On 8/9/23 10:53 PM, Greg KH wrote:
> On Wed, Aug 09, 2023 at 04:08:52PM -0600, Jens Axboe wrote:
>> On 8/9/23 6:56 AM, Sweet Tea Dorminy wrote:
>>> blk_crypto_profile_init() calls lockdep_register_key(), which warns and
>>> does not register if the provided memory is a static object.
>>> blk-crypto-fallback currently has a static blk_crypto_profile and calls
>>> blk_crypto_profile_init() thereupon, resulting in the warning and
>>> failure to register.
>>>
>>> Fortunately it is simple enough to use a dynamically allocated profile
>>> and make lockdep function correctly.
>>>
>>> Fixes: 2fb48d88e77f ("blk-crypto: use dynamic lock class for blk_crypto_profile::lock")
>>> Cc: stable@xxxxxxxxxxxxxxx
>>> Signed-off-by: Sweet Tea Dorminy <sweettea-kernel@xxxxxxxxxx>
>>
>> The offending commit went into 6.5, so there should be no need for a
>> stable tag on this one. But I can edit that while applying, waiting on
>> Eric to ack it.
>
> That commit has been backported to stable releases, so it would be nice
> to keep it there so our tools automatically pick it up properly. Once
> the authorship name is fixed up of course.
But that stable tag should not be necessary? If stable has backported a
commit, surely it'll pick a commit that has that in Fixes? Otherwise
that seems broken and implies that people need to potentially check
every commit for a stable presence.
I can keep the tag, just a bit puzzled as to why that would be
necessary.
The authorship is fine, but looks like the patch needs changes anyway as
per Eric.
--
Jens Axboe
