From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

Basically everything was done but I removed all element
initializers from the trailing entries to make it clear
the entire last entry should be zero filled.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 arch/x86_64/kernel/vsyscall.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c
index 2433d6f..c0e2b48 100644
--- a/arch/x86_64/kernel/vsyscall.c
+++ b/arch/x86_64/kernel/vsyscall.c
@@ -235,13 +235,13 @@ static ctl_table kernel_table2[] = {
 	  .data = &sysctl_vsyscall, .maxlen = sizeof(int), .mode = 0644,
 	  .strategy = vsyscall_sysctl_nostrat,
 	  .proc_handler = vsyscall_sysctl_change },
-	{ 0, }
+	{}
 };
 
 static ctl_table kernel_root_table2[] = {
 	{ .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555,
 	  .child = kernel_table2 },
-	{ 0 },
+	{}
 };
 
 #endif
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 arch/x86_64/mm/init.c |   22 ++++++++++++++++------
 1 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 65aa66c..a04535d 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -711,15 +711,25 @@ int kern_addr_valid(unsigned long addr)
 extern int exception_trace, page_fault_trace;
 
 static ctl_table debug_table2[] = {
-	{ 99, "exception-trace", &exception_trace, sizeof(int), 0644, NULL,
-	  proc_dointvec },
-	{ 0, }
+	{
+		.ctl_name	= 99,
+		.procname	= "exception-trace",
+		.data		= &exception_trace,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{}
 }; 
 
 static ctl_table debug_root_table2[] = { 
-	{ .ctl_name = CTL_DEBUG, .procname = "debug", .mode = 0555, 
-	   .child = debug_table2 }, 
-	{ 0 }, 
+	{
+		.ctl_name = CTL_DEBUG,
+		.procname = "debug",
+		.mode = 0555,
+		.child = debug_table2
+	}, 
+	{}
 }; 
 
 static __init int x8664_sysctl_init(void)
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

In the binary sysctl interface the hpet driver was claiming to
be the cdrom driver.  This is a no-no so remove support for the
binary interface.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 drivers/char/hpet.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index 20dc3be..81be1db 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -703,7 +703,7 @@ int hpet_control(struct hpet_task *tp, unsigned int cmd, unsigned long arg)
 
 static ctl_table hpet_table[] = {
 	{
-	 .ctl_name = 1,
+	 .ctl_name = CTL_UNNUMBERED,
 	 .procname = "max-user-freq",
 	 .data = &hpet_max_freq,
 	 .maxlen = sizeof(int),
@@ -715,7 +715,7 @@ static ctl_table hpet_table[] = {
 
 static ctl_table hpet_root[] = {
 	{
-	 .ctl_name = 1,
+	 .ctl_name = CTL_UNNUMBERED,
 	 .procname = "hpet",
 	 .maxlen = 0,
 	 .mode = 0555,
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

The real time clock driver was using the binary number reserved
for cdroms in the sysctl binary number interface, which is a no-no.
So since the sysctl binary interface is wrong remove it.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 drivers/char/rtc.c |    6 ++----
 1 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c
index 664f36c..df11289 100644
--- a/drivers/char/rtc.c
+++ b/drivers/char/rtc.c
@@ -282,7 +282,7 @@ irqreturn_t rtc_interrupt(int irq, void *dev_id)
  */
 static ctl_table rtc_table[] = {
 	{
-		.ctl_name	= 1,
+		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "max-user-freq",
 		.data		= &rtc_max_user_freq,
 		.maxlen		= sizeof(int),
@@ -294,9 +294,8 @@ static ctl_table rtc_table[] = {
 
 static ctl_table rtc_root[] = {
 	{
-		.ctl_name	= 1,
+		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "rtc",
-		.maxlen		= 0,
 		.mode		= 0555,
 		.child		= rtc_table,
 	},
@@ -307,7 +306,6 @@ static ctl_table dev_root[] = {
 	{
 		.ctl_name	= CTL_DEV,
 		.procname	= "dev",
-		.maxlen		= 0,
 		.mode		= 0555,
 		.child		= rtc_root,
 	},
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 drivers/net/wireless/arlan-proc.c |    2 +-
 include/linux/sysctl.h            |    1 +
 2 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/drivers/net/wireless/arlan-proc.c b/drivers/net/wireless/arlan-proc.c
index 5fa9854..20499a6 100644
--- a/drivers/net/wireless/arlan-proc.c
+++ b/drivers/net/wireless/arlan-proc.c
@@ -1216,7 +1216,7 @@ static ctl_table arlan_table[MAX_ARLANS + 1] =
 static ctl_table arlan_root_table[] =
 {
 	{
-		.ctl_name	= 254,
+		.ctl_name	= CTL_ARLAN,
 		.procname	= "arlan",
 		.maxlen		= 0,
 		.mode		= 0555,
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 56d0161..f4ba72e 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -70,6 +70,7 @@ enum
 	CTL_BUS=8,		/* Busses */
 	CTL_ABI=9,		/* Binary emulation */
 	CTL_CPU=10,		/* CPU stuff (speed scaling, etc) */
+	CTL_ARLAN=254,		/* arlan wireless driver */
 	CTL_SUNRPC=7249,	/* sunrpc debug */
 	CTL_PM=9899,		/* frv power management */
 	CTL_FRV=9898,		/* frv specific sysctls */
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 drivers/parport/procfs.c |  264 +++++++++++++++++++++++++++++++++-------------
 1 files changed, 189 insertions(+), 75 deletions(-)

diff --git a/drivers/parport/procfs.c b/drivers/parport/procfs.c
index 2e744a2..5337789 100644
--- a/drivers/parport/procfs.c
+++ b/drivers/parport/procfs.c
@@ -233,12 +233,12 @@ static int do_hardware_modes (ctl_table *table, int write,
 	return copy_to_user(result, buffer, len) ? -EFAULT : 0;
 }
 
-#define PARPORT_PORT_DIR(child) { 0, NULL, NULL, 0, 0555, child }
-#define PARPORT_PARPORT_DIR(child) { DEV_PARPORT, "parport", \
-                                     NULL, 0, 0555, child }
-#define PARPORT_DEV_DIR(child) { CTL_DEV, "dev", NULL, 0, 0555, child }
-#define PARPORT_DEVICES_ROOT_DIR  { DEV_PARPORT_DEVICES, "devices", \
-                                    NULL, 0, 0555, NULL }
+#define PARPORT_PORT_DIR(CHILD) { .ctl_name = 0, .procname = NULL, .mode = 0555, .child = CHILD }
+#define PARPORT_PARPORT_DIR(CHILD) { .ctl_name = DEV_PARPORT, .procname = "parport", \
+                                     .mode = 0555, .child = CHILD }
+#define PARPORT_DEV_DIR(CHILD) { .ctl_name = CTL_DEV, .procname = "dev", .mode = 0555, .child = CHILD }
+#define PARPORT_DEVICES_ROOT_DIR  { .ctl_name = DEV_PARPORT_DEVICES, .procname = "devices", \
+                                    .mode = 0555, .child = NULL }
 
 static const unsigned long parport_min_timeslice_value =
 PARPORT_MIN_TIMESLICE_VALUE;
@@ -263,50 +263,118 @@ struct parport_sysctl_table {
 };
 
 static const struct parport_sysctl_table parport_sysctl_template = {
-	NULL,
+	.sysctl_header = NULL,
         {
-		{ DEV_PARPORT_SPINTIME, "spintime",
-		  NULL, sizeof(int), 0644, NULL,
-		  &proc_dointvec_minmax, NULL, NULL,
-		  (void*) &parport_min_spintime_value,
-		  (void*) &parport_max_spintime_value },
-		{ DEV_PARPORT_BASE_ADDR, "base-addr",
-		  NULL, 0, 0444, NULL,
-		  &do_hardware_base_addr },
-		{ DEV_PARPORT_IRQ, "irq",
-		  NULL, 0, 0444, NULL,
-		  &do_hardware_irq },
-		{ DEV_PARPORT_DMA, "dma",
-		  NULL, 0, 0444, NULL,
-		  &do_hardware_dma },
-		{ DEV_PARPORT_MODES, "modes",
-		  NULL, 0, 0444, NULL,
-		  &do_hardware_modes },
+		{
+			.ctl_name	= DEV_PARPORT_SPINTIME,
+			.procname	= "spintime",
+			.data		= NULL,
+			.maxlen		= sizeof(int),
+			.mode		= 0644,
+			.proc_handler	= &proc_dointvec_minmax,
+			.extra1		= (void*) &parport_min_spintime_value,
+			.extra2		= (void*) &parport_max_spintime_value
+		},
+		{
+			.ctl_name	= DEV_PARPORT_BASE_ADDR,
+			.procname	= "base-addr",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_hardware_base_addr
+		},
+		{
+			.ctl_name	= DEV_PARPORT_IRQ,
+			.procname	= "irq",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_hardware_irq
+		},
+		{
+			.ctl_name	= DEV_PARPORT_DMA,
+			.procname	= "dma",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_hardware_dma
+		},
+		{
+			.ctl_name	= DEV_PARPORT_MODES,
+			.procname	= "modes",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_hardware_modes
+		},
 		PARPORT_DEVICES_ROOT_DIR,
 #ifdef CONFIG_PARPORT_1284
-		{ DEV_PARPORT_AUTOPROBE, "autoprobe",
-		  NULL, 0, 0444, NULL,
-		  &do_autoprobe },
-		{ DEV_PARPORT_AUTOPROBE + 1, "autoprobe0",
-		 NULL, 0, 0444, NULL,
-		 &do_autoprobe },
-		{ DEV_PARPORT_AUTOPROBE + 2, "autoprobe1",
-		  NULL, 0, 0444, NULL,
-		  &do_autoprobe },
-		{ DEV_PARPORT_AUTOPROBE + 3, "autoprobe2",
-		  NULL, 0, 0444, NULL,
-		  &do_autoprobe },
-		{ DEV_PARPORT_AUTOPROBE + 4, "autoprobe3",
-		  NULL, 0, 0444, NULL,
-		  &do_autoprobe },
+		{
+			.ctl_name	= DEV_PARPORT_AUTOPROBE,
+			.procname	= "autoprobe",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_autoprobe
+		},
+		{
+			.ctl_name	= DEV_PARPORT_AUTOPROBE + 1,
+			.procname	= "autoprobe0",
+			.data		= NULL,
+			.maxlen		= 0,
+			.maxlen		= 0444,
+			.proc_handler	=  &do_autoprobe
+		},
+		{
+			.ctl_name	= DEV_PARPORT_AUTOPROBE + 2,
+			.procname	= "autoprobe1",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_autoprobe
+		},
+		{
+			.ctl_name	= DEV_PARPORT_AUTOPROBE + 3,
+			.procname	= "autoprobe2",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_autoprobe
+		},
+		{
+			.ctl_name	= DEV_PARPORT_AUTOPROBE + 4,
+			.procname	= "autoprobe3",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_autoprobe
+		},
 #endif /* IEEE 1284 support */
-		{0}
+		{}
 	},
-	{ {DEV_PARPORT_DEVICES_ACTIVE, "active", NULL, 0, 0444, NULL,
-	  &do_active_device }, {0}},
-	{ PARPORT_PORT_DIR(NULL), {0}},
-	{ PARPORT_PARPORT_DIR(NULL), {0}},
-	{ PARPORT_DEV_DIR(NULL), {0}}
+	{
+		{
+			.ctl_name	= DEV_PARPORT_DEVICES_ACTIVE,
+			.procname	= "active",
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0444,
+			.proc_handler	= &do_active_device
+		},
+		{}
+	},
+	{
+		PARPORT_PORT_DIR(NULL),
+		{}
+	},
+	{
+		PARPORT_PARPORT_DIR(NULL),
+		{}
+	},
+	{
+		PARPORT_DEV_DIR(NULL),
+		{}
+	}
 };
 
 struct parport_device_sysctl_table
@@ -322,19 +390,46 @@ struct parport_device_sysctl_table
 
 static const struct parport_device_sysctl_table
 parport_device_sysctl_template = {
-	NULL,
+	.sysctl_header = NULL,
+	{
+		{
+			.ctl_name 	= DEV_PARPORT_DEVICE_TIMESLICE,
+			.procname 	= "timeslice",
+			.data		= NULL,
+			.maxlen		= sizeof(int),
+			.mode		= 0644,
+			.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
+			.extra1		= (void*) &parport_min_timeslice_value,
+			.extra2		= (void*) &parport_max_timeslice_value
+		},
+	},
+	{
+		{
+			.ctl_name	= 0,
+			.procname	= NULL,
+			.data		= NULL,
+			.maxlen		= 0,
+			.mode		= 0555,
+			.child		= NULL
+		},
+		{}
+	},
 	{
-		{ DEV_PARPORT_DEVICE_TIMESLICE, "timeslice",
-		  NULL, sizeof(int), 0644, NULL,
-		  &proc_doulongvec_ms_jiffies_minmax, NULL, NULL,
-		  (void*) &parport_min_timeslice_value,
-		  (void*) &parport_max_timeslice_value },
+		PARPORT_DEVICES_ROOT_DIR,
+		{}
+	},
+	{
+		PARPORT_PORT_DIR(NULL),
+		{}
 	},
-	{ {0, NULL, NULL, 0, 0555, NULL}, {0}},
-	{ PARPORT_DEVICES_ROOT_DIR, {0}},
-	{ PARPORT_PORT_DIR(NULL), {0}},
-	{ PARPORT_PARPORT_DIR(NULL), {0}},
-	{ PARPORT_DEV_DIR(NULL), {0}}
+	{
+		PARPORT_PARPORT_DIR(NULL),
+		{}
+	},
+	{
+		PARPORT_DEV_DIR(NULL),
+		{}
+	}
 };
 
