* ptr_ring_unconsume: memory corruption potential?
@ 2025-09-24 3:58 Michael S. Tsirkin
2025-09-24 4:03 ` Michael S. Tsirkin
0 siblings, 1 reply; 2+ messages in thread
From: Michael S. Tsirkin @ 2025-09-24 3:58 UTC (permalink / raw)
To: Jason Wang; +Cc: netdev
Jason, guys,
reading ptr ring code, I noticed:
static inline void ptr_ring_unconsume(struct ptr_ring *r, void **batch, int n,
void (*destroy)(void *))
{
unsigned long flags;
int head;
spin_lock_irqsave(&r->consumer_lock, flags);
spin_lock(&r->producer_lock);
if (!r->size)
goto done;
/*
* Clean out buffered entries (for simplicity). This way following code
* can test entries for NULL and if not assume they are valid.
*/
head = r->consumer_head - 1;
while (likely(head >= r->consumer_tail))
r->queue[head--] = NULL;
__ptr_ring_update(r, head);
r->consumer_tail = r->consumer_head;
...
Does not look like this will DTRT if r->consumer_head == 0 .
In fact it looks like it will go off corrupting memory.
Why isn't this a concern?
--
MST
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: ptr_ring_unconsume: memory corruption potential?
2025-09-24 3:58 ptr_ring_unconsume: memory corruption potential? Michael S. Tsirkin
@ 2025-09-24 4:03 ` Michael S. Tsirkin
0 siblings, 0 replies; 2+ messages in thread
From: Michael S. Tsirkin @ 2025-09-24 4:03 UTC (permalink / raw)
To: Jason Wang; +Cc: netdev
On Tue, Sep 23, 2025 at 11:58:40PM -0400, Michael S. Tsirkin wrote:
> Jason, guys,
> reading ptr ring code, I noticed:
>
Pls ignore. ENOTENOUGHCOFFEE. It's a nop.
> static inline void ptr_ring_unconsume(struct ptr_ring *r, void **batch, int n,
> void (*destroy)(void *))
> {
> unsigned long flags;
> int head;
>
> spin_lock_irqsave(&r->consumer_lock, flags);
> spin_lock(&r->producer_lock);
>
> if (!r->size)
> goto done;
>
> /*
> * Clean out buffered entries (for simplicity). This way following code
> * can test entries for NULL and if not assume they are valid.
> */
> head = r->consumer_head - 1;
> while (likely(head >= r->consumer_tail))
> r->queue[head--] = NULL;
> __ptr_ring_update(r, head);
> r->consumer_tail = r->consumer_head;
>
>
> ...
>
>
>
> Does not look like this will DTRT if r->consumer_head == 0 .
> In fact it looks like it will go off corrupting memory.
>
> Why isn't this a concern?
>
> --
> MST
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2025-09-24 4:03 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-09-24 3:58 ptr_ring_unconsume: memory corruption potential? Michael S. Tsirkin
2025-09-24 4:03 ` Michael S. Tsirkin
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).