[LTP] [PATCH v2] sched: add sched sysctl sanity test

[LTP] [PATCH v2] sched: add sched sysctl sanity test

[Cyril Hrubis]

Currently the test fails due to kernel bug, I will send patch to LKML
later on.

The problem with kernel is that sysctl_sched_rt_period is unsigned int
but it's processed with proc_dointvec() which means that you are allowed
to write negative values into the variable even though documentation
says it shouldn't be possible and the kernel code asserts that rt_period
is > 0.

Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
---

- Fixed a few whitespaces
- Added second kernel commit into metadata

 runtest/sched                                 |   2 +
 testcases/kernel/sched/sysctl/.gitignore      |   1 +
 testcases/kernel/sched/sysctl/Makefile        |   7 +
 .../kernel/sched/sysctl/proc_sched_rt01.c     | 122 ++++++++++++++++++
 4 files changed, 132 insertions(+)
 create mode 100644 testcases/kernel/sched/sysctl/.gitignore
 create mode 100644 testcases/kernel/sched/sysctl/Makefile
 create mode 100644 testcases/kernel/sched/sysctl/proc_sched_rt01.c

diff --git a/runtest/sched b/runtest/sched
index 172fe4174..3457114f4 100644
--- a/runtest/sched
+++ b/runtest/sched
@@ -16,3 +16,5 @@ sched_cli_serv run_sched_cliserv.sh
 sched_stress sched_stress.sh
 
 autogroup01 autogroup01
+
+proc_sched_rt01
diff --git a/testcases/kernel/sched/sysctl/.gitignore b/testcases/kernel/sched/sysctl/.gitignore
new file mode 100644
index 000000000..29b859b81
--- /dev/null
+++ b/testcases/kernel/sched/sysctl/.gitignore
@@ -0,0 +1 @@
+proc_sched_rt01
diff --git a/testcases/kernel/sched/sysctl/Makefile b/testcases/kernel/sched/sysctl/Makefile
new file mode 100644
index 000000000..18896b6f2
--- /dev/null
+++ b/testcases/kernel/sched/sysctl/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0-or-later
+
+top_srcdir		?= ../../../..
+
+include $(top_srcdir)/include/mk/testcases.mk
+
+include $(top_srcdir)/include/mk/generic_leaf_target.mk
diff --git a/testcases/kernel/sched/sysctl/proc_sched_rt01.c b/testcases/kernel/sched/sysctl/proc_sched_rt01.c
new file mode 100644
index 000000000..203846ae4
--- /dev/null
+++ b/testcases/kernel/sched/sysctl/proc_sched_rt01.c
@@ -0,0 +1,122 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (C) Cyril Hrubis <chrubis@suse.cz>
+ */
+
+/*\
+ * [Description]
+ *
+ * Sanity tests for the /proc/sys/kernel/sched_r* files.
+ *
+ * - The sched_rt_period_us range is 1 to INT_MAX
+ *   try invalid values and check for EINVAL
+ *
+ * - The sched_rt_runtime_us range is -1 to INT_MAX
+ *   try invalid values and check for EINVAL
+ *
+ * - The sched_rt_runtime_us must be less or equal to sched_rt_period_us
+ *
+ * - Reset sched_rr_timeslice_ms to default value by writing -1 and check that
+ *   we get the default value on next read.
+ *
+ * This is a regression test for a commits:
+ *
+ *   commit c1fc6484e1fb7cc2481d169bfef129a1b0676abe
+ *   Author: Cyril Hrubis <chrubis@suse.cz>
+ *   Date:   Wed Aug 2 17:19:06 2023 +0200
+ *
+ *       sched/rt: sysctl_sched_rr_timeslice show default timeslice after reset
+ *
+ *  commit 079be8fc630943d9fc70a97807feb73d169ee3fc
+ *  Author: Cyril Hrubis <chrubis@suse.cz>
+ *  Date:   Mon Oct 2 13:55:51 2023 +0200
+ *
+ *       sched/rt: Disallow writing invalid values to sched_rt_period_us
+ */
+
+#include <stdio.h>
+#include "tst_test.h"
+
+#define RT_PERIOD_US "/proc/sys/kernel/sched_rt_period_us"
+#define RT_RUNTIME_US "/proc/sys/kernel/sched_rt_runtime_us"
+#define RR_TIMESLICE_MS "/proc/sys/kernel/sched_rr_timeslice_ms"
+
+static int period_fd;
+static int runtime_fd;
+
+static void rr_timeslice_ms_reset(void)
+{
+	long timeslice_ms;
+
+	SAFE_FILE_PRINTF(RR_TIMESLICE_MS, "-1");
+	SAFE_FILE_SCANF(RR_TIMESLICE_MS, "%li", &timeslice_ms);
+
+	TST_EXP_EXPR(timeslice_ms > 0,
+		"timeslice_ms > 0 after reset to default");
+}
+
+static void rt_period_us_einval(void)
+{
+	TST_EXP_FAIL(write(period_fd, "0", 2), EINVAL,
+		"echo 0 > "RT_PERIOD_US);
+	TST_EXP_FAIL(write(period_fd, "-1", 2), EINVAL,
+		"echo -1 > "RT_PERIOD_US);
+}
+
+static void rt_runtime_us_einval(void)
+{
+	TST_EXP_FAIL(write(runtime_fd, "-2", 2), EINVAL,
+		"echo -2 > "RT_RUNTIME_US);
+}
+
+static void rt_runtime_us_le_period_us(void)
+{
+	int period_us;
+	char buf[32];
+
+	SAFE_FILE_SCANF(RT_PERIOD_US, "%i", &period_us);
+
+	sprintf(buf, "%i", period_us+1);
+
+	TST_EXP_FAIL(write(runtime_fd, buf, strlen(buf)), EINVAL,
+		"echo rt_period_us+1 > "RT_RUNTIME_US);
+}
+
+static void verify_sched_proc(void)
+{
+	rr_timeslice_ms_reset();
+	rt_period_us_einval();
+	rt_runtime_us_einval();
+	rt_runtime_us_le_period_us();
+}
+
+static void setup(void)
+{
+	period_fd = open(RT_PERIOD_US, O_RDWR);
+	runtime_fd = open(RT_RUNTIME_US, O_RDWR);
+}
+
+static void cleanup(void)
+{
+	if (period_fd > 0)
+		SAFE_CLOSE(period_fd);
+
+	if (runtime_fd > 0)
+		SAFE_CLOSE(runtime_fd);
+}
+
+static struct tst_test test = {
+	.needs_root = 1,
+	.setup = setup,
+	.cleanup = cleanup,
+	.test_all = verify_sched_proc,
+	.tags = (struct tst_tag []) {
+		{"linux-git", "c1fc6484e1fb"},
+		{"linux-git", "079be8fc6309"},
+		{}
+	},
+	.needs_kconfigs = (const char *[]) {
+		"CONFIG_SYSCTL",
+		NULL
+	},
+};
-- 
2.41.0