 struct parport_default_sysctl_table
@@ -351,28 +446,47 @@ extern int parport_default_spintime;
 
 static struct parport_default_sysctl_table
 parport_default_sysctl_table = {
-	NULL,
+	.sysctl_header	= NULL,
+	{
+		{
+			.ctl_name	= DEV_PARPORT_DEFAULT_TIMESLICE,
+			.procname	= "timeslice",
+			.data		= &parport_default_timeslice,
+			.maxlen		= sizeof(parport_default_timeslice),
+			.mode		= 0644,
+			.proc_handler	= &proc_doulongvec_ms_jiffies_minmax,
+			.extra1		= (void*) &parport_min_timeslice_value,
+			.extra2		= (void*) &parport_max_timeslice_value
+		},
+		{
+			.ctl_name	= DEV_PARPORT_DEFAULT_SPINTIME,
+			.procname	= "spintime",
+			.data		= &parport_default_spintime,
+			.maxlen		= sizeof(parport_default_spintime),
+			.mode		= 0644,
+			.proc_handler	= &proc_dointvec_minmax,
+			.extra1		= (void*) &parport_min_spintime_value,
+			.extra2		= (void*) &parport_max_spintime_value
+		},
+		{}
+	},
 	{
-		{ DEV_PARPORT_DEFAULT_TIMESLICE, "timeslice",
-		  &parport_default_timeslice,
-		  sizeof(parport_default_timeslice), 0644, NULL,
-		  &proc_doulongvec_ms_jiffies_minmax, NULL, NULL,
-		  (void*) &parport_min_timeslice_value,
-		  (void*) &parport_max_timeslice_value },
-		{ DEV_PARPORT_DEFAULT_SPINTIME, "spintime",
-		  &parport_default_spintime,
-		  sizeof(parport_default_spintime), 0644, NULL,
-		  &proc_dointvec_minmax, NULL, NULL,
-		  (void*) &parport_min_spintime_value,
-		  (void*) &parport_max_spintime_value },
-		{0}
+		{
+			.ctl_name	= DEV_PARPORT_DEFAULT,
+			.procname	= "default",
+			.mode		= 0555,
+			.child		= parport_default_sysctl_table.vars
+		},
+		{}
 	},
-	{ { DEV_PARPORT_DEFAULT, "default", NULL, 0, 0555,
-	    parport_default_sysctl_table.vars },{0}},
 	{
-	PARPORT_PARPORT_DIR(parport_default_sysctl_table.default_dir), 
-	{0}},
-	{ PARPORT_DEV_DIR(parport_default_sysctl_table.parport_dir), {0}}
+		PARPORT_PARPORT_DIR(parport_default_sysctl_table.default_dir), 
+		{}
+	},
+	{
+		PARPORT_DEV_DIR(parport_default_sysctl_table.parport_dir), 
+		{}
+	}
 };
 
 
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

Will converting the coda sysctl initializers I discovered that
it is yet another user of sysctl that was stomping CTL_KERN.
So off with it's sys_sysctl support since it wasn't done
in a supportable way.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 fs/coda/sysctl.c |   58 ++++++++++++++++++++++++++++++++++++++++++++---------
 1 files changed, 48 insertions(+), 10 deletions(-)

diff --git a/fs/coda/sysctl.c b/fs/coda/sysctl.c
index 1c82e9a..df682e2 100644
--- a/fs/coda/sysctl.c
+++ b/fs/coda/sysctl.c
@@ -32,8 +32,6 @@
 
 static struct ctl_table_header *fs_table_header;
 
-#define FS_CODA         1       /* Coda file system */
-
 #define CODA_TIMEOUT    3       /* timeout on upcalls to become intrble */
 #define CODA_HARD       5       /* mount type "hard" or "soft" */
 #define CODA_VFS 	 6       /* vfs statistics */
@@ -184,17 +182,57 @@ static int coda_cache_inv_stats_get_info( char * buffer, char ** start,
 }
 
 static ctl_table coda_table[] = {
- 	{CODA_TIMEOUT, "timeout", &coda_timeout, sizeof(int), 0644, NULL, &proc_dointvec},
- 	{CODA_HARD, "hard", &coda_hard, sizeof(int), 0644, NULL, &proc_dointvec},
- 	{CODA_VFS, "vfs_stats", NULL, 0, 0644, NULL, &do_reset_coda_vfs_stats},
- 	{CODA_CACHE_INV, "cache_inv_stats", NULL, 0, 0644, NULL, &do_reset_coda_cache_inv_stats},
- 	{CODA_FAKE_STATFS, "fake_statfs", &coda_fake_statfs, sizeof(int), 0600, NULL, &proc_dointvec},
-	{ 0 }
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "timeout",
+		.data		= &coda_timeout,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "hard",
+		.data		= &coda_hard,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "vfs_stats",
+		.data		= NULL,
+		.maxlen		= 0,
+		.mode		= 0644,
+		.proc_handler	= &do_reset_coda_vfs_stats
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "cache_inv_stats",
+		.data		= NULL,
+		.maxlen		= 0,
+		.mode		= 0644,
+		.proc_handler	= &do_reset_coda_cache_inv_stats
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "fake_statfs",
+		.data		= &coda_fake_statfs,
+		.maxlen		= sizeof(int),
+		.mode		= 0600,
+		.proc_handler	= &proc_dointvec
+	},
+	{}
 };
 
 static ctl_table fs_table[] = {
-       {FS_CODA, "coda",    NULL, 0, 0555, coda_table},
-       {0}
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "coda",
+		.mode		= 0555,
+		.child		= coda_table
+	},
+	{}
 };
 
 
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

Putting ntfs-debug under FS_NRINODE was not a kosher thing to do
so don't give it any binary number.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 fs/ntfs/sysctl.c |   24 ++++++++++++++++--------
 1 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/fs/ntfs/sysctl.c b/fs/ntfs/sysctl.c
index 1c23138..bc217de 100644
--- a/fs/ntfs/sysctl.c
+++ b/fs/ntfs/sysctl.c
@@ -33,20 +33,28 @@
 #include "sysctl.h"
 #include "debug.h"
 
-#define FS_NTFS	1
-
 /* Definition of the ntfs sysctl. */
 static ctl_table ntfs_sysctls[] = {
-	{ FS_NTFS, "ntfs-debug",		/* Binary and text IDs. */
-	  &debug_msgs,sizeof(debug_msgs),	/* Data pointer and size. */
-	  0644,	NULL, &proc_dointvec },		/* Mode, child, proc handler. */
-	{ 0 }
+	{
+		.ctl_name	= CTL_UNUMBERED,	/* Binary and text IDs. */
+		.procname	= "ntfs-debug",
+		.data		= &debug_msgs,		/* Data pointer and size. */
+		.maxlen		= sizeof(debug_msgs),
+		.mode		= 0644,			/* Mode, proc handler. */
+		.proc_handler	= &proc_dointvec
+	},
+	{}
 };
 
 /* Define the parent directory /proc/sys/fs. */
 static ctl_table sysctls_root[] = {
-	{ CTL_FS, "fs", NULL, 0, 0555, ntfs_sysctls },
-	{ 0 }
+	{
+		.ctl_name	= CTL_FS,
+		.procname	= "fs",
+		.mode		= 0555,
+		.child		= ntfs_sysctls
+	},
+	{}
 };
 
 /* Storage for the sysctls header. */
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

ocfs2 was did not have the binary number it uses under CTL_FS
registered in sysctl.h.  Register it to avoid future conflicts,
and change the name of the definition to be in line with the
rest of the sysctl numbers.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 fs/ocfs2/cluster/nodemanager.c |    4 ++--
 fs/ocfs2/cluster/nodemanager.h |    3 +--
 include/linux/sysctl.h         |    1 +
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c
index b17333a..df763c7 100644
--- a/fs/ocfs2/cluster/nodemanager.c
+++ b/fs/ocfs2/cluster/nodemanager.c
@@ -55,7 +55,7 @@ static ctl_table ocfs2_nm_table[] = {
 
 static ctl_table ocfs2_mod_table[] = {
 	{
-		.ctl_name	= KERN_OCFS2_NM,
+		.ctl_name	= FS_OCFS2_NM,
 		.procname	= "nm",
 		.data		= NULL,
 		.maxlen		= 0,
@@ -67,7 +67,7 @@ static ctl_table ocfs2_mod_table[] = {
 
 static ctl_table ocfs2_kern_table[] = {
 	{
-		.ctl_name	= KERN_OCFS2,
+		.ctl_name	= FS_OCFS2,
 		.procname	= "ocfs2",
 		.data		= NULL,
 		.maxlen		= 0,
diff --git a/fs/ocfs2/cluster/nodemanager.h b/fs/ocfs2/cluster/nodemanager.h
index 8fb23ca..0705221 100644
--- a/fs/ocfs2/cluster/nodemanager.h
+++ b/fs/ocfs2/cluster/nodemanager.h
@@ -33,8 +33,7 @@
 #include <linux/configfs.h>
 #include <linux/rbtree.h>
 
-#define KERN_OCFS2		988
-#define KERN_OCFS2_NM		1
+#define FS_OCFS2_NM		1
 
 const char *o2nm_get_hb_ctl_path(void);
 
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index f4ba72e..63e1bac 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -813,6 +813,7 @@ enum
 	FS_AIO_NR=18,	/* current system-wide number of aio requests */
 	FS_AIO_MAX_NR=19,	/* system-wide maximum number of aio requests */
 	FS_INOTIFY=20,	/* inotify submenu */
+	FS_OCFS2=988,	/* ocfs2 */
 };
 
 /* /proc/sys/fs/quota/ */
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 init/main.c     |    3 +++
 kernel/sysctl.c |    3 ---
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/init/main.c b/init/main.c
index 8b4a7d7..8af5c6e 100644
--- a/init/main.c
+++ b/init/main.c
@@ -691,6 +691,9 @@ static void __init do_basic_setup(void)
 #ifdef CONFIG_SYSCTL
 	sysctl_init();
 #endif
+#ifdef CONFIG_PROC_FS
+	init_irq_proc();
+#endif
 
 	do_initcalls();
 }
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 600b333..7420761 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1172,8 +1172,6 @@ static ctl_table dev_table[] = {
 	{ .ctl_name = 0 }
 };
 
-extern void init_irq_proc (void);
-
 static DEFINE_SPINLOCK(sysctl_lock);
 
 /* called under sysctl_lock */
@@ -1219,7 +1217,6 @@ void __init sysctl_init(void)
 {
 #ifdef CONFIG_PROC_SYSCTL
 	register_proc_table(root_table, proc_sys_root, &root_table_header);
-	init_irq_proc();
 #endif
 }
 
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

This is just a simple cleanup to keep kernel/sysctl.c
from getting to crowded with special cases, and by
keeping all of the utsname logic to together it makes
the code a little more readable.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 kernel/Makefile         |    1 +
 kernel/sysctl.c         |  115 -------------------------------------
 kernel/utsname_sysctl.c |  146 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 147 insertions(+), 115 deletions(-)

diff --git a/kernel/Makefile b/kernel/Makefile
index 14f4d45..d286c44 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -48,6 +48,7 @@ obj-$(CONFIG_SECCOMP) += seccomp.o
 obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
 obj-$(CONFIG_RELAY) += relay.o
 obj-$(CONFIG_UTS_NS) += utsname.o
+obj-$(CONFIG_SYSCTL) += utsname_sysctl.o
 obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
 obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 7420761..a8c0a03 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -135,13 +135,6 @@ static int parse_table(int __user *, int, void __user *, size_t __user *,
 		void __user *, size_t, ctl_table *);
 #endif
 
-static int proc_do_uts_string(ctl_table *table, int write, struct file *filp,
-		  void __user *buffer, size_t *lenp, loff_t *ppos);
-
-static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen,
-		  void __user *oldval, size_t __user *oldlenp,
-		  void __user *newval, size_t newlen);
-
 #ifdef CONFIG_SYSVIPC
 static int sysctl_ipc_data(ctl_table *table, int __user *name, int nlen,
 		  void __user *oldval, size_t __user *oldlenp,
@@ -174,27 +167,6 @@ extern ctl_table inotify_table[];
 int sysctl_legacy_va_layout;
 #endif
 
-static void *get_uts(ctl_table *table, int write)
-{
-	char *which = table->data;
-#ifdef CONFIG_UTS_NS
-	struct uts_namespace *uts_ns = current->nsproxy->uts_ns;
-	which = (which - (char *)&init_uts_ns) + (char *)uts_ns;
-#endif
-	if (!write)
-		down_read(&uts_sem);
-	else
-		down_write(&uts_sem);
-	return which;
-}
-
-static void put_uts(ctl_table *table, int write, void *which)
-{
-	if (!write)
-		up_read(&uts_sem);
-	else
-		up_write(&uts_sem);
-}
 
