This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
[patch] testsuite: Fix inferior-died.exp racy FAILs [Re: RFC: fix crash when inferior exits during "continue"]
On Wed, 15 Feb 2012 21:13:25 +0100, Tom Tromey wrote:
> 2012-02-15 Tom Tromey <tromey@redhat.com>
>
> * gdb.base/inferior-died.c: New file.
> * gdb.base/inferior-died.exp: New file.
It has racy FAILs:
continue
Continuing.
-[Inferior 2 (process 17329) exited normally]
-(gdb) PASS: gdb.base/inferior-died.exp: continue
-testcase ./gdb.base/inferior-died.exp completed in 2 seconds
+[Inferior 2 (process 17248) exited normally]
+(gdb) [Inferior 1 (process 17243) exited with code 0140]
+FAIL: gdb.base/inferior-died.exp: continue (timeout)
+testcase ./gdb.base/inferior-died.exp completed in 11 seconds
I will check in this fix in some days.
Thanks,
Jan
gdb/testsuite/
2012-02-19 Jan Kratochvil <jan.kratochvil@redhat.com>
Fix racy FAILs.
* gdb.base/inferior-died.c (main): Add return of 0.
* gdb.base/inferior-died.exp (continue): Fix expectation of
asynchronous events.
(p 1): New test.
--- a/gdb/testsuite/gdb.base/inferior-died.c
+++ b/gdb/testsuite/gdb.base/inferior-died.c
@@ -33,4 +33,5 @@ int main()
function ();
else
waitpid (child, NULL, 0);
+ return 0;
}
--- a/gdb/testsuite/gdb.base/inferior-died.exp
+++ b/gdb/testsuite/gdb.base/inferior-died.exp
@@ -53,4 +53,19 @@ gdb_breakpoint $srcfile:$line
gdb_continue_to_breakpoint "breakpoint"
gdb_test "inferior 2" "Switching to inferior 2.*"
-gdb_test "continue" "exited normally.*"
+
+# The inferior 1 exit may come unexpectedly in any moment.
+set test "continue"
+set seen 0
+gdb_test_multiple $test $test {
+ -re "($gdb_prompt |\\\[Inferior \[^\r\n\]* exited normally\\\])" {
+ incr seen
+ if {$seen < 3} {
+ exp_continue
+ }
+ pass $test
+ }
+}
+
+# Internal error may show up after all the messages above.
+gdb_test "p 1" " = 1"