Re: [PATCH v5 4/9] block: add storage element management ioctls

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

On 2026/10/12 16:57, Christoph Hellwig wrote:
> On Mon, Oct 12, 2026 at 04:56:18PM +0200, Damien Le Moal wrote:
>>>
>>> Sashiko is obviously confused here about partitions that aren't supported
>>> on zoned devices.  The conventional zones argument isn't entirely
>>> stupid, but also not a great one.  Maybe we should add a check to just
>>> reject the ioctls if conventional zones are present, as that would make
>>> a mess of the depop anyway?
>>
>> Not sure sure it makes sense. 3 things can happen:
>> 1) The conv zone goes offline on remove
>> 2) The conv zone goes read-only on remove
>> 3) The conv zone content is totally unknown on restore since that is a
>> destructive op.
>>
>> So the "data loss" comment makes sense only in the case (2). All other cases
>> lose the data anyway. As we make it clear that the user is responsible for sync
>> & no io before issuing a remove ioctl, I am not sure if we really need to do
>> anything special here. And we cannot really do anything special since we do not
>> know which zones are going to be affected. So either we do nothing and let the
>> user deal with potentially cached data not physically accessible anymore (not
>> great), or like here invalidate everything.
> 
> Maybe add a comment?

Will do.

-- 
Damien Le Moal
Western Digital Research