Skip to contentSkip to conversation

kernel/sysctl: support setting sysctl parameters from kernel command line

RFC Superseded Vlastimil Babka 7 participants 18 messages Mar 25, 2020 – Mar 30, 2020

Subsystems (5)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

RFC v2 1/2 kernel/sysctl: support setting sysctl parameters from kernel command line

A recently proposed patch to add vm_swappiness command line parameter in addition to existing sysctl [1] made me wonder why we don't have a general support for passing sysctl parameters via command line. Googling found only somebody else wondering the same [2], but I haven't found any prior discussion with reasons why not to do this.

Settings the vm_swappiness issue aside (the underlying issue might be solved in a different way), quick search of kernel-parameters.txt shows there are already some that exist as both sysctl and kernel parameter - hung_task_panic, nmi_watchdog, numa_zonelist_order, traceoff_on_warning. A general mechanism would remove the need to add more of those one-offs and might be handy in situations where configuration by e.g. /etc/sysctl.d/ is impractical. Also after 61a47c1ad3a4 ("sysctl: Remove the sysctl system call") the only way to set sysctl is via procfs, so this would eventually allow small systems to be built without CONFIG_PROC_SYSCTL and still be able to change sysctl parameters.

Hence, this patch adds a new parse_args() pass that looks for parameters prefixed by 'sysctl.' and searches for them in the sysctl ctl_tables. When found, the respective proc handler is invoked. The search is just a naive linear one, to avoid using the whole procfs layer. It should be acceptable, as the cost depends on number of sysctl. parameters passed.

The main limitation of avoiding the procfs layer is however that sysctls dynamically registered by register_sysctl_table() or register_sysctl_paths() cannot currently be set by this method.

The processing is hooked right before the init process is loaded, as some handlers might be more complicated than simple setters and might need some subsystems to be initialized. At the moment the init process can be started and eventually execute a process writing to /proc/sys/ then it should be also fine to do that from the kernel.

[1] https://lore.kernel.org/linux-doc/BL0PR02MB560167492CA4094C91589930E9FC0@BL0PR02MB5601.namprd02.prod.outlook.com/
[2] https://unix.stackexchange.com/questions/558802/how-to-set-sysctl-using-kernel-command-line-parameter

---
v2: - handle any nesting level of parameter name
- add Documentation/admin-guide/kernel-parameters.txt blurb
- alias support for legacy one-off parameters, with first conversion (patch 2)
- still no support for dynamically registed sysctls

 .../admin-guide/kernel-parameters.txt         |  9 +++
 include/linux/sysctl.h                        |  1 +
 init/main.c                                   | 21 +++++++
 kernel/sysctl.c                               | 62 +++++++++++++++++++
 4 files changed, 93 insertions(+)
Documentation/admin-guide/kernel-parameters.txt +9 -0
index c07815d230bc..5076e288f93f 100644
@@ -4793,6 +4793,15 @@
 
 	switches=	[HW,M68k]
 
+	sysctl.*=	[KNL]
+			Set a sysctl parameter right before loading the init
+			process, as if the value was written to the respective
+			/proc/sys/... file. Currently a subset of sysctl
+			parameters is supported that is not registered
+			dynamically. Unrecognized parameters and invalid values
+			are reported in the kernel log.
+			Example: sysctl.vm.swappiness=40
+
 	sysfs.deprecated=0|1 [KNL]
 			Enable/disable old style sysfs layout for old udev
 			on older distributions. When this option is enabled
include/linux/sysctl.h +1 -0
index 02fa84493f23..62ae963a5c0c 100644
@@ -206,6 +206,7 @@ struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
 void unregister_sysctl_table(struct ctl_table_header * table);
 
 extern int sysctl_init(void);
+int process_sysctl_arg(char *param, char *val, const char *unused, void *arg);
 
 extern struct ctl_table sysctl_mount_point[];
 
init/main.c +21 -0
index ee4947af823f..74a094c6b8b9 100644
@@ -1345,6 +1345,25 @@ void __weak free_initmem(void)
 	free_initmem_default(POISON_FREE_INITMEM);
 }
 
+static void do_sysctl_args(void)
+{
+#ifdef CONFIG_SYSCTL
+	size_t len = strlen(saved_command_line) + 1;
+	char *command_line;
+
+	command_line = kzalloc(len, GFP_KERNEL);
+	if (!command_line)
+		panic("%s: Failed to allocate %zu bytes\n", __func__, len);
+
+	strcpy(command_line, saved_command_line);
+
+	parse_args("Setting sysctl args", command_line,
+		   NULL, 0, -1, -1, NULL, process_sysctl_arg);
+
+	kfree(command_line);
+#endif
+}
+
 static int __ref kernel_init(void *unused)
 {
 	int ret;
@@ -1367,6 +1386,8 @@ static int __ref kernel_init(void *unused)
 
 	rcu_end_inkernel_boot();
 
+	do_sysctl_args();
+
 	if (ramdisk_execute_command) {
 		ret = run_init_process(ramdisk_execute_command);
 		if (!ret)
kernel/sysctl.c +62 -0

68 lines not on this page · raw message

Signature · 1 line
2.25.1
#1 lore raw

RFC v2 2/2 kernel/sysctl: support handling command line aliases

We can now handle sysctl parameters on kernel command line, but historically some parameters introduced their own command line equivalent, which we don't want to remove for compatibility reasons. We can however convert them to the generic infrastructure with a table translating the legacy command line parameters to their sysctl names, and removing the one-off param handlers.

This patch adds the support and makes the first conversion to demonstrate it, on the (deprecated) numa_zonelist_order parameter.

---
 kernel/sysctl.c | 39 +++++++++++++++++++++++++++++++++++----
 mm/page_alloc.c |  9 ---------
 2 files changed, 35 insertions(+), 13 deletions(-)
kernel/sysctl.c +35 -4
index 18c7f5606d55..fd72853396f9 100644
@@ -1971,6 +1971,22 @@ static struct ctl_table dev_table[] = {
 	{ }
 };
 
+struct sysctl_alias {
+	char *kernel_param;
+	char *sysctl_param;
+};
+
+/*
+ * Historically some settings had both sysctl and a command line parameter.
+ * With the generic sysctl. parameter support, we can handle them at a single
+ * place and only keep the historical name for compatibility. This is not meant
+ * to add brand new aliases.
+ */
+static struct sysctl_alias sysctl_aliases[] = {
+	{"numa_zonelist_order",		"vm.numa_zonelist_order" },
+	{ }
+};
+
 int __init sysctl_init(void)
 {
 	struct ctl_table_header *hdr;
@@ -1980,6 +1996,18 @@ int __init sysctl_init(void)
 	return 0;
 }
 
