Re: [PATCH v2 9/9] ata: Annotate the code that uses the host lock

[Damien Le Moal <dlemoal@kernel.org>]

On 2026/05/28 3:54, Bart Van Assche wrote:
> On 5/27/26 11:51 AM, Damien Le Moal wrote:
>> On 2026/05/22 2:33, Bart Van Assche wrote:
>>> Annotate all functions and also their direct and indirect callers with
>>> __must_hold() that have a lockdep_assert_held() statement in their
>>> function body for the host lock. As one can see in the comments added by
>>> this patch, locking is missing from the following two functions:
>>>   * Some of the ata_port_freeze() callers.
>>>   * nv_do_interrupt().
>>
>> What do you mean with locking is missing ? Do you mean we have locking problems
>> and bugs ? Clarify please.
> 
> Yes. It means that there is a call chain from the code where I wrote 
> that locking is missing to a lockdep_assert_held() statement for that
> particular synchronization object.

OK. So this shows that improving lock annotations does find issues. Nice!

So please at the very least mentions that in the cover letter, with lockdep
splats or compiler warnings showing the issues. We need to fix that before
applying your patches. Ideally, your series should start with fix patches for
the issues you found. But if you cannot do it, we can look into it.



-- 
Damien Le Moal
Western Digital Research