Re: [PATCH net-next v7 01/15] mm: page_frag: add a test module for page_frag

On 2024/6/18 22:45, wang wei wrote:

quoted

+
+static struct objpool_head ptr_pool;
+static int nr_objs = 512;
+static atomic_t nthreads;
+static struct completion wait;
+static struct page_frag_cache test_frag;
+
+static int nr_test = 5120000;
+module_param(nr_test, int, 0600);

"S_IRUSR | S_IWUSR" is better than "0600".

Yes, it is better.
But as we do the testing in module init, it seems we could just
use module_param(nr_test, int, 0) instead.

quoted

+MODULE_PARM_DESC(nr_test, "number of iterations to test");
+
+static bool test_align;
+module_param(test_align, bool, 0600);
+MODULE_PARM_DESC(test_align, "use align API for testing");
+
+static int test_alloc_len = 2048;
+module_param(test_alloc_len, int, 0600);
+MODULE_PARM_DESC(test_alloc_len, "alloc len for testing");
+
+static int test_push_cpu;
+module_param(test_push_cpu, int, 0600);
+MODULE_PARM_DESC(test_push_cpu, "test cpu for pushing fragment");
+
+static int test_pop_cpu;
+module_param(test_pop_cpu, int, 0600);
+MODULE_PARM_DESC(test_pop_cpu, "test cpu for popping fragment");
+
+static int page_frag_pop_thread(void *arg)
+{
+	struct objpool_head *pool = arg;
+	int nr = nr_test;
+
+	pr_info("page_frag pop test thread begins on cpu %d\n",
+		smp_processor_id());
+
+	while (nr > 0) {
+		void *obj = objpool_pop(pool);
+
+		if (obj) {
+			nr--;
+			page_frag_free(obj);
+		} else {
+			cond_resched();
+		}
+	}
+
+	if (atomic_dec_and_test(&nthreads))
+		complete(&wait);
+
+	pr_info("page_frag pop test thread exits on cpu %d\n",
+		smp_processor_id());
+
+	return 0;
+}
+
+static int page_frag_push_thread(void *arg)
+{
+	struct objpool_head *pool = arg;
+	int nr = nr_test;
+
+	pr_info("page_frag push test thread begins on cpu %d\n",
+		smp_processor_id());
+
+	while (nr > 0) {
+		void *va;
+		int ret;
+
+		if (test_align)
+			va = page_frag_alloc_align(&test_frag, test_alloc_len,
+ GFP_KERNEL, SMP_CACHE_BYTES);

Every page fragment max size is PAGE_FRAG_CACHE_MAX_SIZE, hence the value of test_alloc_len needs to be checked.

yes, that needs to be checked.
limit the test_alloc_len to PGAE_SIZE seems better, as we may fail back to
order 0 page.