[PATCH 1/1] MM: swapfile, fix crash on double swapon

Double swapon on a device causes a crash:
BUG: unable to handle kernel NULL pointer dereference at (null)
IP: [<ffffffff810af160>] sys_swapon+0x1f0/0xc60
PGD 1dc0b067 PUD 1dc09067 PMD 0
Oops: 0000 [#1] SMP
last sysfs file:
CPU 1
Modules linked in:
Pid: 562, comm: swapon Tainted: G W 2.6.32-rc5-mm1_64 #867
RIP: 0010:[<ffffffff810af160>] [<ffffffff810af160>] sys_swapon+0x1f0/0xc60
....

It is due to swap_info_struct->first_swap_extent.list not being
initialized. ->next is NULL in such a situation and
destroy_swap_extents fails to iterate over the list with the BUG
above.

Introduced by swap_info-include-first_swap_extent.patch. Revert the
INIT_LIST_HEAD move.

Signed-off-by: Jiri Slaby <jirislaby.RemoveThis@gmail.com>
Cc: Hugh Dickins <hugh.dickins.RemoveThis@tiscali.co.uk>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu.RemoveThis@jp.fujitsu.com>
Cc: Rik van Riel <riel.RemoveThis@redhat.com>
---
mm/swapfile.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/mm/swapfile.c b/mm/swapfile.c
index 93e71cf..26ef6a2 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -1313,7 +1313,6 @@ add_swap_extent(struct swap_info_struct *sis, unsigned long start_page,
if (start_page == 0) {
se = &sis->first_swap_extent;
sis->curr_swap_extent = se;
- INIT_LIST_HEAD(&se->list);
se->start_page = 0;
se->nr_pages = nr_pages;
se->start_block = start_block;
@@ -1769,6 +1768,7 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags)
kfree(p);
goto out;
}
+ INIT_LIST_HEAD(&p->first_swap_extent.list);
if (type >= nr_swapfiles) {
p->type = type;
swap_info[type] = p;
--
1.6.4.2

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo.RemoveThis@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

On Mon, 2 Nov 2009, Jiri Slaby wrote:

> Double swapon on a device causes a crash:
> BUG: unable to handle kernel NULL pointer dereference at (null)

Thanks a lot for finding that: it doesn't just happen with a double
swapon of the same device, it happens with most kinds of error in
the swapon sequence. I thought I was being nice and tidy moving
that initialization, but actually I was just being careless.

> IP: [<ffffffff810af160>] sys_swapon+0x1f0/0xc60
> PGD 1dc0b067 PUD 1dc09067 PMD 0
> Oops: 0000 [#1] SMP
> last sysfs file:
> CPU 1
> Modules linked in:
> Pid: 562, comm: swapon Tainted: G W 2.6.32-rc5-mm1_64 #867
> RIP: 0010:[<ffffffff810af160>] [<ffffffff810af160>] sys_swapon+0x1f0/0xc60
> ...
>
> It is due to swap_info_struct->first_swap_extent.list not being
> initialized. ->next is NULL in such a situation and
> destroy_swap_extents fails to iterate over the list with the BUG
> above.
>
> Introduced by swap_info-include-first_swap_extent.patch. Revert the
> INIT_LIST_HEAD move.
>
> Signed-off-by: Jiri Slaby <jirislaby.TakeThisOut@gmail.com>
> Cc: Hugh Dickins <hugh.dickins.TakeThisOut@tiscali.co.uk>
> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu.TakeThisOut@jp.fujitsu.com>
> Cc: Rik van Riel <riel.TakeThisOut@redhat.com>

Acked-by: Hugh Dickins <hugh.dickins.TakeThisOut@tiscali.co.uk>

> ---
> mm/swapfile.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 93e71cf..26ef6a2 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -1313,7 +1313,6 @@ add_swap_extent(struct swap_info_struct *sis, unsigned long start_page,
> if (start_page == 0) {
> se = &sis->first_swap_extent;
> sis->curr_swap_extent = se;
> - INIT_LIST_HEAD(&se->list);
> se->start_page = 0;
> se->nr_pages = nr_pages;
> se->start_block = start_block;
> @@ -1769,6 +1768,7 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags)
> kfree(p);
> goto out;
> }
> + INIT_LIST_HEAD(&p->first_swap_extent.list);
> if (type >= nr_swapfiles) {
> p->type = type;
> swap_info[type] = p;
> --
> 1.6.4.2
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo.TakeThisOut@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/