The patch is against 3.0
The arg of kjournald/kjournald2 is void *, and points to type
journal_t. We should convert it to journal_t * explicitly in
the kjournald/kjournald2 function body.
Signed-off-by: Wang Sheng-Hui <[email protected]>
---
fs/jbd/journal.c | 2 +-
fs/jbd2/journal.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index e2d4285..32956fd 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -116,7 +116,7 @@ static void commit_timeout(unsigned long __data)
static int kjournald(void *arg)
{
- journal_t *journal = arg;
+ journal_t *journal = (journal_t *)arg;
transaction_t *transaction;
/*
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index 0dfa5b5..c4f4bfc 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -129,7 +129,7 @@ static void commit_timeout(unsigned long __data)
static int kjournald2(void *arg)
{
- journal_t *journal = arg;
+ journal_t *journal = (journal_t *)arg;
transaction_t *transaction;
/*
--
1.7.1
09:58, Wang Sheng-Hui wrote:
> The patch is against 3.0
>
> The arg of kjournald/kjournald2 is void *, and points to type
> journal_t. We should convert it to journal_t * explicitly in
> the kjournald/kjournald2 function body.
>
Why?
Implicit convertion from void * to foo * is ok. Did the compiler
complain about this to you?
> Signed-off-by: Wang Sheng-Hui <[email protected]>
> ---
> fs/jbd/journal.c | 2 +-
> fs/jbd2/journal.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
> index e2d4285..32956fd 100644
> --- a/fs/jbd/journal.c
> +++ b/fs/jbd/journal.c
> @@ -116,7 +116,7 @@ static void commit_timeout(unsigned long __data)
>
> static int kjournald(void *arg)
> {
> - journal_t *journal = arg;
> + journal_t *journal = (journal_t *)arg;
> transaction_t *transaction;
>
> /*
> diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
> index 0dfa5b5..c4f4bfc 100644
> --- a/fs/jbd2/journal.c
> +++ b/fs/jbd2/journal.c
> @@ -129,7 +129,7 @@ static void commit_timeout(unsigned long __data)
>
> static int kjournald2(void *arg)
> {
> - journal_t *journal = arg;
> + journal_t *journal = (journal_t *)arg;
> transaction_t *transaction;
>
> /*
On 2011年08月03日 10:04, Li Zefan wrote:
> 09:58, Wang Sheng-Hui wrote:
>> The patch is against 3.0
>>
>> The arg of kjournald/kjournald2 is void *, and points to type
>> journal_t. We should convert it to journal_t * explicitly in
>> the kjournald/kjournald2 function body.
>>
>
> Why?
>
> Implicit convertion from void * to foo * is ok. Did the compiler
> complain about this to you?
I remember any * can be assigned directly to void * in ANSI C, but
void * should be converted to specific point * type. Right?
And I checked the code of kswapd, in which explicitly conversion is
taken on the arg *. I think it should do so in journal.c too.
>
>> Signed-off-by: Wang Sheng-Hui <[email protected]>
>> ---
>> fs/jbd/journal.c | 2 +-
>> fs/jbd2/journal.c | 2 +-
>> 2 files changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
>> index e2d4285..32956fd 100644
>> --- a/fs/jbd/journal.c
>> +++ b/fs/jbd/journal.c
>> @@ -116,7 +116,7 @@ static void commit_timeout(unsigned long __data)
>>
>> static int kjournald(void *arg)
>> {
>> - journal_t *journal = arg;
>> + journal_t *journal = (journal_t *)arg;
>> transaction_t *transaction;
>>
>> /*
>> diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
>> index 0dfa5b5..c4f4bfc 100644
>> --- a/fs/jbd2/journal.c
>> +++ b/fs/jbd2/journal.c
>> @@ -129,7 +129,7 @@ static void commit_timeout(unsigned long __data)
>>
>> static int kjournald2(void *arg)
>> {
>> - journal_t *journal = arg;
>> + journal_t *journal = (journal_t *)arg;
>> transaction_t *transaction;
>>
>> /*
Wang Sheng-Hui wrote:
> On 2011年08月03日 10:04, Li Zefan wrote:
>> 09:58, Wang Sheng-Hui wrote:
>>> The patch is against 3.0
>>>
>>> The arg of kjournald/kjournald2 is void *, and points to type
>>> journal_t. We should convert it to journal_t * explicitly in
>>> the kjournald/kjournald2 function body.
>>>
>>
>> Why?
>>
>> Implicit convertion from void * to foo * is ok. Did the compiler
>> complain about this to you?
>
> I remember any * can be assigned directly to void * in ANSI C, but
> void * should be converted to specific point * type. Right?
>
No.
> And I checked the code of kswapd, in which explicitly conversion is
> taken on the arg *. I think it should do so in journal.c too.
>
Please check more similar code.
On Wed, Aug 3, 2011 at 10:17 AM, Wang Sheng-Hui <[email protected]> wrote:
>
> I remember any * can be assigned directly to void * in ANSI C, but
> void * should be converted to specific point * type. Right?
>
C99 6.3.2.3
A pointer to void may be converted to or from a pointer to any
incomplete or object
type. A pointer to any incomplete or object type may be converted to a
pointer to void
and back again; the result shall compare equal to the original pointer.
On 2011年08月03日 10:52, Américo Wang wrote:
> On Wed, Aug 3, 2011 at 10:17 AM, Wang Sheng-Hui <[email protected]> wrote:
>>
>> I remember any * can be assigned directly to void * in ANSI C, but
>> void * should be converted to specific point * type. Right?
>>
>
> C99 6.3.2.3
Got it. Thanks,
>
> A pointer to void may be converted to or from a pointer to any
> incomplete or object
> type. A pointer to any incomplete or object type may be converted to a
> pointer to void
> and back again; the result shall compare equal to the original pointer.
Hi,
On Tue, Aug 2, 2011 at 10:17 PM, Wang Sheng-Hui <[email protected]> wrote:
> And I checked the code of kswapd, in which explicitly conversion is
> taken on the arg *. I think it should do so in journal.c too.
>
because it is done wrong somewhere is no argument to do it wrong here ...
- Arnaud
On Wed, 3 Aug 2011, Li Zefan wrote:
> 09:58, Wang Sheng-Hui wrote:
> > The patch is against 3.0
> >
> > The arg of kjournald/kjournald2 is void *, and points to type
> > journal_t. We should convert it to journal_t * explicitly in
> > the kjournald/kjournald2 function body.
> >
>
> Why?
>
> Implicit convertion from void * to foo * is ok. Did the compiler
> complain about this to you?
>
Not only is it perfectly OK to rely on the implicit conversion, a lot of
work has gone into actively removing such unneeded explicit casts over the
years, so let's not introduce new ones.
--
Jesper Juhl <[email protected]> http://www.chaosbits.net/
Don't top-post http://www.catb.org/jargon/html/T/top-post.html
Plain text mails only, please.