lkml.org 
[lkml]   [2014]   [Jan]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Subject[PATCH] mm: slub: do not VM_BUG_ON_PAGE() for temporary on-stack pages
From
Date

From: Dave Hansen <dave.hansen@linux.intel.com>

This patch:

commit 309381feaee564281c3d9e90fbca8963bb7428ad
Author: Sasha Levin <sasha.levin@oracle.com>
Date: Thu Jan 23 15:52:54 2014 -0800
Subject: mm: dump page when hitting a VM_BUG_ON using VM_BUG_ON_PAGE

added a bunch of VM_BUG_ON_PAGE() calls. But, most of the ones
in the slub code are for _temporary_ 'struct page's which are
declared on the stack and likely have lots of gunk in them.
Dumping their contents out will just confuse folks looking at
bad_page() output. Plus, if we try to page_to_pfn() on them or
soemthing, we'll probably oops anyway.

Turn them back in to VM_BUG_ON()s.

I believe this is 3.14 material.

Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Pekka Enberg <penberg@kernel.org>

---

b/mm/slub.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)

diff -puN mm/slub.c~mm-slub-do-not-VM_BUG_ON-stack-pages mm/slub.c
--- a/mm/slub.c~mm-slub-do-not-VM_BUG_ON-stack-pages 2014-01-28 09:17:49.869780639 -0800
+++ b/mm/slub.c 2014-01-28 09:19:27.905212000 -0800
@@ -1559,7 +1559,7 @@ static inline void *acquire_slab(struct
new.freelist = freelist;
}

- VM_BUG_ON_PAGE(new.frozen, &new);
+ VM_BUG_ON(new.frozen);
new.frozen = 1;

if (!__cmpxchg_double_slab(s, page,
@@ -1812,7 +1812,7 @@ static void deactivate_slab(struct kmem_
set_freepointer(s, freelist, prior);
new.counters = counters;
new.inuse--;
- VM_BUG_ON_PAGE(!new.frozen, &new);
+ VM_BUG_ON(!new.frozen);

} while (!__cmpxchg_double_slab(s, page,
prior, counters,
@@ -1840,7 +1840,7 @@ redo:

old.freelist = page->freelist;
old.counters = page->counters;
- VM_BUG_ON_PAGE(!old.frozen, &old);
+ VM_BUG_ON(!old.frozen);

/* Determine target state of the slab */
new.counters = old.counters;
@@ -1952,7 +1952,7 @@ static void unfreeze_partials(struct kme

old.freelist = page->freelist;
old.counters = page->counters;
- VM_BUG_ON_PAGE(!old.frozen, &old);
+ VM_BUG_ON(!old.frozen);

new.counters = old.counters;
new.freelist = old.freelist;
@@ -2225,7 +2225,7 @@ static inline void *get_freelist(struct
counters = page->counters;

new.counters = counters;
- VM_BUG_ON_PAGE(!new.frozen, &new);
+ VM_BUG_ON(!new.frozen);

new.inuse = page->objects;
new.frozen = freelist != NULL;
@@ -2319,7 +2319,7 @@ load_freelist:
* page is pointing to the page from which the objects are obtained.
* That page must be frozen for per cpu allocations to work.
*/
- VM_BUG_ON_PAGE(!c->page->frozen, c->page);
+ VM_BUG_ON(!c->page->frozen);
c->freelist = get_freepointer(s, freelist);
c->tid = next_tid(c->tid);
local_irq_restore(flags);
_

\
 
 \ /
  Last update: 2014-01-28 19:01    [W:0.018 / U:0.260 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site