[JAUNTY SRU] UBUNTU: SAUCE: remove AppArmor debug check for calls from interrupt context

Tim Gardner tim.gardner at canonical.com
Tue Jul 28 12:54:10 UTC 2009


John Johansen wrote:
> SRU JUSTIFICATION
> 
> IMPACT: AppArmor will cause failures when socket_sendmsg is called from
> interrupt context.  These failures have been reported to affect bluetooth
> for some people in Jaunty.
> 
> TEST: This works in testing
> 
> 
> From b926d008f661730a264d474ea8e57ea4d6fa57bb Mon Sep 17 00:00:00 2001
> From: John Johansen <john.johansen at canonical.com>
> Date: Tue, 28 Jul 2009 05:36:37 -0700
> Subject: [PATCH] UBUNTU: SAUCE: remove AppArmor debug check for calls from
> interrupt context
> 
> BugLink: https://bugs.edge.launchpad.net/ubuntu/jaunty/+source/linux/+bug/35
> 
> launchpad bug #350789, is caused by AppArmor network permission hooks
> being called from interrupt context.  Replace the debug check with
> short circuiting the check for calls from interupt context.
> 
> Signed-off-by: John Johansen <john.johansen at canonical.com>
> ---
>  security/apparmor/main.c |   10 ++++------
>  1 files changed, 4 insertions(+), 6 deletions(-)
> 
> diff --git a/security/apparmor/main.c b/security/apparmor/main.c
> index 5f9c1cd..485a296 100644
> --- a/security/apparmor/main.c
> +++ b/security/apparmor/main.c
> @@ -859,13 +859,11 @@ int aa_revalidate_sk(struct sock *sk, char *operation)
>         struct aa_profile *profile;
>         int error = 0;
> 
> -       /* this is some debugging code to flush out the network hooks that
> -          that are called in interrupt context */
> -       if (in_interrupt()) {
> -               printk("AppArmor Debug: Hook being called from interrupt
> context\n");
> -               dump_stack();
> +       /* aa_revalidate_sk should not be called from interrupt context
> +        * don't mediate these calls as they are not task related
> +        */
> +       if (in_interrupt())
>                 return 0;
> -       }
> 
>         profile = aa_get_profile(current);
>         if (profile)

ACK

-- 
Tim Gardner tim.gardner at canonical.com




More information about the kernel-team mailing list