Fix multiple dumps from being generated (#80708)

Issue: #78956

After a core dump is generated because of a unhandled managed exception
abort() is called but a SIGSEGV is generated in libpthread.so which is
caught by the runtime and a second core dump is generated. The fix is
to uninstall/uninitialize all the signal handlers, not just SIGABORT.

Co-authored-by: Mike McLaughlin <[email protected]>

Author: github-actions[bot]

src/coreclr/pal/src/exception/signal.cpp

@@ -294,22 +294,6 @@ void SEHCleanupSignals()
     }
 }
 
-/*++
-Function :
-    SEHCleanupAbort()
-
-    Restore default SIGABORT signal handlers
-
-    (no parameters, no return value)
---*/
-void SEHCleanupAbort()
-{
-    if (g_registered_signal_handlers)
-    {
-        restore_signal(SIGABRT, &g_previous_sigabrt);
-    }
-}
-
 /* internal function definitions **********************************************/
 
 /*++

src/coreclr/pal/src/include/pal/signal.hpp

@@ -117,14 +117,4 @@ Function :
 --*/
 void SEHCleanupSignals();
 
-/*++
-Function :
-    SEHCleanupAbort()
-
-    Restore default SIGABORT signal handlers
-
-    (no parameters, no return value)
---*/
-void SEHCleanupAbort();
-
 #endif /* _PAL_SIGNAL_HPP_ */

src/coreclr/pal/src/thread/process.cpp

@@ -2677,8 +2677,9 @@ PROCAbort(int signal)
 
     PROCCreateCrashDumpIfEnabled(signal);
 
-    // Restore the SIGABORT handler to prevent recursion
-    SEHCleanupAbort();
+    // Restore all signals; the SIGABORT handler to prevent recursion and
+    // the others to prevent multiple core dumps from being generated.
+    SEHCleanupSignals();
 
     // Abort the process after waiting for the core dump to complete
     abort();