Re: [PATCH v2 2/4] coredump: ensure the fpu state is flushed for proper multi-threaded core dump

[Oleg Nesterov <oleg@redhat.com>]

On 05/10, Suresh Siddha wrote:
>
> --- a/fs/exec.c
> +++ b/fs/exec.c
> @@ -1930,8 +1930,21 @@ static int coredump_wait(int exit_code, struct core_state *core_state)
>  		core_waiters = zap_threads(tsk, mm, core_state, exit_code);
>  	up_write(&mm->mmap_sem);
>
> -	if (core_waiters > 0)
> +	if (core_waiters > 0) {
> +		struct core_thread *ptr;
> +
>  		wait_for_completion(&core_state->startup);
> +		/*
> +		 * Wait for all the threads to become inactive, so that
> +		 * all the thread context (extended register state, like
> +		 * fpu etc) gets copied to the memory.
> +		 */
> +		ptr = core_state->dumper.next;
> +		while (ptr != NULL) {
> +			wait_task_inactive(ptr->task, 0);
> +			ptr = ptr->next;
> +		}
> +	}

OK, but this adds the unnecessary penalty if we are not going to dump
the core.

Perhaps it makes sense to create a separate helper and call it from
do_coredump() right before "retval = binfmt->core_dump(&cprm)" ?

This also increases the chance that wait_task_inactive() won't actually
wait.

Oleg.