Skip to contentSkip to conversation

sched/rt: Disallow writing invalid values to sched_rt_period_us

PATCH Superseded Cyril Hrubis 3 participants 6 messages Sep 1, 2023 – Sep 22, 2023 Scheduler

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 sched/rt: Disallow writing invalid values to sched_rt_period_us

The validation of the value written to sched_rt_period_us was broken because:

- the sysclt_sched_rt_period is declared as unsigned int
- parsed by proc_do_intvec()
- the range is asserted after the value parsed by proc_do_intvec()

Because of this negative values written to the file were written into a unsigned integer that were later on interpreted as large positive integers which did passed the check:

if (sysclt_sched_rt_period <= 0)

	return EINVAL;

This commit fixes the parsing by setting explicit range for both perid_us and runtime_us into the sched_rt_sysctls table and processes the values with proc_dointvec_minmax() instead.

Alternatively if we wanted to use full range of unsigned int for the period value we would have to split the proc_handler and use proc_douintvec() for it however even the Documentation/scheduller/sched-rt-group.rst describes the range as 1 to INT_MAX.

As far as I can tell the only problem this causes is that the sysctl file allows writing negative values which when read back may confuse userspace.

There is also a LTP test being submitted for these sysctl files at:

http://patchwork.ozlabs.org/project/ltp/patch/[email protected]/

---
 kernel/sched/rt.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
kernel/sched/rt.c +5 -4
index 0597ba0f85ff..aed3d55de2dd 100644
@@ -37,6 +37,8 @@ static struct ctl_table sched_rt_sysctls[] = {
 		.maxlen         = sizeof(unsigned int),
 		.mode           = 0644,
 		.proc_handler   = sched_rt_handler,
+		.extra1         = SYSCTL_ONE,
+		.extra2         = SYSCTL_INT_MAX,
 	},
 	{
 		.procname       = "sched_rt_runtime_us",
@@ -44,6 +46,8 @@ static struct ctl_table sched_rt_sysctls[] = {
 		.maxlen         = sizeof(int),
 		.mode           = 0644,
 		.proc_handler   = sched_rt_handler,
+		.extra1         = SYSCTL_NEG_ONE,
+		.extra2         = SYSCTL_INT_MAX,
 	},
 	{
 		.procname       = "sched_rr_timeslice_ms",
@@ -2985,9 +2989,6 @@ static int sched_rt_global_constraints(void)
 #ifdef CONFIG_SYSCTL
 static int sched_rt_global_validate(void)
 {
-	if (sysctl_sched_rt_period <= 0)
-		return -EINVAL;
-
 	if ((sysctl_sched_rt_runtime != RUNTIME_INF) &&
 		((sysctl_sched_rt_runtime > sysctl_sched_rt_period) ||
 		 ((u64)sysctl_sched_rt_runtime *
@@ -3018,7 +3019,7 @@ static int sched_rt_handler(struct ctl_table *table, int write, void *buffer,
 	old_period = sysctl_sched_rt_period;
 	old_runtime = sysctl_sched_rt_runtime;
 
-	ret = proc_dointvec(table, write, buffer, lenp, ppos);
+	ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
 
 	if (!ret && write) {
 		ret = sched_rt_global_validate();
Signature · 1 line
2.41.0
#1 lore raw
Quoted 51 lines

The validation of the value written to sched_rt_period_us was broken because:

- the sysclt_sched_rt_period is declared as unsigned int
- parsed by proc_do_intvec()
- the range is asserted after the value parsed by proc_do_intvec()

Because of this negative values written to the file were written into a unsigned integer that were later on interpreted as large positive integers which did passed the check:

if (sysclt_sched_rt_period <= 0)

	return EINVAL;

This commit fixes the parsing by setting explicit range for both perid_us and runtime_us into the sched_rt_sysctls table and processes the values with proc_dointvec_minmax() instead.

Alternatively if we wanted to use full range of unsigned int for the period value we would have to split the proc_handler and use proc_douintvec() for it however even the Documentation/scheduller/sched-rt-group.rst describes the range as 1 to INT_MAX.

As far as I can tell the only problem this causes is that the sysctl file allows writing negative values which when read back may confuse userspace.

There is also a LTP test being submitted for these sysctl files at:

http://patchwork.ozlabs.org/project/ltp/patch/[email protected]/

---
 kernel/sched/rt.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
kernel/sched/rt.c +2 -0
index 0597ba0f85ff..aed3d55de2dd 100644
@@ -37,6 +37,8 @@ static struct ctl_table sched_rt_sysctls[] = {
 		.maxlen         = sizeof(unsigned int),
 		.mode           = 0644,
 		.proc_handler   = sched_rt_handler,
+		.extra1         = SYSCTL_ONE,
+		.extra2         = SYSCTL_INT_MAX,
 	},
 	{
 		.procname       = "sched_rt_runtime_us",

kernel/sched/rt.c static struct ctl_table sched_rt_sysctls[] = { · line 50

Quoted 5 lines
kernel/sched/rt.c +2 -0
@@ -44,6 +46,8 @@ static struct ctl_table sched_rt_sysctls[] = {
 		.maxlen         = sizeof(int),
 		.mode           = 0644,
 		.proc_handler   = sched_rt_handler,
+		.extra1         = SYSCTL_NEG_ONE,
+		.extra2         = SYSCTL_INT_MAX,

Documentation/scheduller/sched-rt-group.rst [1] specifies this as values from -1 to (INT_MAX - 1), I guess due int range. Looking into proc_dointvec_minmax() [2] even INT_MAX would pass the check. I suppose we can do nothing about that, because there is no value in sysctl_vals[] which would represent INT_MAX - 1.

