Re: [PATCH 0/5] xen-netback: Fine-tuning for three function implementations

On Sat, 2016-01-02 at 18:50 +0100, SF Markus Elfring wrote:

A few update suggestions were taken into account
from static source code analysis.

While static analysis can be useful, I don't think these
specific conversions are generally useful.

Perhaps it would be more useful to convert the string
duplication or snprintf logic to kstrdup/kasprintf

This:

	if (num_queues == 1) {
		xspath = kzalloc(strlen(dev->otherend) + 1, GFP_KERNEL);
		if (!xspath) {
			xenbus_dev_fatal(dev, -ENOMEM,
					 "reading ring references");
			return -ENOMEM;
		}
		strcpy(xspath, dev->otherend);
	} else {
		xspathsize = strlen(dev->otherend) + xenstore_path_ext_size;
		xspath = kzalloc(xspathsize, GFP_KERNEL);
		if (!xspath) {
			xenbus_dev_fatal(dev, -ENOMEM,
					 "reading ring references");
			return -ENOMEM;
		}
		snprintf(xspath, xspathsize, "%s/queue-%u", dev->otherend,
			 queue->id);
	}

could be simplified to something like:

	if (num_queues == 1)
		xspath = kstrdup(dev->otherend, GFP_KERNEL);
	else
		xspath = kasprintf(GFP_KERNEL, "%s/queue-%u",
				   dev->otherend, queue->id);
	if (!xspath)
		etc...