mini_racer_rs 1.5.2 → 1.6.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: ac6808f9735e55f734d5279ac4a61df43c360de9fd62475417fb3a86eed95419
4
- data.tar.gz: 775a078c2e03d715bce13eca7deac20a9df74be55adec4d1942717042ffdf29d
3
+ metadata.gz: 83f21cde14bd9acd57ee2a35e4536767bbd5c7dd6d4d4fe497c3e95043c9a6fc
4
+ data.tar.gz: f755934e6ee5581273b9b2b77c311890a160e313b06206da81284f746117fef9
5
5
  SHA512:
6
- metadata.gz: 9b4a2ee6af6bf08c29484eaab6228ae78b05c17e9c57457b12c339d013047855371b600b1e1f870e62a02d2f04bf75f69e779aab07f094c73179fea3f132634d
7
- data.tar.gz: 2662edd8f467161d2f4d7c93a29e3033feddac65411ce5e5118590aa8c216c09b36f52b259d0aa88285b6876839a13df59bc4d80b48df9133a4203beb337c0d5
6
+ metadata.gz: c8a1fcacea4a5020de89ad8b7005a3692382c73115618ada1d941543a709afe5564637239ec9965ba5bcaf10083383c1dc87ba43116bfc9c2aff20911f6ab0eb
7
+ data.tar.gz: 6765a38eb4439310ac203693c7ba3e2350955e0bf2ce317d61865d4b2d4140e4fcbf86bd3db0e85c8b30bd6e4bd44c2fb23e0d4816955f5ce9b3e3342984bbb6
data/CHANGELOG.md CHANGED
@@ -1,5 +1,26 @@
1
1
  # Changelog
2
2
 
3
+ ## [1.6.0](https://github.com/andymai/mini_racer_rs/compare/v1.5.3...v1.6.0) (2026-10-24)
4
+
5
+
6
+ ### Features
7
+
8
+ * add MiniRacerRs.code_cache ([#164](https://github.com/andymai/mini_racer_rs/issues/164)) ([9238e61](https://github.com/andymai/mini_racer_rs/commit/9238e61e4ae9ab4ae16d50c0035ee7bf611ea258))
9
+
10
+
11
+ ### Performance
12
+
13
+ * **ext:** copy a large eval source without the GVL ([#163](https://github.com/andymai/mini_racer_rs/issues/163)) ([25fa05e](https://github.com/andymai/mini_racer_rs/commit/25fa05e009086eca64641a699fb5190fb876c69c))
14
+
15
+ ## [1.5.3](https://github.com/andymai/mini_racer_rs/compare/v1.5.2...v1.5.3) (2026-10-24)
16
+
17
+
18
+ ### Performance
19
+
20
+ * **engine:** cache a large first sighting that runs as long as it compiles ([#159](https://github.com/andymai/mini_racer_rs/issues/159)) ([47cc0cc](https://github.com/andymai/mini_racer_rs/commit/47cc0ccd76c5d39ec3c9005b848e5846e213d618))
21
+ * **engine:** cache a script again after its builder's next call ([#160](https://github.com/andymai/mini_racer_rs/issues/160)) ([8f632e0](https://github.com/andymai/mini_racer_rs/commit/8f632e065ec3506687d69e8e13f0df56b2e21064))
22
+ * **engine:** cache scripts from 128 KiB and keep 64 caches ([#158](https://github.com/andymai/mini_racer_rs/issues/158)) ([0b77e64](https://github.com/andymai/mini_racer_rs/commit/0b77e640d311db4c22d711a884be5aeadbcc0e0b))
23
+
3
24
  ## [1.5.2](https://github.com/andymai/mini_racer_rs/compare/v1.5.1...v1.5.2) (2026-10-24)
4
25
 
5
26
 
data/COMPATIBILITY.md CHANGED
@@ -103,6 +103,7 @@ What mini_racer_rs adds under `MiniRacerRs`. An internal id is not for callers.
103
103
  | `imethod:MiniRacerRs::Pool#with` | done | n/a |
104
104
  | `module:MiniRacerRs` | done | done |
105
105
  | `module:MiniRacerRs::ForkHook` | internal | n/a |
106
+ | `smethod:MiniRacerRs.code_cache` | done | n/a |
106
107
  | `smethod:MiniRacerRs.counters` | done | n/a |
107
108
  | `smethod:MiniRacerRs.heap_limit` | done | n/a |
108
109
  | `smethod:MiniRacerRs.heap_limit=` | done | n/a |
data/Cargo.lock CHANGED
@@ -593,7 +593,7 @@ checksum = "cf8baf1c55e62ffcace7a9f06f4bd9cd3f0c4beb022d3b367256b91b87513d98"
593
593
 
594
594
  [[package]]
595
595
  name = "mini_racer_engine"
596
- version = "1.5.2"
596
+ version = "1.6.0"
597
597
  dependencies = [
598
598
  "deno_core_icudata",
599
599
  "libc",
@@ -606,7 +606,7 @@ dependencies = [
606
606
 
607
607
  [[package]]
608
608
  name = "mini_racer_rs"
609
- version = "1.5.2"
609
+ version = "1.6.0"
610
610
  dependencies = [
611
611
  "magnus",
612
612
  "mini_racer_engine",
@@ -617,7 +617,7 @@ dependencies = [
617
617
 