+char *sysctl_find_alias(char *param)
+{
+	struct sysctl_alias *alias;
+
+	for (alias = &sysctl_aliases[0]; alias->kernel_param != NULL; alias++) {
+		if (strcmp(alias->kernel_param, param) == 0)
+			return alias->sysctl_param;
+	}
+
+	return NULL;
+}
+
 /* Set sysctl value passed on kernel command line. */
 int process_sysctl_arg(char *param, char *val,
 			       const char *unused, void *arg)
@@ -1990,10 +2018,13 @@ int process_sysctl_arg(char *param, char *val,
 	loff_t ppos = 0;
 	struct ctl_table *ctl, *found = NULL;
 
-	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
-		return 0;
-
-	param += sizeof("sysctl.") - 1;
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1) == 0) {
+		param += sizeof("sysctl.") - 1;
+	} else {
+		param = sysctl_find_alias(param);
+		if (!param)
+			return 0;
+	}
 
 	remaining = param;
 	ctl = &sysctl_base_table[0];
mm/page_alloc.c +0 -9
index 3c4eb750a199..de7a134b1b8a 100644
@@ -5460,15 +5460,6 @@ static int __parse_numa_zonelist_order(char *s)
 	return 0;
 }
 
-static __init int setup_numa_zonelist_order(char *s)
-{
-	if (!s)
-		return 0;
-
-	return __parse_numa_zonelist_order(s);
-}
-early_param("numa_zonelist_order", setup_numa_zonelist_order);
-
 char numa_zonelist_order[] = "Node";
 
 /*
Signature · 1 line
2.25.1
#2 lore raw

Re: [RFC v2 2/2] kernel/sysctl: support handling command line aliases

Both patches look really great to me. I haven't really checked all the details but from a quick glance they both seem ok.

I would just add a small clarification here. Unless I am mistaken early_param is called earlier than it would be now. But that shouldn't cause any problems because the underlying implementation is just a noop for backward compatibility.

Thanks a lot this looks like a very nice improvement.

On Wed 25-03-20 13:03:45, Vlastimil Babka wrote:
[...]

Quoted 14 lines
mm/page_alloc.c +0 -9
-static __init int setup_numa_zonelist_order(char *s)
-{
-	if (!s)
-		return 0;
-
-	return __parse_numa_zonelist_order(s);
-}
-early_param("numa_zonelist_order", setup_numa_zonelist_order);
-
 char numa_zonelist_order[] = "Node";
 
 /*
Signature · 1 line
2.25.1
Signature · 2 lines
Michal Hocko
SUSE Labs
#3 lore raw

Re: [RFC v2 2/2] kernel/sysctl: support handling command line aliases

Quoted 2 lines · Michal Hocko

Both patches look really great to me. I haven't really checked all the details but from a quick glance they both seem ok.

Thanks.

Quoted 4 lines

I would just add a small clarification here. Unless I am mistaken early_param is called earlier than it would be now. But that shouldn't cause any problems because the underlying implementation is just a noop for backward compatibility.

Yeah, indeed worth noting somewhere explicitly. The conversion can't be done blindly, one has to consider whether the delay compared to early_param can be a disadvantage or not. For example the nmi_watchdog parameter is probably best left as it is?

Quoted 19 lines

Thanks a lot this looks like a very nice improvement.

On Wed 25-03-20 13:03:45, Vlastimil Babka wrote:
[...]

Quoted 14 lines
mm/page_alloc.c +0 -9
-static __init int setup_numa_zonelist_order(char *s)
-{
-	if (!s)
-		return 0;
-
-	return __parse_numa_zonelist_order(s);
-}
-early_param("numa_zonelist_order", setup_numa_zonelist_order);
-
 char numa_zonelist_order[] = "Node";
 
 /*
Signature · 1 line
2.25.1
#4 lore raw

Re: [RFC v2 2/2] kernel/sysctl: support handling command line aliases

Quoted 11 lines · Vlastimil Babka
Quoted 2 lines · Michal Hocko

Both patches look really great to me. I haven't really checked all the details but from a quick glance they both seem ok.

Thanks.

Quoted 4 lines

I would just add a small clarification here. Unless I am mistaken early_param is called earlier than it would be now. But that shouldn't cause any problems because the underlying implementation is just a noop for backward compatibility.

Quoted 4 lines

Yeah, indeed worth noting somewhere explicitly. The conversion can't be done blindly, one has to consider whether the delay compared to early_param can be a disadvantage or not. For example the nmi_watchdog parameter is probably best left as it is?

I wouldn't mind moving nmi_watchdog timeout initialization to later. If there is a usecase to rely on an early initialization then the patch can be reverted but I struggle to think of anything reasonable. If the early init code needs a lonter timeout to prevent from false positives then there is clearly a bug to be better fixed. And a necessary shorter timeout sounds quite exotic to me TBH.

Signature · 2 lines
Michal Hocko
SUSE Labs
#5 lore raw

Re: [RFC v2 2/2] kernel/sysctl: support handling command line aliases

In reply to Vlastimil Babka
Quoted 20 lines · Vlastimil Babka

We can now handle sysctl parameters on kernel command line, but historically some parameters introduced their own command line equivalent, which we don't want to remove for compatibility reasons. We can however convert them to the generic infrastructure with a table translating the legacy command line parameters to their sysctl names, and removing the one-off param handlers.

This patch adds the support and makes the first conversion to demonstrate it, on the (deprecated) numa_zonelist_order parameter.

---
 kernel/sysctl.c | 39 +++++++++++++++++++++++++++++++++++----
 mm/page_alloc.c |  9 ---------
 2 files changed, 35 insertions(+), 13 deletions(-)
kernel/sysctl.c +0 -0
index 18c7f5606d55..fd72853396f9 100644

kernel/sysctl.c static struct ctl_table dev_table[] = { · line 1975

Quoted 5 lines
kernel/sysctl.c +2 -0
@@ -1971,6 +1971,22 @@ static struct ctl_table dev_table[] = {
 	{ }
 };
 
+struct sysctl_alias {
+	char *kernel_param;

const char ...

kernel/sysctl.c static struct ctl_table dev_table[] = { · line 1985

Quoted 10 lines
kernel/sysctl.c +10 -0
+	char *sysctl_param;
+};
+
+/*
+ * Historically some settings had both sysctl and a command line parameter.
+ * With the generic sysctl. parameter support, we can handle them at a single
+ * place and only keep the historical name for compatibility. This is not meant
+ * to add brand new aliases.
+ */
+static struct sysctl_alias sysctl_aliases[] = {

static const ...

