Re: [PATCH v2 bpf-next 3/3] bpf: tcp: Add bpf_cubic example

On Wed, Jan 22, 2020 at 01:53:45PM -0800, Andrii Nakryiko wrote:

On Tue, Jan 21, 2020 at 10:42 PM Martin KaFai Lau [off-list ref] wrote:

quoted

This patch adds a bpf_cubic example.  Some highlights:
1. CONFIG_HZ .kconfig map is used.
2. In bictcp_update(), calculation is changed to use usec
   resolution (i.e. USEC_PER_JIFFY) instead of using jiffies.
   Thus, usecs_to_jiffies() is not used in the bpf_cubic.c.
3. In bitctcp_update() [under tcp_friendliness], the original
   "while (ca->ack_cnt > delta)" loop is changed to the equivalent
   "ca->ack_cnt / delta" operation.

Signed-off-by: Martin KaFai Lau <redacted>
---

just my few cents below...

[...]

quoted

+static void test_cubic(void)
+{
+       struct bpf_cubic *cubic_skel;
+       struct bpf_link *link;
+
+       cubic_skel = bpf_cubic__open_and_load();
+       if (CHECK(!cubic_skel, "bpf_cubic__open_and_load", "failed\n"))
+               return;
+
+       link = bpf_map__attach_struct_ops(cubic_skel->maps.cubic);

we should probably teach bpftool and libbpf to generate a link for
struct_ops map and also auto-attach it as part of skeleton's attach...
I'll add it if noone gets to it sooner

It is in my TODO list.  I am happy to be beaten though ;)

quoted

+       if (CHECK(IS_ERR(link), "bpf_map__attach_struct_ops", "err:%ld\n",
+                 PTR_ERR(link))) {
+               bpf_cubic__destroy(cubic_skel);
+               return;
+       }
+
+       do_test("bpf_cubic");
+
+       bpf_link__destroy(link);
+       bpf_cubic__destroy(cubic_skel);
+}
+

[...]

quoted

+
+extern unsigned long CONFIG_HZ __kconfig __weak;

you probably don't want __weak, if CONFIG_HZ is not defined in
Kconfig, then something wrong is going on, probably, so it's better to
error out early

Thanks for pointing it out.  Will do.