
loop (perform_multi_line_analysis, 31.7% Ir) inline across cgu and
|
||||
crate boundaries (memchr/aho-corasick calls visible in profile);
|
||||
expect 2-8% median speedup. No semantics change; auto-acceptable if
|
||||
golden replay passes.
|
||||
- **Patch**: [profile.release] lto = true, codegen-units = 1 in
|
||||
targets/tokei/workspace/Cargo.toml only.
|
||||
|
||||
## Outcomes (ledger is authoritative: results/ledger.sqlite)
|
||||
|
||||
| run_id | class | speedup median | 95% CI | verdict |
|
||||
|---|---|---|---|---|
|
||||
| phase0-comrak-001 | 1 build-config | 0.9965 | [0.9826, 1.0110] | rejected-bench (null) |
|
||||
| phase0-comrak-002 | 2 allocator | **1.0459** | **[1.0333, 1.0577]** | needs-human-review |
|
||||
| phase0-tokei-001 | 1 build-config | 1.0161 | [1.0026, 1.0262] | rejected-bench (< 2% threshold) |
|
||||
|
||||
- comrak-002 gates: 848 upstream tests green on patched tree; golden
|
||||
replay byte-identical; pristine ASan+LSan clean; patched ASan-proper
|
||||
clean; LSan flags 484 B in onig (analysis: LSan × custom-allocator
|
||||
reachability false positive — see case-study-comrak.md). SPEC §8
|
||||
routes any sanitizer flag to human review, so that is the verdict.
|
||||
- tokei-001 is a *real but small* effect — CI excludes 1.0 but its
|
||||
lower bound (+0.26%) is under the 2% bar. Correctly rejected; logged
|
||||
so nobody re-grinds it.
|
||||
- Phase 0 exit criteria (≥1 verified win ≥10%) **not yet met**: best
|
||||
verified candidate is +4.6% pending human review. Next cheapest
|
||||
hypotheses per profile: comrak class 3 (allocation churn — arena the
|
||||
inline parser's Vec growth, RawVec::finish_grow 2.3% + memcpy 3.1%);
|
||||
tokei class 3 (memset 8.5% suggests per-file buffer zeroing) and
|
||||
class 5 (perform_multi_line_analysis 31.7% single frame).
|
||||
@@ -0,0 +1,87 @@
|
||||
# Phase 0 case study — comrak (CommonMark/GFM renderer)
|
||||
|
||||
**Result: +4.6% median wall-time speedup (95% CI [+3.3%, +5.8%]) from a
|
||||
2-line allocator swap, byte-identical output, upstream tests green,
|
||||
verdict `needs-human-review` on a LeakSanitizer reachability flag.**
|
||||
|
||||
## Target
|
||||
|
||||
- comrak v0.53.0, commit `45c1995f`, BSD-2-Clause, built `--release`
|
||||
(upstream profile already ships `lto = true`, `codegen-units = 1`).
|
||||
- Workload: the target's own benchmark (benches/bench.sh) — the Pro Git
|
||||
book, all translations, 11.0 MB concatenated markdown, single stream
|
||||
through stdin, HTML to /dev/null. Corpus pinned:
|
||||
`corpora/comrak/MANIFEST.sha256` (progit @ `61833a52`).
|
||||
|
||||
## Environment
|
||||
|
||||
Remote 4-core container, Intel Xeon @ 2.10GHz, kernel 6.18.5, governor
|
||||
and turbo state unavailable/uncontrollable, SMT off, ASLR system
|
||||
default, benches pinned with `taskset -c 2`, 30 measured + 3 warm-up
|
||||
runs per side, interleaved ABAB, bootstrap 95% CI (10k resamples).
|
||||
|
||||
**A/A calibration, same session, pinned** (results/calibration/):
|
||||
median 1.0006, 95% CI [0.9932, 1.0080] — null, as required. Unpinned
|
||||
A/A for contrast: CI [0.9376, 1.0080] — 5x noisier; pinning is what
|
||||
makes a 2% threshold meaningful in this container.
|
||||
|
||||
## Loop
|
||||
|
||||
1. **Profile** (callgrind; perf unavailable in container — Ir counts,
|
||||
not cycles): allocator symbols total **26.7% of instructions**
|
||||
(`_int_malloc` 9.99%, `_int_free` 6.85%, `malloc` 3.92%,
|
||||
`malloc_consolidate` 3.48%, `free` 2.49%). Parser frames
|
||||
(`parse_inline` 6.7%, `process_line` 6.5%) come *after* the
|
||||
allocator. Full ranking: results/comrak/hotspots.txt.