Quoted 52 lines
kernel/sysctl.c +23 -4
+	{"numa_zonelist_order",		"vm.numa_zonelist_order" },
+	{ }
+};
+
 int __init sysctl_init(void)
 {
 	struct ctl_table_header *hdr;
@@ -1980,6 +1996,18 @@ int __init sysctl_init(void)
 	return 0;
 }
 
+char *sysctl_find_alias(char *param)
+{
+	struct sysctl_alias *alias;
+
+	for (alias = &sysctl_aliases[0]; alias->kernel_param != NULL; alias++) {
+		if (strcmp(alias->kernel_param, param) == 0)
+			return alias->sysctl_param;
+	}
+
+	return NULL;
+}
+
 /* Set sysctl value passed on kernel command line. */
 int process_sysctl_arg(char *param, char *val,
 			       const char *unused, void *arg)
@@ -1990,10 +2018,13 @@ int process_sysctl_arg(char *param, char *val,
 	loff_t ppos = 0;
 	struct ctl_table *ctl, *found = NULL;
 
-	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
-		return 0;
-
-	param += sizeof("sysctl.") - 1;
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1) == 0) {
+		param += sizeof("sysctl.") - 1;
+	} else {
+		param = sysctl_find_alias(param);
+		if (!param)
+			return 0;
+	}
 
 	remaining = param;
 	ctl = &sysctl_base_table[0];
mm/page_alloc.c +0 -0
index 3c4eb750a199..de7a134b1b8a 100644
@@ -5460,15 +5460,6 @@ static int __parse_numa_zonelist_order(char *s)
 	return 0;
 }
 

mm/page_alloc.c static int __parse_numa_zonelist_order(char *s) · line 5463

Quoted 10 lines
mm/page_alloc.c +0 -9
-static __init int setup_numa_zonelist_order(char *s)
-{
-	if (!s)
-		return 0;
-
-	return __parse_numa_zonelist_order(s);
-}
-early_param("numa_zonelist_order", setup_numa_zonelist_order);
-
 char numa_zonelist_order[] = "Node";

Nice. :) Effectively: -9 lines +1 line for the using aliasing. I think it would be worth identifying the specific requirements for a sysctl alias to be safe to use, and likely in a comment before the alias table:

- boot param parsing must be identical to the sysctl parsing
- temporal changes must be tolerable: i.e. early_param() runs earlier than when the sysctl-in-boot-param runs -- must the variable be set before the code's other __init functions run?
- must be for a non-module code (since we don't have the dynamic support yet)

As it turns out, "numa_zonelist_order" has literally no effect on anything -- it's a parsed but ignored setting:

static int __parse_numa_zonelist_order(char *s)

{
	/*
	 * We used to support different zonlists modes but they turned
	 * out to be just not useful. Let's keep the warning in place
	 * if somebody still use the cmd line parameter so that we do
	 * not fail it silently
	 */
	if (!(*s == 'd' || *s == 'D' || *s == 'n' || *s == 'N')) {
		pr_warn("Ignoring unsupported numa_zonelist_order value: %s\n", s);
		return -EINVAL;
	}
	return 0;
}

But anyway, do you have a way to generate a list of potential candidates?

Signature · 1 line
Kees Cook
#6 lore raw

Re: [RFC v2 2/2] kernel/sysctl: support handling command line aliases

In reply to Vlastimil Babka
Quoted 43 lines · Vlastimil Babka

We can now handle sysctl parameters on kernel command line, but historically some parameters introduced their own command line equivalent, which we don't want to remove for compatibility reasons. We can however convert them to the generic infrastructure with a table translating the legacy command line parameters to their sysctl names, and removing the one-off param handlers.

This patch adds the support and makes the first conversion to demonstrate it, on the (deprecated) numa_zonelist_order parameter.

---
 kernel/sysctl.c | 39 +++++++++++++++++++++++++++++++++++----
 mm/page_alloc.c |  9 ---------
 2 files changed, 35 insertions(+), 13 deletions(-)
kernel/sysctl.c +16 -0
index 18c7f5606d55..fd72853396f9 100644
@@ -1971,6 +1971,22 @@ static struct ctl_table dev_table[] = {
 	{ }
 };
 
+struct sysctl_alias {
+	char *kernel_param;
+	char *sysctl_param;
+};
+
+/*
+ * Historically some settings had both sysctl and a command line parameter.
+ * With the generic sysctl. parameter support, we can handle them at a single
+ * place and only keep the historical name for compatibility. This is not meant
+ * to add brand new aliases.
+ */
+static struct sysctl_alias sysctl_aliases[] = {
+	{"numa_zonelist_order",		"vm.numa_zonelist_order" },
+	{ }
+};
+
 int __init sysctl_init(void)
 {
 	struct ctl_table_header *hdr;

kernel/sysctl.c int __init sysctl_init(void) · line 1999

Quoted 4 lines
kernel/sysctl.c +1 -0
@@ -1980,6 +1996,18 @@ int __init sysctl_init(void)
 	return 0;
 }
 
+char *sysctl_find_alias(char *param)

This function should probably be declared static?

Quoted 55 lines
kernel/sysctl.c +18 -4
+{
+	struct sysctl_alias *alias;
+
+	for (alias = &sysctl_aliases[0]; alias->kernel_param != NULL; alias++) {
+		if (strcmp(alias->kernel_param, param) == 0)
+			return alias->sysctl_param;
+	}
+
+	return NULL;
+}
+
 /* Set sysctl value passed on kernel command line. */
 int process_sysctl_arg(char *param, char *val,
 			       const char *unused, void *arg)
@@ -1990,10 +2018,13 @@ int process_sysctl_arg(char *param, char *val,
 	loff_t ppos = 0;
 	struct ctl_table *ctl, *found = NULL;
 
-	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
-		return 0;
-
-	param += sizeof("sysctl.") - 1;
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1) == 0) {
+		param += sizeof("sysctl.") - 1;
+	} else {
+		param = sysctl_find_alias(param);
+		if (!param)
+			return 0;
+	}
 
 	remaining = param;
 	ctl = &sysctl_base_table[0];
mm/page_alloc.c +0 -9
index 3c4eb750a199..de7a134b1b8a 100644
@@ -5460,15 +5460,6 @@ static int __parse_numa_zonelist_order(char *s)
 	return 0;
 }
 
-static __init int setup_numa_zonelist_order(char *s)
-{
-	if (!s)
-		return 0;
-
-	return __parse_numa_zonelist_order(s);
-}
-early_param("numa_zonelist_order", setup_numa_zonelist_order);
-
 char numa_zonelist_order[] = "Node";
 
 /*
Signature · 1 line
2.25.1
#7 lore raw
In reply to Vlastimil Babka
Quoted 89 lines · Vlastimil Babka

A recently proposed patch to add vm_swappiness command line parameter in addition to existing sysctl [1] made me wonder why we don't have a general support for passing sysctl parameters via command line. Googling found only somebody else wondering the same [2], but I haven't found any prior discussion with reasons why not to do this.

