Re: [RFC PATCH] nvmet: Make blksize_shift configurable

[Chaitanya Kulkarni <chaitanyak@nvidia.com>]

On 4/18/25 02:08, Richard Weinberger wrote:
> +static ssize_t nvmet_ns_blksize_shift_store(struct config_item *item,
> +		const char *page, size_t count)
> +{
> +	struct nvmet_ns *ns = to_nvmet_ns(item);
> +	u32 shift;
> +	int ret;
> +
> +	ret = kstrtou32(page, 0, &shift);
> +	if (ret)
> +		return ret;
> +
> +	mutex_lock(&ns->subsys->lock);
> +	if (ns->enabled) {
> +		pr_err("the ns:%d is already enabled.\n", ns->nsid);
> +		mutex_unlock(&ns->subsys->lock);
> +		return -EINVAL;
> +	}
> +	ns->blksize_shift = shift;
> +	mutex_unlock(&ns->subsys->lock);
> +
> +	return count;
> +}

before we overwrite the old value in ns->blksize_shift don't we need
to make sure new value shift is in the acceptable range ? in case value
is not within the acceptable range then we need to return -EINVAL ?

Also, do we need to #define min and max values for the acceptable
range?

-ck