 #ifdef CONFIG_SYSVIPC
 static void *get_ipc(ctl_table *table, int write)
@@ -275,51 +247,6 @@ static ctl_table root_table[] = {
 
 static ctl_table kern_table[] = {
 	{
-		.ctl_name	= KERN_OSTYPE,
-		.procname	= "ostype",
-		.data		= init_uts_ns.name.sysname,
-		.maxlen		= sizeof(init_uts_ns.name.sysname),
-		.mode		= 0444,
-		.proc_handler	= &proc_do_uts_string,
-		.strategy	= &sysctl_uts_string,
-	},
-	{
-		.ctl_name	= KERN_OSRELEASE,
-		.procname	= "osrelease",
-		.data		= init_uts_ns.name.release,
-		.maxlen		= sizeof(init_uts_ns.name.release),
-		.mode		= 0444,
-		.proc_handler	= &proc_do_uts_string,
-		.strategy	= &sysctl_uts_string,
-	},
-	{
-		.ctl_name	= KERN_VERSION,
-		.procname	= "version",
-		.data		= init_uts_ns.name.version,
-		.maxlen		= sizeof(init_uts_ns.name.version),
-		.mode		= 0444,
-		.proc_handler	= &proc_do_uts_string,
-		.strategy	= &sysctl_uts_string,
-	},
-	{
-		.ctl_name	= KERN_NODENAME,
-		.procname	= "hostname",
-		.data		= init_uts_ns.name.nodename,
-		.maxlen		= sizeof(init_uts_ns.name.nodename),
-		.mode		= 0644,
-		.proc_handler	= &proc_do_uts_string,
-		.strategy	= &sysctl_uts_string,
-	},
-	{
-		.ctl_name	= KERN_DOMAINNAME,
-		.procname	= "domainname",
-		.data		= init_uts_ns.name.domainname,
-		.maxlen		= sizeof(init_uts_ns.name.domainname),
-		.mode		= 0644,
-		.proc_handler	= &proc_do_uts_string,
-		.strategy	= &sysctl_uts_string,
-	},
-	{
 		.ctl_name	= KERN_PANIC,
 		.procname	= "panic",
 		.data		= &panic_timeout,
@@ -1746,21 +1673,6 @@ int proc_dostring(ctl_table *table, int write, struct file *filp,
 			       buffer, lenp, ppos);
 }
 
-/*
- *	Special case of dostring for the UTS structure. This has locks
- *	to observe. Should this be in kernel/sys.c ????
- */
-
-static int proc_do_uts_string(ctl_table *table, int write, struct file *filp,
-		  void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	int r;
-	void *which;
-	which = get_uts(table, write);
-	r = _proc_do_string(which, table->maxlen,write,filp,buffer,lenp, ppos);
-	put_uts(table, write, which);
-	return r;
-}
 
 static int do_proc_dointvec_conv(int *negp, unsigned long *lvalp,
 				 int *valp,
@@ -2379,12 +2291,6 @@ int proc_dostring(ctl_table *table, int write, struct file *filp,
 	return -ENOSYS;
 }
 
-static int proc_do_uts_string(ctl_table *table, int write, struct file *filp,
-		void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
 #ifdef CONFIG_SYSVIPC
 static int proc_do_ipc_string(ctl_table *table, int write, struct file *filp,
 		void __user *buffer, size_t *lenp, loff_t *ppos)
@@ -2602,21 +2508,6 @@ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen,
 }
 
 
-/* The generic string strategy routine: */
-static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen,
-		  void __user *oldval, size_t __user *oldlenp,
-		  void __user *newval, size_t newlen)
-{
-	struct ctl_table uts_table;
-	int r, write;
-	write = newval && newlen;
-	memcpy(&uts_table, table, sizeof(uts_table));
-	uts_table.data = get_uts(table, write);
-	r = sysctl_string(&uts_table, name, nlen,
-		oldval, oldlenp, newval, newlen);
-	put_uts(table, write, uts_table.data);
-	return r;
-}
 
 #ifdef CONFIG_SYSVIPC
 /* The generic sysctl ipc data routine. */
@@ -2723,12 +2614,6 @@ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen,
 	return -ENOSYS;
 }
 
-static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen,
-		  void __user *oldval, size_t __user *oldlenp,
-		  void __user *newval, size_t newlen)
-{
-	return -ENOSYS;
-}
 static int sysctl_ipc_data(ctl_table *table, int __user *name, int nlen,
 		void __user *oldval, size_t __user *oldlenp,
 		void __user *newval, size_t newlen)
diff --git a/kernel/utsname_sysctl.c b/kernel/utsname_sysctl.c
new file mode 100644
index 0000000..324aa13
--- /dev/null
+++ b/kernel/utsname_sysctl.c
@@ -0,0 +1,146 @@
+/*
+ *  Copyright (C) 2007
+ *
+ *  Author: Eric Biederman <ebiederm@xmision.com>
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License as
+ *  published by the Free Software Foundation, version 2 of the
+ *  License.
+ */
+
+#include <linux/module.h>
+#include <linux/uts.h>
+#include <linux/utsname.h>
+#include <linux/version.h>
+#include <linux/sysctl.h>
+
+static void *get_uts(ctl_table *table, int write)
+{
+	char *which = table->data;
+#ifdef CONFIG_UTS_NS
+	struct uts_namespace *uts_ns = current->nsproxy->uts_ns;
+	which = (which - (char *)&init_uts_ns) + (char *)uts_ns;
+#endif
+	if (!write)
+		down_read(&uts_sem);
+	else
+		down_write(&uts_sem);
+	return which;
+}
+
+static void put_uts(ctl_table *table, int write, void *which)
+{
+	if (!write)
+		up_read(&uts_sem);
+	else
+		up_write(&uts_sem);
+}
+
+#ifdef CONFIG_PROC_FS
+/*
+ *	Special case of dostring for the UTS structure. This has locks
+ *	to observe. Should this be in kernel/sys.c ????
+ */
+static int proc_do_uts_string(ctl_table *table, int write, struct file *filp,
+		  void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+	struct ctl_table uts_table;
+	int r;
+	memcpy(&uts_table, table, sizeof(uts_table));
+	uts_table.data = get_uts(table, write);
+	r = proc_dostring(&uts_table,write,filp,buffer,lenp, ppos);
+	put_uts(table, write, uts_table.data);
+	return r;
+}
+#else
+#define proc_do_uts_string NULL
+#endif
+
+
+#ifdef CONFIG_SYSCTL_SYSCALL
+/* The generic string strategy routine: */
+static int sysctl_uts_string(ctl_table *table, int __user *name, int nlen,
+		  void __user *oldval, size_t __user *oldlenp,
+		  void __user *newval, size_t newlen)
+{
+	struct ctl_table uts_table;
+	int r, write;
+	write = newval && newlen;
+	memcpy(&uts_table, table, sizeof(uts_table));
+	uts_table.data = get_uts(table, write);
+	r = sysctl_string(&uts_table, name, nlen,
+		oldval, oldlenp, newval, newlen);
+	put_uts(table, write, uts_table.data);
+	return r;
+}
+#else
+#define sysctl_uts_string NULL
+#endif
+
+static struct ctl_table uts_kern_table[] = {
+	{
+		.ctl_name	= KERN_OSTYPE,
+		.procname	= "ostype",
+		.data		= init_uts_ns.name.sysname,
+		.maxlen		= sizeof(init_uts_ns.name.sysname),
+		.mode		= 0444,
+		.proc_handler	= proc_do_uts_string,
+		.strategy	= sysctl_uts_string,
+	},
+	{
+		.ctl_name	= KERN_OSRELEASE,
+		.procname	= "osrelease",
+		.data		= init_uts_ns.name.release,
+		.maxlen		= sizeof(init_uts_ns.name.release),
+		.mode		= 0444,
+		.proc_handler	= proc_do_uts_string,
+		.strategy	= sysctl_uts_string,
+	},
+	{
+		.ctl_name	= KERN_VERSION,
+		.procname	= "version",
+		.data		= init_uts_ns.name.version,
+		.maxlen		= sizeof(init_uts_ns.name.version),
+		.mode		= 0444,
+		.proc_handler	= proc_do_uts_string,
+		.strategy	= sysctl_uts_string,
+	},
+	{
+		.ctl_name	= KERN_NODENAME,
+		.procname	= "hostname",
+		.data		= init_uts_ns.name.nodename,
+		.maxlen		= sizeof(init_uts_ns.name.nodename),
+		.mode		= 0644,
+		.proc_handler	= proc_do_uts_string,
+		.strategy	= sysctl_uts_string,
+	},
+	{
+		.ctl_name	= KERN_DOMAINNAME,
+		.procname	= "domainname",
+		.data		= init_uts_ns.name.domainname,
+		.maxlen		= sizeof(init_uts_ns.name.domainname),
+		.mode		= 0644,
+		.proc_handler	= proc_do_uts_string,
+		.strategy	= sysctl_uts_string,
+	},
+	{}
+};
+
+static struct ctl_table uts_root_table[] = {
+	{
+		.ctl_name	= CTL_KERN,
+		.procname	= "kernel",
+		.mode		= 0555,
+		.child		= uts_kern_table,
+	},
+	{}
+};
+
+static int __init utsname_sy

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

This is just a simple cleanup to keep kernel/sysctl.c
from getting to crowded with special cases, and by
keeping all of the ipc logic to together it makes
the code a little more readable.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 init/Kconfig     |    6 ++
 ipc/Makefile     |    1 +
 ipc/ipc_sysctl.c |  182 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 kernel/sysctl.c  |  174 ---------------------------------------------------
 4 files changed, 189 insertions(+), 174 deletions(-)

diff --git a/init/Kconfig b/init/Kconfig
index a3f83e2..33bc38d 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -116,6 +116,12 @@ config SYSVIPC
 	  section 6.4 of the Linux Programmer's Guide, available from
 	  <http://www.tldp.org/guides.html>.
 
+config SYSVIPC_SYSCTL
+	bool
+	depends on SYSVIPC
+	depends on SYSCTL
+	default y
+
 config IPC_NS
 	bool "IPC Namespaces"
 	depends on SYSVIPC
diff --git a/ipc/Makefile b/ipc/Makefile
index 0a6d626..b93bba6 100644
--- a/ipc/Makefile
+++ b/ipc/Makefile
@@ -4,6 +4,7 @@
 
 obj-$(CONFIG_SYSVIPC_COMPAT) += compat.o
 obj-$(CONFIG_SYSVIPC) += util.o msgutil.o msg.o sem.o shm.o
+obj-$(CONFIG_SYSVIPC_SYSCTL) += ipc_sysctl.o
 obj_mq-$(CONFIG_COMPAT) += compat_mq.o
 obj-$(CONFIG_POSIX_MQUEUE) += mqueue.o msgutil.o $(obj_mq-y)
 
