Skip to contentSkip to conversation

fs/sysctl: fix potential page fault while unregistering sysctl table

PATCH Superseded Danilo Krummrich 2 participants 6 messages Feb 27, 2018

Subsystems (3)

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 fs/sysctl: fix potential page fault while unregistering sysctl table

proc_sys_link_fill_cache() does not take currently unregistering sysctl tables into account, which might result into a page fault in sysctl_follow_link() - add a check to fix it.

---
 fs/proc/proc_sysctl.c | 3 +++
 1 file changed, 3 insertions(+)
fs/proc/proc_sysctl.c +3 -0
index c5cbbdff3c3d..a0b6c647835e 100644
@@ -709,6 +709,9 @@ static bool proc_sys_link_fill_cache(struct file *file,
 	bool ret = true;
 	head = sysctl_head_grab(head);
 
+	if (IS_ERR(head))
+		return false;
+
 	if (S_ISLNK(table->mode)) {
 		/* It is not an error if we can not follow the link ignore it */
 		int err = sysctl_follow_link(&head, &table);
Signature · 1 line
2.14.1
#1 lore raw

PATCH 2/2 fs/sysctl: remove redundant link check in proc_sys_link_fill_cache()

proc_sys_link_fill_cache() does not need to check whether we're called for a link - it's already done by scan().

---
 fs/proc/proc_sysctl.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
fs/proc/proc_sysctl.c +5 -6
index a0b6c647835e..7e7d9facb842 100644
@@ -707,17 +707,16 @@ static bool proc_sys_link_fill_cache(struct file *file,
 				    struct ctl_table *table)
 {
 	bool ret = true;
+	int err = 0;
 	head = sysctl_head_grab(head);
 
 	if (IS_ERR(head))
 		return false;
 
-	if (S_ISLNK(table->mode)) {
-		/* It is not an error if we can not follow the link ignore it */
-		int err = sysctl_follow_link(&head, &table);
-		if (err)
-			goto out;
-	}
+	/* It is not an error if we can not follow the link ignore it */
+	sysctl_follow_link(&head, &table);
+	if (err)
+		goto out;
 
 	ret = proc_sys_fill_cache(file, ctx, head, table);
 out:
Signature · 1 line
2.14.1
#2 lore raw

Re: [PATCH 2/2] fs/sysctl: remove redundant link check in proc_sys_link_fill_cache()

Quoted 20 lines · Danilo Krummrich

proc_sys_link_fill_cache() does not need to check whether we're called for a link - it's already done by scan().

---
 fs/proc/proc_sysctl.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
fs/proc/proc_sysctl.c +1 -0
index a0b6c647835e..7e7d9facb842 100644
@@ -707,17 +707,16 @@ static bool proc_sys_link_fill_cache(struct file *file,
                                    struct ctl_table *table)
 {
        bool ret = true;
+       int err = 0;
        head = sysctl_head_grab(head);
 
        if (IS_ERR(head))

fs/proc/proc_sysctl.c static bool proc_sys_link_fill_cache(struct file *file, · line 717

Quoted 10 lines
fs/proc/proc_sysctl.c +2 -6
                return false;
 
-       if (S_ISLNK(table->mode)) {
-               /* It is not an error if we can not follow the link ignore it */
-               int err = sysctl_follow_link(&head, &table);
-               if (err)
-                       goto out;
-       }
+       /* It is not an error if we can not follow the link ignore it */
+       sysctl_follow_link(&head, &table);

Shouldn't this be err = sysctl_follow_link... ? Otherwise I don't see where err is used.

-Kees

Quoted 8 lines
fs/proc/proc_sysctl.c +2 -1
+       if (err)
+               goto out;
 
        ret = proc_sys_fill_cache(file, ctx, head, table);
 out:
--

2.14.1

Signature · 2 lines
Kees Cook
Pixel Security
#3 lore raw

Re: [PATCH 2/2] fs/sysctl: remove redundant link check in proc_sys_link_fill_cache()

Quoted 39 lines · Kees Cook
Quoted 31 lines · Danilo Krummrich

proc_sys_link_fill_cache() does not need to check whether we're called for a link - it's already done by scan().