Settings the vm_swappiness issue aside (the underlying issue might be solved in a different way), quick search of kernel-parameters.txt shows there are already some that exist as both sysctl and kernel parameter - hung_task_panic, nmi_watchdog, numa_zonelist_order, traceoff_on_warning. A general mechanism would remove the need to add more of those one-offs and might be handy in situations where configuration by e.g. /etc/sysctl.d/ is impractical. Also after 61a47c1ad3a4 ("sysctl: Remove the sysctl system call") the only way to set sysctl is via procfs, so this would eventually allow small systems to be built without CONFIG_PROC_SYSCTL and still be able to change sysctl parameters.

Hence, this patch adds a new parse_args() pass that looks for parameters prefixed by 'sysctl.' and searches for them in the sysctl ctl_tables. When found, the respective proc handler is invoked. The search is just a naive linear one, to avoid using the whole procfs layer. It should be acceptable, as the cost depends on number of sysctl. parameters passed.

The main limitation of avoiding the procfs layer is however that sysctls dynamically registered by register_sysctl_table() or register_sysctl_paths() cannot currently be set by this method.

The processing is hooked right before the init process is loaded, as some handlers might be more complicated than simple setters and might need some subsystems to be initialized. At the moment the init process can be started and eventually execute a process writing to /proc/sys/ then it should be also fine to do that from the kernel.

[1] https://lore.kernel.org/linux-doc/BL0PR02MB560167492CA4094C91589930E9FC0@BL0PR02MB5601.namprd02.prod.outlook.com/
[2] https://unix.stackexchange.com/questions/558802/how-to-set-sysctl-using-kernel-command-line-parameter

---
v2: - handle any nesting level of parameter name
- add Documentation/admin-guide/kernel-parameters.txt blurb
- alias support for legacy one-off parameters, with first conversion (patch 2)
- still no support for dynamically registed sysctls

 .../admin-guide/kernel-parameters.txt         |  9 +++
 include/linux/sysctl.h                        |  1 +
 init/main.c                                   | 21 +++++++
 kernel/sysctl.c                               | 62 +++++++++++++++++++
 4 files changed, 93 insertions(+)
Documentation/admin-guide/kernel-parameters.txt +9 -0
index c07815d230bc..5076e288f93f 100644
@@ -4793,6 +4793,15 @@
 
 	switches=	[HW,M68k]
 
+	sysctl.*=	[KNL]
+			Set a sysctl parameter right before loading the init
+			process, as if the value was written to the respective
+			/proc/sys/... file. Currently a subset of sysctl
+			parameters is supported that is not registered
+			dynamically. Unrecognized parameters and invalid values
+			are reported in the kernel log.
+			Example: sysctl.vm.swappiness=40
+
 	sysfs.deprecated=0|1 [KNL]
 			Enable/disable old style sysfs layout for old udev
 			on older distributions. When this option is enabled
include/linux/sysctl.h +1 -0
index 02fa84493f23..62ae963a5c0c 100644
@@ -206,6 +206,7 @@ struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
 void unregister_sysctl_table(struct ctl_table_header * table);
 
 extern int sysctl_init(void);
+int process_sysctl_arg(char *param, char *val, const char *unused, void *arg);
 
 extern struct ctl_table sysctl_mount_point[];
 
init/main.c +1 -0
index ee4947af823f..74a094c6b8b9 100644
@@ -1345,6 +1345,25 @@ void __weak free_initmem(void)
 	free_initmem_default(POISON_FREE_INITMEM);
 }
 
+static void do_sysctl_args(void)

init/main.c void __weak free_initmem(void) · line 1358

Quoted 10 lines
init/main.c +10 -0
+{
+#ifdef CONFIG_SYSCTL
+	size_t len = strlen(saved_command_line) + 1;
+	char *command_line;
+
+	command_line = kzalloc(len, GFP_KERNEL);
+	if (!command_line)
+		panic("%s: Failed to allocate %zu bytes\n", __func__, len);
+
+	strcpy(command_line, saved_command_line);

No need to open-code this:

	char *command_line;

	command_line = kstrdup(saved_command_line, GFP_KERNEL);
	if (!command_line)
		panic("%s: Failed to allocate %zu bytes\n", __func__, len);
Quoted 42 lines
init/main.c +10 -0
+
+	parse_args("Setting sysctl args", command_line,
+		   NULL, 0, -1, -1, NULL, process_sysctl_arg);
+
+	kfree(command_line);
+#endif
+}
+
 static int __ref kernel_init(void *unused)
 {
 	int ret;
@@ -1367,6 +1386,8 @@ static int __ref kernel_init(void *unused)
 
 	rcu_end_inkernel_boot();
 
+	do_sysctl_args();
+
 	if (ramdisk_execute_command) {
 		ret = run_init_process(ramdisk_execute_command);
 		if (!ret)
kernel/sysctl.c +14 -0
index ad5b88a53c5a..18c7f5606d55 100644
@@ -1980,6 +1980,68 @@ int __init sysctl_init(void)
 	return 0;
 }
 
+/* Set sysctl value passed on kernel command line. */
+int process_sysctl_arg(char *param, char *val,
+			       const char *unused, void *arg)
+{
+	size_t count;
+	char *remaining;
+	int err;
+	loff_t ppos = 0;
+	struct ctl_table *ctl, *found = NULL;
+
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;
+
+	param += sizeof("sysctl.") - 1;

kernel/sysctl.c int __init sysctl_init(void) · line 2006

Quoted 10 lines
kernel/sysctl.c +10 -0
+
+	remaining = param;
+	ctl = &sysctl_base_table[0];
+
+	while(ctl->procname != 0) {
+		int len = strlen(ctl->procname);
+		if (strncmp(remaining, ctl->procname, len)) {
+			ctl++;
+			continue;
+		}

I think you need to validate that "len" is within "remaining" here first.

kernel/sysctl.c int __init sysctl_init(void) · line 2009

Quoted 3 lines
kernel/sysctl.c +3 -0
+		if (ctl->child) {
+			if (remaining[len] == '.') {
+				remaining += len + 1;

And that "len + 1" is still valid.

Quoted 12 lines
kernel/sysctl.c +12 -0
+				ctl = ctl->child;
+				continue;
+			}
+		} else {
+			if (remaining[len] == '\0') {
+				found = ctl;
+				break;
+			}
+		}
+		ctl++;
+	}
+

kernel/sysctl.c int __init sysctl_init(void) · line 2031

Quoted 10 lines
kernel/sysctl.c +10 -0
+	if (!found) {
+		pr_warn("Unknown sysctl param '%s' on command line", param);
+		return 0;
+	}
+
+	if (!(found->mode & 0200)) {
+		pr_warn("Cannot set sysctl '%s=%s' from command line - not writable",
+			param, val);
+		return 0;
+	}

Oh yes; good call about this writable mode test.

