From: Stephen Bertram via ltp <ltp@lists.linux.it>
To: ltp@lists.linux.it
Cc: Stephen Bertram <sbertram@redhat.com>
Subject: [LTP] [PATCH v2] cachestat02: Harden parallel worker execution.
Date: Thu, 19 Feb 2026 08:50:01 -0500 [thread overview]
Message-ID: <20260219135019.1140234-1-sbertram@redhat.com> (raw)
This does not happen all the time but when running in parallel
sometimes one of the test instances will throw an error:
TFAIL: cs->nr_cache + cs->nr_evicted (297) != num_pages (512)
When LTP runs several instances of the cachestat tests in parallel, they share resources:
cachestat02: Uses a fixed shm name "myfile.bin". All instances use the same POSIX shared memory object.
That leads to:
Races on creation/removal of the same file/shm.
cachestat() reporting stats for a shared or partially overwritten file instead of the one the current process wrote.
nr_cache + nr_evicted not matching num_pages because the measured range is not the one written by this process.
The change is to create a unique file/shm space for each instance using the pid of each cachestat() process.
Harden parallel worker execution for cachestat02.
Signed-off-by: Stephen Bertram <sbertram@redhat.com>
---
testcases/kernel/syscalls/cachestat/cachestat02.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/testcases/kernel/syscalls/cachestat/cachestat02.c b/testcases/kernel/syscalls/cachestat/cachestat02.c
index 72428ee83..d6505c4a4 100644
--- a/testcases/kernel/syscalls/cachestat/cachestat02.c
+++ b/testcases/kernel/syscalls/cachestat/cachestat02.c
@@ -17,10 +17,9 @@
#include <stdlib.h>
#include "cachestat.h"
-#define FILENAME "myfile.bin"
-
static int page_size;
static char *page_data;
+static char shm_name[64];
static struct cachestat *cs;
static struct cachestat_range *cs_range;
@@ -32,7 +31,8 @@ static void test_cached_pages(const int num_pages)
memset(cs, 0, sizeof(struct cachestat));
- fd = shm_open(FILENAME, O_RDWR | O_CREAT, 0600);
+ snprintf(shm_name, sizeof(shm_name), "/cachestat_%d.bin", getpid());
+ fd = shm_open(shm_name, O_RDWR | O_CREAT | O_EXCL, 0600);
if (fd < 0)
tst_brk(TBROK | TERRNO, "shm_open error");
@@ -53,7 +53,7 @@ static void test_cached_pages(const int num_pages)
TST_EXP_EQ_LI(cs->nr_cache + cs->nr_evicted, num_pages);
SAFE_CLOSE(fd);
- shm_unlink(FILENAME);
+ shm_unlink(shm_name);
}
static void run(void)
--
2.52.0
--
Mailing list info: https://lists.linux.it/listinfo/ltp
next reply other threads:[~2026-02-19 13:50 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-19 13:50 Stephen Bertram via ltp [this message]
2026-02-20 10:55 ` [LTP] [PATCH v2] cachestat02: Harden parallel worker execution Cyril Hrubis
2026-02-24 8:44 ` Andrea Cervesato via ltp
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=20260219135019.1140234-1-sbertram@redhat.com \
--to=ltp@lists.linux.it \
--cc=sbertram@redhat.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