|
||||
2. **Attempt 1 — class 1, build config** (`target-cpu=native`):
|
||||
hypothesis logged first; golden replay byte-identical; bench median
|
||||
0.9965, 95% CI [0.9826, 1.0110] → **rejected-bench** (null). Ledger
|
||||
`phase0-comrak-001`.
|
||||
3. **Attempt 2 — class 2, allocator swap** (mimalloc as
|
||||
`#[global_allocator]` in the CLI binary only; library consumers
|
||||
unaffected; patch: results/phase0/comrak-mimalloc.patch):
|
||||
- Golden replay: stdout sha256 identical to pinned golden.
|
||||
- Upstream test suite on the patched tree: green (see ledger row).
|
||||
- Bench, interleaved A/B vs pristine-built baseline: **median
|
||||
speedup 1.0459, 95% CI [1.0333, 1.0577]** — CI lower bound 3.3%
|
||||
clears the 2% accept threshold.
|
||||
- Sanitizers (nightly ASan on the full corpus run):
|
||||
- pristine binary: clean, output hash = golden.
|
||||
- patched binary: ASan-proper clean with `detect_leaks=0`;
|
||||
**LeakSanitizer reports 484 B / 3 allocations in
|
||||
`onig_new`/`onig_compile`** (oniguruma, syntect's C regex
|
||||
engine). Analysis: these allocations exist identically in the
|
||||
pristine binary (LSan exit 0 there) — the report appears because
|
||||
LSan cannot scan mimalloc-owned memory for roots, so a
|
||||
still-reachable global compiled regex looks unreachable. Almost
|
||||
certainly a false positive of the LSan × custom-allocator
|
||||
interaction, **not** a leak introduced by the patch.
|
||||
- Per SPEC §8 ("anything the sanitizers flag, even as warnings"):
|
||||
verdict **needs-human-review**, not auto-accept. Ledger
|
||||
`phase0-comrak-002`.
|
||||
|
||||
## ROI (illustrative, config/pricing.toml rates)
|
||||
|
||||
For a hypothetical fleet spending 100 cores 24/7 on this workload shape,
|
||||
at $0.04/core-hr (public-cloud example rate — replace per engagement):
|
||||
|
||||
- cores returned: median **4.4**, 95% CI [3.2, 5.5]
|
||||
(`fleet × (1 − 1/speedup)`)
|
||||
- dollars/yr: median **$1,540**, 95% CI [$1,130, $1,910]
|
||||
|
||||
Caveats printed with the number, per policy: single workload shape
|
||||
(large-document batch rendering); Ir-based profile, not cycles;
|
||||
container hardware without governor control; speedup CI is wall-time on
|
||||
pinned single-core execution; ISA = generic x86-64 (no target-cpu
|
||||
narrowing — attempt 1 showed it buys nothing here anyway).
|
||||
|
||||
## Reproduction
|
||||
|
||||
```
|
||||
just pin-check comrak # corpus manifest verify
|
||||
cargo run -p bench-runner -- --config config/accept.toml compare \
|
||||
--baseline "taskset -c 2 sh -c '<pristine comrak> < corpora/comrak/progit.md > /dev/null'" \
|
||||
--candidate "taskset -c 2 sh -c '<patched comrak> < corpora/comrak/progit.md > /dev/null'"
|
||||
```
|
||||
|
||||
Patch: `results/phase0/comrak-mimalloc.patch` applied at `45c1995f`.
|
||||
@@ -0,0 +1,39 @@
|
||||
diff --git a/Cargo.toml b/Cargo.toml
|
||||
index bd1dfe9..fcc70f3 100644
|
||||
--- a/Cargo.toml
|
||||
+++ b/Cargo.toml
|
||||
@@ -69,7 +69,7 @@ phf_codegen = "0.13"
|
||||
|
||||
[features]
|
||||
default = ["cli", "syntect-onig", "bon"]
|
||||
-cli = ["clap", "bon", "shell-words", "xdg", "fmt2io", "shortcodes", "phoenix_heex", "attributes"]
|
||||
+cli = ["clap", "bon", "shell-words", "xdg", "fmt2io", "shortcodes", "phoenix_heex", "attributes", "mimalloc"]
|
||||
shortcodes = ["emojis"]
|
||||
phoenix_heex = []
|
||||
attributes = []
|
||||
@@ -107,6 +107,7 @@ clap = { version = "4", optional = true, features = [
|
||||
"string",
|
||||
"wrap_help",
|
||||
] }
|
||||
+mimalloc = { version = "0.1", optional = true }
|
||||
|
||||
[[example]]
|
||||
name = "syntect"
|
||||
diff --git a/src/main.rs b/src/main.rs
|
||||
index 6c5b2a6..a439fb7 100644
|
||||
--- a/src/main.rs
|
||||
+++ b/src/main.rs
|
||||
@@ -10,6 +10,13 @@ use std::{boxed::Box, io::BufWriter};
|
||||
|
||||
use clap::{Parser, ValueEnum};
|
||||
|
||||
+// The CLI churns many small AST allocations; mimalloc services them with
|
||||
+// far fewer instructions than glibc malloc. Library consumers are
|
||||
+// unaffected: this only applies to the comrak binary.