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

Re: [LTP] [PATCH v2] sched: add sched sysctl sanity test

[Cyril Hrubis]

Hi!
> Currently the test fails due to kernel bug, I will send patch to LKML
> later on.

Should have removed this line, the patch has been pulled into mainline
today.

> The problem with kernel is that sysctl_sched_rt_period is unsigned int
> but it's processed with proc_dointvec() which means that you are allowed
> to write negative values into the variable even though documentation
> says it shouldn't be possible and the kernel code asserts that rt_period
> is > 0.
> 
> Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
> ---
> 
> - Fixed a few whitespaces
> - Added second kernel commit into metadata
> 
>  runtest/sched                                 |   2 +
>  testcases/kernel/sched/sysctl/.gitignore      |   1 +
>  testcases/kernel/sched/sysctl/Makefile        |   7 +
>  .../kernel/sched/sysctl/proc_sched_rt01.c     | 122 ++++++++++++++++++
>  4 files changed, 132 insertions(+)
>  create mode 100644 testcases/kernel/sched/sysctl/.gitignore
>  create mode 100644 testcases/kernel/sched/sysctl/Makefile
>  create mode 100644 testcases/kernel/sched/sysctl/proc_sched_rt01.c
> 
> diff --git a/runtest/sched b/runtest/sched
> index 172fe4174..3457114f4 100644
> --- a/runtest/sched
> +++ b/runtest/sched
> @@ -16,3 +16,5 @@ sched_cli_serv run_sched_cliserv.sh
>  sched_stress sched_stress.sh
>  
>  autogroup01 autogroup01
> +
> +proc_sched_rt01
> diff --git a/testcases/kernel/sched/sysctl/.gitignore b/testcases/kernel/sched/sysctl/.gitignore
> new file mode 100644
> index 000000000..29b859b81
> --- /dev/null
> +++ b/testcases/kernel/sched/sysctl/.gitignore
> @@ -0,0 +1 @@
> +proc_sched_rt01
> diff --git a/testcases/kernel/sched/sysctl/Makefile b/testcases/kernel/sched/sysctl/Makefile
> new file mode 100644
> index 000000000..18896b6f2
> --- /dev/null
> +++ b/testcases/kernel/sched/sysctl/Makefile
> @@ -0,0 +1,7 @@
> +# SPDX-License-Identifier: GPL-2.0-or-later
> +
> +top_srcdir		?= ../../../..
> +
> +include $(top_srcdir)/include/mk/testcases.mk
> +
> +include $(top_srcdir)/include/mk/generic_leaf_target.mk
> diff --git a/testcases/kernel/sched/sysctl/proc_sched_rt01.c b/testcases/kernel/sched/sysctl/proc_sched_rt01.c
> new file mode 100644
> index 000000000..203846ae4
> --- /dev/null
> +++ b/testcases/kernel/sched/sysctl/proc_sched_rt01.c
> @@ -0,0 +1,122 @@
> +// SPDX-License-Identifier: GPL-2.0-or-later
> +/*
> + * Copyright (C) Cyril Hrubis <chrubis@suse.cz>
> + */
> +
> +/*\
> + * [Description]
> + *
> + * Sanity tests for the /proc/sys/kernel/sched_r* files.
> + *
> + * - The sched_rt_period_us range is 1 to INT_MAX
> + *   try invalid values and check for EINVAL
> + *
> + * - The sched_rt_runtime_us range is -1 to INT_MAX
> + *   try invalid values and check for EINVAL
> + *
