[ASan][test-only] Allow Darwin test duplicate_os_log_reports to retry

[DanBlackwell]

Currently this test is flaky. I believe that it may be because the logs have not flushed to disk before the log show command is run.

This patch allows the test to retry reading the log twice more with a 5 second sleep inbetween.

[llvmbot]

@llvm/pr-subscribers-compiler-rt-sanitizer

Author: Dan Blackwell (DanBlackwell)

Changes

Currently this test is flaky. I believe that it may be because the logs have not flushed to disk before the log show command is run.

This patch allows the test to retry reading the log twice more with a 5 second sleep inbetween.

---

Full diff: https://github.com/llvm/llvm-project/pull/158662.diff

1 Files Affected:

• (modified) compiler-rt/test/asan/TestCases/Darwin/duplicate_os_log_reports.cpp (+7-2)

diff --git a/compiler-rt/test/asan/TestCases/Darwin/duplicate_os_log_reports.cpp b/compiler-rt/test/asan/TestCases/Darwin/duplicate_os_log_reports.cpp
index 0091ebc09205c..5a0353bfb1b31 100644
--- a/compiler-rt/test/asan/TestCases/Darwin/duplicate_os_log_reports.cpp
+++ b/compiler-rt/test/asan/TestCases/Darwin/duplicate_os_log_reports.cpp
@@ -8,8 +8,13 @@
 // RUN: FileCheck %s --check-prefixes CHECK,CHECK-PROC -input-file=%t.process_output.txt
 
 // Check syslog output. We filter recent system logs based on PID to avoid
-// getting the logs of previous test runs.
-// RUN: log show --debug --last 5m  --predicate "processID == ${TEST_PID}" --style syslog > %t.process_syslog_output.txt
+// getting the logs of previous test runs. Make some reattempts in case there
+// is a delay.
+// RUN: for I in {1..3}; do \
+// RUN:   log show --debug --last $((SECONDS + 30))s --predicate "processID == ${TEST_PID}" --style syslog > %t.process_syslog_output.txt; \
+// RUN:   if grep -q "use-after-poison" %t.process_syslog_output.txt; then break; fi; \
+// RUN:   sleep 5; \
+// RUN: done
 // RUN: FileCheck %s -input-file=%t.process_syslog_output.txt
 #include <cassert>
 #include <cstdio>

[wrotki]

Looks good. We also have this deflake script you could use, but I think it won't allow steps between reattempts.