Quoted 19 lines
kernel/sysctl.c +13 -0
+
+	count = strlen(val);
+	err = found->proc_handler(found, 1, val, &count, &ppos);
+
+	if (err)
+		pr_warn("Error %d setting sysctl '%s=%s' from command line",
+			err, param, val);
+
+	pr_debug("Set sysctl '%s=%s' from command line", param, val);
+
+	return 0;
+}
+
 #endif /* CONFIG_SYSCTL */
 
 /*
Signature · 1 line
2.25.1

Outside of the nits and missing bounds check, I like it! :)

Signature · 1 line
Kees Cook
#8 lore raw
Quoted 20 lines · Kees Cook
Quoted 7 lines
init/main.c +1 -0
@@ -1345,6 +1345,25 @@ void __weak free_initmem(void)
 	free_initmem_default(POISON_FREE_INITMEM);
 }
 
+static void do_sysctl_args(void)

init/main.c void __weak free_initmem(void) · line 1358

Quoted 10 lines
init/main.c +10 -0
+{
+#ifdef CONFIG_SYSCTL
+	size_t len = strlen(saved_command_line) + 1;
+	char *command_line;
+
+	command_line = kzalloc(len, GFP_KERNEL);
+	if (!command_line)
+		panic("%s: Failed to allocate %zu bytes\n", __func__, len);
+
+	strcpy(command_line, saved_command_line);

No need to open-code this:

Quoted 6 lines
	char *command_line;

	command_line = kstrdup(saved_command_line, GFP_KERNEL);
	if (!command_line)
		panic("%s: Failed to allocate %zu bytes\n", __func__, len);

Ah, right. I admit I basically copy_pasted some other parse_args user.

Quoted 27 lines
Quoted 27 lines
kernel/sysctl.c +19 -0
index ad5b88a53c5a..18c7f5606d55 100644
@@ -1980,6 +1980,68 @@ int __init sysctl_init(void)
 	return 0;
 }
 
+/* Set sysctl value passed on kernel command line. */
+int process_sysctl_arg(char *param, char *val,
+			       const char *unused, void *arg)
+{
+	size_t count;
+	char *remaining;
+	int err;
+	loff_t ppos = 0;
+	struct ctl_table *ctl, *found = NULL;
+
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;
+
+	param += sizeof("sysctl.") - 1;
+
+	remaining = param;
+	ctl = &sysctl_base_table[0];
+
+	while(ctl->procname != 0) {
Quoted 8 lines

kernel/sysctl.c int __init sysctl_init(void) · line 2006

Quoted 5 lines
kernel/sysctl.c +5 -0
+		int len = strlen(ctl->procname);
+		if (strncmp(remaining, ctl->procname, len)) {
+			ctl++;
+			continue;
+		}

I think you need to validate that "len" is within "remaining" here first.

My reasoning was that if remaining terminates too early, the null byte would be different from non-null byte in ctl->procname and thus strncmp will return it as different?
And the reason I used len in strncmp there is only so it doesn't compare the terminating null, because remaning can continue with ".foo" instead.

Quoted 5 lines

kernel/sysctl.c int __init sysctl_init(void) · line 2009

Quoted 3 lines
kernel/sysctl.c +3 -0
+		if (ctl->child) {
+			if (remaining[len] == '.') {
+				remaining += len + 1;

And that "len + 1" is still valid.

And since we passed strncmp(..., len), remaining[len] might be null byte, but then we can still compare it with '.'.

But C strings are full of landmines.

#9 lore raw
In reply to Vlastimil Babka
Quoted 129 lines · Vlastimil Babka

A recently proposed patch to add vm_swappiness command line parameter in addition to existing sysctl [1] made me wonder why we don't have a general support for passing sysctl parameters via command line. Googling found only somebody else wondering the same [2], but I haven't found any prior discussion with reasons why not to do this.

Settings the vm_swappiness issue aside (the underlying issue might be solved in a different way), quick search of kernel-parameters.txt shows there are already some that exist as both sysctl and kernel parameter - hung_task_panic, nmi_watchdog, numa_zonelist_order, traceoff_on_warning. A general mechanism would remove the need to add more of those one-offs and might be handy in situations where configuration by e.g. /etc/sysctl.d/ is impractical. Also after 61a47c1ad3a4 ("sysctl: Remove the sysctl system call") the only way to set sysctl is via procfs, so this would eventually allow small systems to be built without CONFIG_PROC_SYSCTL and still be able to change sysctl parameters.

Hence, this patch adds a new parse_args() pass that looks for parameters prefixed by 'sysctl.' and searches for them in the sysctl ctl_tables. When found, the respective proc handler is invoked. The search is just a naive linear one, to avoid using the whole procfs layer. It should be acceptable, as the cost depends on number of sysctl. parameters passed.

The main limitation of avoiding the procfs layer is however that sysctls dynamically registered by register_sysctl_table() or register_sysctl_paths() cannot currently be set by this method.

The processing is hooked right before the init process is loaded, as some handlers might be more complicated than simple setters and might need some subsystems to be initialized. At the moment the init process can be started and eventually execute a process writing to /proc/sys/ then it should be also fine to do that from the kernel.

[1] https://lore.kernel.org/linux-doc/BL0PR02MB560167492CA4094C91589930E9FC0@BL0PR02MB5601.namprd02.prod.outlook.com/
[2] https://unix.stackexchange.com/questions/558802/how-to-set-sysctl-using-kernel-command-line-parameter

---
v2: - handle any nesting level of parameter name
- add Documentation/admin-guide/kernel-parameters.txt blurb
- alias support for legacy one-off parameters, with first conversion (patch 2)
- still no support for dynamically registed sysctls

 .../admin-guide/kernel-parameters.txt         |  9 +++
 include/linux/sysctl.h                        |  1 +
 init/main.c                                   | 21 +++++++
 kernel/sysctl.c                               | 62 +++++++++++++++++++
 4 files changed, 93 insertions(+)
Documentation/admin-guide/kernel-parameters.txt +9 -0
index c07815d230bc..5076e288f93f 100644
@@ -4793,6 +4793,15 @@
 
 	switches=	[HW,M68k]
 
