Re: [External] Re: [PATCH] bpf: avoid grabbing spin_locks of all cpus when no free elems

[Feng Zhou <zhoufeng.zf@bytedance.com>]

在 2022/5/20 上午12:12, Alexei Starovoitov 写道:
> On Wed, May 18, 2022 at 8:12 PM Feng Zhou <zhoufeng.zf@bytedance.com> wrote:
>> 在 2022/5/19 上午4:39, Yonghong Song 写道:
>>>
>>> On 5/17/22 11:57 PM, Feng Zhou wrote:
>>>> 在 2022/5/18 下午2:32, Alexei Starovoitov 写道:
>>>>> On Tue, May 17, 2022 at 11:27 PM Feng zhou
>>>>> <zhoufeng.zf@bytedance.com> wrote:
>>>>>> From: Feng Zhou <zhoufeng.zf@bytedance.com>
>>>>>>
>>>>>> We encountered bad case on big system with 96 CPUs that
>>>>>> alloc_htab_elem() would last for 1ms. The reason is that after the
>>>>>> prealloc hashtab has no free elems, when trying to update, it will
>>>>>> still
>>>>>> grab spin_locks of all cpus. If there are multiple update users, the
>>>>>> competition is very serious.
>>>>>>
>>>>>> So this patch add is_empty in pcpu_freelist_head to check freelist
>>>>>> having free or not. If having, grab spin_lock, or check next cpu's
>>>>>> freelist.
>>>>>>
>>>>>> Before patch: hash_map performance
>>>>>> ./map_perf_test 1
>>> could you explain what parameter '1' means here?
>> This code is here:
>> samples/bpf/map_perf_test_user.c
>> samples/bpf/map_perf_test_kern.c
>> parameter '1' means testcase flag, test hash_map's performance
>> parameter '2048' means test hash_map's performance when free=0.
>> testcase flag '2048' is added by myself to reproduce the problem phenomenon.
> Please convert it to selftests/bpf/bench,
> so that everyone can reproduce the issue you're seeing
> and can assess whether it's a real issue or a corner case.
>
> Also please avoid adding indent in the patch.
> Instead of
>   if (!s->extralist.is_empty) {
>    .. churn
>
> do
>
>   if (s->extralist.is_empty)

Ok, will do. Thanks.