All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] fs/proc/kcore.c: remove check of list iterator against head past the loop body
@ 2022-03-31 22:37 Jakob Koschel
  2022-03-31 23:48 ` Andrew Morton
  0 siblings, 1 reply; 5+ messages in thread
From: Jakob Koschel @ 2022-03-31 22:37 UTC (permalink / raw)
  To: Mike Rapoport
  Cc: David Hildenbrand, Andrew Morton, Oscar Salvador, linux-kernel,
	linux-fsdevel, Brian Johannesmeyer, Cristiano Giuffrida, Bos,
	H.J.,
	Jakob Koschel

When list_for_each_entry() completes the iteration over the whole list
without breaking the loop, the iterator value will be a bogus pointer
computed based on the head element.

While it is safe to use the pointer to determine if it was computed
based on the head element, either with list_entry_is_head() or
&pos->member == head, using the iterator variable after the loop should
be avoided.

In preparation to limit the scope of a list iterator to the list
traversal loop, use a dedicated pointer to point to the found element [1].

Link: https://lore.kernel.org/all/CAHk-=wgRr_D8CB-D9Kg-c=EHreAsk5SqXPwr9Y7k9sA6cWXJ6w@mail.gmail.com/ [1]
Signed-off-by: Jakob Koschel <jakobkoschel@gmail.com>
---
 fs/proc/kcore.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c
index 982e694aae77..344edcb2addd 100644
--- a/fs/proc/kcore.c
+++ b/fs/proc/kcore.c
@@ -316,6 +316,7 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 	size_t page_offline_frozen = 1;
 	size_t phdrs_len, notes_len;
 	struct kcore_list *m;
+	struct kcore_list *iter;
 	size_t tsz;
 	int nphdr;
 	unsigned long start;
@@ -479,10 +480,13 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 		 * the previous entry, search for a matching entry.
 		 */
 		if (!m || start < m->addr || start >= m->addr + m->size) {
-			list_for_each_entry(m, &kclist_head, list) {
-				if (start >= m->addr &&
-				    start < m->addr + m->size)
+			m = NULL;
+			list_for_each_entry(iter, &kclist_head, list) {
+				if (start >= iter->addr &&
+				    start < iter->addr + iter->size) {
+					m = iter;
 					break;
+				}
 			}
 		}
 
@@ -492,12 +496,11 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 			page_offline_freeze();
 		}
 
-		if (&m->list == &kclist_head) {
+		if (!m) {
 			if (clear_user(buffer, tsz)) {
 				ret = -EFAULT;
 				goto out;
 			}
-			m = NULL;	/* skip the list anchor */
 			goto skip;
 		}
 

base-commit: f82da161ea75dc4db21b2499e4b1facd36dab275
-- 
2.25.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH] fs/proc/kcore.c: remove check of list iterator against head past the loop body
  2022-03-31 22:37 [PATCH] fs/proc/kcore.c: remove check of list iterator against head past the loop body Jakob Koschel
@ 2022-03-31 23:48 ` Andrew Morton
  2022-04-01  7:19   ` Jakob Koschel
  0 siblings, 1 reply; 5+ messages in thread
From: Andrew Morton @ 2022-03-31 23:48 UTC (permalink / raw)
  To: Jakob Koschel
  Cc: Mike Rapoport, David Hildenbrand, Oscar Salvador, linux-kernel,
	linux-fsdevel, Brian Johannesmeyer, Cristiano Giuffrida, Bos,
	H.J.

On Fri,  1 Apr 2022 00:37:00 +0200 Jakob Koschel <jakobkoschel@gmail.com> wrote:

> When list_for_each_entry() completes the iteration over the whole list
> without breaking the loop, the iterator value will be a bogus pointer
> computed based on the head element.
> 
> While it is safe to use the pointer to determine if it was computed
> based on the head element, either with list_entry_is_head() or
> &pos->member == head, using the iterator variable after the loop should
> be avoided.
> 
> In preparation to limit the scope of a list iterator to the list
> traversal loop, use a dedicated pointer to point to the found element [1].
> 
> ...
>

Speaking of limiting scope...

--- a/fs/proc/kcore.c~fs-proc-kcorec-remove-check-of-list-iterator-against-head-past-the-loop-body-fix
+++ a/fs/proc/kcore.c
@@ -316,7 +316,6 @@ read_kcore(struct file *file, char __use
 	size_t page_offline_frozen = 1;
 	size_t phdrs_len, notes_len;
 	struct kcore_list *m;
-	struct kcore_list *iter;
 	size_t tsz;
 	int nphdr;
 	unsigned long start;
@@ -480,6 +479,8 @@ read_kcore(struct file *file, char __use
 		 * the previous entry, search for a matching entry.
 		 */
 		if (!m || start < m->addr || start >= m->addr + m->size) {
+			struct kcore_list *iter;
+
 			m = NULL;
 			list_for_each_entry(iter, &kclist_head, list) {
 				if (start >= iter->addr &&
_


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] fs/proc/kcore.c: remove check of list iterator against head past the loop body
  2022-03-31 23:48 ` Andrew Morton
@ 2022-04-01  7:19   ` Jakob Koschel
  2022-04-04 22:55     ` Andrew Morton
  0 siblings, 1 reply; 5+ messages in thread
From: Jakob Koschel @ 2022-04-01  7:19 UTC (permalink / raw)
  To: Andrew Morton
  Cc: Mike Rapoport, David Hildenbrand, Oscar Salvador, linux-kernel,
	linux-fsdevel, Brian Johannesmeyer, Cristiano Giuffrida, Bos,
	H.J.


