[Bug dap/30680] [gdb/dap] ThreadSanitizer: data race gdbsupport/event-pipe.h:44 in event_pipe::is_open() const

["vries at gcc dot gnu.org" <sourceware-bugzilla@sourceware.org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=30680

--- Comment #6 from Tom de Vries <vries at gcc dot gnu.org> ---
(In reply to Tom Tromey from comment #4)
> >     #2 sigchld_handler /data/vries/gdb/src/gdb/linux-nat.c:4238 (gdb+0x92f381)
> 
> This is odd because all the DAP-related threads should be started
> with SIGCHLD blocked.

I've written an assert to try to catch this situation without tsan:
...
diff --git a/gdb/linux-nat.c b/gdb/linux-nat.c
index 250a8f43282..d6d0ba0f415 100644
--- a/gdb/linux-nat.c
+++ b/gdb/linux-nat.c
@@ -69,6 +69,7 @@
 #include "gdbsupport/scope-exit.h"
 #include "gdbsupport/gdb-sigmask.h"
 #include "gdbsupport/common-debug.h"
+#include "run-on-main-thread.h"
 #include <unordered_map>

 /* This comment documents high-level logic of this file.
@@ -4227,6 +4228,7 @@ linux_nat_target::supports_disable_randomization ()
 static void
 sigchld_handler (int signo)
 {
+  gdb_assert (is_main_thread ());
   int old_errno = errno;

   if (debug_linux_nat)
diff --git a/gdb/run-on-main-thread.c b/gdb/run-on-main-thread.c
index c7d9de0afc8..f9388e33633 100644
--- a/gdb/run-on-main-thread.c
+++ b/gdb/run-on-main-thread.c
@@ -94,12 +94,27 @@ run_on_main_thread (std::function<void ()> &&func)
   serial_event_set (runnable_event);
 }

+#if CXX_STD_THREAD
+static void
+initialize_main_thread_var (void)
+{
+  static bool initialized = false;
+
+  if (initialized)
+    return;
+  initialized = true;
+
+  main_thread = std::this_thread::get_id ();
+}
+#endif
+
 /* See run-on-main-thread.h.  */

 bool
 is_main_thread ()
 {
 #if CXX_STD_THREAD
+  initialize_main_thread_var ();
   return std::this_thread::get_id () == main_thread;
 #else--
   return true;
@@ -111,7 +126,7 @@ void
 _initialize_run_on_main_thread ()
 {
 #if CXX_STD_THREAD
-  main_thread = std::this_thread::get_id ();
+  initialize_main_thread_var ();
 #endif
   runnable_event = make_serial_event ();
   add_file_handler (serial_event_fd (runnable_event), run_events, nullptr,
...
and it does trigger.

-- 
You are receiving this mail because:
You are on the CC list for the bug.