Re: [PATCH v3 10/13] block: fail zone management operations to read-only and offline zones

On 2026/08/11 4:41, Hannes Reinecke wrote:
> On 8/11/26 12:58 PM, Damien Le Moal wrote:
>> Any zone management operation targeting a zone that is in the read-only
>> or offline condition will fail. So there is no point in issuing such
>> BIO. Modify the check in submit_bio_noacct() to use the new helper
>> function bdev_check_zone_mgmt() to check that a zone is sequential (as was
>> checked before) and also that the zone is not offline nor read-only.
>>
>> Signed-off-by: Damien Le Moal <[email protected]>
>> Reviewed-by: Bart Van Assche <[email protected]>
>> ---
>>   block/blk-core.c  |  7 +++++--
>>   block/blk-zoned.c | 24 ++++++++++++++++++++++++
>>   block/blk.h       |  6 ++++++
>>   3 files changed, 35 insertions(+), 2 deletions(-)
>>
> Hmm. But that checks the in-kernel zone state only,
> so I would expect that we still can see failures if
> the zone state has not been updated (yet) while
> the command has been issued.
> What harm would it do to just run the command and
> deal with the results?

NCQ errors are painful with SATA due to the queue drain they cause. So I am all
for avoiding them as much as possible. This is not the hot path and this check
is not a heavy one, so I added it.

This is a minor case anyway since these errors can only come from a bad
application behavior given that a head depop that offline zones is always user
triggered. So a well-behaving application must do a report zones right after the
depop to note the zones that went offline and let them be.

> 
> Cheers,
> 
> Hannes


-- 
Damien Le Moal
Western Digital Research