Stephen Rothwell wrote:
> Hi Jiri,
>
> Today's linux-next merge of the livepatching tree got conflicts in:
>
> include/linux/init_task.h
> include/linux/sched.h
>
> between commit:
>
> e4e55b47ed9a ("LSM: Revive security_task_alloc() hook and per "struct
> task_struct" security blob.")
>
> from the security tree and commit:
>
> d83a7cb375ee ("livepatch: change to a per-task consistency model")
>
> from the livepatching tree.
>
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging. You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
Thank you for fixing.
This is an append to "struct task_struct" race which is inevitable.
>
> --
> Cheers,
> Stephen Rothwell
>
> diff --cc include/linux/init_task.h
> index 926f2f553cc5,5a791055b176..000000000000
> --- a/include/linux/init_task.h
> +++ b/include/linux/init_task.h
> @@@ -294,7 -296,7 +302,8 @@@ extern struct cred init_cred
> INIT_VTIME(tsk) \
> INIT_NUMA_BALANCING(tsk) \
> INIT_KASAN(tsk) \
> + INIT_TASK_SECURITY \
> + INIT_LIVEPATCH(tsk) \
> }
>
>
> diff --cc include/linux/sched.h
> index 9ac23308c82c,e11032010318..000000000000
> --- a/include/linux/sched.h
> +++ b/include/linux/sched.h
> @@@ -1042,10 -1038,9 +1042,13 @@@ struct task_struct
> /* A live task holds one reference: */
> atomic_t stack_refcount;
> #endif
> +#ifdef CONFIG_SECURITY
> + /* Used by LSM modules for access restriction: */
> + void *security;
> +#endif
> + #ifdef CONFIG_LIVEPATCH
Just a comment to livepatch change. It seems that addition of
/* A comment line for this variable: */
line and indentation alignment is expected due to
"Unscheduled maintenance for sched.h" work.
> + int patch_state;
> + #endif
> /* CPU-specific state of this task: */
> struct thread_struct thread;
>
>
