Skip to contentSkip to conversation

munlockall: add test case that verifies memory has been unlocked

PATCH In review Dennis Brendel 4 participants 8 messages Mar 6, 2024 – Mar 7, 2024 1 message not in the archive

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 munlockall: add test case that verifies memory has been unlocked

Changes compared to v2:
- replacing munlockall01.c as suggested
- respecting line length of 100 chars
- use tst_brk(TBROK | TERRNO, ...) to print symbolic errno value
- remove braces in the final error/success conditional
- use the suggested success message

---
 .../kernel/syscalls/munlockall/munlockall01.c | 148 ++++--------------
 1 file changed, 31 insertions(+), 117 deletions(-)
testcases/kernel/syscalls/munlockall/munlockall01.c +31 -117
index 51f731b65..2745a1aee 100644
@@ -1,134 +1,48 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
 /*
- * Copyright (c) Wipro Technologies Ltd, 2002.  All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
- *
- * You should have received a copy of the GNU General Public License along
- * with this program; if not, write the Free Software Foundation, Inc.,
- * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
- *
+ * Copyright Red Hat
+ * Author: Dennis Brendel <[email protected]>
  */
-/**************************************************************************
- *
- *    TEST IDENTIFIER	: munlockall01
- *
- *    EXECUTED BY	: root / superuser
- *
- *    TEST TITLE	: Basic test for munlockall(2)
- *
- *    TEST CASE TOTAL	: 1
- *
- *    AUTHOR		: sowmya adiga<[email protected]>
- *
- *    SIGNALS
- * 	Uses SIGUSR1 to pause before test if option set.
- * 	(See the parse_opts(3) man page).
- *
- *    DESCRIPTION
- *	This is a phase I test for the munlockall(2) system call.
- *	It is intended to provide a limited exposure of the system call.
- *
- * 	Setup:
- *	  Setup signal handling.
- *	  Pause for SIGUSR1 if option specified.
- *
- * 	Test:
- *        Execute system call
- *	  Check return code, if system call failed (return=-1)
- *	  Log the errno and Issue a FAIL message.
- *	  Otherwise, Issue a PASS message.
- *
- * 	Cleanup:
- * 	  Print errno log and/or timing stats if options given
- *
- * USAGE:  <for command-line>
- *  munlockall01 [-c n] [-e] [-i n] [-I x] [-p x] [-t]
- *		where,		-c n : Run n copies concurrently
- *	               		-e   : Turn on errno logging.
- *				-h   : Show this help screen
- *				-i n : Execute test n times.
- *				-I x : Execute test for x seconds.
- *				-p   : Pause for SIGUSR1 before starting
- *                      	-P x : Pause for x seconds between iterations.
- *                       	 t   : Turn on syscall timing.
- *
- * RESTRICTIONS
- * Must be root/superuser to run it.
- *****************************************************************************/
-#include <errno.h>
-#include <sys/mman.h>
-#include "test.h"
 
-void setup();
-void cleanup();
+/*\
+ * [Description]
+ *
+ * Verify that munlockall(2) unlocks all previously locked memory
+ */
 
-char *TCID = "munlockall01";
-int TST_TOTAL = 1;
+#include <sys/mman.h>
 
-#if !defined(UCLINUX)
+#include "tst_test.h"
 
-int main(int ac, char **av)
+static void verify_munlockall(void)
 {
-	int lc;
-
-	tst_parse_opts(ac, av, NULL, NULL);
+	unsigned long size = 0;
 
-	setup();
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 
-	/* check looping state */
-	for (lc = 0; TEST_LOOPING(lc); lc++) {
+	if (size != 0UL)
+		tst_brk(TBROK, "Locked memory after init should be 0 but is %ld", size);
 
-		tst_count = 0;
+	if (mlockall(MCL_CURRENT | MCL_FUTURE) != 0)
+		tst_brk(TBROK | TERRNO, "Could not lock memory using mlockall()");
 