> + * - The sched_rt_runtime_us must be less or equal to sched_rt_period_us
> + *
> + * - Reset sched_rr_timeslice_ms to default value by writing -1 and check that
> + *   we get the default value on next read.
> + *
> + * This is a regression test for a commits:
> + *
> + *   commit c1fc6484e1fb7cc2481d169bfef129a1b0676abe
> + *   Author: Cyril Hrubis <chrubis@suse.cz>
> + *   Date:   Wed Aug 2 17:19:06 2023 +0200
> + *
> + *       sched/rt: sysctl_sched_rr_timeslice show default timeslice after reset
> + *
> + *  commit 079be8fc630943d9fc70a97807feb73d169ee3fc
> + *  Author: Cyril Hrubis <chrubis@suse.cz>
> + *  Date:   Mon Oct 2 13:55:51 2023 +0200
> + *
> + *       sched/rt: Disallow writing invalid values to sched_rt_period_us
> + */
> +
> +#include <stdio.h>
> +#include "tst_test.h"
> +
> +#define RT_PERIOD_US "/proc/sys/kernel/sched_rt_period_us"
> +#define RT_RUNTIME_US "/proc/sys/kernel/sched_rt_runtime_us"
> +#define RR_TIMESLICE_MS "/proc/sys/kernel/sched_rr_timeslice_ms"
> +
> +static int period_fd;
> +static int runtime_fd;
> +
> +static void rr_timeslice_ms_reset(void)
> +{
> +	long timeslice_ms;
> +
> +	SAFE_FILE_PRINTF(RR_TIMESLICE_MS, "-1");
> +	SAFE_FILE_SCANF(RR_TIMESLICE_MS, "%li", &timeslice_ms);
> +
> +	TST_EXP_EXPR(timeslice_ms > 0,
> +		"timeslice_ms > 0 after reset to default");
> +}
> +
> +static void rt_period_us_einval(void)
> +{
> +	TST_EXP_FAIL(write(period_fd, "0", 2), EINVAL,
> +		"echo 0 > "RT_PERIOD_US);
> +	TST_EXP_FAIL(write(period_fd, "-1", 2), EINVAL,
> +		"echo -1 > "RT_PERIOD_US);
> +}
> +
> +static void rt_runtime_us_einval(void)
> +{
> +	TST_EXP_FAIL(write(runtime_fd, "-2", 2), EINVAL,
> +		"echo -2 > "RT_RUNTIME_US);
> +}
> +
> +static void rt_runtime_us_le_period_us(void)
> +{
> +	int period_us;
> +	char buf[32];
> +
> +	SAFE_FILE_SCANF(RT_PERIOD_US, "%i", &period_us);
> +
> +	sprintf(buf, "%i", period_us+1);
> +
> +	TST_EXP_FAIL(write(runtime_fd, buf, strlen(buf)), EINVAL,
> +		"echo rt_period_us+1 > "RT_RUNTIME_US);
> +}
> +
> +static void verify_sched_proc(void)
> +{
> +	rr_timeslice_ms_reset();
> +	rt_period_us_einval();
> +	rt_runtime_us_einval();
> +	rt_runtime_us_le_period_us();
> +}
> +
> +static void setup(void)
> +{
> +	period_fd = open(RT_PERIOD_US, O_RDWR);
> +	runtime_fd = open(RT_RUNTIME_US, O_RDWR);
> +}
> +
> +static void cleanup(void)
> +{
> +	if (period_fd > 0)
> +		SAFE_CLOSE(period_fd);
> +
> +	if (runtime_fd > 0)
> +		SAFE_CLOSE(runtime_fd);
> +}
> +
> +static struct tst_test test = {
> +	.needs_root = 1,
> +	.setup = setup,
> +	.cleanup = cleanup,
> +	.test_all = verify_sched_proc,
> +	.tags = (struct tst_tag []) {
> +		{"linux-git", "c1fc6484e1fb"},
> +		{"linux-git", "079be8fc6309"},
> +		{}
> +	},
> +	.needs_kconfigs = (const char *[]) {
> +		"CONFIG_SYSCTL",
> +		NULL
> +	},
> +};
> -- 
> 2.41.0
> 
> 
> -- 
> Mailing list info: https://lists.linux.it/listinfo/ltp