And you specify in LTP test range: from -1 to INT_MAX.

But even much shorter value than INT_MAX fails:

$ echo 1234567 > /proc/sys/kernel/sched_rt_runtime_us
sh: echo: write error: Invalid argument

Ranges in /proc/sys/kernel/sched_rt_period_us works as expected.

[1] https://www.kernel.org/doc/html/latest/scheduler/sched-rt-group.html#system-wide-settings
[2] https://elixir.bootlin.com/linux/latest/source/kernel/sysctl.c#L843

Quoted 22 lines
kernel/sched/rt.c +1 -4
 	},
 	{
 		.procname       = "sched_rr_timeslice_ms",
@@ -2985,9 +2989,6 @@ static int sched_rt_global_constraints(void)
 #ifdef CONFIG_SYSCTL
 static int sched_rt_global_validate(void)
 {
-	if (sysctl_sched_rt_period <= 0)
-		return -EINVAL;
-
 	if ((sysctl_sched_rt_runtime != RUNTIME_INF) &&
 		((sysctl_sched_rt_runtime > sysctl_sched_rt_period) ||
 		 ((u64)sysctl_sched_rt_runtime *
@@ -3018,7 +3019,7 @@ static int sched_rt_handler(struct ctl_table *table, int write, void *buffer,
 	old_period = sysctl_sched_rt_period;
 	old_runtime = sysctl_sched_rt_runtime;
 
-	ret = proc_dointvec(table, write, buffer, lenp, ppos);
+	ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
 
 	if (!ret && write) {
 		ret = sched_rt_global_validate();
#2 lore raw

Hi!

Quoted 11 lines

Documentation/scheduller/sched-rt-group.rst [1] specifies this as values from -1 to (INT_MAX - 1), I guess due int range. Looking into proc_dointvec_minmax() [2] even INT_MAX would pass the check. I suppose we can do nothing about that, because there is no value in sysctl_vals[] which would represent INT_MAX - 1.

And you specify in LTP test range: from -1 to INT_MAX.

But even much shorter value than INT_MAX fails:

$ echo 1234567 > /proc/sys/kernel/sched_rt_runtime_us
sh: echo: write error: Invalid argument

That is because runtime_us must be < period_us by definition, since runtime_us defines how much time is allocated from the period_us. I guess that this is not described good enough in the kernel docs.

Signature · 2 lines
Cyril Hrubis
[email protected]
#3 lore raw
Quoted 8 lines

Hi!

Quoted 6 lines

Documentation/scheduller/sched-rt-group.rst [1] specifies this as values from -1 to (INT_MAX - 1), I guess due int range. Looking into proc_dointvec_minmax() [2] even INT_MAX would pass the check. I suppose we can do nothing about that, because there is no value in sysctl_vals[] which would represent INT_MAX - 1.

And you specify in LTP test range: from -1 to INT_MAX.

Quoted 8 lines
Quoted 4 lines

But even much shorter value than INT_MAX fails:

$ echo 1234567 > /proc/sys/kernel/sched_rt_runtime_us
sh: echo: write error: Invalid argument

That is because runtime_us must be < period_us by definition, since runtime_us defines how much time is allocated from the period_us. I guess that this is not described good enough in the kernel docs.

Thanks for clarification. +1 for docs update.

Kind regards,
Petr

#4 lore raw
In reply to Cyril Hrubis
Quoted 8 lines · Cyril Hrubis

Hi!

Quoted 7 lines

Documentation/scheduller/sched-rt-group.rst [1] specifies this as values from -1 to (INT_MAX - 1), I guess due int range. Looking into proc_dointvec_minmax() [2] even INT_MAX would pass the check. I suppose we can do nothing about that, because there is no value in sysctl_vals[] which would represent INT_MAX - 1.

And you specify in LTP test range: from -1 to INT_MAX.

Quoted 8 lines
Quoted 4 lines

But even much shorter value than INT_MAX fails:

$ echo 1234567 > /proc/sys/kernel/sched_rt_runtime_us
sh: echo: write error: Invalid argument

That is because runtime_us must be < period_us by definition, since runtime_us defines how much time is allocated from the period_us. I guess that this is not described good enough in the kernel docs.

Mind adding a second patch to your series, clarifying Documentation/scheduler/sched-rt-group.rst?

Thanks,

Ingo

#5 lore raw

Hi!

Quoted 12 lines
Quoted 12 lines
Quoted 11 lines

Documentation/scheduller/sched-rt-group.rst [1] specifies this as values from -1 to (INT_MAX - 1), I guess due int range. Looking into proc_dointvec_minmax() [2] even INT_MAX would pass the check. I suppose we can do nothing about that, because there is no value in sysctl_vals[] which would represent INT_MAX - 1.

And you specify in LTP test range: from -1 to INT_MAX.

But even much shorter value than INT_MAX fails:

$ echo 1234567 > /proc/sys/kernel/sched_rt_runtime_us
sh: echo: write error: Invalid argument
Quoted 6 lines
Quoted 3 lines

That is because runtime_us must be < period_us by definition, since runtime_us defines how much time is allocated from the period_us. I guess that this is not described good enough in the kernel docs.

Mind adding a second patch to your series, clarifying Documentation/scheduler/sched-rt-group.rst?

Yes, that is the plan. I've been working on LTP release for past two weeks so this has been postponed. I will get to it hopefully next week.

Signature · 2 lines
Cyril Hrubis
[email protected]