Re: [PATCH 12/18] io_uring: add support for pre-mapped user IO buffers

[Jens Axboe]

On 2/7/19 3:38 PM, Jeff Moyer wrote:
> Hi, Jens,
> 
> Jens Axboe <ax...@kernel.dk> writes:
> 
>> For now, buffers must not be file backed. If file backed buffers are
>> passed in, the registration will fail with -1/EOPNOTSUPP. This
>> restriction may be relaxed in the future.
> 
> [...]
> 
>> +            down_write(&current->mm->mmap_sem);
>> +            pret = get_user_pages_longterm(ubuf, nr_pages, FOLL_WRITE,
>> +                                            pages, vmas);
>> +            if (pret == nr_pages) {
>> +                    /* don't support file backed memory */
>> +                    for (j = 0; j < nr_pages; j++) {
>> +                            struct vm_area_struct *vma = vmas[j];
>> +
>> +                            if (vma->vm_file) {
>> +                                    ret = -EOPNOTSUPP;
>> +                                    break;
>> +                            }
>> +                    }
> 
> Unfortunately, this suffers the same problem as FOLL_ANON.  Huge pages
> are backed by hugetlbfs, and vma->vm_file will be filled in.
> 
> I guess you could check is_file_hugepages(vma->vm_file):
> 
>         if (vma->vm_file &&
>             !is_file_hugepages(vma->vm_file)) {
>                 ret = -EOPNOTSUPP;
>                 break;
>        }
> 
> That works for me.

Thanks, that looks better. Fixed!

-- 
Jens Axboe