2015-05-11 10:47:29

by Sudip Mukherjee

[permalink] [raw]
Subject: [PATCH] staging: lustre: remove unused variable

there variables were not used anywhere and was showing as build warning.

Signed-off-by: Sudip Mukherjee <[email protected]>
---
drivers/staging/lustre/lustre/lov/lov_request.c | 4 +---
drivers/staging/lustre/lustre/obdclass/acl.c | 1 -
2 files changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/staging/lustre/lustre/lov/lov_request.c b/drivers/staging/lustre/lustre/lov/lov_request.c
index f6e1314..d24ba8e 100644
--- a/drivers/staging/lustre/lustre/lov/lov_request.c
+++ b/drivers/staging/lustre/lustre/lov/lov_request.c
@@ -74,10 +74,8 @@ void lov_finish_set(struct lov_request_set *set)
kfree(req);
}

- if (set->set_pga) {
- int len = set->set_oabufs * sizeof(*set->set_pga);
+ if (set->set_pga)
OBD_FREE_LARGE(set->set_pga, len);
- }
if (set->set_lockh)
lov_llh_put(set->set_lockh);

diff --git a/drivers/staging/lustre/lustre/obdclass/acl.c b/drivers/staging/lustre/lustre/obdclass/acl.c
index 194c48a..bc3fc47 100644
--- a/drivers/staging/lustre/lustre/obdclass/acl.c
+++ b/drivers/staging/lustre/lustre/obdclass/acl.c
@@ -120,7 +120,6 @@ static int lustre_ext_acl_xattr_reduce_space(ext_acl_xattr_header **header,
{
int ext_count = le32_to_cpu((*header)->a_count);
int ext_size = CFS_ACL_XATTR_SIZE(ext_count, ext_acl_xattr);
- int old_size = CFS_ACL_XATTR_SIZE(old_count, ext_acl_xattr);
ext_acl_xattr_header *new;

if (unlikely(old_count <= ext_count))
--
1.8.1.2


2015-05-11 11:31:16

by Dan Carpenter

[permalink] [raw]
Subject: Re: [PATCH] staging: lustre: remove unused variable

On Mon, May 11, 2015 at 04:17:10PM +0530, Sudip Mukherjee wrote:
> there variables were not used anywhere and was showing as build warning.
>
> Signed-off-by: Sudip Mukherjee <[email protected]>
> ---
> drivers/staging/lustre/lustre/lov/lov_request.c | 4 +---
> drivers/staging/lustre/lustre/obdclass/acl.c | 1 -
> 2 files changed, 1 insertion(+), 4 deletions(-)
>
> diff --git a/drivers/staging/lustre/lustre/lov/lov_request.c b/drivers/staging/lustre/lustre/lov/lov_request.c
> index f6e1314..d24ba8e 100644
> --- a/drivers/staging/lustre/lustre/lov/lov_request.c
> +++ b/drivers/staging/lustre/lustre/lov/lov_request.c
> @@ -74,10 +74,8 @@ void lov_finish_set(struct lov_request_set *set)
> kfree(req);
> }
>
> - if (set->set_pga) {
> - int len = set->set_oabufs * sizeof(*set->set_pga);
> + if (set->set_pga)
> OBD_FREE_LARGE(set->set_pga, len);
^^^
It's used right here?

regards,
dan carpenter

2015-05-11 11:44:06

by Sudip Mukherjee

[permalink] [raw]
Subject: Re: [PATCH] staging: lustre: remove unused variable

On Mon, May 11, 2015 at 02:31:02PM +0300, Dan Carpenter wrote:
> On Mon, May 11, 2015 at 04:17:10PM +0530, Sudip Mukherjee wrote:
> > - if (set->set_pga) {
> > - int len = set->set_oabufs * sizeof(*set->set_pga);
> > + if (set->set_pga)
> > OBD_FREE_LARGE(set->set_pga, len);
> ^^^
> It's used right here?

oops.. sorry .. but why i didnot get a build failure after this?

ok, got it ...

#define OBD_FREE_LARGE(ptr, size) \
do { \
kvfree(ptr); \
} while (0)

the variable is actually not used. I will make a v2 and send
tomorrow. today is a day of silly mistakes :(

regards
sudip

>
> regards,
> dan carpenter
>

2015-05-11 11:54:13

by Oleg Drokin

[permalink] [raw]
Subject: Re: [PATCH] staging: lustre: remove unused variable

Hello!
On May 11, 2015, at 7:43 AM, Sudip Mukherjee wrote:

> On Mon, May 11, 2015 at 02:31:02PM +0300, Dan Carpenter wrote:
>> On Mon, May 11, 2015 at 04:17:10PM +0530, Sudip Mukherjee wrote:
>>> - if (set->set_pga) {
>>> - int len = set->set_oabufs * sizeof(*set->set_pga);
>>> + if (set->set_pga)
>>> OBD_FREE_LARGE(set->set_pga, len);
>> ^^^
>> It's used right here?
>
> oops.. sorry .. but why i didnot get a build failure after this?
>
> ok, got it ...
>
> #define OBD_FREE_LARGE(ptr, size) \
> do { \
> kvfree(ptr); \
> } while (0)
>
> the variable is actually not used. I will make a v2 and send
> tomorrow. today is a day of silly mistakes :(

It's probably best to wait for a bit at this point.

Julia is going to send in a bunch of patches actually getting rid of
those macros.
And then whatever stuff surfaces would be easier to see.

Thanks!

Bye,
Oleg