[PATCH 31/34] nfs: convert put_page() to put_user_page*()

Calum Mackay calum.mackay at oracle.com
Sun Aug 4 23:28:01 UTC 2019


On 03/08/2019 2:41 am, John Hubbard wrote:
> On 8/2/19 6:27 PM, Calum Mackay wrote:
>> On 02/08/2019 3:20 am, john.hubbard at gmail.com wrote:
> ...
>> Since it's static, and only called twice, might it be better to change its two callers [nfs_direct_{read,write}_schedule_iovec()] to call put_user_pages() directly, and remove nfs_direct_release_pages() entirely?
>>
>> thanks,
>> calum.
>>
>>
>>>      void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo,
>>>
>   
> Hi Calum,
> 
> Absolutely! Is it OK to add your reviewed-by, with the following incremental
> patch made to this one?

Thanks John; looks good.

Reviewed-by: Calum Mackay <calum.mackay at oracle.com>

> 
> diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
> index b00b89dda3c5..c0c1b9f2c069 100644
> --- a/fs/nfs/direct.c
> +++ b/fs/nfs/direct.c
> @@ -276,11 +276,6 @@ ssize_t nfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
>          return nfs_file_direct_write(iocb, iter);
>   }
>   
> -static void nfs_direct_release_pages(struct page **pages, unsigned int npages)
> -{
> -       put_user_pages(pages, npages);
> -}
> -
>   void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo,
>                                struct nfs_direct_req *dreq)
>   {
> @@ -510,7 +505,7 @@ static ssize_t nfs_direct_read_schedule_iovec(struct nfs_direct_req *dreq,
>                          pos += req_len;
>                          dreq->bytes_left -= req_len;
>                  }
> -               nfs_direct_release_pages(pagevec, npages);
> +               put_user_pages(pagevec, npages);
>                  kvfree(pagevec);
>                  if (result < 0)
>                          break;
> @@ -933,7 +928,7 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq,
>                          pos += req_len;
>                          dreq->bytes_left -= req_len;
>                  }
> -               nfs_direct_release_pages(pagevec, npages);
> +               put_user_pages(pagevec, npages);
>                  kvfree(pagevec);
>                  if (result < 0)
>                          break;
> 
> 
> 
> thanks,
> 


More information about the amd-gfx mailing list