diff --git a/ipc/ipc_sysctl.c b/ipc/ipc_sysctl.c
new file mode 100644
index 0000000..9018009
--- /dev/null
+++ b/ipc/ipc_sysctl.c
@@ -0,0 +1,182 @@
+/*
+ *  Copyright (C) 2007
+ *
+ *  Author: Eric Biederman <ebiederm@xmision.com>
+ *
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License as
+ *  published by the Free Software Foundation, version 2 of the
+ *  License.
+ */
+
+#include <linux/module.h>
+#include <linux/ipc.h>
+#include <linux/nsproxy.h>
+#include <linux/sysctl.h>
+
+#ifdef CONFIG_IPC_NS
+static void *get_ipc(ctl_table *table)
+{
+	char *which = table->data;
+	struct ipc_namespace *ipc_ns = current->nsproxy->ipc_ns;
+	which = (which - (char *)&init_ipc_ns) + (char *)ipc_ns;
+	return which;
+}
+#else
+#define get_ipc(T) ((T)->data)
+#endif
+
+#ifdef CONFIG_PROC_FS
+static int proc_ipc_dointvec(ctl_table *table, int write, struct file *filp,
+	void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+	struct ctl_table ipc_table;
+	memcpy(&ipc_table, table, sizeof(ipc_table));
+	ipc_table.data = get_ipc(table);
+
+	return proc_dointvec(&ipc_table, write, filp, buffer, lenp, ppos);
+}
+
+static int proc_ipc_doulongvec_minmax(ctl_table *table, int write,
+	struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+	struct ctl_table ipc_table;
+	memcpy(&ipc_table, table, sizeof(ipc_table));
+	ipc_table.data = get_ipc(table);
+
+	return proc_doulongvec_minmax(&ipc_table, write, filp, buffer,
+					lenp, ppos);
+}
+
+#else
+#define proc_ipc_do_ulongvec_minmax NULL
+#define proc_ipc_do_intvec	    NULL
+#endif
+
+#ifdef CONFIG_SYSCTL_SYSCALL
+/* The generic sysctl ipc data routine. */
+static int sysctl_ipc_data(ctl_table *table, int __user *name, int nlen,
+		void __user *oldval, size_t __user *oldlenp,
+		void __user *newval, size_t newlen)
+{
+	size_t len;
+	void *data;
+
+	/* Get out of I don't have a variable */
+	if (!table->data || !table->maxlen)
+		return -ENOTDIR;
+
+	data = get_ipc(table);
+	if (!data)
+		return -ENOTDIR;
+
+	if (oldval && oldlenp) {
+		if (get_user(len, oldlenp))
+			return -EFAULT;
+		if (len) {
+			if (len > table->maxlen)
+				len = table->maxlen;
+			if (copy_to_user(oldval, data, len))
+				return -EFAULT;
+			if (put_user(len, oldlenp))
+				return -EFAULT;
+		}
+	}
+
+	if (newval && newlen) {
+		if (newlen > table->maxlen)
+			newlen = table->maxlen;
+
+		if (copy_from_user(data, newval, newlen))
+			return -EFAULT;
+	}
+	return 1;
+}
+#else
+#define sysctl_ipc_data NULL
+#endif
+
+static struct ctl_table ipc_kern_table[] = {
+	{
+		.ctl_name	= KERN_SHMMAX,
+		.procname	= "shmmax",
+		.data		= &init_ipc_ns.shm_ctlmax,
+		.maxlen		= sizeof (init_ipc_ns.shm_ctlmax),
+		.mode		= 0644,
+		.proc_handler	= proc_ipc_doulongvec_minmax,
+		.strategy	= sysctl_ipc_data,
+	},
+	{
+		.ctl_name	= KERN_SHMALL,
+		.procname	= "shmall",
+		.data		= &init_ipc_ns.shm_ctlall,
+		.maxlen		= sizeof (init_ipc_ns.shm_ctlall),
+		.mode		= 0644,
+		.proc_handler	= proc_ipc_doulongvec_minmax,
+		.strategy	= sysctl_ipc_data,
+	},
+	{
+		.ctl_name	= KERN_SHMMNI,
+		.procname	= "shmmni",
+		.data		= &init_ipc_ns.shm_ctlmni,
+		.maxlen		= sizeof (init_ipc_ns.shm_ctlmni),
+		.mode		= 0644,
+		.proc_handler	= proc_ipc_dointvec,
+		.strategy	= sysctl_ipc_data,
+	},
+	{
+		.ctl_name	= KERN_MSGMAX,
+		.procname	= "msgmax",
+		.data		= &init_ipc_ns.msg_ctlmax,
+		.maxlen		= sizeof (init_ipc_ns.msg_ctlmax),
+		.mode		= 0644,
+		.proc_handler	= proc_ipc_dointvec,
+		.strategy	= sysctl_ipc_data,
+	},
+	{
+		.ctl_name	= KERN_MSGMNI,
+		.procname	= "msgmni",
+		.data		= &init_ipc_ns.msg_ctlmni,
+		.maxlen		= sizeof (init_ipc_ns.msg_ctlmni),
+		.mode		= 0644,
+		.proc_handler	= proc_ipc_dointvec,
+		.strategy	= sysctl_ipc_data,
+	},
+	{
+		.ctl_name	= KERN_MSGMNB,
+		.procname	=  "msgmnb",
+		.data		= &init_ipc_ns.msg_ctlmnb,
+		.maxlen		= sizeof (init_ipc_ns.msg_ctlmnb),
+		.mode		= 0644,
+		.proc_handler	= proc_ipc_dointvec,
+		.strategy	= sysctl_ipc_data,
+	},
+	{
+		.ctl_name	= KERN_SEM,
+		.procname	= "sem",
+		.data		= &init_ipc_ns.sem_ctls,
+		.maxlen		= 4*sizeof (int),
+		.mode		= 0644,
+		.proc_handler	= proc_ipc_dointvec,
+		.strategy	= sysctl_ipc_data,
+	},
+	{}
+};
+
+static struct ctl_table ipc_root_table[] = {
+	{
+		.ctl_name	= CTL_KERN,
+		.procname	= "kernel",
+		.mode		= 0555,
+		.child		= ipc_kern_table,
+	},
+	{}
+};
+
+static int __init ipc_sysctl_init(void)
+{
+	register_sysctl_table(ipc_root_table, 0);
+	return 0;
+}
+
+__initcall(ipc_sysctl_init);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index a8c0a03..6e2e608 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -90,12 +90,6 @@ extern char modprobe_path[];
 #ifdef CONFIG_CHR_DEV_SG
 extern int sg_big_buff;
 #endif
-#ifdef CONFIG_SYSVIPC
-static int proc_ipc_dointvec(ctl_table *table, int write, struct file *filp,
-		void __user *buffer, size_t *lenp, loff_t *ppos);
-static int proc_ipc_doulongvec_minmax(ctl_table *table, int write, struct file *filp,
-		void __user *buffer, size_t *lenp, loff_t *ppos);
-#endif
 
 #ifdef __sparc__
 extern char reboot_command [];
@@ -135,11 +129,6 @@ static int parse_table(int __user *, int, void __user *, size_t __user *,
 		void __user *, size_t, ctl_table *);
 #endif
 
-#ifdef CONFIG_SYSVIPC
-static int sysctl_ipc_data(ctl_table *table, int __user *name, int nlen,
-		  void __user *oldval, size_t __user *oldlenp,
-		  void __user *newval, size_t newlen);
-#endif
 
 #ifdef CONFIG_PROC_SYSCTL
 static int proc_do_cad_pid(ctl_table *table, int write, struct file *filp,
@@ -168,17 +157,6 @@ int sysctl_legacy_va_layout;
 #endif
 
 
-#ifdef CONFIG_SYSVIPC
-static void *get_ipc(ctl_table *table, int write)
-{
-	char *which = table->data;
-	struct ipc_namespace *ipc_ns = current->nsproxy->ipc_ns;
-	which = (which - (char *)&init_ipc_ns) + (char *)ipc_ns;
-	return which;
-}
-#else
-#define get_ipc(T,W) ((T)->data)
-#endif
 
 /* /proc declarations: */
 
@@ -400,71 +378,6 @@ static ctl_table kern_table[] = {
 		.proc_handler	= &proc_dointvec,
 	},
 #endif
-#ifdef CONFIG_SYSVIPC
-	{
-		.ctl_name	= KERN_SHMMAX,
-		.procname	= "shmmax",
-		.data		= &init_ipc_ns.shm_ctlmax,
-		.maxlen		= sizeof (init_ipc_ns.shm_ctlmax),
-		.mode		= 0644,
-		.proc_handler	= &proc_ipc_doulongvec_minmax,
-		.strategy	= sysctl_ipc_data,
-	},
-	{
-		.ctl_name	= KERN_SHMALL,
-		.procname	= "shmall",
-		.data		= &init_ipc_ns.shm_ctlall,
-		.maxlen		= sizeof (init_ipc_ns.shm_ctlall),
-		.mode		= 0644,
-		.proc_handler	= &proc_ipc_doulongvec_minmax,
-		.strategy	= sysctl_ipc_data,
-	},
-	{
-		.ctl_name	= KERN_SHMMNI,
-		.procname	= "shmmni",
-		.data		= &init_ipc_ns.shm_ctlmni,
-		.maxlen		= sizeof (init_ipc_ns.shm_ctlmni),
-		.mode		= 0644,
-		.proc_handler	= &proc_ipc_dointvec,
-		.strategy	= sysctl_ipc_data,
-	},
-	{
-		.ctl_name	= KERN_MSGMAX,
-		.procname	= "msgmax",
-		.data		= &init_ipc_ns.msg_ctlmax,
-		.maxlen		= sizeof (init_ipc_ns.msg_ctlmax),
-		.mode		= 0644,
-		.proc_handler	= &proc_ipc_dointvec,
-		.strategy	= sysctl_ipc_data,
-	},
-	{
-		.ctl_name	= KERN_MSGMNI,
-		.procname	= "msgmni",
-		.data		= &init_ipc_ns.msg_ctlmni,
-		.maxlen		= sizeof (init_ipc_ns.msg_ctlmni),
-		.mode		= 0644,
-		.proc_handler	= &proc_ipc_dointvec,
-		.strategy	= sysctl_ipc_data,
-	},
-	{
-		.ctl_name	= KERN_MSGMNB,
-		.procname	=  "msgmnb",
-		.data		= &init_ipc_ns.msg_ctlmnb,
-		.maxlen		= sizeof (init_ipc_ns.msg_ctlmnb),
-		.mode		= 0644,
-		.proc_handler	= &proc_ipc_dointvec,
-		.strategy	= sysctl_ipc_data,
-	},
-	{
-		.ctl_name	= KERN_SEM,
-		.procname	= "sem",
-		.data		= &init_ipc_ns.sem_ctls,
-		.maxlen		= 4*sizeof (int),
-		.mode		= 0644,
-		.proc_handler	= &proc_ipc_dointvec,
-		.strategy	= sysctl_ipc_data,
-	},
-#endif
 #ifdef CONFIG_MAGIC_SYSRQ
 	{
 		.ctl_name	= KERN_SYSRQ,
@@ -2240,27 +2153,6 @@ int proc_dointvec_ms_jiffies(ctl_table *table, int write, struct file *filp,
 				do_proc_dointvec_ms_jiffies_conv, NULL);
 }
 
-#ifdef CONFIG_SYSVIPC
-static int proc_ipc_dointvec(ctl_table *table, int write, struct file *filp,
-	void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	void *which;
-	which = get_ipc(table, write);
-	return __do_proc_dointvec(which, table, write, filp, buffer,
-			lenp, ppos, NULL, NULL);
-}
-
-static int proc_ipc_doulongve

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

binfmt_misc has a mount point in the middle of the sysctl and
that mount point is created as a proc_generic directory.

Doing it that way gets in the way of cleaning up the sysctl
proc support as it continues the existence of a horrible hack.
So instead simply create the directory as an ordinary sysctl
directory.   At least that removes the magic special case.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 fs/proc/root.c  |    4 ----
 kernel/sysctl.c |   13 +++++++++++++
 2 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/fs/proc/root.c b/fs/proc/root.c
index 64d242b..8059e92 100644
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -74,10 +74,6 @@ void __init proc_root_init(void)
 #ifdef CONFIG_SYSCTL
 	proc_sys_root = proc_mkdir("sys", NULL);
 #endif
-#if defined(CONFIG_BINFMT_MISC) || defined(CONFIG_BINFMT_MISC_MODULE)
-	proc_mkdir("sys/fs", NULL);
-	proc_mkdir("sys/fs/binfmt_misc", NULL);
-#endif
 	proc_root_fs = proc_mkdir("fs", NULL);
 	proc_root_driver = proc_mkdir("driver", NULL);
 	proc_mkdir("fs/nfsd", NULL); /* somewhere for the nfsd filesystem to be mounted */
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 6e2e608..8da6647 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -142,6 +142,7 @@ static struct ctl_table_header root_table_header =
 static ctl_table kern_table[];
 static ctl_table vm_table[];
 static ctl_table fs_table[];
+static ctl_table binfmt_misc_table[];
 static ctl_table debug_table[];
 static ctl_table dev_table[];
 extern ctl_table random_table[];
@@ -1001,6 +1002,18 @@ static ctl_table fs_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+#if defined(CONFIG_BINFMT_MISC) || defined(CONFIG_BINFMT_MISC_MODULE)
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "binfmt_misc",
+		.mode		= 0555,
+		.child		= binfmt_misc_table,
+	},
+#endif
+	{ .ctl_name = 0 }
+};
+
+static ctl_table binfmt_misc_table[] = {
 	{ .ctl_name = 0 }
 };
 
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

There are currently no users in the kernel for CTL_ANY and it
only has effect on the binary interface which is practically
unused.

So this complicates sysctl lookups for no good reason so just remove it.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 include/linux/sysctl.h |    1 -
 kernel/sysctl.c        |    2 +-
 2 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 63e1bac..c99e4bb 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -53,7 +53,6 @@ struct __sysctl_args {
 
 /* For internal pattern-matching use only: */
 #ifdef __KERNEL__
-#define CTL_ANY		-1	/* Matches any name */
 #define CTL_NONE	0
 #define CTL_UNNUMBERED	CTL_NONE	/* sysctl without a binary number */
 #endif
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 8da6647..e655b11 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1166,7 +1166,7 @@ repeat:
 	for ( ; table->ctl_name || table->procname; table++) {
 		if (!table->ctl_name)
 			continue;
-		if (n == table->ctl_name || table->ctl_name == CTL_ANY) {
+		if (n == table->ctl_name) {
 			int error;
 			if (table->child) {
 				if (ctl_perm(table, 001))
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

parse_table has support for calling a strategy routine
when descending into a directory.  To date no one has
used this functionality and the /proc/sys interface has
no analog to it.

So no one is using this functionality kill it and make
the binary sysctl code easier to follow.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 kernel/sysctl.c |    8 --------
 1 files changed, 0 insertions(+), 8 deletions(-)

diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index e655b11..2c3703d 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1171,14 +1171,6 @@ repeat:
 			if (table->child) {
 				if (ctl_perm(table, 001))
 					return -EPERM;
-				if (table->strategy) {
-					error = table->strategy(
-						table, name, nlen,
-						oldval, oldlenp,
-						newval, newlen);
-					if (error)
-						return error;
-				}
 				name++;
 				nlen--;
 				table = table->child;
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

The semantic effect of insert_at_head is that it would allow
new registered sysctl entries to override existing sysctl entries
of the same name.  Which is pain for caching and the proc interface
never implemented.

