2011-02-09 14:44:06

by Ryota Ozaki

Subject: [PATCH] mm: Fix out-of-date comments which refers non-existent functions

From: Ryota Ozaki <[email protected]>

do_file_page and do_no_page don't exist anymore, but some comments
still refers them. The patch fixes them by replacing them with
existing ones.

Signed-off-by: Ryota Ozaki <[email protected]>
---
mm/memory.c | 6 +++---
1 files changed, 3 insertions(+), 3 deletions(-)

I'm not familiar with the codes very much, so the fix may be wrong.

diff --git a/mm/memory.c b/mm/memory.c
index 31250fa..3fbf32a 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2115,10 +2115,10 @@ EXPORT_SYMBOL_GPL(apply_to_page_range);
* handle_pte_fault chooses page fault handler according to an entry
* which was read non-atomically. Before making any commitment, on
* those architectures or configurations (e.g. i386 with PAE) which
- * might give a mix of unmatched parts, do_swap_page and do_file_page
+ * might give a mix of unmatched parts, do_swap_page and do_nonlinear_fault
* must check under lock before unmapping the pte and proceeding
* (but do_wp_page is only called after already making such a check;
- * and do_anonymous_page and do_no_page can safely check later on).
+ * and do_anonymous_page can safely check later on).
*/
static inline int pte_unmap_same(struct mm_struct *mm, pmd_t *pmd,
pte_t *page_table, pte_t orig_pte)
@@ -2316,7 +2316,7 @@ reuse:
* bit after it clear all dirty ptes, but before a racing
* do_wp_page installs a dirty pte.
*
- * do_no_page is protected similarly.
+ * __do_fault is protected similarly.
*/
if (!page_mkwrite) {
wait_on_page_locked(dirty_page);
--
1.7.2.3

2011-02-10 00:04:33

by Kamezawa Hiroyuki

Subject: Re: [PATCH] mm: Fix out-of-date comments which refers non-existent functions

On Wed, 9 Feb 2011 23:42:17 +0900
Ryota Ozaki <[email protected]> wrote:

> From: Ryota Ozaki <[email protected]>
>
> do_file_page and do_no_page don't exist anymore, but some comments
> still refers them. The patch fixes them by replacing them with
> existing ones.
>
> Signed-off-by: Ryota Ozaki <[email protected]>

Acked-by: KAMEZAWA Hiroyuki <[email protected]>

It seems there are other ones ;)
==
Searched full:do_no_page (Results 1 - 3 of 3) sorted by relevancy

/linux-2.6-git/arch/alpha/include/asm/
H A D cacheflush.h 66 /* This is used only in do_no_page and do_swap_page. */
/linux-2.6-git/arch/avr32/mm/
H A D cache.c 116 * This one is called from do_no_page(), do_swap_page() and install_page().
/linux-2.6-git/mm/
H A D memory.c 2121 * and do_anonymous_page and do_no_page can safely check later on).
2319 * do_no_page is protected similarly.





> ---
> mm/memory.c | 6 +++---
> 1 files changed, 3 insertions(+), 3 deletions(-)
>
> I'm not familiar with the codes very much, so the fix may be wrong.
>
> diff --git a/mm/memory.c b/mm/memory.c
> index 31250fa..3fbf32a 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -2115,10 +2115,10 @@ EXPORT_SYMBOL_GPL(apply_to_page_range);
> * handle_pte_fault chooses page fault handler according to an entry
> * which was read non-atomically. Before making any commitment, on
> * those architectures or configurations (e.g. i386 with PAE) which
> - * might give a mix of unmatched parts, do_swap_page and do_file_page
> + * might give a mix of unmatched parts, do_swap_page and do_nonlinear_fault
> * must check under lock before unmapping the pte and proceeding
> * (but do_wp_page is only called after already making such a check;
> - * and do_anonymous_page and do_no_page can safely check later on).
> + * and do_anonymous_page can safely check later on).
> */
> static inline int pte_unmap_same(struct mm_struct *mm, pmd_t *pmd,
> pte_t *page_table, pte_t orig_pte)
> @@ -2316,7 +2316,7 @@ reuse:
> * bit after it clear all dirty ptes, but before a racing
> * do_wp_page installs a dirty pte.
> *
> - * do_no_page is protected similarly.
> + * __do_fault is protected similarly.
> */
> if (!page_mkwrite) {
> wait_on_page_locked(dirty_page);
> --
> 1.7.2.3
>
>

2011-02-10 03:58:44

by Ryota Ozaki

Subject: Re: [PATCH] mm: Fix out-of-date comments which refers non-existent functions

On Thu, Feb 10, 2011 at 8:58 AM, KAMEZAWA Hiroyuki
<[email protected]> wrote:
> On Wed, ?9 Feb 2011 23:42:17 +0900
> Ryota Ozaki <[email protected]> wrote:
>
>> From: Ryota Ozaki <[email protected]>
>>
>> do_file_page and do_no_page don't exist anymore, but some comments
>> still refers them. The patch fixes them by replacing them with
>> existing ones.
>>
>> Signed-off-by: Ryota Ozaki <[email protected]>
>
> Acked-by: KAMEZAWA Hiroyuki <[email protected]>

Thanks, Kamezawa-san.

