The patch titled
Subject: CMA: page_isolation: check buddy before accessing it
has been added to the -mm tree. Its filename is
cma-page_isolation-check-buddy-before-access-it.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/cma-page_isolation-check-buddy-before-access-it.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/cma-page_isolation-check-buddy-before-access-it.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Hui Zhu <zhuhui@xxxxxxxxxx>
Subject: CMA: page_isolation: check buddy before accessing it
I had an issue:
[ 214.294917] Unable to handle kernel NULL pointer dereference at virtual address 0000082a
[ 214.303013] pgd = cc970000
[ 214.305721] [0000082a] *pgd=00000000
[ 214.309316] Internal error: Oops: 5 [#1] PREEMPT SMP ARM
[ 214.335704] PC is at get_pageblock_flags_group+0x5c/0xb0
[ 214.341030] LR is at unset_migratetype_isolate+0x148/0x1b0
[ 214.346523] pc : [<c00cc9a0>] lr : [<c0109874>] psr: 80000093
[ 214.346523] sp : c7029d00 ip : 00000105 fp : c7029d1c
[ 214.358005] r10: 00000001 r9 : 0000000a r8 : 00000004
[ 214.363231] r7 : 60000013 r6 : 000000a4 r5 : c0a357e4 r4 : 00000000
[ 214.369761] r3 : 00000826 r2 : 00000002 r1 : 00000000 r0 : 0000003f
[ 214.376291] Flags: Nzcv IRQs off FIQs on Mode SVC_32 ISA ARM Segment user
[ 214.383516] Control: 10c5387d Table: 2cb7006a DAC: 00000015
[ 214.949720] Backtrace:
[ 214.952192] [<c00cc944>] (get_pageblock_flags_group+0x0/0xb0) from [<c0109874>] (unset_migratetype_isolate+0x148/0x1b0)
[ 214.962978] r7:60000013 r6:c0a357c0 r5:c0a357e4 r4:c1555000
[ 214.968693] [<c010972c>] (unset_migratetype_isolate+0x0/0x1b0) from [<c0109adc>] (undo_isolate_page_range+0xd0/0xdc)
[ 214.979222] [<c0109a0c>] (undo_isolate_page_range+0x0/0xdc) from [<c00d097c>] (__alloc_contig_range+0x254/0x34c)
[ 214.989398] r9:000abc00 r8:c7028000 r7:000b1f53 r6:000b3e00 r5:00000005
r4:c7029db4
[ 214.997308] [<c00d0728>] (__alloc_contig_range+0x0/0x34c) from [<c00d0a88>] (alloc_contig_range+0x14/0x18)
[ 215.006973] [<c00d0a74>] (alloc_contig_range+0x0/0x18) from [<c0398148>] (dma_alloc_from_contiguous_addr+0x1ac/0x304)
This issue is because when calling unset_migratetype_isolate() to unset a
part of CMA memory, it try to access the buddy page to get its status:
if (order >= pageblock_order) {
page_idx = page_to_pfn(page) & ((1 << MAX_ORDER) - 1);
buddy_idx = __find_buddy_index(page_idx, order);
buddy = page + (buddy_idx - page_idx);
if (!is_migrate_isolate_page(buddy)) {
But the begin addr of this part of CMA memory is very close to a part of
memory that is reserved at boot time (not in buddy system). So add a
check before accessing it.
Signed-off-by: Hui Zhu <zhuhui@xxxxxxxxxx>
Suggested-by: Laura Abbott <labbott@xxxxxxxxxx>
Suggested-by: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/page_isolation.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff -puN mm/page_isolation.c~cma-page_isolation-check-buddy-before-access-it mm/page_isolation.c
--- a/mm/page_isolation.c~cma-page_isolation-check-buddy-before-access-it
+++ a/mm/page_isolation.c
@@ -101,7 +101,8 @@ void unset_migratetype_isolate(struct pa
buddy_idx = __find_buddy_index(page_idx, order);
buddy = page + (buddy_idx - page_idx);
- if (!is_migrate_isolate_page(buddy)) {
+ if (pfn_valid_within(page_to_pfn(buddy))
+ && !is_migrate_isolate_page(buddy)) {
__isolate_free_page(page, order);
kernel_map_pages(page, (1 << order), 1);
set_page_refcounted(page);
_
Patches currently in -mm which might be from zhuhui@xxxxxxxxxx are
cma-page_isolation-check-buddy-before-access-it.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
