linux-kernel - Re: [PATCH 15/22] Do not disable interrupts in free_page

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Fri, 24 Apr 2009 12:33:12 +0100
From:	Mel Gorman <mel@....ul.ie>
To:	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>
Cc:	Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
	cl@...ux-foundation.org, npiggin@...e.de,
	linux-kernel@...r.kernel.org, ming.m.lin@...el.com,
	yanmin_zhang@...ux.intel.com, peterz@...radead.org,
	penberg@...helsinki.fi, Lee Schermerhorn <Lee.Schermerhorn@...com>,
	Dave Hansen <dave@...ux.vnet.ibm.com>
Subject: Re: [PATCH 15/22] Do not disable interrupts in free_page_mlock()

On Fri, Apr 24, 2009 at 09:33:50AM +0900, KOSAKI Motohiro wrote:
> > > @@ -157,14 +157,9 @@ static inline void mlock_migrate_page(struct page *newpage, struct page *page)
> > >   */
> > >  static inline void free_page_mlock(struct page *page)
> > >  {
> > > -	if (unlikely(TestClearPageMlocked(page))) {
> > > -		unsigned long flags;
> > > -
> > > -		local_irq_save(flags);
> > > -		__dec_zone_page_state(page, NR_MLOCK);
> > > -		__count_vm_event(UNEVICTABLE_MLOCKFREED);
> > > -		local_irq_restore(flags);
> > > -	}
> > > +	__ClearPageMlocked(page);
> > > +	__dec_zone_page_state(page, NR_MLOCK);
> > > +	__count_vm_event(UNEVICTABLE_MLOCKFREED);
> > >  }
> > 
> > The conscientuous reviewer runs around and checks for free_page_mlock()
> > callers in other .c files which might be affected.
> > 
> > Only there are no such callers.
> > 
> > The reviewer's job would be reduced if free_page_mlock() wasn't
> > needlessly placed in a header file!
> 
> very sorry.
> 
> How about this?
> 
> =============================================
> Subject: [PATCH] move free_page_mlock() to page_alloc.c
> 
> Currently, free_page_mlock() is only called from page_alloc.c.
> Thus, we can move it to page_alloc.c.
> 

Looks good, but here is a version rebased on top of the patch series
where it would be easier to merge with "Do not disable interrupts in
free_page_mlock()".

I do note why it might be in the header though - it keeps all the
CONFIG_HAVE_MLOCKED_PAGE_BIT-related helper functions together making it
easier to find them. Lee, was that the intention?

=======
From: KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>

Move free_page_mlock() from mm/internal.h to mm/page_alloc.c

Currently, free_page_mlock() is only called from page_alloc.c. This patch
moves it from a header to to page_alloc.c.

[mel@....ul.ie: Rebase on top of page allocator patches]
Cc: Lee Schermerhorn <Lee.Schermerhorn@...com>
Cc: Christoph Lameter <cl@...ux-foundation.org>
Cc: Pekka Enberg <penberg@...helsinki.fi>
Cc: Dave Hansen <dave@...ux.vnet.ibm.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>
---
 mm/internal.h   |   13 -------------
 mm/page_alloc.c |   16 ++++++++++++++++
 2 files changed, 16 insertions(+), 13 deletions(-)

diff --git a/mm/internal.h b/mm/internal.h
index 58ec1bc..4b1672a 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -150,18 +150,6 @@ static inline void mlock_migrate_page(struct page *newpage, struct page *page)
 	}
 }
 
-/*
- * free_page_mlock() -- clean up attempts to free and mlocked() page.
- * Page should not be on lru, so no need to fix that up.
- * free_pages_check() will verify...
- */
-static inline void free_page_mlock(struct page *page)
-{
-	__ClearPageMlocked(page);
-	__dec_zone_page_state(page, NR_MLOCK);
-	__count_vm_event(UNEVICTABLE_MLOCKFREED);
-}
-
 #else /* CONFIG_HAVE_MLOCKED_PAGE_BIT */
 static inline int is_mlocked_vma(struct vm_area_struct *v, struct page *p)
 {
@@ -170,7 +158,6 @@ static inline int is_mlocked_vma(struct vm_area_struct *v, struct page *p)
 static inline void clear_page_mlock(struct page *page) { }
 static inline void mlock_vma_page(struct page *page) { }
 static inline void mlock_migrate_page(struct page *new, struct page *old) { }
-static inline void free_page_mlock(struct page *page) { }
 
 #endif /* CONFIG_HAVE_MLOCKED_PAGE_BIT */
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f08b4cb..3db5f57 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -433,6 +433,22 @@ static inline int page_is_buddy(struct page *page, struct page *buddy,
 	return 0;
 }
 
+#ifdef CONFIG_HAVE_MLOCKED_PAGE_BIT
+/*
+ * free_page_mlock() -- clean up attempts to free and mlocked() page.
+ * Page should not be on lru, so no need to fix that up.
+ * free_pages_check() will verify...
+ */
+static inline void free_page_mlock(struct page *page)
+{
+	__ClearPageMlocked(page);
+	__dec_zone_page_state(page, NR_MLOCK);
+	__count_vm_event(UNEVICTABLE_MLOCKFREED);
+}
+#else
+static inline void free_page_mlock(struct page *page) { }
+#endif /* CONFIG_HAVE_MLOCKED_PAGE_BIT */
+
 /*
  * Freeing function for a buddy system allocator.
  *

-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/