On Wed, 27 Jan 2016, Mathieu Desnoyers wrote:
> +/*
> + * sys_getcpu_cache - setup getcpu cache for caller thread
> + */
> +SYSCALL_DEFINE2(getcpu_cache, int32_t __user **, cpu_cachep, int, flags)
> +{
> + int32_t __user *cpu_cache;
> +
> + if (unlikely(flags))
> + return -EINVAL;
> + /* Check if cpu_cache is already registered. */
> + if (current->cpu_cache) {
> + if (put_user(current->cpu_cache, cpu_cachep))
> + return -EFAULT;
> + return 0;
> + }
This is really odd. How is the caller supposed to differentiate between:
1) Get the installed cpucache pointer
2) Set the cpucache pointer
We really want clearly seperated functionality here.
getcpu_cache(ptr, GET_CACHEP);
and
getcpu_cache(ptr, SET_CACHEP);
Returns -EBUSY if current->cpu_cache is already set, except we allow
replacing the pointer.
Thanks,
tglx
--
To unsubscribe from this list: send the line "unsubscribe linux-api" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
