From: Pedro Alves <pedro@palves.net>
Date: Mon, 26 Jun 2023 18:19:04 +0000 (+0100)
Subject: Fix gdb.threads/multiple-successive-infcall.exp on Cygwin (hack)
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=7ef0394da8282d99fcc09871cf3b9d275e1896c0;p=thirdparty%2Fbinutils-gdb.git

Fix gdb.threads/multiple-successive-infcall.exp on Cygwin (hack)

Need to retest, maybe it fixed itself on Windows 11, or something,
meanwhile...

Change-Id: I62becd91b41dc2bede8c35c1b4f792ecbbd048dc
---

diff --git a/gdb/testsuite/gdb.threads/multiple-successive-infcall.exp b/gdb/testsuite/gdb.threads/multiple-successive-infcall.exp
index 2ead754b8f3..1fed87bd94d 100644
--- a/gdb/testsuite/gdb.threads/multiple-successive-infcall.exp
+++ b/gdb/testsuite/gdb.threads/multiple-successive-infcall.exp
@@ -122,3 +122,11 @@ for {set thread 1} {$thread <= $thread_count} {incr thread} {
 }
 
 gdb_assert {$testing_pthread == 4} "tested all pthreads"
+
+# Cygwin very often hangs closing GDB if we don't do this?  But WHY???
+# I don't really know.  I seems like GDB never sees stdin close.  But
+# then, DejaGnu sends kill after some seconds of timeout, and that
+# fails to kill gdb too.  On top of that, if I do "kill -9 GDBPID" on
+# another command line, GDB does die, and the testrun finishes...  For
+# now, force-kill so that the testsuite doesn't hang...
+gdb_test "with confirm off -- kill" ".*"