2012-04-09 23:22:53

by Jim Rees

[permalink] [raw]
Subject: [PATCH] fix page number calculation bug for block layout decode buffer

Signed-off-by: Jim Rees <[email protected]>
Suggested-by: Andy Adamson <[email protected]>
---
fs/nfs/blocklayout/blocklayout.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 48cfac3..576da5b 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -872,7 +872,7 @@ nfs4_blk_get_deviceinfo(struct nfs_server *server, const struct nfs_fh *fh,
* GETDEVICEINFO's maxcount
*/
max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
- max_pages = max_resp_sz >> PAGE_SHIFT;
+ max_pages = (max_resp_sz + PAGE_SIZE - 1) >> PAGE_SHIFT;
dprintk("%s max_resp_sz %u max_pages %d\n",
__func__, max_resp_sz, max_pages);

--
1.7.5.4



2012-04-10 00:00:18

by Fred Isaman

[permalink] [raw]
Subject: Re: [PATCH] fix page number calculation bug for block layout decode buffer





On Apr 9, 2012, at 7:25 PM, Chuck Lever <[email protected]> wrote:

> Hi-
>
> Since all the layouts (and I'm guessing the non-pNFS path too) seem to need this calculation, would it make sense to create an inline function in a header somewhere?
>

nfs_page_array_len(0, size) would work.

Fred

2012-04-10 00:18:54

by NeilBrown

[permalink] [raw]
Subject: Re: [PATCH] fix page number calculation bug for block layout decode buffer

On Mon, 9 Apr 2012 19:22:07 -0400 Jim Rees <[email protected]> wrote:

> Signed-off-by: Jim Rees <[email protected]>
> Suggested-by: Andy Adamson <[email protected]>
> ---
> fs/nfs/blocklayout/blocklayout.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
> index 48cfac3..576da5b 100644
> --- a/fs/nfs/blocklayout/blocklayout.c
> +++ b/fs/nfs/blocklayout/blocklayout.c
> @@ -872,7 +872,7 @@ nfs4_blk_get_deviceinfo(struct nfs_server *server, const struct nfs_fh *fh,
> * GETDEVICEINFO's maxcount
> */
> max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
> - max_pages = max_resp_sz >> PAGE_SHIFT;
> + max_pages = (max_resp_sz + PAGE_SIZE - 1) >> PAGE_SHIFT;

max_pages = DIV_ROUND_UP(max_resp_sz, PAGE_SIZE);

??

compiler should optimise the constant-divide into a shift, and it is more
"obviously correct" this way.

NeilBrown


> dprintk("%s max_resp_sz %u max_pages %d\n",
> __func__, max_resp_sz, max_pages);
>

Attachments:
signature.asc (828.00 B)

2012-04-09 23:25:54

by Chuck Lever III

[permalink] [raw]
Subject: Re: [PATCH] fix page number calculation bug for block layout decode buffer

Hi-

Since all the layouts (and I'm guessing the non-pNFS path too) seem to need this calculation, would it make sense to create an inline function in a header somewhere?

On Apr 9, 2012, at 7:22 PM, Jim Rees wrote:

> Signed-off-by: Jim Rees <[email protected]>
> Suggested-by: Andy Adamson <[email protected]>
> ---
> fs/nfs/blocklayout/blocklayout.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
> index 48cfac3..576da5b 100644
> --- a/fs/nfs/blocklayout/blocklayout.c
> +++ b/fs/nfs/blocklayout/blocklayout.c
> @@ -872,7 +872,7 @@ nfs4_blk_get_deviceinfo(struct nfs_server *server, const struct nfs_fh *fh,
> * GETDEVICEINFO's maxcount
> */
> max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz;
> - max_pages = max_resp_sz >> PAGE_SHIFT;
> + max_pages = (max_resp_sz + PAGE_SIZE - 1) >> PAGE_SHIFT;
> dprintk("%s max_resp_sz %u max_pages %d\n",
> __func__, max_resp_sz, max_pages);
>
> --
> 1.7.5.4
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html

--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com