I have done an audit and discovered that none of the current
users of register_sysctl care as (excpet for directories) they
do not register duplicate sysctl entries.

So this patch simply removes the support for overriding
existing entries in the sys_sysctl interface since no one
uses it or cares and it makes future enhancments harder.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 arch/arm/kernel/isa.c                        |    2 +-
 arch/frv/kernel/pm.c                         |    2 +-
 arch/frv/kernel/sysctl.c                     |    2 +-
 arch/ia64/kernel/crash.c                     |    2 +-
 arch/ia64/kernel/perfmon.c                   |    2 +-
 arch/ia64/sn/kernel/xpc_main.c               |    2 +-
 arch/mips/au1000/common/power.c              |    2 +-
 arch/mips/lasat/sysctl.c                     |    2 +-
 arch/powerpc/kernel/idle.c                   |    2 +-
 arch/ppc/kernel/ppc_htab.c                   |    2 +-
 arch/s390/appldata/appldata_base.c           |    4 ++--
 arch/s390/kernel/debug.c                     |    2 +-
 arch/s390/mm/cmm.c                           |    2 +-
 arch/sh64/kernel/traps.c                     |    2 +-
 arch/x86_64/ia32/ia32_binfmt.c               |    2 +-
 arch/x86_64/kernel/vsyscall.c                |    2 +-
 arch/x86_64/mm/init.c                        |    2 +-
 drivers/cdrom/cdrom.c                        |    2 +-
 drivers/char/hpet.c                          |    2 +-
 drivers/char/ipmi/ipmi_poweroff.c            |    2 +-
 drivers/char/rtc.c                           |    2 +-
 drivers/macintosh/mac_hid.c                  |    2 +-
 drivers/md/md.c                              |    2 +-
 drivers/net/wireless/arlan-proc.c            |    2 +-
 drivers/parport/procfs.c                     |    6 +++---
 drivers/scsi/scsi_sysctl.c                   |    2 +-
 fs/coda/sysctl.c                             |    2 +-
 fs/dquot.c                                   |    2 +-
 fs/lockd/svc.c                               |    2 +-
 fs/nfs/sysctl.c                              |    2 +-
 fs/ntfs/sysctl.c                             |    2 +-
 fs/ocfs2/cluster/nodemanager.c               |    2 +-
 fs/xfs/linux-2.6/xfs_sysctl.c                |    2 +-
 include/linux/sysctl.h                       |    4 ++--
 ipc/ipc_sysctl.c                             |    2 +-
 ipc/mqueue.c                                 |    2 +-
 kernel/sysctl.c                              |    9 ++-------
 kernel/utsname_sysctl.c                      |    2 +-
 net/appletalk/sysctl_net_atalk.c             |    2 +-
 net/ax25/sysctl_net_ax25.c                   |    2 +-
 net/bridge/br_netfilter.c                    |    2 +-
 net/core/neighbour.c                         |    2 +-
 net/dccp/sysctl.c                            |    2 +-
 net/decnet/dn_dev.c                          |    2 +-
 net/decnet/sysctl_net_decnet.c               |    2 +-
 net/ipv4/devinet.c                           |    4 ++--
 net/ipv4/ipvs/ip_vs_ctl.c                    |    2 +-
 net/ipv4/ipvs/ip_vs_lblc.c                   |    2 +-
 net/ipv4/ipvs/ip_vs_lblcr.c                  |    2 +-
 net/ipv4/netfilter/ip_conntrack_proto_sctp.c |    2 +-
 net/ipv4/netfilter/ip_conntrack_standalone.c |    2 +-
 net/ipv4/netfilter/ip_queue.c                |    2 +-
 net/ipv6/addrconf.c                          |    4 ++--
 net/ipv6/netfilter/ip6_queue.c               |    2 +-
 net/ipv6/sysctl_net_ipv6.c                   |    2 +-
 net/ipx/sysctl_net_ipx.c                     |    2 +-
 net/irda/irsysctl.c                          |    2 +-
 net/llc/sysctl_net_llc.c                     |    2 +-
 net/netfilter/nf_conntrack_standalone.c      |    2 +-
 net/netfilter/nf_sysctl.c                    |    2 +-
 net/netrom/sysctl_net_netrom.c               |    2 +-
 net/rose/sysctl_net_rose.c                   |    2 +-
 net/rxrpc/sysctl.c                           |    2 +-
 net/sctp/sysctl.c                            |    2 +-
 net/sunrpc/sysctl.c                          |    2 +-
 net/sunrpc/xprtsock.c                        |    2 +-
 net/unix/sysctl_net_unix.c                   |    2 +-
 net/x25/sysctl_net_x25.c                     |    2 +-
 68 files changed, 75 insertions(+), 80 deletions(-)

diff --git a/arch/arm/kernel/isa.c b/arch/arm/kernel/isa.c
index 54bbd9f..50a30bc 100644
--- a/arch/arm/kernel/isa.c
+++ b/arch/arm/kernel/isa.c
@@ -70,5 +70,5 @@ register_isa_ports(unsigned int membase, unsigned int portbase, unsigned int por
 	isa_membase = membase;
 	isa_portbase = portbase;
 	isa_portshift = portshift;
-	isa_sysctl_header = register_sysctl_table(ctl_bus, 0);
+	isa_sysctl_header = register_sysctl_table(ctl_bus);
 }
diff --git a/arch/frv/kernel/pm.c b/arch/frv/kernel/pm.c
index aa50333..c57ce3f 100644
--- a/arch/frv/kernel/pm.c
+++ b/arch/frv/kernel/pm.c
@@ -455,7 +455,7 @@ static struct ctl_table pm_dir_table[] =
  */
 static int __init pm_init(void)
 {
-	register_sysctl_table(pm_dir_table, 0);
+	register_sysctl_table(pm_dir_table);
 	return 0;
 }
 
diff --git a/arch/frv/kernel/sysctl.c b/arch/frv/kernel/sysctl.c
index 577ad16..3e9d7e0 100644
--- a/arch/frv/kernel/sysctl.c
+++ b/arch/frv/kernel/sysctl.c
@@ -216,7 +216,7 @@ static struct ctl_table frv_dir_table[] =
  */
 static int __init frv_sysctl_init(void)
 {
-	register_sysctl_table(frv_dir_table, 0);
+	register_sysctl_table(frv_dir_table);
 	return 0;
 }
 
diff --git a/arch/ia64/kernel/crash.c b/arch/ia64/kernel/crash.c
index bc2f64d..8a2cb76 100644
--- a/arch/ia64/kernel/crash.c
+++ b/arch/ia64/kernel/crash.c
@@ -214,7 +214,7 @@ machine_crash_setup(void)
 	if((ret = register_die_notifier(&kdump_init_notifier_nb)) != 0)
 		return ret;
 #ifdef CONFIG_SYSCTL
-	register_sysctl_table(sys_table, 0);
+	register_sysctl_table(sys_table);
 #endif
 	return 0;
 }
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 8c679ab..8a15377 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -6727,7 +6727,7 @@ pfm_init(void)
 	/*
 	 * create /proc/sys/kernel/perfmon (for debugging purposes)
 	 */
-	pfm_sysctl_header = register_sysctl_table(pfm_sysctl_root, 0);
+	pfm_sysctl_header = register_sysctl_table(pfm_sysctl_root);
 
 	/*
 	 * initialize all our spinlocks
diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c
index e04f7b5..68355ef 100644
--- a/arch/ia64/sn/kernel/xpc_main.c
+++ b/arch/ia64/sn/kernel/xpc_main.c
@@ -1241,7 +1241,7 @@ xpc_init(void)
 	snprintf(xpc_part->bus_id, BUS_ID_SIZE, "part");
 	snprintf(xpc_chan->bus_id, BUS_ID_SIZE, "chan");
 
-	xpc_sysctl = register_sysctl_table(xpc_sys_dir, 0);
+	xpc_sysctl = register_sysctl_table(xpc_sys_dir);
 
 	/*
 	 * The first few fields of each entry of xpc_partitions[] need to
diff --git a/arch/mips/au1000/common/power.c b/arch/mips/au1000/common/power.c
index 31256b8..3901e8e 100644
--- a/arch/mips/au1000/common/power.c
+++ b/arch/mips/au1000/common/power.c
@@ -461,7 +461,7 @@ static struct ctl_table pm_dir_table[] = {
  */
 static int __init pm_init(void)
 {
-	register_sysctl_table(pm_dir_table, 0);
+	register_sysctl_table(pm_dir_table);
 	return 0;
 }
 
diff --git a/arch/mips/lasat/sysctl.c b/arch/mips/lasat/sysctl.c
index c04e82f..699ab18 100644
--- a/arch/mips/lasat/sysctl.c
+++ b/arch/mips/lasat/sysctl.c
@@ -432,7 +432,7 @@ static int __init lasat_register_sysctl(void)
 	struct ctl_table_header *lasat_table_header;
 
 	lasat_table_header =
-		register_sysctl_table(lasat_root_table, 0);
+		register_sysctl_table(lasat_root_table);
 
 	return 0;
 }
diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c
index 8b27bb1..6e7f509 100644
--- a/arch/powerpc/kernel/idle.c
+++ b/arch/powerpc/kernel/idle.c
@@ -125,7 +125,7 @@ static ctl_table powersave_nap_sysctl_root[] = {
 static int __init
 register_powersave_nap_sysctl(void)
 {
-	register_sysctl_table(powersave_nap_sysctl_root, 0);
+	register_sysctl_table(powersave_nap_sysctl_root);
 
 	return 0;
 }
diff --git a/arch/ppc/kernel/ppc_htab.c b/arch/ppc/kernel/ppc_htab.c
index 77b20ff..0a7e42d 100644
--- a/arch/ppc/kernel/ppc_htab.c
+++ b/arch/ppc/kernel/ppc_htab.c
@@ -457,7 +457,7 @@ static ctl_table htab_sysctl_root[] = {
 static int __init
 register_ppc_htab_sysctl(void)
 {
-	register_sysctl_table(htab_sysctl_root, 0);
+	register_sysctl_table(htab_sysctl_root);
 
 	return 0;
 }
diff --git a/arch/s390/appldata/appldata_base.c b/arch/s390/appldata/appldata_base.c
index cdc4109..b1ff93a 100644
--- a/arch/s390/appldata/appldata_base.c
+++ b/arch/s390/appldata/appldata_base.c
@@ -506,7 +506,7 @@ int appldata_register_ops(struct appldata_ops *ops)
 
 	ops->ctl_table[3].ctl_name = 0;
 
-	ops->sysctl_header = register_sysctl_table(ops->ctl_table,0);
+	ops->sysctl_header = register_sysctl_table(ops->ctl_table);
 
 	P_INFO("%s-ops registered!\n", ops->name);
 	return 0;
@@ -606,7 +606,7 @@ static int __init appldata_init(void)
 	/* Register cpu hotplug notifier */
 	register_hotcpu_notifier(&appldata_nb);
 
-	appldata_sysctl_header = register_sysctl_table(appldata_dir_table, 0);
+	appldata_sysctl_header = register_sysctl_table(appldata_dir_table);
 #ifdef MODULE
 	appldata_dir_table[0].de->owner = THIS_MODULE;
 	appldata_table[0].de->owner = THIS_MODULE;
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index d38cb27..00f0382 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -1053,7 +1053,7 @@ __init debug_init(void)
 {
 	int rc = 0;
 
-	s390dbf_sysctl_header = register_sysctl_table(s390dbf_dir_table, 0);
+	s390dbf_sysctl_header = register_sysctl_table(s390dbf_dir_table);
 	down(&debug_lock);
 	deb

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

The current logic to walk through the list of sysctl table
headers is slightly painful and implement in a way it cannot
be used by code outside sysctl.c

I am in the process of implementing a version of the sysctl
proc support that instead of using the proc generic non-caching
monster, just uses the existing sysctl data structure as
backing store for building the dcache entries and for doing
directory reads.   To use the existing data structures however
I need a way to get at them.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 include/linux/sysctl.h |    4 +++
 kernel/sysctl.c        |   57 +++++++++++++++++++++++++++++++++++------------
 2 files changed, 46 insertions(+), 15 deletions(-)

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 6113f3b..81ee9ea 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -923,6 +923,10 @@ enum
 #ifdef __KERNEL__
 #include <linux/list.h>
 
+/* For the /proc/sys support */
+extern struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev);
+extern void sysctl_head_finish(struct ctl_table_header *prev);
+
 extern void sysctl_init(void);
 
 typedef struct ctl_table ctl_table;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 5beee1f..ca2831a 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1066,6 +1066,42 @@ static void start_unregistering(struct ctl_table_header *p)
 	list_del_init(&p->ctl_entry);
 }
 