-- 
Cyril Hrubis
chrubis@suse.cz

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

Re: [LTP] [PATCH v2] sched: add sched sysctl sanity test

[Cyril Hrubis]

Hi!
Ping?

-- 
Cyril Hrubis
chrubis@suse.cz

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

Re: [LTP] [PATCH v2] sched: add sched sysctl sanity test

[Petr Vorel]

Hi Cyril,

It looks like the second fix 079be8fc6309 ("sched/rt: Disallow writing invalid
values to sched_rt_period_us") is not in stable, shouldn't be there?

> Currently the test fails due to kernel bug, I will send patch to LKML
> later on.

> The problem with kernel is that sysctl_sched_rt_period is unsigned int
> but it's processed with proc_dointvec() which means that you are allowed
> to write negative values into the variable even though documentation
> says it shouldn't be possible and the kernel code asserts that rt_period
> is > 0.

> Signed-off-by: Cyril Hrubis <chrubis@suse.cz>
> ---

> - Fixed a few whitespaces
> - Added second kernel commit into metadata
nit: metadata of the git commit looks ugly, but we would need to agree on better
solution.

>  runtest/sched                                 |   2 +
>  testcases/kernel/sched/sysctl/.gitignore      |   1 +
>  testcases/kernel/sched/sysctl/Makefile        |   7 +
>  .../kernel/sched/sysctl/proc_sched_rt01.c     | 122 ++++++++++++++++++
>  4 files changed, 132 insertions(+)
>  create mode 100644 testcases/kernel/sched/sysctl/.gitignore
>  create mode 100644 testcases/kernel/sched/sysctl/Makefile
>  create mode 100644 testcases/kernel/sched/sysctl/proc_sched_rt01.c

> diff --git a/runtest/sched b/runtest/sched
> index 172fe4174..3457114f4 100644
> --- a/runtest/sched
> +++ b/runtest/sched
> @@ -16,3 +16,5 @@ sched_cli_serv run_sched_cliserv.sh
>  sched_stress sched_stress.sh

>  autogroup01 autogroup01
> +
> +proc_sched_rt01

This should obviously be:
proc_sched_rt01 proc_sched_rt01

And I would move it above sched_stress.sh.

The rest LGTM, thanks!
Reviewed-by: Petr Vorel <pvorel@suse.cz>

Kind regards,
Petr

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

Re: [LTP] [PATCH v2] sched: add sched sysctl sanity test

[Cyril Hrubis]

Hi!
> It looks like the second fix 079be8fc6309 ("sched/rt: Disallow writing invalid
> values to sched_rt_period_us") is not in stable, shouldn't be there?