+	sysctl.*=	[KNL]
+			Set a sysctl parameter right before loading the init
+			process, as if the value was written to the respective
+			/proc/sys/... file. Currently a subset of sysctl
+			parameters is supported that is not registered
+			dynamically. Unrecognized parameters and invalid values
+			are reported in the kernel log.
+			Example: sysctl.vm.swappiness=40
+
 	sysfs.deprecated=0|1 [KNL]
 			Enable/disable old style sysfs layout for old udev
 			on older distributions. When this option is enabled
include/linux/sysctl.h +1 -0
index 02fa84493f23..62ae963a5c0c 100644
@@ -206,6 +206,7 @@ struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
 void unregister_sysctl_table(struct ctl_table_header * table);
 
 extern int sysctl_init(void);
+int process_sysctl_arg(char *param, char *val, const char *unused, void *arg);
 
 extern struct ctl_table sysctl_mount_point[];
 
init/main.c +21 -0
index ee4947af823f..74a094c6b8b9 100644
@@ -1345,6 +1345,25 @@ void __weak free_initmem(void)
 	free_initmem_default(POISON_FREE_INITMEM);
 }
 
+static void do_sysctl_args(void)
+{
+#ifdef CONFIG_SYSCTL
+	size_t len = strlen(saved_command_line) + 1;
+	char *command_line;
+
+	command_line = kzalloc(len, GFP_KERNEL);
+	if (!command_line)
+		panic("%s: Failed to allocate %zu bytes\n", __func__, len);
+
+	strcpy(command_line, saved_command_line);
+
+	parse_args("Setting sysctl args", command_line,
+		   NULL, 0, -1, -1, NULL, process_sysctl_arg);
+
+	kfree(command_line);
+#endif
+}
+
 static int __ref kernel_init(void *unused)
 {
 	int ret;
@@ -1367,6 +1386,8 @@ static int __ref kernel_init(void *unused)
 
 	rcu_end_inkernel_boot();
 
+	do_sysctl_args();
+
 	if (ramdisk_execute_command) {
 		ret = run_init_process(ramdisk_execute_command);
 		if (!ret)
kernel/sysctl.c +2 -0
index ad5b88a53c5a..18c7f5606d55 100644
@@ -1980,6 +1980,68 @@ int __init sysctl_init(void)
 	return 0;
 }
 
+/* Set sysctl value passed on kernel command line. */
+int process_sysctl_arg(char *param, char *val,

kernel/sysctl.c int __init sysctl_init(void) · line 1994

Quoted 10 lines
kernel/sysctl.c +10 -0
+			       const char *unused, void *arg)
+{
+	size_t count;
+	char *remaining;
+	int err;
+	loff_t ppos = 0;
+	struct ctl_table *ctl, *found = NULL;
+
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;

Is there any way we can use a slash separated path. I know in practice there are not any sysctl names that don't have a '.' in them but why should we artifically limit ourselves?

I guess as long as we don't mind not being able to set sysctls that have a '.' in them it doesn't matter.

kernel/sysctl.c int __init sysctl_init(void) · line 2001

Quoted 7 lines
kernel/sysctl.c +7 -0
+
+	param += sizeof("sysctl.") - 1;
+
+	remaining = param;
+	ctl = &sysctl_base_table[0];
+
+	while(ctl->procname != 0) {
              ^^^^^^^^^^^^^^^^^^

Please either test "while(ctl->procname)" or "while(ctl->procname != NULL)" testing against 0 makes it look like procname is an integer. The style in the kernel is to test against NULL, to make it clear when something is a pointer.

kernel/sysctl.c int __init sysctl_init(void) · line 2002

Quoted 1 line
kernel/sysctl.c +1 -0
+		int len = strlen(ctl->procname);

You should have done "strchr(remaining)" and figured out if there is another '.' and only compared up to that dot. Probably skipping this entry entirely if the two lengths don't match.

Quoted 7 lines
kernel/sysctl.c +7 -0
+		if (strncmp(remaining, ctl->procname, len)) {
+			ctl++;
+			continue;
+		}
+		if (ctl->child) {
+			if (remaining[len] == '.') {
+				remaining += len + 1;

kernel/sysctl.c int __init sysctl_init(void) · line 2019

Quoted 10 lines
kernel/sysctl.c +10 -0
+				ctl = ctl->child;
+				continue;
+			}
+		} else {
+			if (remaining[len] == '\0') {
+				found = ctl;
+				break;
+			}
+		}
+		ctl++;

There should be exactly one match for a name a table. If you get here the code should break, not continue on.

Quoted 14 lines
kernel/sysctl.c +14 -0
+	}
+
+	if (!found) {
+		pr_warn("Unknown sysctl param '%s' on command line", param);
+		return 0;
+	}
+
+	if (!(found->mode & 0200)) {
+		pr_warn("Cannot set sysctl '%s=%s' from command line - not writable",
+			param, val);
+		return 0;
+	}
+
+	count = strlen(val);

kernel/sysctl.c int __init sysctl_init(void) · line 2043

Quoted 10 lines
kernel/sysctl.c +10 -0
+	err = found->proc_handler(found, 1, val, &count, &ppos);
+
+	if (err)
+		pr_warn("Error %d setting sysctl '%s=%s' from command line",
+			err, param, val);
+
+	pr_debug("Set sysctl '%s=%s' from command line", param, val);
+
+	return 0;
+}

You really should be able to have this code live in fs/proc/proc_sysctl.c and utilize lookup_entry.

That should give you the ability to lookup any sysctl. If kernel/sysctl.c is compiled into the kernel proc_sysctl.c is compiled into the kernel. Systems that don't select CONFIG_PROC_SYSCTL won't have any sysctl tables installed at all so they do not make sense to consider or design for.

Further it will be faster to lookup the sysctls using the code from proc_sysctl.c as it constructs an rbtree of all of the entries in a directory. The code might as well take advantage of that for large directories.

Arguably the main sysctl tables in kernel/sysctl.c should be split up so that things are more localized and there is less global state exported throughout the kernel. I certainly don't want to discourage anyone from doing that just so their sysctl can be used on the command line.

Hmm. There is a big gotcha in here and I think it should be mentioned. This code only works because no one has done set_fs(KERNEL_DS). Which means this only works with strings that are kernel addresses essentially by mistake. A big fat comment documenting why it is safe to pass in kernel addresses to a function that takes a "char __user*" pointer would be very good.

Eric

#10 lore raw
Quoted 6 lines · Eric W. Biederman

Hmm. There is a big gotcha in here and I think it should be mentioned. This code only works because no one has done set_fs(KERNEL_DS). Which means this only works with strings that are kernel addresses essentially by mistake. A big fat comment documenting why it is safe to pass in kernel addresses to a function that takes a "char __user*" pointer would be very good.