+void sysctl_head_finish(struct ctl_table_header *head)
+{
+	if (!head)
+		return;
+	spin_lock(&sysctl_lock);
+	unuse_table(head);
+	spin_unlock(&sysctl_lock);
+}
+
+struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev)
+{
+	struct ctl_table_header *head;
+	struct list_head *tmp;
+	spin_lock(&sysctl_lock);
+	if (prev) {
+		tmp = &prev->ctl_entry;
+		unuse_table(prev);
+		goto next;
+	}
+	tmp = &root_table_header.ctl_entry;
+	for (;;) {
+		head = list_entry(tmp, struct ctl_table_header, ctl_entry);
+
+		if (!use_table(head))
+			goto next;
+		spin_unlock(&sysctl_lock);
+		return head;
+	next:
+		tmp = tmp->next;
+		if (tmp == &root_table_header.ctl_entry)
+			break;
+	}
+	spin_unlock(&sysctl_lock);
+	return NULL;
+}
+
 void __init sysctl_init(void)
 {
 #ifdef CONFIG_PROC_SYSCTL
@@ -1077,6 +1113,7 @@ void __init sysctl_init(void)
 int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp,
 	       void __user *newval, size_t newlen)
 {
+	struct ctl_table_header *head;
 	struct list_head *tmp;
 	int error = -ENOTDIR;
 
@@ -1087,26 +1124,16 @@ int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *ol
 		if (!oldlenp || get_user(old_len, oldlenp))
 			return -EFAULT;
 	}
-	spin_lock(&sysctl_lock);
-	tmp = &root_table_header.ctl_entry;
-	do {
-		struct ctl_table_header *head =
-			list_entry(tmp, struct ctl_table_header, ctl_entry);
 
-		if (!use_table(head))
-			continue;
-
-		spin_unlock(&sysctl_lock);
+	for (head = sysctl_head_next(NULL); head; head = sysctl_head_next(head)) {
 
 		error = parse_table(name, nlen, oldval, oldlenp, 
 					newval, newlen, head->ctl_table);
-
-		spin_lock(&sysctl_lock);
-		unuse_table(head);
-		if (error != -ENOTDIR)
+		if (error != -ENOTDIR) {
+			sysctl_head_finish(head);
 			break;
-	} while ((tmp = tmp->next) != &root_table_header.ctl_entry);
-	spin_unlock(&sysctl_lock);
+		}
+	}
 	return error;
 }
 
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 include/linux/sysctl.h |    2 ++
 kernel/sysctl.c        |   10 +++++-----
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 81ee9ea..20c23b5 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -924,8 +924,10 @@ enum
 #include <linux/list.h>
 
 /* For the /proc/sys support */
+struct ctl_table;
 extern struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev);
 extern void sysctl_head_finish(struct ctl_table_header *prev);
+extern int sysctl_perm(struct ctl_table *table, int op);
 
 extern void sysctl_init(void);
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index ca2831a..ec5e4a1 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1154,7 +1154,7 @@ asmlinkage long sys_sysctl(struct __sysctl_args __user *args)
 #endif /* CONFIG_SYSCTL_SYSCALL */
 
 /*
- * ctl_perm does NOT grant the superuser all rights automatically, because
+ * sysctl_perm does NOT grant the superuser all rights automatically, because
  * some sysctl variables are readonly even to root.
  */
 
@@ -1169,7 +1169,7 @@ static int test_perm(int mode, int op)
 	return -EACCES;
 }
 
-static inline int ctl_perm(ctl_table *table, int op)
+int sysctl_perm(ctl_table *table, int op)
 {
 	int error;
 	error = security_sysctl(table, op);
@@ -1196,7 +1196,7 @@ repeat:
 		if (n == table->ctl_name) {
 			int error;
 			if (table->child) {
-				if (ctl_perm(table, 001))
+				if (sysctl_perm(table, 001))
 					return -EPERM;
 				name++;
 				nlen--;
@@ -1225,7 +1225,7 @@ int do_sysctl_strategy (ctl_table *table,
 		op |= 004;
 	if (newval) 
 		op |= 002;
-	if (ctl_perm(table, op))
+	if (sysctl_perm(table, op))
 		return -EPERM;
 
 	if (table->strategy) {
@@ -1495,7 +1495,7 @@ static ssize_t do_rw_proc(int write, struct file * file, char __user * buf,
 			goto out;
 		error = -EPERM;
 		op = (write ? 002 : 004);
-		if (ctl_perm(table, op))
+		if (sysctl_perm(table, op))
 			goto out;
 		
 		/* careful: calling conventions are nasty here */
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

With this change the sysctl inodes can be cached and
nothing needs to be done when removing a sysctl table.

For a costk of 2K code we will save about 4K of static tables
(when we remove de from ctl_table) and 70K in proc_dir_entries
that we will not allocate, or about half that on a 32bit arch.

The speed feels about the same, even though we can now cache
the sysctl dentries :(

We get the core advantage that we don't need to have a
1 to 1 mapping between ctl table entries and proc files.
Making it possible to have /proc/sys vary depending on
the namespace you are in.  The currently merged namespaces
don't have an issue here but the network namespace under
/proc/sys/net needs to have different directories depending
on which network adapters are visible.  By simply being
a cache different directories being visible depending
on who you are is trivial to implement.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 fs/proc/Makefile      |    2 +-
 fs/proc/inode.c       |    1 +
 fs/proc/internal.h    |    2 +
 fs/proc/proc_sysctl.c |  477 +++++++++++++++++++++++++++++++++++++++++++++++++
 fs/proc/root.c        |   10 +-
 init/main.c           |    4 -
 kernel/sysctl.c       |  182 -------------------
 7 files changed, 484 insertions(+), 194 deletions(-)

diff --git a/fs/proc/Makefile b/fs/proc/Makefile
index f6c7762..a6b3a8f 100644
--- a/fs/proc/Makefile
+++ b/fs/proc/Makefile
@@ -8,7 +8,7 @@ proc-y			:= nommu.o task_nommu.o
 proc-$(CONFIG_MMU)	:= mmu.o task_mmu.o
 
 proc-y       += inode.o root.o base.o generic.o array.o \
-		proc_tty.o proc_misc.o
+		proc_tty.o proc_misc.o proc_sysctl.o
 
 proc-$(CONFIG_PROC_KCORE)	+= kcore.o
 proc-$(CONFIG_PROC_VMCORE)	+= vmcore.o
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index e26945b..0ea8265 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -161,6 +161,7 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino,
 	if (!inode)
 		goto out_ino;
 
+	PROC_I(inode)->fd = 0;
 	PROC_I(inode)->pde = de;
 	if (de) {
 		if (de->mode) {
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 987c773..3c9a305 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -11,6 +11,8 @@
 
 #include <linux/proc_fs.h>
 
+extern int proc_sys_init(void);
+
 struct vmalloc_info {
 	unsigned long	used;
 	unsigned long	largest_chunk;
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
new file mode 100644
index 0000000..08a2e66
--- /dev/null
+++ b/fs/proc/proc_sysctl.c
@@ -0,0 +1,477 @@
+/*
+ * /proc/sys support
+ */
+
+#include <linux/sysctl.h>
+#include <linux/proc_fs.h>
+#include <linux/security.h>
+#include "internal.h"
+
+static struct dentry_operations proc_sys_dentry_operations;
+static const struct file_operations proc_sys_file_operations;
+static struct inode_operations proc_sys_inode_operations;
+
+static void proc_sys_refresh_inode(struct inode *inode, struct ctl_table *table)
+{
+	/* Refresh the cached information bits in the inode */
+	if (table) {
+		inode->i_uid = 0;
+		inode->i_gid = 0;
+		inode->i_mode = table->mode;
+		if (table->proc_handler) {
+			inode->i_mode |= S_IFREG;
+			inode->i_nlink = 1;
+		} else {
+			inode->i_mode |= S_IFDIR;
+			inode->i_nlink = 0;	/* It is too hard to figure out */
+		}
+	}
+}
+
+static struct inode *proc_sys_make_inode(struct inode *dir, struct ctl_table *table)
+{
+	struct inode *inode;
+	struct proc_inode *dir_ei, *ei;
+	int depth;
+
+	inode = new_inode(dir->i_sb);
+	if (!inode)
+		goto out;
+
+	/* A directory is always one deeper than it's parent */
+	dir_ei = PROC_I(dir);
+	depth = dir_ei->fd + 1;
+
+	ei = PROC_I(inode);
+	ei->fd = depth;
+	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
+	inode->i_op = &proc_sys_inode_operations;
+	inode->i_fop = &proc_sys_file_operations;
+	proc_sys_refresh_inode(inode, table);
+out:
+	return inode;
+}
+
+static struct dentry *proc_sys_ancestor(struct dentry *dentry, int depth)
+{
+	for (;;) {
+		struct proc_inode *ei;
+
+		ei = PROC_I(dentry->d_inode);
+		if (ei->fd == depth)
+			break; /* found */
+
+		dentry = dentry->d_parent;
+	}
+	return dentry;
+}
+
+static struct ctl_table *proc_sys_lookup_table_one(struct ctl_table *table,
+							struct qstr *name)
+{
+	int len;
+	for ( ; table->ctl_name || table->procname; table++) {
+
+		if (!table->procname)
+			continue;
+
+		len = strlen(table->procname);
+		if (len != name->len)
+			continue;
+
+		if (memcmp(table->procname, name->name, len) != 0)
+			continue;
+
+		/* I have a match */
+		return table;
+	}
+	return NULL;
+}
+
+static struct ctl_table *proc_sys_lookup_table(struct dentry *dentry,
+						struct ctl_table *table)
+{
+	struct dentry *ancestor;
+	struct proc_inode *ei;
+	int depth, i;
+
+	ei = PROC_I(dentry->d_inode);
+	depth = ei->fd;
+
+	if (depth == 0)
+		return table;
+
+	for (i = 1; table && (i <= depth); i++) {
+		ancestor = proc_sys_ancestor(dentry, i);
+		table = proc_sys_lookup_table_one(table, &ancestor->d_name);
+		if (table)
+			table = table->child;
+	}
+	return table;
+
+}
+static struct ctl_table *proc_sys_lookup_entry(struct dentry *dparent,
+						struct qstr *name,
+						struct ctl_table *table)
+{
+	table = proc_sys_lookup_table(dparent, table);
+	if (table)
+		table = proc_sys_lookup_table_one(table, name);
+	return table;
+}
+
+static struct ctl_table *do_proc_sys_lookup(struct dentry *parent,
+						struct qstr *name,
+						struct ctl_table_header **ptr)
+{
+	struct ctl_table_header *head;
+	struct ctl_table *table;
+
+	for (head = sysctl_head_next(NULL); head; head = sysctl_head_next(head)) {
+		table = proc_sys_lookup_entry(parent, name, head->ctl_table);
+		if (table)
+			break;
+	}
+	*ptr = head;
+	return table;
+}
+
+static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry,
+					struct nameidata *nd)
+{
+	struct ctl_table_header *head;
+	struct inode *inode;
+	struct dentry *err;
+	struct ctl_table *table;
+
+	err = ERR_PTR(-ENOENT);
+	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
+	if (!table)
+		goto out;
+
+	err = ERR_PTR(-ENOMEM);
+	inode = proc_sys_make_inode(dir, table);
+	if (!inode)
+		goto out;
+
+	err = NULL;
+	dentry->d_op = &proc_sys_dentry_operations;
+	d_add(dentry, inode);
+
+out:
+	sysctl_head_finish(head);
+	return err;
+}
+
+static ssize_t proc_sys_read(struct file *filp, char __user *buf,
+				size_t count, loff_t *ppos)
+{
+	struct dentry *dentry = filp->f_dentry;
+	struct ctl_table_header *head;
+	struct ctl_table *table;
+	ssize_t error, res;
+
+	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
+	/* Has the sysctl entry disappeared on us? */
+	error = -ENOENT;
+	if (!table)
+		goto out;
+
+	/* Has the sysctl entry been replaced by a directory? */
+	error = -EISDIR;
+	if (!table->proc_handler)
+		goto out;
+
+	/*
+	 * At this point we know that the sysctl was not unregistered
+	 * and won't be until we finish.
+	 */
+	error = -EPERM;
+	if (sysctl_perm(table, MAY_READ))
+		goto out;
+
+	/* careful: calling conventions are nasty here */
+	res = count;
+	error = table->proc_handler(table, 0, filp, buf, &res, ppos);
+	if (!error)
+		error = res;
+out:
+	sysctl_head_finish(head);
+
+	return error;
+}
+
+static ssize_t proc_sys_write(struct file *filp, const char __user *buf,
+				size_t count, loff_t *ppos)
+{
+	struct dentry *dentry = filp->f_dentry;
+	struct ctl_table_header *head;
+	struct ctl_table *table;
+	ssize_t error, res;
+
+	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
+	/* Has the sysctl entry disappeared on us? */
+	error = -ENOENT;
+	if (!table)
+		goto out;
+
+	/* Has the sysctl entry been replaced by a directory? */
+	error = -EISDIR;
+	if (!table->proc_handler)
+		goto out;
+
+	/*
+	 * At this point we know that the sysctl was not unregistered
+	 * and won't be until we finish.
+	 */
+	error = -EPERM;
+	if (sysctl_perm(table, MAY_WRITE))
+		goto out;
+
+	/* careful: calling conventions are nasty here */
+	res = count;
+	error = table->proc_handler(table, 1, filp, buf, &res, ppos);
+	if (!error)
+		error = res;
+out:
+	sysctl_head_finish(head);
+
+	return error;
+}
+
+
+static int proc_sys_fill_cache(struct file *filp, void *dirent,
+				filldir_t filldir, struct ctl_table *table)
+{
+	struct ctl_table_header *head;
+	struct ctl_table *child_table = NULL;
+	struct dentry *child, *dir = filp->f_path.dentry;
+	struct inode *inode;
+	struct qstr qname;
+	ino_t ino = 0;
+	unsigned type = DT_UNKNOWN;
+	int ret;
+
+	qname.name = table->procname;
+	qname.len  = strlen(table->procname);
+	qname.hash = full_name_hash(qname.name, qname.len);
+
+	/* Suppress duplicates.
+	 * Only fill a directory entry if it is the value that
+	 * an ordinary lookup of that name returns.  Hide all
+	 * others.
+	 *
+	 * If we ever cache this translation in the dcache
+	 * I should do a dcache lookup first.  But for now
+	 * it is just simpler not to.
+	 */
+	ret = 0;
+	child_table = do_proc_sys_lookup(dir, &qname, &head);
+	sysctl_head_finish(head);
+	if (child_table != table)
+		return 0;
+
+	child = d_lookup(dir, &qname);
+	if (!child) {
+		struct dentry *new;
+		new = d_alloc(dir, &qname);
+		if (new) {
+			inode = proc_sys_make_inode(dir->d_inode, table);
+			if (!inode)
+				child = ERR_PTR(-ENOMEM);
+			else {
+				new->d_op = &proc_sys_dentry_operations;
+				d_add(new, inode);
+			}
+			if (child)
+				dput(new);
+			else
+				child = new;
+		}
+	}
+	if (!child || IS_ERR(child) || !child->d_inode)
+		goto end_instantiate;
+	inode = child->d_inode;
+	if (inode) {
+		ino  = inode->i_ino;
+		type = inode->i_mode >> 12;
+	}
+	dput(child);
+end_instantiate:
+	if (!ino)
+		ino= find_inode_number(dir,