-		TEST(munlockall());
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 
-		/* check return code */
-		if (TEST_RETURN == -1) {
-			tst_resm(TFAIL | TTERRNO, "munlockall() Failed with"
-				 " return=%ld", TEST_RETURN);
-		} else {
-			tst_resm(TPASS, "munlockall() passed with"
-				 " return=%ld ", TEST_RETURN);
+	if (size == 0UL)
+		tst_brk(TBROK, "Locked memory after mlockall() should be greater than 0, "
+			       "but is %ld", size);
 
-		}
-	}
+	if (munlockall() != 0)
+		tst_brk(TBROK | TERRNO, "Could not unlock memory using munlockall()");
 
-	/* cleanup and exit */
-	cleanup();
-	tst_exit();
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 
+	if (size != 0UL)
+		tst_res(TFAIL, "Locked memory after munlockall() should be 0 but is %ld", size);
+	else
+		tst_res(TPASS, "Memory successfully locked and unlocked");
 }
 
-#else
-
-int main(void)
-{
-	tst_resm(TINFO, "test is not available on uClinux");
-	tst_exit();
-}
-
-#endif /* if !defined(UCLINUX) */
-
-/* setup() - performs all ONE TIME setup for this test. */
-void setup(void)
-{
-	tst_require_root();
-
-	tst_sig(NOFORK, DEF_HANDLER, cleanup);
-
-	TEST_PAUSE;
-}
-
-/*
- * cleanup() - performs all ONE TIME cleanup for this test at
- *		completion or premature exit.
- */
-void cleanup(void)
-{
-}
+static struct tst_test test = {
+	.test_all = verify_munlockall,
+};
Signature · 5 lines
2.44.0


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
Reviewed-by on v3 1/1
#1 lore raw

Hi Dennis,

LGTM, thanks!

I would just explicitly note the test was rewritten from scratch (can be added before merge).

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
Reviewed-by on v3 1/1
#2 lore raw
In reply to Dennis Brendel

Hi Dennis,

Good to see your patch soon, nice work!

Next time remember to commit the patch with Signed-off-by.
git commit -sm xxxx

Signature · 6 lines
Regards,
Li Wang


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#3 lore raw

Hi Dennis, Li,

Quoted 6 lines

Hi Dennis,

Good to see your patch soon, nice work!

Next time remember to commit the patch with Signed-off-by.
git commit -sm xxxx

+1 (planning to add that before merge).

Also these comments "Changes compared to v2:" should be below "---", that way we can read them but they will not endup in the git commit message when we apply the patch.

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp

A message that is not in the archive (sent off-list or not yet received): [email protected]

#5 lore raw

Hi Dennis,

Quoted 6 lines

Hi Petr and Li,

Thank you very much for your valuable feedback!

What's left to do for me? Re-submit as v4 with sign-off and proper commit message?

v3 is ready to be merged, I'll add your SBT and cleanup the commit message, while I'll be adding my and Li's RBT.

I'm just waiting a little bit because Cyril looked into this patchset, he might have look (I merge tonight).

Kind regards,
Petr

Quoted 21 lines

Thanks,
Dennis

Quoted 17 lines · Petr Vorel

Hi Dennis, Li,

Quoted 6 lines

Hi Dennis,

Good to see your patch soon, nice work!

Next time remember to commit the patch with Signed-off-by.
git commit -sm xxxx

+1 (planning to add that before merge).

Also these comments "Changes compared to v2:" should be below "---", that way we can read them but they will not endup in the git commit message when we apply the patch.

Kind regards,
Petr

Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp
#6 lore raw

Hi Petr and Dennis,

As this is the first patch Dennis(thanks!!) writes down for LTP.

I would suggest sending a patch V4 (including SBT, Cyril's new reviews) to make you familiar with the workflow of the community:).

Quoted 46 lines · Petr Vorel

Hi Dennis,

Quoted 6 lines

Hi Petr and Li,

Thank you very much for your valuable feedback!

What's left to do for me? Re-submit as v4 with sign-off and proper commit message?

v3 is ready to be merged, I'll add your SBT and cleanup the commit message, while I'll be adding my and Li's RBT.

I'm just waiting a little bit because Cyril looked into this patchset, he might have look (I merge tonight).

