From 7644e8bae18a683e4a93c22d27787929e8cfe74a Mon Sep 17 00:00:00 2001
From: Martin Vogel <martin.vogel.tech@gmail.com>
Date: Fri, 2 Oct 2026 21:19:51 +0200
Subject: [PATCH] ci: upload failing suite logs when a test job fails

The job log of a red test job carries a summary of each failing suite,
not its log, and the only artifact was shard-manifest.txt. A failure
whose summary was not enough could only be understood by reproducing it.

The four jobs in .github/workflows/_test.yml that run the parallel
harness (test-unix, test-windows, test-diag, test-lsan-macos) get one
more step, "Failing suite logs". It runs only when the job failed and
uploads build/c/test-logs/failed/: the logs of the suites that did not
end green, plus results.txt, as the harness leaves them there. One red
suite is typically a few kilobytes; a whole run's logs are about 25 MB.
Retention is 7 days. The artifact names follow the shard-manifest ones
(suite-logs-<os>-<compiler or msystem>-<job index>, suite-logs-diag,
suite-logs-lsan-macos), and the step uses the upload-artifact reference
the file already pins.

A job that failed outside the suite waves has no such directory; the
step then warns and uploads nothing. The step cannot change a job's
result: it is skipped on a green job, and a failed job stays failed
whatever the upload does. Triggers, permissions, job dependencies,
timeouts, the matrix and the shard-completeness download pattern are
unchanged.

Signed-off-by: Martin Vogel <martin.vogel.tech@gmail.com>
---
 .github/workflows/_test.yml | 50 +++++++++++++++++++++++++++++++++++++
 1 file changed, 50 insertions(+)

diff --git a/.github/workflows/_test.yml b/.github/workflows/_test.yml
index 88ed9498a..0a1f0c488 100644
--- a/.github/workflows/_test.yml
+++ b/.github/workflows/_test.yml
@@ -256,6 +256,23 @@ jobs:
           # shard-completeness job still gates the cross-shard union.
           if-no-files-found: warn
 
+      # The job log carries a red suite's SUMMARY; these are the complete logs
+      # behind it, so a failure can be read instead of re-run. The harness
+      # copies the log of every suite that did not end green, plus results.txt,
+      # into test-logs/failed/ (scripts/suite-failure-report.sh) — typically a
+      # few kilobytes for a red suite, not the ~25 MB of a whole run's logs.
+      # Runs only on a failed job. warn, not error: a job that failed OUTSIDE
+      # the suite wave (contract step, build, prod-binary guards) has no such
+      # directory, and a second error would only bury the first.
+      - name: Failing suite logs
+        if: failure()
+        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
+        with:
+          name: suite-logs-${{ matrix.os }}-${{ matrix.cc }}-${{ strategy.job-index }}
+          path: build/c/test-logs/failed/
+          retention-days: 7
+          if-no-files-found: warn
+
       - name: Compiler cache (save)
         if: always() && (matrix.shard == null || startsWith(matrix.shard, '1/'))
         uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
@@ -375,6 +392,17 @@ jobs:
           scripts/test.sh CC=clang-21 CXX=clang++-21 \
             SANITIZE="-fsanitize=address,undefined -fno-omit-frame-pointer -fno-optimize-sibling-calls"
 
+      # Same step as in test-unix: the complete logs of the suites that did
+      # not end green, only when the job failed.
+      - name: Failing suite logs
+        if: failure()
+        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
+        with:
+          name: suite-logs-diag
+          path: build/c/test-logs/failed/
+          retention-days: 7
+          if-no-files-found: warn
+
       - name: Compiler cache (save)
         if: always()
         uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
@@ -462,6 +490,17 @@ jobs:
           LLVM_PREFIX="$(brew --prefix llvm@22)"
           scripts/test.sh CC="$LLVM_PREFIX/bin/clang" CXX="$LLVM_PREFIX/bin/clang++"
 
+      # Same step as in test-unix: the complete logs of the suites that did
+      # not end green, only when the job failed.
+      - name: Failing suite logs
+        if: failure()
+        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
+        with:
+          name: suite-logs-lsan-macos
+          path: build/c/test-logs/failed/
+          retention-days: 7
+          if-no-files-found: warn
+
       - name: Compiler cache (save)
         if: always()
         uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
@@ -636,6 +675,17 @@ jobs:
           # shard-completeness job still gates the cross-shard union.
           if-no-files-found: warn
 
+      # Same step as in test-unix: the complete logs of the suites that did
+      # not end green, only when the job failed.
+      - name: Failing suite logs
+        if: failure()
+        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7.0.1
+        with:
+          name: suite-logs-${{ matrix.os }}-${{ matrix.msystem }}-${{ strategy.job-index }}
+          path: build/c/test-logs/failed/
+          retention-days: 7
+          if-no-files-found: warn
+
       # Shard 1 (or the unsharded job) saves the compiler cache; every shard
       # builds the identical runner, so one cache carries the full object set.
       - name: Compiler cache (save)