>
> It seems there are other ones ;)
> ==
> ? ?Searched full:do_no_page (Results 1 - 3 of 3) sorted by relevancy
>
> ?/linux-2.6-git/arch/alpha/include/asm/
> H A D ? cacheflush.h ? ?66 /* This is used only in do_no_page and do_swap_page. */
> ?/linux-2.6-git/arch/avr32/mm/
> H A D ? cache.c ? ? ? ? 116 * This one is called from do_no_page(), do_swap_page() and install_page().
> ?/linux-2.6-git/mm/
> H A D ? memory.c ? ? ? ?2121 * and do_anonymous_page and do_no_page can safely check later on).
> 2319 * do_no_page is protected similarly.

Nice catch :-) Cloud I assemble all fixes into one patch?

ozaki-r

>
>
>
>
>
>> ---
>> ?mm/memory.c | ? ?6 +++---
>> ?1 files changed, 3 insertions(+), 3 deletions(-)
>>
>> I'm not familiar with the codes very much, so the fix may be wrong.
>>
>> diff --git a/mm/memory.c b/mm/memory.c
>> index 31250fa..3fbf32a 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -2115,10 +2115,10 @@ EXPORT_SYMBOL_GPL(apply_to_page_range);
>> ? * handle_pte_fault chooses page fault handler according to an entry
>> ? * which was read non-atomically. ?Before making any commitment, on
>> ? * those architectures or configurations (e.g. i386 with PAE) which
>> - * might give a mix of unmatched parts, do_swap_page and do_file_page
>> + * might give a mix of unmatched parts, do_swap_page and do_nonlinear_fault
>> ? * must check under lock before unmapping the pte and proceeding
>> ? * (but do_wp_page is only called after already making such a check;
>> - * and do_anonymous_page and do_no_page can safely check later on).
>> + * and do_anonymous_page can safely check later on).
>> ? */
>> ?static inline int pte_unmap_same(struct mm_struct *mm, pmd_t *pmd,
>> ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? pte_t *page_table, pte_t orig_pte)
>> @@ -2316,7 +2316,7 @@ reuse:
>> ? ? ? ? ? ? ? ?* bit after it clear all dirty ptes, but before a racing
>> ? ? ? ? ? ? ? ?* do_wp_page installs a dirty pte.
>> ? ? ? ? ? ? ? ?*
>> - ? ? ? ? ? ? ?* do_no_page is protected similarly.
>> + ? ? ? ? ? ? ?* __do_fault is protected similarly.
>> ? ? ? ? ? ? ? ?*/
>> ? ? ? ? ? ? ? if (!page_mkwrite) {
>> ? ? ? ? ? ? ? ? ? ? ? wait_on_page_locked(dirty_page);
>> --
>> 1.7.2.3
>>
>>
>
>

2011-02-10 04:23:50

by Kamezawa Hiroyuki

Subject: Re: [PATCH] mm: Fix out-of-date comments which refers non-existent functions

On Thu, 10 Feb 2011 12:58:21 +0900
Ryota Ozaki <[email protected]> wrote:

> On Thu, Feb 10, 2011 at 8:58 AM, KAMEZAWA Hiroyuki
> <[email protected]> wrote:
> > On Wed,  9 Feb 2011 23:42:17 +0900
> > Ryota Ozaki <[email protected]> wrote:
> >
> >> From: Ryota Ozaki <[email protected]>
> >>
> >> do_file_page and do_no_page don't exist anymore, but some comments
> >> still refers them. The patch fixes them by replacing them with
> >> existing ones.
> >>
> >> Signed-off-by: Ryota Ozaki <[email protected]>
> >
> > Acked-by: KAMEZAWA Hiroyuki <[email protected]>
>
> Thanks, Kamezawa-san.
>
> >
> > It seems there are other ones ;)
> > ==
> >    Searched full:do_no_page (Results 1 - 3 of 3) sorted by relevancy
> >
> >  /linux-2.6-git/arch/alpha/include/asm/
> > H A D   cacheflush.h    66 /* This is used only in do_no_page and do_swap_page. */
> >  /linux-2.6-git/arch/avr32/mm/
> > H A D   cache.c         116 * This one is called from do_no_page(), do_swap_page() and install_page().
> >  /linux-2.6-git/mm/
> > H A D   memory.c        2121 * and do_anonymous_page and do_no_page can safely check later on).
> > 2319 * do_no_page is protected similarly.
>
> Nice catch :-) Cloud I assemble all fixes into one patch?
>
> ozaki-r
>

Yes, I think it's allowed.