Kind regards,
Petr

Quoted 21 lines

Thanks,
Dennis

Quoted 17 lines · Petr Vorel

Hi Dennis, Li,

Quoted 6 lines

Hi Dennis,

Good to see your patch soon, nice work!

Next time remember to commit the patch with Signed-off-by.
git commit -sm xxxx

+1 (planning to add that before merge).

Also these comments "Changes compared to v2:" should be below "---", that way we can read them but they will not endup in the git commit message when we apply the patch.

Kind regards,
Petr

Signature · 6 lines
Regards,
Li Wang


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#7 lore raw
In reply to Dennis Brendel

Hi!

Quoted 36 lines
testcases/kernel/syscalls/munlockall/munlockall01.c +15 -12
+/*\
+ * [Description]
+ *
+ * Verify that munlockall(2) unlocks all previously locked memory
+ */
 
-char *TCID = "munlockall01";
-int TST_TOTAL = 1;
+#include <sys/mman.h>
 
-#if !defined(UCLINUX)
+#include "tst_test.h"
 
-int main(int ac, char **av)
+static void verify_munlockall(void)
 {
-	int lc;
-
-	tst_parse_opts(ac, av, NULL, NULL);
+	unsigned long size = 0;
 
-	setup();
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 
-	/* check looping state */
-	for (lc = 0; TEST_LOOPING(lc); lc++) {
+	if (size != 0UL)
+		tst_brk(TBROK, "Locked memory after init should be 0 but is %ld", size);
 
-		tst_count = 0;
+	if (mlockall(MCL_CURRENT | MCL_FUTURE) != 0)
+		tst_brk(TBROK | TERRNO, "Could not lock memory using mlockall()");
 
-		TEST(munlockall());
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 

testcases/kernel/syscalls/munlockall/munlockall01.c line 33

Quoted 10 lines
testcases/kernel/syscalls/munlockall/munlockall01.c +3 -7
-		/* check return code */
-		if (TEST_RETURN == -1) {
-			tst_resm(TFAIL | TTERRNO, "munlockall() Failed with"
-				 " return=%ld", TEST_RETURN);
-		} else {
-			tst_resm(TPASS, "munlockall() passed with"
-				 " return=%ld ", TEST_RETURN);
+	if (size == 0UL)
+		tst_brk(TBROK, "Locked memory after mlockall() should be greater than 0, "
+			       "but is %ld", size);

This line can be shorter:

	tst_brk(TBROK, "After mlockall() locked memory should be >0");

We already checked that size is 0 so no need to print it.

testcases/kernel/syscalls/munlockall/munlockall01.c line 36

Quoted 4 lines
testcases/kernel/syscalls/munlockall/munlockall01.c +2 -2
-		}
-	}
+	if (munlockall() != 0)
+		tst_brk(TBROK | TERRNO, "Could not unlock memory using munlockall()");

We are testing the munlockall() syscall here so it would be better to use the TST_EXP_PASS() macro.

Quoted 48 lines
testcases/kernel/syscalls/munlockall/munlockall01.c +9 -30
-	/* cleanup and exit */
-	cleanup();
-	tst_exit();
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 
+	if (size != 0UL)
+		tst_res(TFAIL, "Locked memory after munlockall() should be 0 but is %ld", size);
+
+	else
+		tst_res(TPASS, "Memory successfully locked and unlocked");
 }
 
-#else
-
-int main(void)
-{
-	tst_resm(TINFO, "test is not available on uClinux");
-	tst_exit();
-}
-
-#endif /* if !defined(UCLINUX) */
-
-/* setup() - performs all ONE TIME setup for this test. */
-void setup(void)
-{
-	tst_require_root();
-
-	tst_sig(NOFORK, DEF_HANDLER, cleanup);
-
-	TEST_PAUSE;
-}
-
-/*
- * cleanup() - performs all ONE TIME cleanup for this test at
- *		completion or premature exit.
- */
-void cleanup(void)
-{
-}
+static struct tst_test test = {
+	.test_all = verify_munlockall,
+};
Signature · 5 lines
2.44.0


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
Signature · 5 lines
Cyril Hrubis
[email protected]

-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
#8 lore raw

Hi Cyril,

Quoted 50 lines · Cyril Hrubis

Hi!