618
618
  [[package]]
619
619
  name = "mini_racer_sourcemap"
620
- version = "1.5.2"
620
+ version = "1.6.0"
621
621
  dependencies = [
622
622
  "memchr",
623
623
  "serde",
@@ -627,7 +627,7 @@ dependencies = [
627
627
 
628
628
  [[package]]
629
629
  name = "mini_racer_sync"
630
- version = "1.5.2"
630
+ version = "1.6.0"
631
631
  dependencies = [
632
632
  "loom",
633
633
  "thiserror",
@@ -635,7 +635,7 @@ dependencies = [
635
635
 
636
636
  [[package]]
637
637
  name = "mini_racer_wire"
638
- version = "1.5.2"
638
+ version = "1.6.0"
639
639
  dependencies = [
640
640
  "criterion",
641
641
  "proptest",
data/Cargo.toml CHANGED
@@ -15,7 +15,7 @@ license = "MIT"
15
15
  publish = false
16
16
  repository = "https://github.com/andymai/mini_racer_rs"
17
17
  rust-version = "1.91"
18
- version = "1.5.2"
18
+ version = "1.6.0"
19
19
 
20
20
  [workspace.dependencies]
21
21
  criterion = {version = "0.8", default-features = false, features = ["cargo_bench_support"]}
data/README.md CHANGED
@@ -208,7 +208,7 @@ The repository tests the `path:` form of the shim with rtlcss in `test/consumers
208
208
 
209
209
  ### Forking servers such as Puma
210
210
 
211
- Before forking, evaluate each bundle of 1 MiB or more twice in the master, then dispose those contexts. Use the renderer's exact source text and file name. The second evaluation builds the process-wide V8 code cache. Matching evaluations compile from it, and forked children inherit it. For Babel's 2.3 MB bundle at commit `5e13c1c`, a forked child's first load took 60.8 ms without a parent load. It took 65.5 ms after one load and 25.0 ms after two. These are medians across 5 trials. Prime a `react-rails` app by building its server renderer twice. Children do not use parent contexts. Only the cache carries over.
211
+ Before forking, evaluate each bundle of 128 KiB or more twice in the master, then dispose those contexts. Use the renderer's exact source text and file name. The second evaluation builds the process-wide V8 code cache, or the first for a bundle of 16 MiB or more whose top-level code runs at least as long as it compiles. Matching evaluations compile from the cache, and forked children inherit it. The context that built a cache rebuilds it once after its next call, so a render in the master also gives children the functions that render compiled. For Babel's 2.3 MB bundle at commit `5e13c1c`, a forked child's first load took 60.8 ms without a parent load. It took 65.5 ms after one load and 25.0 ms after two. These are medians across 5 trials. Prime a `react-rails` app by building its server renderer twice. Children do not use parent contexts. Only the cache carries over.
212
212
 
213
213
  For Ruby's `fork`, `IO.popen("-")`, or `Process.daemon`, mini_racer_rs waits for context runners, pauses its worker pool and waits for running tasks. An async `WebAssembly.compile` in a new default-mode child completes on the child's own pool. On Ubuntu 24.04 with glibc 2.39, 20 of 20 mini_racer 0.22.1 children aborted with signal 6 after React rendering; all mini_racer_rs children exited cleanly. mini_racer 0.22.2 has the same fork handling. On Fedora, `test/stress/fork_busy.rb` aborted with mini_racer 0.22.2 in 3 of 3 runs, after 5, 14 and 19 forks, when an evaluating parent thread raised `TypeError`. mini_racer_rs completed 300 forks per run on Fedora and Ubuntu.
214
214
 
@@ -230,7 +230,7 @@ With `mini_racer` 0.22.2, each worker compiles the bundle from source on its fir
230
230
 
231
231
  Results are medians of 5 runs. Every workload produced identical output in both gems. The eight library workloads use one run of the Ruby 3.4 extension from the published mini_racer_rs 1.5.1 `x86_64-linux-gnu` gem against mini_racer 0.22.2. The 44 MB production bundle load uses a second run of the same gem against mini_racer 0.22.2. Benchmarks ran on Ruby 3.4.2 and Linux with an AMD Ryzen 9 7950X3D. The governor was `performance`. Both gems were pinned to CPUs 0 through 7.
232
232
 
233
- The per-call and memory figures below come from the first run of that extension on the same machine, except the production bundle row, which comes from the second run. One watchdog thread serves the whole process, making a call with `timeout:` fast. Scripts of 1 MiB or more use a process-wide V8 code cache. Contexts loading the same large script share one source copy.
233
+ The per-call and memory figures below come from the first run of that extension on the same machine, except the production bundle row, which comes from the second run. One watchdog thread serves the whole process, making a call with `timeout:` fast. Scripts of 128 KiB or more use a process-wide V8 code cache. Contexts loading the same script of 1 MiB or more share one source copy.
234
234
 
235
235
  | Measure | mini_racer 0.22.2 | mini_racer_rs |
236
236
  |---|---:|---:|
@@ -251,6 +251,7 @@ impl ContextHandle {
251
251
  process::ensure_initialized();
252
252
  // Read here, on the caller's thread, not first on a runner's.
253
253
  let _ = process::code_cache::enabled();
254
+ let _ = process::code_cache::first_override();
254
255
  let _ = process::fast_calls();
255
256
  let generation = fork::current();
256
257
  let block = Arc::new(ControlBlock::new(config));