Hi,
On 1/16/2023 6:51 AM, Martin KaFai Lau wrote:
> On 1/13/23 1:15 AM, Tonghao Zhang wrote:
>>
>>
>>> On Jan 13, 2023, at 9:53 AM, Martin KaFai Lau <martin.lau@xxxxxxxxx> wrote:
>>>
>>> On 1/11/23 1:29 AM, tong@xxxxxxxxxxxxx wrote:
>>>> + /*
>>>> + * The lock may be taken in both NMI and non-NMI contexts.
>>>> + * There is a false lockdep warning (inconsistent lock state),
>>>> + * if lockdep enabled. The potential deadlock happens when the
>>>> + * lock is contended from the same cpu. map_locked rejects
>>>> + * concurrent access to the same bucket from the same CPU.
>>>> + * When the lock is contended from a remote cpu, we would
>>>> + * like the remote cpu to spin and wait, instead of giving
>>>> + * up immediately. As this gives better throughput. So replacing
>>>> + * the current raw_spin_lock_irqsave() with trylock sacrifices
>>>> + * this performance gain. atomic map_locked is necessary.
>>>> + * lockdep_off is invoked temporarily to fix the false warning.
>>>> + */
>>>> + lockdep_off();
>>>> raw_spin_lock_irqsave(&b->raw_lock, flags);
>>>> - *pflags = flags;
>>>> + lockdep_on();
>>>
>>> I am not very sure about the lockdep_off/on. Other than the false warning
>>> when using the very same htab map by both NMI and non-NMI context, I think
>>> the lockdep will still be useful to catch other potential issues. The commit
>>> c50eb518e262 ("bpf: Use separate lockdep class for each hashtab") has
>>> already solved this false alarm when NMI happens on one map and non-NMI
>>> happens on another map.
>>>
>>> Alexei, what do you think? May be only land the patch 1 fix for now.
>> Hi Martin
>> Patch 2 is used for patch 1 to test whether there is a deadlock. We should
>> apply this two patches.
>
> It is too noisy for test_progs that developers routinely run. Lets continue to
> explore other ways (or a different test) without this false positive splat.
> Patch 1 was applied as already mentioned in the earlier reply.
> .
Sorry for the late reply. We had discussed with lockdep experts [0]. They
suggested to add a virtual lockdep class for map_locked, because map_locked acts
like a spin_trylock. So how about using lockdep_off() and lockdep_on() for
raw_lock to get rid of these false alarms and adding the lockdep annotation for
map_locked to do the lockdep check ?
[0]: https://lore.kernel.org/bpf/Y4ZABpDSs4%2FuRutC@Boquns-Mac-mini.local/