---
 fs/proc/proc_sysctl.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
fs/proc/proc_sysctl.c +1 -2
index a0b6c647835e..7e7d9facb842 100644
@@ -707,17 +707,16 @@ static bool proc_sys_link_fill_cache(struct file 
 *file,
                                    struct ctl_table *table)
 {
        bool ret = true;
+       int err = 0;
        head = sysctl_head_grab(head);
 
        if (IS_ERR(head))
                return false;
 
-       if (S_ISLNK(table->mode)) {
-               /* It is not an error if we can not follow the link 

ignore it */

fs/proc/proc_sysctl.c +1 -4
-               int err = sysctl_follow_link(&head, &table);
-               if (err)
-                       goto out;
-       }
+       /* It is not an error if we can not follow the link ignore it 
Quoted 2 lines

*/
+ sysctl_follow_link(&head, &table);

Shouldn't this be err = sysctl_follow_link... ? Otherwise I don't see where err is used.

Quoted 2 lines

-Kees

Of course, thanks.

Quoted 8 lines
Quoted 8 lines
fs/proc/proc_sysctl.c +2 -1
+       if (err)
+               goto out;
 
        ret = proc_sys_fill_cache(file, ctx, head, table);
 out:
--

2.14.1

#4 lore raw
In reply to Danilo Krummrich
Quoted 14 lines · Danilo Krummrich

proc_sys_link_fill_cache() does not take currently unregistering sysctl tables into account, which might result into a page fault in sysctl_follow_link() - add a check to fix it.

---
 fs/proc/proc_sysctl.c | 3 +++
 1 file changed, 3 insertions(+)
fs/proc/proc_sysctl.c +0 -0
index c5cbbdff3c3d..a0b6c647835e 100644

fs/proc/proc_sysctl.c static bool proc_sys_link_fill_cache(struct file *file, · line 709

Quoted 1 line
fs/proc/proc_sysctl.c +0 -0
@@ -709,6 +709,9 @@ static bool proc_sys_link_fill_cache(struct file *file,
        bool ret = true;

Nothing appears to actually change "ret" in this function. It should likely be dropped too.

fs/proc/proc_sysctl.c static bool proc_sys_link_fill_cache(struct file *file, · line 714

Quoted 5 lines
fs/proc/proc_sysctl.c +3 -0
        head = sysctl_head_grab(head);
 
+       if (IS_ERR(head))
+               return false;
+

This looks sensible. I'd drop the blank line between sysctl_head_grab and the IS_ERR, though.

How are you testing this change?

Thanks!

-Kees

Quoted 6 lines
fs/proc/proc_sysctl.c +0 -1
        if (S_ISLNK(table->mode)) {
                /* It is not an error if we can not follow the link ignore it */
                int err = sysctl_follow_link(&head, &table);
--

2.14.1

Signature · 2 lines
Kees Cook
Pixel Security
#5 lore raw
Quoted 11 lines · Kees Cook
Quoted 9 lines · Danilo Krummrich

proc_sys_link_fill_cache() does not take currently unregistering sysctl tables into account, which might result into a page fault in sysctl_follow_link() - add a check to fix it.

---
 fs/proc/proc_sysctl.c | 3 +++
 1 file changed, 3 insertions(+)
Quoted 11 lines

fs/proc/proc_sysctl.c static bool proc_sys_link_fill_cache(struct file · line 710

Quoted 7 lines
fs/proc/proc_sysctl.c +0 -0
index c5cbbdff3c3d..a0b6c647835e 100644
@@ -709,6 +709,9 @@ static bool proc_sys_link_fill_cache(struct file 
 *file,
        bool ret = true;

Nothing appears to actually change "ret" in this function. It should likely be dropped too.

proc_sys_fill_cache() potentially changes "ret".

Quoted 9 lines

fs/proc/proc_sysctl.c static bool proc_sys_link_fill_cache(struct file *file, · line 714

Quoted 5 lines
fs/proc/proc_sysctl.c +3 -0
        head = sysctl_head_grab(head);
 
+       if (IS_ERR(head))
+               return false;
+

This looks sensible. I'd drop the blank line between sysctl_head_grab and the IS_ERR, though.

I'll do that.

Quoted 2 lines

How are you testing this change?

Honestly, not at all. Actually, I never run in such a page fault. I spotted it by accident while reading the code.

Quoted 11 lines

Thanks!

-Kees

Quoted 7 lines
fs/proc/proc_sysctl.c +0 -0
        if (S_ISLNK(table->mode)) {
                /* It is not an error if we can not follow the link 

ignore it */

                int err = sysctl_follow_link(&head, &table);

--
2.14.1