> On 1. Apr 2022, at 01:48, Andrew Morton <akpm@linux-foundation.org> wrote:
> 
> On Fri,  1 Apr 2022 00:37:00 +0200 Jakob Koschel <jakobkoschel@gmail.com> wrote:
> 
>> When list_for_each_entry() completes the iteration over the whole list
>> without breaking the loop, the iterator value will be a bogus pointer
>> computed based on the head element.
>> 
>> While it is safe to use the pointer to determine if it was computed
>> based on the head element, either with list_entry_is_head() or
>> &pos->member == head, using the iterator variable after the loop should
>> be avoided.
>> 
>> In preparation to limit the scope of a list iterator to the list
>> traversal loop, use a dedicated pointer to point to the found element [1].
>> 
>> ...
>> 
> 
> Speaking of limiting scope...

Fair point :-)

I see you have applied this already to the -mm tree. Shall I still move the iterator?
The hope is to remove the 'iter' variable altogether when there are no uses after
the loop anymore.

> 
> --- a/fs/proc/kcore.c~fs-proc-kcorec-remove-check-of-list-iterator-against-head-past-the-loop-body-fix
> +++ a/fs/proc/kcore.c
> @@ -316,7 +316,6 @@ read_kcore(struct file *file, char __use
> 	size_t page_offline_frozen = 1;
> 	size_t phdrs_len, notes_len;
> 	struct kcore_list *m;
> -	struct kcore_list *iter;
> 	size_t tsz;
> 	int nphdr;
> 	unsigned long start;
> @@ -480,6 +479,8 @@ read_kcore(struct file *file, char __use
> 		 * the previous entry, search for a matching entry.
> 		 */
> 		if (!m || start < m->addr || start >= m->addr + m->size) {
> +			struct kcore_list *iter;
> +
> 			m = NULL;
> 			list_for_each_entry(iter, &kclist_head, list) {
> 				if (start >= iter->addr &&
> _
> 

	Jakob

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] fs/proc/kcore.c: remove check of list iterator against head past the loop body
  2022-04-01  7:19   ` Jakob Koschel
@ 2022-04-04 22:55     ` Andrew Morton
  2022-04-05  8:48       ` Jakob Koschel
  0 siblings, 1 reply; 5+ messages in thread
From: Andrew Morton @ 2022-04-04 22:55 UTC (permalink / raw)
  To: Jakob Koschel
  Cc: Mike Rapoport, David Hildenbrand, Oscar Salvador, linux-kernel,
	linux-fsdevel, Brian Johannesmeyer, Cristiano Giuffrida, Bos,
	H.J.

On Fri, 1 Apr 2022 09:19:57 +0200 Jakob Koschel <jakobkoschel@gmail.com> wrote:

> > Speaking of limiting scope...
> 
> Fair point :-)
> 
> I see you have applied this already to the -mm tree. Shall I still move the iterator?
> The hope is to remove the 'iter' variable altogether when there are no uses after
> the loop anymore.

I don't really understand the question.

My plan is to merge your patch with my fixlet immediately prior to
sending upstream.


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] fs/proc/kcore.c: remove check of list iterator against head past the loop body
  2022-04-04 22:55     ` Andrew Morton
@ 2022-04-05  8:48       ` Jakob Koschel
  0 siblings, 0 replies; 5+ messages in thread
From: Jakob Koschel @ 2022-04-05  8:48 UTC (permalink / raw)
  To: Andrew Morton
  Cc: Mike Rapoport, David Hildenbrand, Oscar Salvador, LKML,
	linux-fsdevel, Brian Johannesmeyer, Cristiano Giuffrida, Bos,
	H.J.



> On 5. Apr 2022, at 00:55, Andrew Morton <akpm@linux-foundation.org> wrote:
> 
> On Fri, 1 Apr 2022 09:19:57 +0200 Jakob Koschel <jakobkoschel@gmail.com> wrote:
> 
>>> Speaking of limiting scope...
>> 
>> Fair point :-)
>> 
>> I see you have applied this already to the -mm tree. Shall I still move the iterator?
>> The hope is to remove the 'iter' variable altogether when there are no uses after
>> the loop anymore.
> 
> I don't really understand the question.

Basically I was asking if I should send a v2 with the change you suggested.
> 
> My plan is to merge your patch with my fixlet immediately prior to
> sending upstream.

ok, great. Even better, so ignore my question.

Thanks,
Jakob

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2022-04-05 12:19 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-03-31 22:37 [PATCH] fs/proc/kcore.c: remove check of list iterator against head past the loop body Jakob Koschel
2022-03-31 23:48 ` Andrew Morton
2022-04-01  7:19   ` Jakob Koschel
2022-04-04 22:55     ` Andrew Morton
2022-04-05  8:48       ` Jakob Koschel

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.