From: Jan Kara <jack@suse.cz>
To: ltp@lists.linux.it
Cc: Jan Kara <jack@suse.cz>, petr.vorel@suse.com
Subject: [LTP] [PATCH v2] fanotify23: Make evictable marks tests more reliable
Date: Thu, 10 Aug 2023 19:59:26 +0200 [thread overview]
Message-ID: <20230810175926.6205-1-jack@suse.cz> (raw)
It has been observed that when fanotify23 test is run in a loop, it
eventually fails as:
fanotify23.c:112: TPASS: FAN_MARK_ADD failed with EEXIST as expected when trying to downgrade to evictable mark
fanotify23.c:75: TPASS: FAN_MARK_REMOVE failed with ENOENT as expected after empty mask
fanotify23.c:156: TPASS: Got no events as expected
fanotify23.c:81: TFAIL: FAN_MARK_REMOVE did not fail with ENOENT as expected after drop_caches: SUCCESS (0)
This is because repeated evictions of caches done by the test reclaim
all freeable slab objects in the system. So when the test calls drop
caches, only the dentry and inode of the test file are there to reclaim.
But for inode to be reclaimed, dentry (which holds inode reference) has
to be freed first and for dentry to be freed it has to first cycle
through the LRU which takes two slab reclaim calls.
Call drop caches twice to make sure dentry has chance to pass through
the LRU and be reclaimed.
Signed-off-by: Jan Kara <jack@suse.cz>
---
testcases/kernel/syscalls/fanotify/fanotify23.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/testcases/kernel/syscalls/fanotify/fanotify23.c b/testcases/kernel/syscalls/fanotify/fanotify23.c
index 89fd4f36a09b..fb812c51e34e 100644
--- a/testcases/kernel/syscalls/fanotify/fanotify23.c
+++ b/testcases/kernel/syscalls/fanotify/fanotify23.c
@@ -160,10 +160,16 @@ static void test_fanotify(void)
}
/*
- * drop_caches should evict inode from cache and remove evictable mark
+ * drop_caches should evict inode from cache and remove evictable mark.
+ * We call drop_caches twice as once the dentries will just cycle
+ * through the LRU without being reclaimed and if there are no other
+ * objects to reclaim, the slab reclaim will just stop instead of
+ * retrying. Note that this relies on how reclaim of fs objects work
+ * for the filesystem but this test is restricted to ext2...
*/
fsync_file(TEST_FILE);
SAFE_FILE_PRINTF(DROP_CACHES_FILE, "3");
+ SAFE_FILE_PRINTF(DROP_CACHES_FILE, "3");
verify_mark_removed(TEST_FILE, "after drop_caches");
--
2.35.3
--
Mailing list info: https://lists.linux.it/listinfo/ltp
next reply other threads:[~2023-08-10 17:59 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-08-10 17:59 Jan Kara [this message]
2023-08-11 12:01 ` [LTP] [PATCH v2] fanotify23: Make evictable marks tests more reliable Amir Goldstein
2023-08-11 13:43 ` Cyril Hrubis
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230810175926.6205-1-jack@suse.cz \
--to=jack@suse.cz \
--cc=ltp@lists.linux.it \
--cc=petr.vorel@suse.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox