Re: [PATCH net-next v3 1/3] selftests/ptr_ring: add benchmark application for ptr_ring

On 2021/7/2 14:43, Jason Wang wrote:

在 2021/7/1 下午8:26, Yunsheng Lin 写道:

quoted

Currently ptr_ring selftest is embedded within the virtio
selftest, which involves some specific virtio operation,
such as notifying and kicking.

As ptr_ring has been used by various subsystems, it deserves
it's owner selftest in order to benchmark different usecase
of ptr_ring, such as page pool and pfifo_fast qdisc.

So add a simple application to benchmark ptr_ring performance.
Currently two test mode is supported:
Mode 0: Both producing and consuming is done in a single thread,
         it is called simple test mode in the test app.
Mode 1: Producing and consuming is done in different thread
         concurrently, also known as SPSC(single-producer/
         single-consumer) test.

The multi-producer/single-consumer test for pfifo_fast case is
not added yet, which can be added if using CAS atomic operation
to enable lockless multi-producer is proved to be better than
using r->producer_lock.

Signed-off-by: Yunsheng Lin <redacted>
---
V3: Remove timestamp sampling, use standard C library as much
     as possible.

[...]

quoted

+static void *produce_worker(void *arg)
+{
+    struct worker_info *info = arg;
+    unsigned long i = 0;
+    int ret;
+
+    while (++i <= info->test_count) {
+        while (__ptr_ring_full(&ring))
+            cpu_relax();
+
+        ret = __ptr_ring_produce(&ring, (void *)i);
+        if (ret) {
+            fprintf(stderr, "produce failed: %d\n", ret);
+            info->error = true;
+            return NULL;
+        }
+    }
+
+    info->error = false;
+
+    return NULL;
+}
+
+static void *consume_worker(void *arg)
+{
+    struct worker_info *info = arg;
+    unsigned long i = 0;
+    int *ptr;
+
+    while (++i <= info->test_count) {
+        while (__ptr_ring_empty(&ring))
+            cpu_relax();

Any reason for not simply use __ptr_ring_consume() here?

No particular reason, just to make sure the ring is
non-empty before doing the enqueuing, we could check
if the __ptr_ring_consume() return NULL to decide
the if the ring is empty. Using __ptr_ring_consume()
here enable testing the correctness and performance of
__ptr_ring_consume() too.

quoted

+
+        ptr = __ptr_ring_consume(&ring);
+        if ((unsigned long)ptr != i) {
+            fprintf(stderr, "consumer failed, ptr: %lu, i: %lu\n",
+                (unsigned long)ptr, i);
+            info->error = true;
+            return NULL;
+        }
+    }
+
+    if (!__ptr_ring_empty(&ring)) {
+        fprintf(stderr, "ring should be empty, test failed\n");
+        info->error = true;
+        return NULL;
+    }
+
+    info->error = false;
+    return NULL;
+}
+

[...]

quoted

+
+    return 0;
+}

diff --git a/tools/testing/selftests/ptr_ring/ptr_ring_test.h b/tools/testing/selftests/ptr_ring/ptr_ring_test.h
new file mode 100644
index 0000000..32bfefb
--- /dev/null
+++ b/tools/testing/selftests/ptr_ring/ptr_ring_test.h

Let's reuse ptr_ring.c in tools/virtio/ringtest. Nothing virt specific there.

It *does* have some virtio specific at the end of ptr_ring.c.
It can be argued that the ptr_ring.c in tools/virtio/ringtest
could be refactored to remove the function related to virtio.

But as mentioned in the previous disscusion [1], the tools/virtio/
seems to have compile error in the latest kernel, it does not seems
right to reuse that. And most of testcase in tools/virtio/ seems
better be in tools/virtio/ringtest instead，so until the testcase
in tools/virtio/ is compile-error-free and moved to tools/testing/
selftests/, it seems better not to reuse it for now.

1. https://patchwork.kernel.org/project/netdevbpf/patch/1624591136-6647-2-git-send-email-linyunsheng@huawei.com/#24278945

Thanks

[...]

.