From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

It isn't needed anymore, all of the users are gone, and all of the
ctl_table initializers have been converted to use explicit names
of the fields they are initializing.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
---
 include/linux/sysctl.h |    1 -
 net/decnet/dn_dev.c    |    5 -----
 net/ipv4/devinet.c     |    5 -----
 net/ipv6/addrconf.c    |    5 -----
 4 files changed, 0 insertions(+), 16 deletions(-)

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 20c23b5..8c2fab5 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -1025,7 +1025,6 @@ struct ctl_table
 	ctl_table *child;
 	proc_handler *proc_handler;	/* Callback for text formatting */
 	ctl_handler *strategy;		/* Callback function for all r/w */
-	struct proc_dir_entry *de;	/* /proc control block */
 	void *extra1;
 	void *extra2;
 };
diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c
index baaa02e..324eb47 100644
--- a/net/decnet/dn_dev.c
+++ b/net/decnet/dn_dev.c
@@ -261,7 +261,6 @@ static void dn_dev_sysctl_register(struct net_device *dev, struct dn_dev_parms *
 	for(i = 0; i < ARRAY_SIZE(t->dn_dev_vars) - 1; i++) {
 		long offset = (long)t->dn_dev_vars[i].data;
 		t->dn_dev_vars[i].data = ((char *)parms) + offset;
-		t->dn_dev_vars[i].de = NULL;
 	}
 
 	if (dev) {
@@ -273,13 +272,9 @@ static void dn_dev_sysctl_register(struct net_device *dev, struct dn_dev_parms *
 	}
 
 	t->dn_dev_dev[0].child = t->dn_dev_vars;
-	t->dn_dev_dev[0].de = NULL;
 	t->dn_dev_conf_dir[0].child = t->dn_dev_dev;
-	t->dn_dev_conf_dir[0].de = NULL;
 	t->dn_dev_proto_dir[0].child = t->dn_dev_conf_dir;
-	t->dn_dev_proto_dir[0].de = NULL;
 	t->dn_dev_root_dir[0].child = t->dn_dev_proto_dir;
-	t->dn_dev_root_dir[0].de = NULL;
 	t->dn_dev_vars[0].extra1 = (void *)dev;
 
 	t->sysctl_header = register_sysctl_table(t->dn_dev_root_dir);
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index b731a0c..8cfcc78 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1573,7 +1573,6 @@ static void devinet_sysctl_register(struct in_device *in_dev,
 		return;
 	for (i = 0; i < ARRAY_SIZE(t->devinet_vars) - 1; i++) {
 		t->devinet_vars[i].data += (char *)p - (char *)&ipv4_devconf;
-		t->devinet_vars[i].de = NULL;
 	}
 
 	if (dev) {
@@ -1595,13 +1594,9 @@ static void devinet_sysctl_register(struct in_device *in_dev,
 
 	t->devinet_dev[0].procname    = dev_name;
 	t->devinet_dev[0].child	      = t->devinet_vars;
-	t->devinet_dev[0].de	      = NULL;
 	t->devinet_conf_dir[0].child  = t->devinet_dev;
-	t->devinet_conf_dir[0].de     = NULL;
 	t->devinet_proto_dir[0].child = t->devinet_conf_dir;
-	t->devinet_proto_dir[0].de    = NULL;
 	t->devinet_root_dir[0].child  = t->devinet_proto_dir;
-	t->devinet_root_dir[0].de     = NULL;
 
 	t->sysctl_header = register_sysctl_table(t->devinet_root_dir);
 	if (!t->sysctl_header)
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 791aaba..6aded83 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3973,7 +3973,6 @@ static void addrconf_sysctl_register(struct inet6_dev *idev, struct ipv6_devconf
 		return;
 	for (i=0; t->addrconf_vars[i].data; i++) {
 		t->addrconf_vars[i].data += (char*)p - (char*)&ipv6_devconf;
-		t->addrconf_vars[i].de = NULL;
 		t->addrconf_vars[i].extra1 = idev; /* embedded; no ref */
 	}
 	if (dev) {
@@ -3996,13 +3995,9 @@ static void addrconf_sysctl_register(struct inet6_dev *idev, struct ipv6_devconf
 	t->addrconf_dev[0].procname = dev_name;
 
 	t->addrconf_dev[0].child = t->addrconf_vars;
-	t->addrconf_dev[0].de = NULL;
 	t->addrconf_conf_dir[0].child = t->addrconf_dev;
-	t->addrconf_conf_dir[0].de = NULL;
 	t->addrconf_proto_dir[0].child = t->addrconf_conf_dir;
-	t->addrconf_proto_dir[0].de = NULL;
 	t->addrconf_root_dir[0].child = t->addrconf_proto_dir;
-	t->addrconf_root_dir[0].de = NULL;
 
 	t->sysctl_header = register_sysctl_table(t->addrconf_root_dir);
 	if (t->sysctl_header == NULL)
-- 
1.4.4.1.g278f

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

Eric W. Biederman wrote:
> 
>> I think it would be fair to say that if they're not in <linux/sysctl.h> they're
>> not architectural, but that doesn't resolve the counterpositive (are there
>> sysctls in <linux/sysctl.h> which aren't architectural?  From the looks of it, I
>> would say yes.)  Non-architectural sysctl numbers should not be exported to
>> userspace, and should eventually be rejected by sys_sysctl.
> 
> This last bit doesn't make much sense.  I believe you are saying all sysctl
> numbers should be per architecture.
> 

With "architectural" I mean "guaranteed to be stable" (as opposed to 
"incidental").  Sorry for the confusion.

	-hpa
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

Eric W. Biederman wrote:
>>>
>> With "architectural" I mean "guaranteed to be stable" (as opposed to
>> "incidental").  Sorry for the confusion.
> 
> Ok.  Then largely we are in agreement.  To implement that the rule is simple.
> If it isn't CTL_UNNUMBERED and the number is in Linus's tree, it is
> our responsibility to never change the meaning of that number.
> 
> If a new sysctl entry is introduced it should be CTL_UNNUMBERED until
> it reaches Linus's tree to avoid conflicts.
> 
> There is simply no point in having any kind of support for numbers
> whose meanings can change.
> 
> Which is why I removed the few cases of binary number duplication I
> found.
> 

Agreed.  *Furthermore*, if the number isn't in <linux/sysctl.h> it 
shouldn't exist anywhere else, either.

	-hpa

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

Eric W. Biederman wrote:
>>
>> Agreed.  *Furthermore*, if the number isn't in <linux/sysctl.h> it shouldn't
>> exist anywhere else, either.
> 
> That would be a good habit.  Feel free to send the patches to ensure that
> is so.
> 
> I'm a practical fix it when it is in my way kind of guy ;)

That's fine.  However, I am wondering if there are things in 
<linux/sysctl.h> which really doesn't need architectural numbers, i.e. 
which should be removed from the binary interface.

	-hpa
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

"H. Peter Anvin" <hpa@zytor.com> writes:

> Eric W. Biederman wrote:
>>
>>> I think it would be fair to say that if they're not in <linux/sysctl.h>
> they're
>>> not architectural, but that doesn't resolve the counterpositive (are there
>>> sysctls in <linux/sysctl.h> which aren't architectural?  From the looks of
> it, I
>>> would say yes.)  Non-architectural sysctl numbers should not be exported to
>>> userspace, and should eventually be rejected by sys_sysctl.
>>
>> This last bit doesn't make much sense.  I believe you are saying all sysctl
>> numbers should be per architecture.
>>
>
> With "architectural" I mean "guaranteed to be stable" (as opposed to
> "incidental").  Sorry for the confusion.

Ok.  Then largely we are in agreement.  To implement that the rule is simple.
If it isn't CTL_UNNUMBERED and the number is in Linus's tree, it is
our responsibility to never change the meaning of that number.

If a new sysctl entry is introduced it should be CTL_UNNUMBERED until
it reaches Linus's tree to avoid conflicts.

There is simply no point in having any kind of support for numbers
whose meanings can change.

Which is why I removed the few cases of binary number duplication I
found.

Eric
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

"H. Peter Anvin" <hpa@zytor.com> writes:

> Eric W. Biederman wrote:
>>>>
>>> With "architectural" I mean "guaranteed to be stable" (as opposed to
>>> "incidental").  Sorry for the confusion.
>>
>> Ok.  Then largely we are in agreement.  To implement that the rule is simple.
>> If it isn't CTL_UNNUMBERED and the number is in Linus's tree, it is
>> our responsibility to never change the meaning of that number.
>>
>> If a new sysctl entry is introduced it should be CTL_UNNUMBERED until
>> it reaches Linus's tree to avoid conflicts.
>>
>> There is simply no point in having any kind of support for numbers
>> whose meanings can change.
>>
>> Which is why I removed the few cases of binary number duplication I
>> found.
>>
>
> Agreed.  *Furthermore*, if the number isn't in <linux/sysctl.h> it shouldn't
> exist anywhere else, either.

That would be a good habit.  Feel free to send the patches to ensure that
is so.

I'm a practical fix it when it is in my way kind of guy ;)

Eric
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

On Tue, 2007-01-16 at 09:39 -0700, Eric W. Biederman wrote:
> From: Eric W. Biederman <ebiederm@xmission.com> - unquoted

These three look OK:

[PATCH 15/59] sysctl: scsi remove unnecessary insert_at_head flag
[PATCH 19/59] sysctl: cdrom remove unnecessary insert_at_head flag
[PATCH 20/59] sysctl: cdrom Don't set de->owner

So you can add an ACK from me.

It would have been nice not to have 56 other irrelevant patches sprayed
over the list and into my inbox, though ...

James


_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

"H. Peter Anvin" <hpa@zytor.com> writes:

> Eric W. Biederman wrote:
>>>
>>> Agreed.  *Furthermore*, if the number isn't in <linux/sysctl.h> it shouldn't
>>> exist anywhere else, either.
>>
>> That would be a good habit.  Feel free to send the patches to ensure that
>> is so.
>>
>> I'm a practical fix it when it is in my way kind of guy ;)
>
> That's fine.  However, I am wondering if there are things in <linux/sysctl.h>
> which really doesn't need architectural numbers, i.e. which should be removed
> from the binary interface.

As a practical measure the 32bit hierarchical numeric namespace is extremely
sparsely populated.  So even if there are things that we never intend to export
again it doesn't hurt to reserve the number for them so we don't confuse something
that thought the number actually meant something.

In the worst case we want to comment out the entry so we keep the number
reserved even if we don't use it.

To see if there are entries we are not currently using just requires
going through the loop.

<pseudo shell>
For DEFINE in sysctl.h ; do
   if ! grep -r DEFINE /usr/src/linux/ ; then
   	echo DEFINE not used.
   fi
done
</pseudo shell>

Eric
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

On Tue, Jan 16, 2007 at 09:39:53AM -0700, Eric W. Biederman wrote:
> From: Eric W. Biederman <ebiederm@xmission.com> - unquoted
> 
> ocfs2 was did not have the binary number it uses under CTL_FS
> registered in sysctl.h.  Register it to avoid future conflicts,
> and change the name of the definition to be in line with the
> rest of the sysctl numbers.