|
||||
+#[cfg(not(target_arch = "wasm32"))]
|
||||
+#[global_allocator]
|
||||
+static GLOBAL: mimalloc::MiMalloc = mimalloc::MiMalloc;
|
||||
+
|
||||
use comrak::options;
|
||||
use comrak::{Arena, Options};
|
||||
#[cfg(feature = "syntect")]
|
||||
@@ -0,0 +1,14 @@
|
||||
diff --git a/Cargo.toml b/Cargo.toml
|
||||
index 8b19522..b64e2b4 100644
|
||||
--- a/Cargo.toml
|
||||
+++ b/Cargo.toml
|
||||
@@ -30,7 +30,8 @@ default = ["cli"]
|
||||
yaml = ["dep:serde_yaml"]
|
||||
|
||||
[profile.release]
|
||||
-lto = "thin"
|
||||
+lto = true
|
||||
+codegen-units = 1
|
||||
panic = "abort"
|
||||
|
||||
[[bin]]
|
||||
@@ -0,0 +1,40 @@
|
||||
--------------------------------------------------------------------------------
|
||||
Profile data file 'results/tokei/callgrind.out' (creator: callgrind-3.22.0)
|
||||
--------------------------------------------------------------------------------
|
||||
I1 cache:
|
||||
D1 cache:
|
||||
LL cache:
|
||||
Timerange: Basic block 0 - 600400493
|
||||
Trigger: Program termination
|
||||
Profiled target: targets/tokei/workspace/target/release/tokei --sort code corpora/tokei/input (PID 17959, part 1)
|
||||
Events recorded: Ir
|
||||
Events shown: Ir
|
||||
Event sort order: Ir
|
||||
Thresholds: 99
|
||||
Include dirs:
|
||||
User annotated:
|
||||
Auto-annotation: on
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
Ir
|
||||
--------------------------------------------------------------------------------
|
||||
1,808,959,061 (100.0%) PROGRAM TOTALS
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
Ir file:function
|
||||
--------------------------------------------------------------------------------
|
||||
572,758,799 (31.66%) ???:tokei::language::syntax::SyntaxCounter::perform_multi_line_analysis [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
172,785,648 ( 9.55%) ???:tokei::language::language_type::LanguageType::parse_lines [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
154,178,149 ( 8.52%) ./string/../sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S:__memset_avx2_unaligned_erms [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
146,199,746 ( 8.08%) ./string/../sysdeps/x86_64/multiarch/memcmp-avx2-movbe.S:__memcmp_avx2_movbe [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
104,071,008 ( 5.75%) ???:tokei::language::syntax::SyntaxCounter::try_perform_single_line_analysis [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
93,030,608 ( 5.14%) ???:memchr::arch::x86_64::avx2::packedpair::Finder::find_impl [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
75,136,592 ( 4.15%) ???:aho_corasick::automaton::try_find_fwd [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
71,820,456 ( 3.97%) ./string/../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S:__memcpy_avx_unaligned_erms [/usr/lib/x86_64-linux-gnu/libc.so.6]
|
||||
50,442,206 ( 2.79%) ???:memchr::arch::x86_64::avx2::memchr::One::find_raw_avx2 [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
28,664,232 ( 1.58%) ???:memchr::memmem::searcher::searcher_kind_avx2 [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
27,626,392 ( 1.53%) ???:<aho_corasick::util::prefilter::Memmem as aho_corasick::util::prefilter::PrefilterI>::find_in [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
26,582,656 ( 1.47%) ???:memchr::arch::x86_64::memchr::memchr_raw::find_avx2 [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
22,263,784 ( 1.23%) ???:memchr::arch::x86_64::avx2::memchr::Three::find_raw [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
20,246,771 ( 1.12%) ???:tokei::language::syntax::SyntaxCounter::perform_multi_line_analysis'2 [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
20,005,440 ( 1.11%) ???:memchr::arch::x86_64::avx2::memchr::One::find_raw [/home/user/peltier.io/targets/tokei/workspace/target/release/tokei]
|
||||
Reference in New Issue
Block a user