jolt/run-tests.janet
Dmitri Sotnikov f0293fb4ee
Parallelize the test gate; cache cold-init tests, drop the benchmark from it (#139)
run-tests.janet runs the same file set as `jpm test` across a pool of worker
processes (one `janet FILE` each, ev-based). The full gate goes from ~790s
serial to ~98s here (8x), and more on CI where the heavy files don't thrash on
swap. CI and the docs point at it; `jpm test` still works serially.

Three things dominated the wall:

- Nine integration tests cold-built a compile ctx (~8s each); switch them to
  api/init-cached so they share the prebuilt image. The cache key already
  fingerprints the ctx-shaping env vars, so the direct-link ones share one DL
  image and the rest share the plain one.
- core-bench's main ran on every gate (~35s of benchmark loops that assert
  nothing); gate it behind JOLT_BENCH=1.
- cli-test spawned `janet src/jolt/main.janet` ~20 times at ~8s cold each
  (340s under parallel load, and it was the whole wall); prefer build/jolt
  (~20ms baked ctx) when present, fall back to from-source for an unbuilt tree.

type-check-test stays on cold init: a snapshot-loaded ctx loses the success
checker's op/msg detail (jolt-vley). jolt-pria tracks caching from-source
startup generally, which would let cli-test drop the build/jolt preference.

Co-authored-by: Yogthos <yogthos@gmail.com>
2026-06-16 14:23:02 +00:00

100 lines
3.6 KiB
Text

#!/usr/bin/env janet
# Parallel test runner — the gate, faster.
#
# `jpm test` runs every file under test/ serially; this runs the same set across
# a pool of worker processes (each test file is an independent `janet FILE`).
# Build the executable FIRST (`jpm build`) so the binary-spawning tests don't
# skip: jpm build && janet run-tests.janet
#
# Worker count is the detected CPU count; override with JOLT_TEST_JOBS. Per-file
# stdout+stderr is captured to a temp file and only printed for failures (and,
# with -v / JOLT_TEST_VERBOSE, for everything). Exits non-zero if any file fails
# and prints the literal "All tests passed." on success (CI greps for it).
(defn- detect-jobs []
(or (when-let [j (os/getenv "JOLT_TEST_JOBS")] (scan-number j))
(try
(let [p (os/spawn ["sh" "-c" "nproc 2>/dev/null || sysctl -n hw.ncpu 2>/dev/null"]
:p {:out :pipe})
out (:read (p :out) :all)]
(os/proc-wait p)
(scan-number (string/trim (or out ""))))
([_] nil))
4))
(defn- find-tests [dir acc]
(each name (sorted (os/dir dir))
(def full (string dir "/" name))
(case (os/stat full :mode)
# support/ holds shared libraries (harness, shims), not standalone tests —
# jpm doesn't run them and neither do we.
:directory (unless (= name "support") (find-tests full acc))
:file (when (string/has-suffix? ".janet" name) (array/push acc full))))
acc)
(def verbose? (or (has-value? (dyn :args) "-v") (os/getenv "JOLT_TEST_VERBOSE")))
(def janet-bin (dyn :executable))
(def tmpdir (string (or (os/getenv "TMPDIR") "/tmp") "/jolt-test-" (os/getpid)))
(os/mkdir tmpdir)
(def files (find-tests "test" @[]))
(def total (length files))
(def jobs (max 1 (min 16 (detect-jobs))))
(def results @[])
(var done 0)
(defn- run-one [file]
(def t0 (os/clock))
(def of (string tmpdir "/" (string/replace-all "/" "_" file) ".out"))
# let the shell own redirection — avoids pipe-buffer deadlock on chatty tests.
(def cmd (string/format "exec '%s' '%s' > '%s' 2>&1" janet-bin file of))
(def rc (os/proc-wait (os/spawn ["sh" "-c" cmd] :p)))
(def dt (- (os/clock) t0))
(array/push results {:file file :rc rc :out of :dt dt})
(++ done)
(printf "[%d/%d] %s %6.2fs %s%s" done total (if (zero? rc) "ok " "FAIL")
dt file (if (zero? rc) "" (string/format " (rc=%d)" rc)))
(flush))
# A :stop sentinel per worker signals end-of-queue — ev/chan-close discards
# buffered items (Janet 1.41), so we can't just close and drain.
(def q (ev/chan (+ total jobs)))
(each f files (ev/give q f))
(loop [_ :range [0 jobs]] (ev/give q :stop))
(defn- worker [&]
(forever
(def f (ev/take q))
(if (= f :stop) (break) (run-one f))))
(printf "running %d test files across %d workers...\n" total jobs)
(flush)
(def wall0 (os/clock))
(def super (ev/chan))
(loop [_ :range [0 jobs]] (ev/go worker nil super))
(loop [_ :range [0 jobs]] (ev/take super))
(def wall (- (os/clock) wall0))
(def failed (filter |(not (zero? ($ :rc))) results))
(def slow (->> results (sorted-by |(- ($ :dt))) (take 5)))
(print "\n--- slowest ---")
(each r slow (printf " %6.2fs %s" (r :dt) (r :file)))
(when verbose?
(each r results
(printf "\n===== %s (rc=%d, %.2fs) =====" (r :file) (r :rc) (r :dt))
(print (slurp (r :out)))))
(unless (empty? failed)
(print "\n--- FAILURES ---")
(each r failed
(printf "\n===== %s (rc=%d) =====" (r :file) (r :rc))
(print (slurp (r :out)))))
(printf "\n%d files, %d failed, %.1fs wall (%d workers)"
total (length failed) wall jobs)
(if (empty? failed)
(do (print "\nAll tests passed.") (os/exit 0))
(do (printf "\n%d test files FAILED." (length failed)) (os/exit 1)))