Thanks,
-Kame

2011-02-10 04:44:12

by Minchan Kim

Subject: Re: [PATCH] mm: Fix out-of-date comments which refers non-existent functions

On Thu, Feb 10, 2011 at 12:58 PM, Ryota Ozaki <[email protected]> wrote:
> On Thu, Feb 10, 2011 at 8:58 AM, KAMEZAWA Hiroyuki
> <[email protected]> wrote:
>> On Wed,  9 Feb 2011 23:42:17 +0900
>> Ryota Ozaki <[email protected]> wrote:
>>
>>> From: Ryota Ozaki <[email protected]>
>>>
>>> do_file_page and do_no_page don't exist anymore, but some comments
>>> still refers them. The patch fixes them by replacing them with
>>> existing ones.
>>>
>>> Signed-off-by: Ryota Ozaki <[email protected]>
>>
>> Acked-by: KAMEZAWA Hiroyuki <[email protected]>
>
> Thanks, Kamezawa-san.
>
>>
>> It seems there are other ones ;)
>> ==
>>    Searched full:do_no_page (Results 1 - 3 of 3) sorted by relevancy
>>
>>  /linux-2.6-git/arch/alpha/include/asm/
>> H A D   cacheflush.h    66 /* This is used only in do_no_page and do_swap_page. */
>>  /linux-2.6-git/arch/avr32/mm/
>> H A D   cache.c         116 * This one is called from do_no_page(), do_swap_page() and install_page().
>>  /linux-2.6-git/mm/
>> H A D   memory.c        2121 * and do_anonymous_page and do_no_page can safely check later on).
>> 2319 * do_no_page is protected similarly.
>
> Nice catch :-) Cloud I assemble all fixes into one patch?
>
>  ozaki-r
>

When you resend the patch, Please Cc Jiri Kosina <[email protected]>


--
Kind regards,
Minchan Kim

2011-02-10 04:56:25

by Ryota Ozaki

Subject: Re: [PATCH] mm: Fix out-of-date comments which refers non-existent functions

On Thu, Feb 10, 2011 at 1:44 PM, Minchan Kim <[email protected]> wrote:
> On Thu, Feb 10, 2011 at 12:58 PM, Ryota Ozaki <[email protected]> wrote:
>> On Thu, Feb 10, 2011 at 8:58 AM, KAMEZAWA Hiroyuki
>> <[email protected]> wrote:
>>> On Wed, ?9 Feb 2011 23:42:17 +0900
>>> Ryota Ozaki <[email protected]> wrote:
>>>
>>>> From: Ryota Ozaki <[email protected]>
>>>>
>>>> do_file_page and do_no_page don't exist anymore, but some comments
>>>> still refers them. The patch fixes them by replacing them with
>>>> existing ones.
>>>>
>>>> Signed-off-by: Ryota Ozaki <[email protected]>
>>>
>>> Acked-by: KAMEZAWA Hiroyuki <[email protected]>
>>
>> Thanks, Kamezawa-san.
>>
>>>
>>> It seems there are other ones ;)
>>> ==
>>> ? ?Searched full:do_no_page (Results 1 - 3 of 3) sorted by relevancy
>>>
>>> ?/linux-2.6-git/arch/alpha/include/asm/
>>> H A D ? cacheflush.h ? ?66 /* This is used only in do_no_page and do_swap_page. */
>>> ?/linux-2.6-git/arch/avr32/mm/
>>> H A D ? cache.c ? ? ? ? 116 * This one is called from do_no_page(), do_swap_page() and install_page().
>>> ?/linux-2.6-git/mm/
>>> H A D ? memory.c ? ? ? ?2121 * and do_anonymous_page and do_no_page can safely check later on).
>>> 2319 * do_no_page is protected similarly.
>>
>> Nice catch :-) Cloud I assemble all fixes into one patch?
>>
>> ?ozaki-r
>>
>
> When you resend the patch, Please Cc Jiri Kosina <[email protected]>

OK, will do.

ozaki-r

>
>
> --
> Kind regards,
> Minchan Kim
>