This looks good - ACK.
	--Mark

--
Mark Fasheh
Senior Software Developer, Oracle
mark.fasheh@oracle.com
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

Ingo Oeser <ioe-lkml@rameria.de> writes:

> Hi Eric,
>
> On Tuesday, 16. January 2007 17:39, Eric W. Biederman wrote:
>> diff --git a/drivers/parport/procfs.c b/drivers/parport/procfs.c
>> index 2e744a2..5337789 100644
>> --- a/drivers/parport/procfs.c
>> +++ b/drivers/parport/procfs.c
>> @@ -263,50 +263,118 @@ struct parport_sysctl_table {
>> +		{
>> +			.ctl_name	= DEV_PARPORT_BASE_ADDR,
>> +			.procname	= "base-addr",
>> +			.data		= NULL,
>> +			.maxlen		= 0,
>> +			.mode		= 0444,
>> +			.proc_handler	= &do_hardware_base_addr
>> +		},
>
> No need to initialize to zero or NULL. Just list any variable, which is NOT zero
> or NULL.

Agreed.  In this case it was left for clarity.

>> +		{
>> +			.ctl_name	= DEV_PARPORT_AUTOPROBE + 1,
>> +			.procname	= "autoprobe0",
>> +			.data		= NULL,
>> +			.maxlen		= 0,
>> +			.maxlen		= 0444,
>> +			.proc_handler	=  &do_autoprobe
>> +		},
>
> Typo here? .mode = 0444 makes mor sense.

Yep looks like it.  On my todo.

Eric
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

On Tue, Jan 16, 2007 at 09:39:10AM -0700, Eric W. Biederman wrote:

Looks ok, for these:

Subject: [PATCH 5/59] sysctl: rose remove unnecessary insert_at_head flag
Subject: [PATCH 6/59] sysctl: netrom remove unnecessary insert_at_head flag
Subject: [PATCH 11/59] sysctl: ax25 remove unnecessary insert_at_head flag
Subject: [PATCH 30/59] sysctl: mips/au1000 Remove sys_sysctl support
Subject: [PATCH 31/59] sysctl: C99 convert the ctl_tables in arch/mips/au1000/common/power.c
Subject: [PATCH 32/59] sysctl: C99 convert arch/mips/lasat/sysctl.c and remove ABI breakage.
Subject: [PATCH 43/59] sysctl: Remove sys_sysctl support from drivers/char/rtc.c
Subject: [PATCH 55/59] sysctl: Remove insert_at_head from register_sysctl

Acked-by: Ralf Baechle <ralf@linux-mips.org>

  Ralf
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

The FRV bits look okay.  I can't test them until I get back from Australia in
Feb.

David
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

Hi Eric,

On Tuesday, 16. January 2007 17:39, Eric W. Biederman wrote:
> diff --git a/drivers/parport/procfs.c b/drivers/parport/procfs.c
> index 2e744a2..5337789 100644
> --- a/drivers/parport/procfs.c
> +++ b/drivers/parport/procfs.c
> @@ -263,50 +263,118 @@ struct parport_sysctl_table {
> +		{
> +			.ctl_name	= DEV_PARPORT_BASE_ADDR,
> +			.procname	= "base-addr",
> +			.data		= NULL,
> +			.maxlen		= 0,
> +			.mode		= 0444,
> +			.proc_handler	= &do_hardware_base_addr
> +		},

No need to initialize to zero or NULL. Just list any variable, which is NOT zero or NULL.

> +		{
> +			.ctl_name	= DEV_PARPORT_AUTOPROBE + 1,
> +			.procname	= "autoprobe0",
> +			.data		= NULL,
> +			.maxlen		= 0,
> +			.maxlen		= 0444,
> +			.proc_handler	=  &do_autoprobe
> +		},

Typo here? .mode = 0444 makes mor sense.

Regards

Ingo Oeser
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

Hi Eric,

On Tuesday, 16. January 2007 17:39, Eric W. Biederman wrote:
> diff --git a/arch/mips/au1000/common/power.c b/arch/mips/au1000/common/power.c
> index b531ab7..31256b8 100644
> --- a/arch/mips/au1000/common/power.c
> +++ b/arch/mips/au1000/common/power.c
> @@ -419,15 +419,41 @@ static int pm_do_freq(ctl_table * ctl, int write, struct file *file,

> +	{
> +		.ctl_name 	= CTL_UNNUMBERED,
> +		.procname	= "suspend",
> +		.data		= NULL,
> +		.maxlen		= 0,
> +		.mode		= 0600,
> +		.proc_handler	= &pm_do_suspend
> +	},

No need for zero initialization for maxlen.

> +	{
> +		.ctl_name	= CTL_UNNUMBERED,
> +		.procname	= "sleep",
> +		.data		= NULL,
> +		.maxlen		= 0,
> +		.mode		= 0600,
> +		.proc_handler	= &pm_do_sleep
> +	},

dito

> +	{
> +		.ctl_name	= CTL_UNNUMBERED,
> +		.procname	= "freq",
> +		.data		= NULL,
> +		.maxlen		= 0,
> +		.mode		= 0600,
> +		.proc_handler	= &pm_do_freq
> +	},
> +	{}
>  };

dito

Regards

Ingo Oeser
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

On Tue, Jan 16, 2007 at 09:39:42AM -0700, Eric W. Biederman wrote:
> From: Eric W. Biederman <ebiederm@xmission.com> - unquoted
> 
> While doing the C99 conversion I notices that the top level sh64
> directory was using the binary number for CTL_KERN.  That is a
> no-no so I removed the support for the sysctl binary interface
> only leaving sysctl /proc support.
> 
> At least the sysctl tables were placed at the end of
> the list so user space did not see this mistake.
> 
> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>

Looks good, thanks Eric.

Acked-by: Paul Mundt <lethal@linux-sh.org>
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

Kirill Korotaev <dev@sw.ru> writes:

> Eric, really good job!
>
> Patches: 1-13, 15-24, 26-32, 34-44, 46-49, 52-55, 57 (all except below)
> Acked-By: Kirill Korotaev <dev@openvz.org>
>
> 14/59 - minor (extra space)
> 25/59 - minor note	
> 33/59 - not sorted sysctl IDs
> 45/59 - typo
> 50/59 - copyright/file note
> 51/59 - copyright/file name/kconfig option notes
>
> 56,58,59/59 - will review tomorrow
>
> another issue I have to think over is removal of de->owner.
> Alexey Dobriyan has sent recently patching fixing /proc <-> modules refcounting.
> I guess w/o these patches your changes are not safe if proc_handler or strategy
> are functions from the module.

sysctl uses the logic in use_table/unuse_table to keep it safe from module
remove while it is in use.  And it does the logic in the generic code
in either do_rw_proc or do_sysctl.  This definitely works on the sys_sysctl path
and it appears to work in the do_rw_proc case, things are a little trickier
there so someone may have missed a race somewhere.  In my rewrite of proc
it works exactly like the binary case so we are good there. 

It is certainly the intention of the sysctl implementation that users
should not have to set de->owner.  So if there is a problem with 
removing de->owner it is a bug in the sysctl implementation not in
the code where it was removed.

Normal proc users definitely have to set de->owner to be safe, but sysctl has
always been it's own thing, with different rules. 

Eric
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

Kirill Korotaev <dev@sw.ru> writes:

> Eric, though I personally don't care much:
> 1. I ask for not setting your authorship/copyright on the code which you just
> copied
>   from other places. Just doesn't look polite IMHO.

I can't claim complete ownership of the code, there was plenty of feed back
and contributions from others but the final form without a big switch
statement is mine.  I certainly can't claim the table, it has been in
that form for years.

If you notice I actually didn't say whose copyright it was :)  just
that I wrote the file.

If there are copyright claims I should include I will be happy to do that.
Mostly I was just trying to find some stupid boiler plate that would work.

> 2. I would propose to not introduce utsname_sysctl.c.
>   both files are too small and minor that I can't see much reasons splitting
> them.

The impact of moving this code out of sysctl.c is a major
simplification, to sysctl.c.  Putting them in their own file means we
can cleanly restrict the code to only be compiled CONFIG_SYSCTL is set.

It is a necessary first step to implementing a per process /proc/sys.

It reorganizes the ipc and utsname sysctl from a terribly fragile
structure to something that is robust and easy to follow.  Code
scattered all throughout sysctl.c was just a disaster.  We had
several instances of having to fix bugs with odd combinations of
CONFIG options, simply because the other spot that needed to be touched
wasn't obvious.

So from my perspective this is an extremely worthwhile change that
will make maintenance easier and is a small first step towards
some nice future functionality.

Eric
_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

On Tue, 2007-01-16 at 09:33 -0700, Eric W. Biederman wrote:
> There has not been much maintenance on sysctl in years, and as a result is
> there is a lot to do to allow future interesting work to happen, and being
> ambitious I'm trying to do it all at once :)

s390 parts look good. Kernels boots and the system controls are still
working. I had to add an #include <linux/uaccess.h> to ipc/ipc_sysctl.c
to get the kernel compiled. That include should be added to patch #51.

Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> for:
[PATCH 33/59] sysctl: s390 move sysctl definitions to sysctl.h
[PATCH 34/59] sysctl: s390 Remove unnecessary use of insert_at_head

and the s390 parts of 
[PATCH 55/59] sysctl: Remove insert_at_head from register_sysctl

-- 
blue skies,
  Martin.

Martin Schwidefsky
Linux for zSeries Development & Services
IBM Deutschland Entwicklung GmbH

"Reality continues to ruin my life." - Calvin.


_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

On Tue, 2007-01-16 at 09:39 -0700, Eric W. Biederman wrote:
> From: Eric W. Biederman <ebiederm@xmission.com> - unquoted
> 
> With unique sysctl binary numbers setting insert_at_head is pointless.
> 
> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>

Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

> ---
>  drivers/char/ipmi/ipmi_poweroff.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/char/ipmi/ipmi_poweroff.c b/drivers/char/ipmi/ipmi_poweroff.c
> index 9d23136..b3ae65e 100644
> --- a/drivers/char/ipmi/ipmi_poweroff.c
> +++ b/drivers/char/ipmi/ipmi_poweroff.c
> @@ -686,7 +686,7 @@ static int ipmi_poweroff_init (void)
>  		printk(KERN_INFO PFX "Power cycle is enabled.\n");
>  
>  #ifdef CONFIG_PROC_FS
> -	ipmi_table_header = register_sysctl_table(ipmi_root_table, 1);
> +	ipmi_table_header = register_sysctl_table(ipmi_root_table, 0);
>  	if (!ipmi_table_header) {
>  		printk(KERN_ERR PFX "Unable to register powercycle sysctl\n");
>  		rv = -ENOMEM;

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

On Tue, 2007-01-16 at 09:39 -0700, Eric W. Biederman wrote:
> From: Eric W. Biederman <ebiederm@xmission.com> - unquoted
> 
> This was partially done already and there was no ABI breakage what
> a relief.
> 
> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>

Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

> ---
>  arch/powerpc/kernel/idle.c |   11 ++++++++---
>  1 files changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c
> index 8994af3..8b27bb1 100644
> --- a/arch/powerpc/kernel/idle.c
> +++ b/arch/powerpc/kernel/idle.c
> @@ -110,11 +110,16 @@ static ctl_table powersave_nap_ctl_table[]={
>  		.mode		= 0644,
>  		.proc_handler	= &proc_dointvec,
>  	},
> -	{ 0, },
> +	{}
>  };
>  static ctl_table powersave_nap_sysctl_root[] = {
> -	{ 1, "kernel", NULL, 0, 0755, powersave_nap_ctl_table, },
> - 	{ 0,},
> +	{
> +		.ctl_name	= CTL_KERN,
> +		.procname	= "kernel",
> +		.mode		= 0755,
> +		.child		= powersave_nap_ctl_table,
> +	},
> +	{}
>  };
>  
>  static int __init

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers

On Tue, 2007-01-16 at 09:39 -0700, Eric W. Biederman wrote:
> From: Eric W. Biederman <ebiederm@xmission.com> - unquoted
> 
> And make the mode of the kernel directory 0555 no one is allowed
> to write to sysctl directories.
> 
> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>

Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>

> ---
>  arch/ppc/kernel/ppc_htab.c |   11 ++++++++---
>  1 files changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/ppc/kernel/ppc_htab.c b/arch/ppc/kernel/ppc_htab.c
> index bd129d3..77b20ff 100644
> --- a/arch/ppc/kernel/ppc_htab.c
> +++ b/arch/ppc/kernel/ppc_htab.c
> @@ -442,11 +442,16 @@ static ctl_table htab_ctl_table[]={
>  		.mode		= 0644,
>  		.proc_handler	= &proc_dol2crvec,
>  	},
> -	{ 0, },
> +	{}
>  };
>  static ctl_table htab_sysctl_root[] = {
> -	{ 1, "kernel", NULL, 0, 0755, htab_ctl_table, },
> - 	{ 0,},
> +	{
> +		.ctl_name	= CTL_KERN,
> +		.procname	= "kernel",
> +		.mode		= 0555,
> +		.child		= htab_ctl_table,
> +	},
> +	{}
>  };
>  
>  static int __init

_______________________________________________
Containers mailing list
Containers@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/containers