Yeah, I was going to mention this too just now as I went looking through one of the handlers and was reminded that the args are marked __user. :P

I suspect we might need to add some __force __user markings or something (with a comment as you say above), to keep sparse from going crazy. :)

Signature · 1 line
Kees Cook
#11 lore raw
In reply to Eric W. Biederman
Quoted 1 line · Eric W. Biederman

Vlastimil Babka <[email protected]> writes:

[...]

Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +2 -0
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;

Is there any way we can use a slash separated path. I know in practice there are not any sysctl names that don't have a '.' in them but why should we artifically limit ourselves?

Because this is the normal userspace interface? Why should it be any different from calling sysctl?
[...]

Quoted 4 lines

Further it will be faster to lookup the sysctls using the code from proc_sysctl.c as it constructs an rbtree of all of the entries in a directory. The code might as well take advantage of that for large directories.

Sounds like a good fit for a follow up patch to me. Let's make this as simple as possible for the initial version. But up to Vlastimil of course.

[...]

Quoted 6 lines

Hmm. There is a big gotcha in here and I think it should be mentioned. This code only works because no one has done set_fs(KERNEL_DS). Which means this only works with strings that are kernel addresses essentially by mistake. A big fat comment documenting why it is safe to pass in kernel addresses to a function that takes a "char __user*" pointer would be very good.

Agreed

Signature · 2 lines
Michal Hocko
SUSE Labs
#12 lore raw
Quoted 10 lines · Michal Hocko
Quoted 1 line · Eric W. Biederman

Vlastimil Babka <[email protected]> writes:

[...]

Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +2 -0
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;

Is there any way we can use a slash separated path. I know in practice there are not any sysctl names that don't have a '.' in them but why should we artifically limit ourselves?

Quoted 2 lines

Because this is the normal userspace interface? Why should it be any different from calling sysctl?

Right. The common method from userspace is dot-separated (which I agree is weird, but it's been like this for ages: see manpages sysctl(8) and sysctl.conf(5) for the details and examples). While "/" is accepted by sysctl, the files shipped in /etc/sysctl.d/ are all using "." separators.

Signature · 1 line
Kees Cook
#13 lore raw
In reply to Michal Hocko
Quoted 10 lines · Michal Hocko
Quoted 1 line · Eric W. Biederman

Vlastimil Babka <[email protected]> writes:

[...]

Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +2 -0
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;

Is there any way we can use a slash separated path. I know in practice there are not any sysctl names that don't have a '.' in them but why should we artifically limit ourselves?

Quoted 3 lines

Because this is the normal userspace interface? Why should it be any different from calling sysctl?
[...]

Why should the kernel command line implement userspace whims? I was thinking something like: "sysctl/kernel/max_lock_depth=2048" doesn't look too bad and it makes things like reusing our kernel internal helpers much easier.

Plus it suggest that we could do the same for sysfs files:

	"sysfs/kernel/fscaps=1"

And the code could be same for both cases except for the filesystem prefix.

Quoted 7 lines
Quoted 4 lines

Further it will be faster to lookup the sysctls using the code from proc_sysctl.c as it constructs an rbtree of all of the entries in a directory. The code might as well take advantage of that for large directories.

Sounds like a good fit for a follow up patch to me. Let's make this as simple as possible for the initial version. But up to Vlastimil of course.

I would argue that reusing proc_sysctl.c:lookup_entry() should make the code simpler, and easier to reason about.

Especially given the bugs in the first version with a sysctl path. A clean separation between separating the path from into pieces and looking up those pieces should make the code more robust.

That plus I want to get very far away from the incorrect idea that you can have sysctls without compiling in proc support. That is not how the code works, that is not how the code is tested.

It is also worth pointing out that:

	proc_mnt = kern_mount(proc_fs_type);
        for_each_sysctl_cmdline() {
        	...
		file = file_open_root(proc_mnt->mnt_root, proc_mnt, sysctl_path, O_WRONLY, 0);
		kernel_write(file, value, value_len);
        }
        kern_umount(proc_mnt);

Is not an unreasonable implementation.

There are problems with a persistent mount of proc in that it forces userspace not to use any proc mount options. But a temporary mount of proc to deal with command line options is not at all unreasonable. Plus it looks like we can have kern_write do all of the kernel/user buffer silliness.

Quoted 10 lines

[...]

Quoted 6 lines

Hmm. There is a big gotcha in here and I think it should be mentioned. This code only works because no one has done set_fs(KERNEL_DS). Which means this only works with strings that are kernel addresses essentially by mistake. A big fat comment documenting why it is safe to pass in kernel addresses to a function that takes a "char __user*" pointer would be very good.

Agreed

Eric

#14 lore raw
Quoted 4 lines · Eric W. Biederman
Quoted 1 line · Eric W. Biederman

plus I want to get very far away from the incorrect idea that you can have sysctls without compiling in proc support. That is not how the code works, that is not how the code is tested.

Agreed.

Quoted 11 lines

It is also worth pointing out that:

	proc_mnt = kern_mount(proc_fs_type);
        for_each_sysctl_cmdline() {
        	...
		file = file_open_root(proc_mnt->mnt_root, proc_mnt, sysctl_path, O_WRONLY, 0);
		kernel_write(file, value, value_len);
        }
        kern_umount(proc_mnt);

Is not an unreasonable implementation.

This:

Quoted 5 lines

There are problems with a persistent mount of proc in that it forces userspace not to use any proc mount options. But a temporary mount of proc to deal with command line options is not at all unreasonable. Plus it looks like we can have kern_write do all of the kernel/user buffer silliness.

Is a bit of tribal knowledge worth documenting for the approach taken forward. Vlastimil can you add a little comment mentioning some of this logic?

Luis

#15 lore raw
In reply to Michal Hocko
Quoted 10 lines · Michal Hocko
Quoted 1 line · Eric W. Biederman

Vlastimil Babka <[email protected]> writes:

[...]

Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +2 -0
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;

Is there any way we can use a slash separated path. I know in practice there are not any sysctl names that don't have a '.' in them but why should we artifically limit ourselves?

Quoted 3 lines

Because this is the normal userspace interface? Why should it be any different from calling sysctl?
[...]

Imho, we should use ".". Kernel developers aren't the ones setting these options, admins are and if I think back to the times doing that as a job at uni I'd be very confused if I learned that I get to set sysctl options through the kernel command but need to use yet another format than what I usually do to set those from the shell. Consistency is most of the times to be preferred imho.

Also, the kernel docs illustrate that the "." syntax is used for other keys as well (e.g. acpi.<option>) and userspace options passed via the kernel command line have standardized on the "." format as well, e.g. systemd appends in the same format (e.g. systemd.unified_cgroup_hierarchy, systemd.unit what have you).

Christian

#16 lore raw
Quoted 15 lines · Christian Brauner
Quoted 10 lines · Michal Hocko
Quoted 1 line · Eric W. Biederman

Vlastimil Babka <[email protected]> writes:

[...]

Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +2 -0
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;

Is there any way we can use a slash separated path. I know in practice there are not any sysctl names that don't have a '.' in them but why should we artifically limit ourselves?

Quoted 3 lines

Because this is the normal userspace interface? Why should it be any different from calling sysctl?
[...]

Quoted 6 lines

Imho, we should use ".". Kernel developers aren't the ones setting these options, admins are and if I think back to the times doing that as a job at uni I'd be very confused if I learned that I get to set sysctl options through the kernel command but need to use yet another format than what I usually do to set those from the shell. Consistency is most of the times to be preferred imho.

Absolutely agreed! Even if sysctl can consume / instead of ., which was a news to me btw, the majority of the usage is with `.'

Signature · 2 lines
Michal Hocko
SUSE Labs
#17 lore raw
In reply to Eric W. Biederman
Quoted 7 lines · Eric W. Biederman
Quoted 6 lines · Vlastimil Babka
kernel/sysctl.c +0 -0
@@ -1980,6 +1980,68 @@ int __init sysctl_init(void)
 	return 0;
 }
 
Quoted 14 lines

kernel/sysctl.c int __init sysctl_init(void) · line 1994

Quoted 12 lines
kernel/sysctl.c +12 -0
+/* Set sysctl value passed on kernel command line. */
+int process_sysctl_arg(char *param, char *val,
+			       const char *unused, void *arg)
+{
+	size_t count;
+	char *remaining;
+	int err;
+	loff_t ppos = 0;
+	struct ctl_table *ctl, *found = NULL;
+
+	if (strncmp(param, "sysctl.", sizeof("sysctl.") - 1))
+		return 0;

Is there any way we can use a slash separated path. I know

We could, but as others explained, people and tools are used to the dot separation, so I think the only sensible options are supporting only dot, or both dot and slash.

Quoted 2 lines

in practice there are not any sysctl names that don't have a '.' in them but why should we artifically limit ourselves?

Existing tools would probably break (or perhaps sysctl(8) is smarter than I think, dunno).

Quoted 2 lines

I guess as long as we don't mind not being able to set sysctls that have a '.' in them it doesn't matter.

Right.

Quoted 9 lines

kernel/sysctl.c int __init sysctl_init(void) · line 2001

Quoted 7 lines
kernel/sysctl.c +7 -0
+
+	param += sizeof("sysctl.") - 1;
+
+	remaining = param;
+	ctl = &sysctl_base_table[0];
+
+	while(ctl->procname != 0) {
              ^^^^^^^^^^^^^^^^^^
Quoted 4 lines

Please either test "while(ctl->procname)" or "while(ctl->procname != NULL)" testing against 0 makes it look like procname is an integer. The style in the kernel is to test against NULL, to make it clear when something is a pointer.

OK

Quoted 5 lines

kernel/sysctl.c int __init sysctl_init(void) · line 2002

Quoted 1 line
kernel/sysctl.c +1 -0
+		int len = strlen(ctl->procname);

You should have done "strchr(remaining)" and figured out if there is another '.' and only compared up to that dot. Probably skipping this entry entirely if the two lengths don't match.

That's also possible, but AFAICS my code works as intended, as I explained in a reply to Kees, and also below:

Quoted 12 lines
Quoted 12 lines
kernel/sysctl.c +12 -0
+		if (strncmp(remaining, ctl->procname, len)) {
+			ctl++;
+			continue;
+		}
+		if (ctl->child) {
+			if (remaining[len] == '.') {
+				remaining += len + 1;
+				ctl = ctl->child;
+				continue;
+			}
+		} else {
+			if (remaining[len] == '\0') {
Quoted 8 lines

kernel/sysctl.c int __init sysctl_init(void) · line 2019

Quoted 5 lines
kernel/sysctl.c +5 -0
+				found = ctl;
+				break;
+			}
+		}
+		ctl++;

There should be exactly one match for a name a table. If you get here the code should break, not continue on.

If there existed e.g. both "vm.swap" and "vm.swappiness" options and user passed "vm.swappiness=10", but the "swap" ctl entry was encountered first, it will succeed the strncmp(), but then realize "swap" was just a prefix of what user specified (remaining[len] is not '\0') and hence continue serching for other matches.

Quoted 28 lines
Quoted 14 lines
kernel/sysctl.c +14 -0
+	}
+
+	if (!found) {
+		pr_warn("Unknown sysctl param '%s' on command line", param);
+		return 0;
+	}
+
+	if (!(found->mode & 0200)) {
+		pr_warn("Cannot set sysctl '%s=%s' from command line - not writable",
+			param, val);
+		return 0;
+	}
+
+	count = strlen(val);

kernel/sysctl.c int __init sysctl_init(void) · line 2043

Quoted 10 lines
kernel/sysctl.c +10 -0
+	err = found->proc_handler(found, 1, val, &count, &ppos);
+
+	if (err)
+		pr_warn("Error %d setting sysctl '%s=%s' from command line",
+			err, param, val);
+
+	pr_debug("Set sysctl '%s=%s' from command line", param, val);
+
+	return 0;
+}

You really should be able to have this code live in fs/proc/proc_sysctl.c and utilize lookup_entry.

Quoted 5 lines

That should give you the ability to lookup any sysctl. If kernel/sysctl.c is compiled into the kernel proc_sysctl.c is compiled into the kernel. Systems that don't select CONFIG_PROC_SYSCTL won't have any sysctl tables installed at all so they do not make sense to consider or design for.

I see. In fact one reason why I tried to avoid the proc stuff was your commit 61a47c1ad3a4 ("sysctl: Remove the sysctl system call") and this part:

Quoted 3 lines

As this removes one of the few uses of the internal kernel mount of proc I hope this allows for even more simplifications of the proc filesystem.

But if you now suggest using the kernel mount then sure, it I don't object make the code simpler and handle all sysctls.

Quoted 9 lines

Further it will be faster to lookup the sysctls using the code from proc_sysctl.c as it constructs an rbtree of all of the entries in a directory. The code might as well take advantage of that for large directories.

Arguably the main sysctl tables in kernel/sysctl.c should be split up so that things are more localized and there is less global state exported throughout the kernel. I certainly don't want to discourage anyone from doing that just so their sysctl can be used on the command line.

Fair point.

Quoted 6 lines

Hmm. There is a big gotcha in here and I think it should be mentioned. This code only works because no one has done set_fs(KERNEL_DS). Which means this only works with strings that are kernel addresses essentially by mistake. A big fat comment documenting why it is safe to pass in kernel addresses to a function that takes a "char __user*" pointer would be very good.

Thanks, didn't realize that.

Quoted 2 lines

Eric