test: slightly reduce the performance/memory overhead for wrapped binaries

Let's drop the quarantine that ASan uses for use-after-free detection,
as it's pointless in wrapped binaries and can consume up to 256 MiB of
memory (with the default configuration). Also, don't keep any stack
traces for allocations & deallocations, which should (slightly) help
with both memory & performance overhead.

(cherry picked from commit 035ba3ea571bad6772cf3731f6b5379ccb08267f)

diff --git a/mkosi/mkosi.sanitizers/mkosi.postinst b/mkosi/mkosi.sanitizers/mkosi.postinst
index 01c5b23fa25055e5c0d319f669a26ea63a3f5b54..420fc83adddd43e1879bb55efc6c849aa2e5b46b 100755 (executable)
--- a/mkosi/mkosi.sanitizers/mkosi.postinst
+++ b/mkosi/mkosi.sanitizers/mkosi.postinst
@@ -137,8 +137,11 @@ for bin in "${wrap[@]}"; do
 # Preload the ASan runtime DSO, otherwise ASan will complain
 export LD_PRELOAD="$ASAN_RT_PATH"
 # Disable LSan to speed things up, since we don't care about leak reports
-# from 'external' binaries
-export ASAN_OPTIONS=detect_leaks=$enable_lsan
+# from 'external' binaries. In addition, disable quarantine (for use-after-free
+# detection) and malloc stack frame collection as we don't care about these in
+# 'external' binaries either, and they just unnecessarily hog up memory & cpu
+# cycles.
+export ASAN_OPTIONS=detect_leaks=$enable_lsan:quarantine_size_mb=0:malloc_context_size=0
 # Set argv[0] to the original binary name without the ".orig" suffix
 exec -a "\$0" -- "${target}.orig" "\$@"
 EOF