[LTP] [PATCH] mem/tunable:min_free_kbytes: fix disabled check_monitor flag

[LTP] [PATCH] mem/tunable:min_free_kbytes: fix disabled check_monitor flag

[Gao Xiang]

From: Xiang Gao <gaoxiang@kylinos.cn>

In min_free_kbytes, check_monitor() is effectively disabled because
the loop condition uses the flag incorrectly.

Fix the condition so the monitor runs during the test and exits after
receiving the signal.

Signed-off-by: Xiang Gao <gaoxiang@kylinos.cn>
---
 testcases/kernel/mem/tunable/min_free_kbytes.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/testcases/kernel/mem/tunable/min_free_kbytes.c b/testcases/kernel/mem/tunable/min_free_kbytes.c
index ae0fe9d90..a62e4ae9d 100644
--- a/testcases/kernel/mem/tunable/min_free_kbytes.c
+++ b/testcases/kernel/mem/tunable/min_free_kbytes.c
@@ -184,7 +184,7 @@ static void check_monitor(void)
 	unsigned long tune;
 	unsigned long memfree;
 
-	while (end) {
+	while (!end) {
 		memfree = SAFE_READ_MEMINFO("MemFree:");
 		tune = TST_SYS_CONF_LONG_GET(MIN_FREE_KBYTES);
 
-- 
2.53.0


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp

Re: [LTP] [PATCH] mem/tunable:min_free_kbytes: fix disabled check_monitor flag

[Andrea Cervesato via ltp]

Hi!

This patch is correct. Indeed, `end` is set to 1 only when signal is received,
otherwise we loop into check_monitor(). Interesting enough we did use this test
for a while now. The monitor that was checking for possible failures was never
used.

We can add before merge:

Fixes: f28b29f16ba7

@Li FYI

--
Andrea Cervesato
SUSE QE Automation Engineer Linux
andrea.cervesato@suse.com

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp

Re: [LTP] [PATCH] mem/tunable:min_free_kbytes: fix disabled check_monitor flag

[Andrea Cervesato via ltp]

Reviewed-by: Andrea Cervesato <andrea.cervesato@suse.com>

--
Andrea Cervesato
SUSE QE Automation Engineer Linux
andrea.cervesato@suse.com

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp

Re: [LTP] [PATCH] mem/tunable:min_free_kbytes: fix disabled check_monitor flag

[Petr Vorel]

Hi,

> Hi!

> This patch is correct. Indeed, `end` is set to 1 only when signal is received,
> otherwise we loop into check_monitor(). Interesting enough we did use this test
> for a while now. The monitor that was checking for possible failures was never
> used.

Good catch, thank you! Patch merged.

> We can add before merge:

> Fixes: f28b29f16ba7
I added 8b7cc630a5 ("mm/tunable: add a new testcase min_free_kbytes")
it was wrong from the start when the test was added in 2012.

> @Li FYI
FYI Li was not Cced, doing it now.

Kind regards,
Petr

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp