Re: [RFC PATCH v2 05/20] tracing/filters: Optimise cpumask vs cpumask filtering when user mask is a single CPU

On 29/07/23 15:34, Steven Rostedt wrote:

On Thu, 20 Jul 2023 17:30:41 +0100
Valentin Schneider [off-list ref] wrote:

quoted

             /* Move along */
             i++;
+
+		/*
+		 * Optimisation: if the user-provided mask has a weight of one
+		 * then we can treat it as a scalar input.
+		 */
+		single = cpumask_weight(pred->mask) == 1;
+		if (single && field->filter_type == FILTER_CPUMASK) {
+			pred->val = cpumask_first(pred->mask);
+			kfree(pred->mask);

Don't we need:
                      pred->mask = NULL;

here, or the free_predicate() will cause a double free?

We do, thanks for spotting this.

-- Steve

quoted

+		}
+