Apparently this way of triggering the msdosfs rename vnode leak

does not bite every time (most commonly observed on the amd64/qemu
runs), so add a race condition catcher.
This commit is contained in:
pooka 2011-03-14 19:05:19 +00:00
parent 4e499ee605
commit bc1269c4f5
1 changed files with 12 additions and 2 deletions

View File

@ -1,4 +1,4 @@
/* $NetBSD: t_renamerace.c,v 1.21 2011/03/06 16:00:16 pooka Exp $ */
/* $NetBSD: t_renamerace.c,v 1.22 2011/03/14 19:05:19 pooka Exp $ */
/*
* Modified for rump and atf from a program supplied
@ -113,8 +113,18 @@ renamerace(const atf_tc_t *tc, const char *mp)
if (FSTYPE_LFS(tc))
abort();
if (FSTYPE_MSDOS(tc))
if (FSTYPE_MSDOS(tc)) {
atf_tc_expect_fail("PR kern/44661");
/*
* XXX: race does not trigger every time at least
* on amd64/qemu.
*/
if (msdosfs_fstest_unmount(tc, mp, 0) != 0) {
rump_pub_vfs_mount_print(mp, 1);
atf_tc_fail_errno("unmount failed");
}
atf_tc_fail("race did not trigger this time");
}
}
static void