No idea to be honest, the second fix about rejecting invalid values is a
minor fix.

> > - Fixed a few whitespaces
> > - Added second kernel commit into metadata
> nit: metadata of the git commit looks ugly, but we would need to agree on better
> solution.

Which metadata exactly? This part is after the scisors and will be cut
when the patch is applied.

> >  runtest/sched                                 |   2 +
> >  testcases/kernel/sched/sysctl/.gitignore      |   1 +
> >  testcases/kernel/sched/sysctl/Makefile        |   7 +
> >  .../kernel/sched/sysctl/proc_sched_rt01.c     | 122 ++++++++++++++++++
> >  4 files changed, 132 insertions(+)
> >  create mode 100644 testcases/kernel/sched/sysctl/.gitignore
> >  create mode 100644 testcases/kernel/sched/sysctl/Makefile
> >  create mode 100644 testcases/kernel/sched/sysctl/proc_sched_rt01.c
> 
> > diff --git a/runtest/sched b/runtest/sched
> > index 172fe4174..3457114f4 100644
> > --- a/runtest/sched
> > +++ b/runtest/sched
> > @@ -16,3 +16,5 @@ sched_cli_serv run_sched_cliserv.sh
> >  sched_stress sched_stress.sh
> 
> >  autogroup01 autogroup01
> > +
> > +proc_sched_rt01
> 
> This should obviously be:
> proc_sched_rt01 proc_sched_rt01
> 
> And I would move it above sched_stress.sh.

Goot catch, will fix.

> The rest LGTM, thanks!
> Reviewed-by: Petr Vorel <pvorel@suse.cz>

-- 
Cyril Hrubis
chrubis@suse.cz

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

Re: [LTP] [PATCH v2] sched: add sched sysctl sanity test

[Petr Vorel]

> Hi!
> > It looks like the second fix 079be8fc6309 ("sched/rt: Disallow writing invalid
> > values to sched_rt_period_us") is not in stable, shouldn't be there?

> No idea to be honest, the second fix about rejecting invalid values is a
> minor fix.

IMHO it's a fix of broken functionality. And Greg stated several times that
stable is not only for security fixes but for other fixes. Therefore I would
vote for adding it.

> > > - Fixed a few whitespaces
> > > - Added second kernel commit into metadata
> > nit: metadata of the git commit looks ugly, but we would need to agree on better
> > solution.

> Which metadata exactly? This part is after the scisors and will be cut
> when the patch is applied.

I don't like formatting after "This is a regression test for a commits:".
It formates like code block (gray background) via using <pre>...</pre> HTML
tags:

	commit c1fc6484e1fb7cc2481d169bfef129a1b0676abe
	Author: Cyril Hrubis <chrubis@suse.cz>
	Date:   Wed Aug 2 17:19:06 2023 +0200

	sched/rt: sysctl_sched_rr_timeslice show default timeslice after reset

	commit 079be8fc630943d9fc70a97807feb73d169ee3fc
	Author: Cyril Hrubis <chrubis@suse.cz>
	Date:   Mon Oct 2 13:55:51 2023 +0200

	sched/rt: Disallow writing invalid values to sched_rt_period_us

(see a real output in metadata.html)

Given we have this in the Tag table below, I found it useless. I mean, we could
have it, but not in docparse comment /*\ ... */, but in normal C comment /* ... */
(for reader of the C source, but not in docparse format). Or, if we want to have
it in docparse output, we could format it as:

 /*\
...
 * This is a regression test for a commits:

 * * c1fc6484e1fb ("sched/rt: sysctl_sched_rr_timeslice show default timeslice after reset")
 * * 079be8fc6309 ("sched/rt: Disallow writing invalid values to sched_rt_period_us")

This is not unique to this commit, we don't have to solve it now (and it's
definitely a blocker for this patch). But it would be good to find a consensus
how the output should look like.

Kind regards,
Petr

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

Re: [LTP] [PATCH v2] sched: add sched sysctl sanity test

[Cyril Hrubis]

Hi!
> The rest LGTM, thanks!
> Reviewed-by: Petr Vorel <pvorel@suse.cz>

Fixed all the problems and pushed, thanks.

-- 
Cyril Hrubis
chrubis@suse.cz

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