Re: [PATCH v2 1/3] io_uring/rsrc: fix folio unpinning

[Pavel Begunkov <asml.silence@gmail.com>]

On 6/25/25 08:53, David Hildenbrand wrote:
...>>   static struct io_mapped_ubuf *io_alloc_imu(struct io_ring_ctx *ctx,
>> @@ -810,7 +813,8 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,
>>       imu->nr_bvecs = nr_pages;
>>       ret = io_buffer_account_pin(ctx, pages, nr_pages, imu, last_hpage);
>>       if (ret) {
>> -        unpin_user_pages(pages, nr_pages);
>> +        for (i = 0; i < nr_pages; i++)
>> +            unpin_user_folio(page_folio(pages[i]), 1);
>>           goto done;
>>       }
> 
> It should fix the issue, but it's a bit suboptimal in the case where we didn't coalesc, but there are folio ranges to coalesc:
> 
> unpin_user_pages() does a per-folio coalescing.
> 
> So in an ideal world, we would cleanly split both paths, and work with folios after we coalesced to use folios, and work with pages, when we didn't coalesc to use folios.

Agreed, but I'm not too much concerned, it's a slow path, users
should be mindful while registering memory. Localised hammering
on the refcount is not great but ultimately shouldn't matter much.

> Then, we can just use unpin_folios() after we coalesced.
> 
> In any case, for a fix this is good enough, but probably we can do better later.
> 
> Acked-by: David Hildenbrand <david@redhat.com>

Thanks for the report!

-- 
Pavel Begunkov