Repository navigation
perf(ci): acelerar el carril lento con cache de libreria nativa, tests grada C y release en paralelo - #582
Merged
Conversation
…s grada C y release en paralelo - Cachea librfirma_crypto.so por hash del puente Java y bootstrap.sh para ahorrar los ~2m de native-image en PRs que no tocan Java - Cambia --include-ignored por --ignored en test-native para ejecutar unicamente las pruebas de grada C sin duplicar unitarios ni conformance - Mueve la compilacion del binario de release a un job paralelo independiente release
sgomez
force-pushed
the
perf/ci-accelerate-native-lane
branch
from
September 8, 2026 21:40
eefe981 to
d01dcf8
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Qué resuelve este PR
El carril lento de CI (
native) tardaba más de 10 minutos (entre 11m 45s y 12m 56s medidos en ejecuciones recientes). Este PR aplica tres optimizaciones clave para reducir su tiempo de pared a ~3–4 minutos:Caché de
librfirma_crypto.soen CI:actions/cachepararfirma-native-bridge/target/lib/rfirmacon clave hash debootstrap.shyrfirma-native-bridge/**.just native, ahorrando los ~2m 15s denative-image.Poda de ejecuciones duplicadas en
test-native:-- --include-ignoredpor-- --ignoreden la recetatest-nativedejustfile.test-nativeahora ejecuta estrictamente las pruebas de grada C (marcadas con#[ignore]: el ciclo completo denative_cycle.rs, los casos del protocolo conRFIRMA_LIB_DIRy las de Maven con-DexcludedGroups=).check-rusty que volverán a correr bajollvm-covencrap-full.Paralelización de la compilación de release:
just build-rust(cargo build --release --features custom-protocol) del final del jobnativea un nuevo job paralelorelease(«Binario de release»).libloadingdinámico en tiempo de ejecución, Rust no enlaza contralibrfirma_crypto.soen compilación ni depende de GraalVM, permitiendo que la compilación de release corra en concurrencia y ahorre ~1m de tiempo de pared ennative.Documentación:
docs/agents/code-host-ci.mdreflejando las flags y la nueva división de jobs.Verificaciones
just check-actions: todas las acciones fijadas por SHA y sin invariantes rotas.cargo test --test adr_citations_resolve: citas de ADR íntegras.just check-repo: todos los candados de repositorio y versiones pasando.just fmt: formato limpio.