From 2bd9a072a9d43d2569104e0c5a0dad899711f1d6 Mon Sep 17 00:00:00 2001
From: Dennis Klein <d.klein@gsi.de>
Date: Wed, 10 Jun 2026 16:14:15 +0200
Subject: [PATCH] test: pre-fill libstdc++ ctype caches before threads exist

- std::ctype<char> caches narrow()/widen() results per character in
  plain char arrays of the global classic-locale facet, written without
  synchronization from header-inlined code (locale_facets.h); two
  threads exercising an uncached character concurrently (e.g. compiling
  a std::regex in Channel::Validate) constitute a true data race that
  ThreadSanitizer rightfully reports
- the stores are real and unsynchronized, so a tsan-instrumented
  libstdc++ cannot help here; instead fill the caches before any thread
  is spawned, which turns every later access into a pure read
- warm the lazily-installed num_put/num_get caches used by stream
  insertion/extraction as well, via a small format/parse round-trip
- wire the warm-up into the gtest runner main() and, via a static
  initializer, into the test device runner
---
 test/CMakeLists.txt           |  1 +
 test/helper/LocaleWarmup.h    | 63 +++++++++++++++++++++++++++++++++++
 test/helper/runTestDevice.cxx |  5 +++
 test/runner.cxx.in            |  2 ++
 4 files changed, 71 insertions(+)
 create mode 100644 test/helper/LocaleWarmup.h

diff --git a/test/CMakeLists.txt b/test/CMakeLists.txt
index 52721c3b..779e826d 100644
--- a/test/CMakeLists.txt
+++ b/test/CMakeLists.txt
@@ -32,6 +32,7 @@ endif()
 add_testhelper(runTestDevice
     SOURCES
     helper/runTestDevice.cxx
+    helper/LocaleWarmup.h
     helper/devices/TestPairLeft.h
     helper/devices/TestPairRight.h
     helper/devices/TestPollIn.h
diff --git a/test/helper/LocaleWarmup.h b/test/helper/LocaleWarmup.h
new file mode 100644
index 00000000..d73dfc42
--- /dev/null
+++ b/test/helper/LocaleWarmup.h
@@ -0,0 +1,63 @@
+/********************************************************************************
+ *    Copyright (C) 2026 GSI Helmholtzzentrum fuer Schwerionenforschung GmbH    *
+ *                                                                              *
+ *              This software is distributed under the terms of the             *
+ *              GNU Lesser General Public Licence (LGPL) version 3,             *
+ *                  copied verbatim in the file "LICENSE"                       *
+ ********************************************************************************/
+
+#ifndef FAIR_MQ_TEST_LOCALEWARMUP_H
+#define FAIR_MQ_TEST_LOCALEWARMUP_H
+
+#include <locale>
+#include <sstream>
+
+namespace fair::mq::test {
+
+/// libstdc++'s std::ctype<char> caches narrow()/widen() results per character
+/// in plain char arrays of the global classic-locale facet -- by design with
+/// unsynchronized writes, emitted from header-inlined code (locale_facets.h).
+/// Whenever two threads exercise an uncached character concurrently (e.g. by
+/// compiling a std::regex), ThreadSanitizer rightfully reports the write as a
+/// data race. The stores are real and unsynchronized, so instrumenting
+/// libstdc++ cannot help; instead, fill the caches before any thread exists,
+/// which turns every later access into a pure read.
+inline void WarmUpLocaleCaches()
+{
+    auto const& ct = std::use_facet<std::ctype<char>>(std::locale::classic());
+    for (int c = 1; c < 256; ++c) {
+        auto const ch = static_cast<char>(c);
+        ct.narrow(ch, '\0');
+        ct.widen(ch);
+    }
+
+    // Only the range overload runs _M_narrow_init(), which sets the
+    // all-cached flag (_M_narrow_ok); the single-char loop above fills the
+    // cache without it. (widen() needs no equivalent: the single-char
+    // overload already runs _M_widen_init().)
+    char from[256];
+    char to[256];
+    for (int c = 0; c < 256; ++c) {
+        from[c] = static_cast<char>(c);
+    }
+    ct.narrow(from, from + 256, '?', to);
+
+    // num_put/num_get caches used by stream insertion/extraction
+    std::ostringstream os;
+    os << 4711 << ' ' << 3.14;
+    std::istringstream is("42 3.14");
+    int i = 0;
+    double d = 0.;
+    is >> i >> d;
+}
+
+/// For binaries that do not own main(): define a namespace-scope instance to
+/// run the warm-up during static initialization, before main().
+struct LocaleWarmup
+{
+    LocaleWarmup() { WarmUpLocaleCaches(); }
+};
+
+}   // namespace fair::mq::test
+
+#endif /* FAIR_MQ_TEST_LOCALEWARMUP_H */
diff --git a/test/helper/runTestDevice.cxx b/test/helper/runTestDevice.cxx
index e94300b2..e757938a 100644
--- a/test/helper/runTestDevice.cxx
+++ b/test/helper/runTestDevice.cxx
@@ -6,6 +6,7 @@
  *                  copied verbatim in the file "LICENSE"                       *
  ********************************************************************************/
 
+#include "LocaleWarmup.h"
 #include "devices/TestPairLeft.h"
 #include "devices/TestPairRight.h"
 #include "devices/TestPollIn.h"
@@ -30,6 +31,10 @@
 
 namespace bpo = boost::program_options;
 
+namespace {
+[[maybe_unused]] fair::mq::test::LocaleWarmup const gLocaleWarmup{};
+}
+
 auto addCustomOptions(bpo::options_description& options) -> void
 {
     options.add_options()
diff --git a/test/runner.cxx.in b/test/runner.cxx.in
index 4d426187..2b67a93a 100644
--- a/test/runner.cxx.in
+++ b/test/runner.cxx.in
@@ -10,6 +10,7 @@
 #include <TestEnvironment.h>
 
 #include <gtest/gtest.h>
+#include <helper/LocaleWarmup.h>
 
 #include <string>
 
@@ -24,6 +25,7 @@ string runTestDevice = "@RUN_TEST_DEVICE@";
 
 int main(int argc, char** argv)
 {
+    fair::mq::test::WarmUpLocaleCaches();
     ::testing::InitGoogleTest(&argc, argv);
     ::testing::FLAGS_gtest_death_test_style = "threadsafe";
     setenv("FAIRMQ_PATH", FAIRMQ_TEST_ENVIRONMENT, 0);