diff mbox series

[v1,net,08/16] sysctl: Add proc_dointvec_jiffies_lockless().

Message ID 20220706052130.16368-9-kuniyu@amazon.com (mailing list archive)
State Superseded
Delegated to: Netdev Maintainers
Headers show
Series sysctl: Fix data-races around ipv4_table. | expand

Checks

Context Check Description
netdev/tree_selection success Clearly marked for net, async
netdev/fixes_present success Fixes tag present in non-next series
netdev/subject_prefix success Link
netdev/cover_letter success Series has a cover letter
netdev/patch_count fail Series longer than 15 patches (and no cover letter)
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 17369 this patch: 17369
netdev/cc_maintainers warning 1 maintainers not CCed: linux-fsdevel@vger.kernel.org
netdev/build_clang success Errors and warnings before: 3291 this patch: 3291
netdev/module_param success Was 0 now: 0
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success Fixes tag looks correct
netdev/build_allmodconfig_warn success Errors and warnings before: 16543 this patch: 16543
netdev/checkpatch success total: 0 errors, 0 warnings, 0 checks, 70 lines checked
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0

Commit Message

Kuniyuki Iwashima July 6, 2022, 5:21 a.m. UTC
A sysctl variable is accessed concurrently, and there is always a chance of
data-race.  So, all readers and writers need some basic protection to avoid
load/store-tearing.

This patch changes proc_dointvec_jiffies() to use READ_ONCE()/WRITE_ONCE()
internally to fix a data-race on the sysctl side.  For now,
proc_dointvec_jiffies() itself is tolerant to a data-race, but we still
need to add annotations on the other subsystem's side.

In case we miss such fixes, this patch converts proc_dointvec_jiffies() to
a wrapper of proc_dointvec_jiffies_lockless().  When we fix a data-race in
the other subsystem, we can explicitly set it as a handler.

Also, this patch removes proc_dointvec_jiffies()'s document and adds
proc_dointvec_jiffies_lockless()'s one so that no one will use
proc_dointvec_jiffies() anymore.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com>
---
 include/linux/sysctl.h |  1 +
 kernel/sysctl.c        | 26 ++++++++++++++++++++------
 2 files changed, 21 insertions(+), 6 deletions(-)
diff mbox series

Patch

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index c23b6beef748..8747dbc721f5 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -89,6 +89,7 @@  PROC_HANDLER(proc_douintvec_lockless);
 PROC_HANDLER(proc_dointvec_minmax_lockless);
 PROC_HANDLER(proc_douintvec_minmax_lockless);
 PROC_HANDLER(proc_doulongvec_minmax_lockless);
+PROC_HANDLER(proc_dointvec_jiffies_lockless);
 
 /*
  * Register a set of sysctl names by calling register_sysctl_table
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 931ab58985f2..11a1ce837623 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1221,10 +1221,15 @@  static int do_proc_dointvec_jiffies_conv(bool *negp, unsigned long *lvalp,
 	if (write) {
 		if (*lvalp > INT_MAX / HZ)
 			return 1;
-		*valp = *negp ? -(*lvalp*HZ) : (*lvalp*HZ);
+
+		if (*negp)
+			WRITE_ONCE(*valp, -(*lvalp * HZ));
+		else
+			WRITE_ONCE(*valp, *lvalp * HZ);
 	} else {
-		int val = *valp;
+		int val = READ_ONCE(*valp);
 		unsigned long lval;
+
 		if (val < 0) {
 			*negp = true;
 			lval = -(unsigned long)val;
@@ -1286,7 +1291,8 @@  static int do_proc_dointvec_ms_jiffies_conv(bool *negp, unsigned long *lvalp,
 }
 
 /**
- * proc_dointvec_jiffies - read a vector of integers as seconds
+ * proc_dointvec_jiffies_lockless - read/write a vector of integers as
+ * seconds locklessly
  * @table: the sysctl table
  * @write: %TRUE if this is a write to the sysctl file
  * @buffer: the user buffer
@@ -1294,17 +1300,23 @@  static int do_proc_dointvec_ms_jiffies_conv(bool *negp, unsigned long *lvalp,
  * @ppos: file position
  *
  * Reads/writes up to table->maxlen/sizeof(unsigned int) integer
- * values from/to the user buffer, treated as an ASCII string. 
+ * values from/to the user buffer, treated as an ASCII string.
  * The values read are assumed to be in seconds, and are converted into
  * jiffies.
  *
  * Returns 0 on success.
  */
+int proc_dointvec_jiffies_lockless(struct ctl_table *table, int write,
+				   void *buffer, size_t *lenp, loff_t *ppos)
+{
+	return do_proc_dointvec(table, write, buffer, lenp, ppos,
+				do_proc_dointvec_jiffies_conv, NULL);
+}
+
 int proc_dointvec_jiffies(struct ctl_table *table, int write,
 			  void *buffer, size_t *lenp, loff_t *ppos)
 {
-    return do_proc_dointvec(table,write,buffer,lenp,ppos,
-		    	    do_proc_dointvec_jiffies_conv,NULL);
+	return proc_dointvec_jiffies_lockless(table, write, buffer, lenp, ppos);
 }
 
 /**
@@ -1552,6 +1564,7 @@  PROC_HANDLER_ENOSYS(proc_douintvec_lockless);
 PROC_HANDLER_ENOSYS(proc_dointvec_minmax_lockless);
 PROC_HANDLER_ENOSYS(proc_douintvec_minmax_lockless);
 PROC_HANDLER_ENOSYS(proc_doulongvec_minmax_lockless);
+PROC_HANDLER_ENOSYS(proc_dointvec_jiffies_lockless);
 
 #endif /* CONFIG_PROC_SYSCTL */
 
@@ -2468,3 +2481,4 @@  EXPORT_SYMBOL(proc_douintvec_lockless);
 EXPORT_SYMBOL(proc_dointvec_minmax_lockless);
 EXPORT_SYMBOL_GPL(proc_douintvec_minmax_lockless);
 EXPORT_SYMBOL(proc_doulongvec_minmax_lockless);
+EXPORT_SYMBOL(proc_dointvec_jiffies_lockless);