Hi all.
First up, sorry for not inlining the patch - trouble with line wrapping.
In 2.6.24-rc8, call_usermodehelper_exec has an exit path that can leave
the helper_lock() call at the top of the routine unbalanced. The
attached patch fixes this issue.
Signed-off-by: Nigel Cunningham <[email protected]>
On Thursday, 17 of January 2008, Nigel Cunningham wrote:
> Hi all.
Hi,
> First up, sorry for not inlining the patch - trouble with line wrapping.
No big deal.
> In 2.6.24-rc8, call_usermodehelper_exec has an exit path that can leave
> the helper_lock() call at the top of the routine unbalanced. The
> attached patch fixes this issue.
Thanks a lot for the patch (reproduced below), I think it's 2.6.24 material.
Andrew?
> Signed-off-by: Nigel Cunningham <[email protected]>
---
From: Nigel Cunningham <[email protected]>
In 2.6.24-rc8, call_usermodehelper_exec has an exit path that can leave
the helper_lock() call at the top of the routine unbalanced. Fix it.
Signed-off-by: Nigel Cunningham <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
---
diff --git a/kernel/kmod.c b/kernel/kmod.c
index c6a4f8a..de27e15 100644
--- a/kernel/kmod.c
+++ b/kernel/kmod.c
@@ -468,8 +468,10 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info,
sub_info->wait = wait;
queue_work(khelper_wq, &sub_info->work);
- if (wait == UMH_NO_WAIT) /* task has freed sub_info */
+ if (wait == UMH_NO_WAIT) { /* task has freed sub_info */
+ helper_unlock();
return 0;
+ }
wait_for_completion(&done);
retval = sub_info->retval;
On Thu, 17 Jan 2008 19:02:43 +0100 "Rafael J. Wysocki" <[email protected]> wrote:
> On Thursday, 17 of January 2008, Nigel Cunningham wrote:
> > Hi all.
>
> Hi,
>
> > First up, sorry for not inlining the patch - trouble with line wrapping.
>
> No big deal.
>
> > In 2.6.24-rc8, call_usermodehelper_exec has an exit path that can leave
> > the helper_lock() call at the top of the routine unbalanced. The
> > attached patch fixes this issue.
>
> Thanks a lot for the patch (reproduced below), I think it's 2.6.24 material.
> Andrew?
>
> > Signed-off-by: Nigel Cunningham <[email protected]>
>
> ---
> From: Nigel Cunningham <[email protected]>
>
> In 2.6.24-rc8, call_usermodehelper_exec has an exit path that can leave
> the helper_lock() call at the top of the routine unbalanced. Fix it.
>
> Signed-off-by: Nigel Cunningham <[email protected]>
> Signed-off-by: Rafael J. Wysocki <[email protected]>
> ---
> diff --git a/kernel/kmod.c b/kernel/kmod.c
> index c6a4f8a..de27e15 100644
> --- a/kernel/kmod.c
> +++ b/kernel/kmod.c
> @@ -468,8 +468,10 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info,
> sub_info->wait = wait;
>
> queue_work(khelper_wq, &sub_info->work);
> - if (wait == UMH_NO_WAIT) /* task has freed sub_info */
> + if (wait == UMH_NO_WAIT) { /* task has freed sub_info */
> + helper_unlock();
> return 0;
> + }
> wait_for_completion(&done);
> retval = sub_info->retval;
>
Yup, I ended up queueing this:
From: Nigel Cunningham <[email protected]>
call_usermodehelper_exec() has an exit path that can leave the
helper_lock() call at the top of the routine unbalanced. The attached
patch fixes this issue.
Signed-off-by: Nigel Cunningham <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
---
kernel/kmod.c | 13 ++++++-------
1 file changed, 6 insertions(+), 7 deletions(-)
diff -puN kernel/kmod.c~fix-unbalanced-helper_lock-in-kernel-kmodc kernel/kmod.c
--- a/kernel/kmod.c~fix-unbalanced-helper_lock-in-kernel-kmodc
+++ a/kernel/kmod.c
@@ -451,13 +451,11 @@ int call_usermodehelper_exec(struct subp
enum umh_wait wait)
{
DECLARE_COMPLETION_ONSTACK(done);
- int retval;
+ int retval = 0;
helper_lock();
- if (sub_info->path[0] == '\0') {
- retval = 0;
+ if (sub_info->path[0] == '\0')
goto out;
- }
if (!khelper_wq || usermodehelper_disabled) {
retval = -EBUSY;
@@ -468,13 +466,14 @@ int call_usermodehelper_exec(struct subp
sub_info->wait = wait;
queue_work(khelper_wq, &sub_info->work);
- if (wait == UMH_NO_WAIT) /* task has freed sub_info */
- return 0;
+ if (wait == UMH_NO_WAIT) /* task has freed sub_info */
+ goto unlock;
wait_for_completion(&done);
retval = sub_info->retval;
- out:
+out:
call_usermodehelper_freeinfo(sub_info);
+unlock:
helper_unlock();
return retval;
}
_