Quoted 36 lines
testcases/kernel/syscalls/munlockall/munlockall01.c +15 -12
+/*\
+ * [Description]
+ *
+ * Verify that munlockall(2) unlocks all previously locked memory
+ */
 
-char *TCID = "munlockall01";
-int TST_TOTAL = 1;
+#include <sys/mman.h>
 
-#if !defined(UCLINUX)
+#include "tst_test.h"
 
-int main(int ac, char **av)
+static void verify_munlockall(void)
 {
-	int lc;
-
-	tst_parse_opts(ac, av, NULL, NULL);
+	unsigned long size = 0;
 
-	setup();
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 
-	/* check looping state */
-	for (lc = 0; TEST_LOOPING(lc); lc++) {
+	if (size != 0UL)
+		tst_brk(TBROK, "Locked memory after init should be 0 but is %ld", size);
 
-		tst_count = 0;
+	if (mlockall(MCL_CURRENT | MCL_FUTURE) != 0)
+		tst_brk(TBROK | TERRNO, "Could not lock memory using mlockall()");
 
-		TEST(munlockall());
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 

testcases/kernel/syscalls/munlockall/munlockall01.c line 33

Quoted 10 lines
testcases/kernel/syscalls/munlockall/munlockall01.c +3 -7
-		/* check return code */
-		if (TEST_RETURN == -1) {
-			tst_resm(TFAIL | TTERRNO, "munlockall() Failed with"
-				 " return=%ld", TEST_RETURN);
-		} else {
-			tst_resm(TPASS, "munlockall() passed with"
-				 " return=%ld ", TEST_RETURN);
+	if (size == 0UL)
+		tst_brk(TBROK, "Locked memory after mlockall() should be greater than 0, "
+			       "but is %ld", size);

This line can be shorter:

Quoted 3 lines
	tst_brk(TBROK, "After mlockall() locked memory should be >0");

We already checked that size is 0 so no need to print it.

Agreed :-)

Quoted 7 lines

testcases/kernel/syscalls/munlockall/munlockall01.c line 36

Quoted 4 lines
testcases/kernel/syscalls/munlockall/munlockall01.c +2 -2
-		}
-	}
+	if (munlockall() != 0)
+		tst_brk(TBROK | TERRNO, "Could not unlock memory using munlockall()");

We are testing the munlockall() syscall here so it would be better to use the TST_EXP_PASS() macro.

The actual purpose of the test was not checking the return value of the syscall (wrapper), but the behavior as reported by the kernel through /proc - but fair enough, it does not do any harm :-)

Quoted 49 lines
Quoted 48 lines
testcases/kernel/syscalls/munlockall/munlockall01.c +9 -30
-	/* cleanup and exit */
-	cleanup();
-	tst_exit();
+	SAFE_FILE_LINES_SCANF("/proc/self/status", "VmLck: %ld", &size);
 
+	if (size != 0UL)
+		tst_res(TFAIL, "Locked memory after munlockall() should be 0 but is %ld", size);
+
+	else
+		tst_res(TPASS, "Memory successfully locked and unlocked");
 }
 
-#else
-
-int main(void)
-{
-	tst_resm(TINFO, "test is not available on uClinux");
-	tst_exit();
-}
-
-#endif /* if !defined(UCLINUX) */
-
-/* setup() - performs all ONE TIME setup for this test. */
-void setup(void)
-{
-	tst_require_root();
-
-	tst_sig(NOFORK, DEF_HANDLER, cleanup);
-
-	TEST_PAUSE;
-}
-
-/*
- * cleanup() - performs all ONE TIME cleanup for this test at
- *		completion or premature exit.
- */
-void cleanup(void)
-{
-}
+static struct tst_test test = {
+	.test_all = verify_munlockall,
+};
Signature · 5 lines
2.44.0


-- 
Mailing list info: https://lists.linux.it/listinfo/ltp
Signature · 1 line
Mailing list info: https://lists.linux.it/listinfo/ltp