On 2019/09/05 16:19, Ming Lei wrote:
> On Thu, Sep 05, 2019 at 01:28:59PM +0900, Damien Le Moal wrote:
>> When elevator_init_mq() is called from blk_mq_init_allocated_queue(),
>> the only information known about the device is the number of hardware
>> queues as the block device scan by the device driver is not completed
>> yet. The device type and the device required features are not set yet,
>> preventing to correctly choose the default elevator most suitable for
>> the device.
>>
>> This currently affects all multi-queue zoned block devices which default
>> to the "none" elevator instead of the required "mq-deadline" elevator.
>> These drives currently include host-managed SMR disks connected to a
>> smartpqi HBA and null_blk block devices with zoned mode enabled.
>> Upcoming NVMe Zoned Namespace devices will also be affected.
>>
>> Fix this by moving the execution of elevator_init_mq() from
>> blk_mq_init_allocated_queue() into __device_add_disk() to allow for the
>> device driver to probe the device characteristics and set attributes
>> of the device request queue prior to the elevator initialization. This
>> initialization is skipped for DM devices using
>> device_add_disk_no_queue_reg() as this also skips the queue
>> registration.
>>
>> Additionally, to make sure that the elevator initialization is never
>> done while requests are in-flight (there should be none when the device
>> driver calls device_add_disk()), freeze and quiesce the device request
>> queue before calling blk_mq_init_sched() in elevator_init_mq().
>>
>> Signed-off-by: Damien Le Moal <damien.lemoal@xxxxxxx>
>> ---
>> block/blk-mq.c | 2 --
>> block/elevator.c | 7 +++++++
>> block/genhd.c | 9 +++++++++
>> 3 files changed, 16 insertions(+), 2 deletions(-)
>>
>> diff --git a/block/blk-mq.c b/block/blk-mq.c
>> index ee4caf0c0807..a37503984206 100644
>> --- a/block/blk-mq.c
>> +++ b/block/blk-mq.c
>> @@ -2902,8 +2902,6 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set,
>> blk_mq_add_queue_tag_set(set, q);
>> blk_mq_map_swqueue(q);
>>
>> - elevator_init_mq(q);
>> -
>> return q;
>>
>> err_hctxs:
>> diff --git a/block/elevator.c b/block/elevator.c
>> index 520d6b224b74..096a670d22d7 100644
>> --- a/block/elevator.c
>> +++ b/block/elevator.c
>> @@ -712,7 +712,14 @@ void elevator_init_mq(struct request_queue *q)
>> if (!e)
>> return;
>>
>> + blk_mq_freeze_queue(q);
>> + blk_mq_quiesce_queue(q);
>> +
>> err = blk_mq_init_sched(q, e);
>> +
>> + blk_mq_unquiesce_queue(q);
>> + blk_mq_unfreeze_queue(q);
>> +
>> if (err) {
>> pr_warn("\"%s\" elevator initialization failed, "
>> "falling back to \"none\"\n", e->elevator_name);
>> diff --git a/block/genhd.c b/block/genhd.c
>> index 54f1f0d381f4..26b31fcae217 100644
>> --- a/block/genhd.c
>> +++ b/block/genhd.c
>> @@ -695,6 +695,15 @@ static void __device_add_disk(struct device *parent, struct gendisk *disk,
>> dev_t devt;
>> int retval;
>>
>> + /*
>> + * The disk queue should now be all set with enough information about
>> + * the device for the elevator code to pick an adequate default
>> + * elevator if one is needed, that is, for devices requesting queue
>> + * registration.
>> + */
>> + if (register_queue)
>> + elevator_init_mq(disk->queue);
>> +
>
> This way is better, but still changes the default elevator to 'none'
> for dm-rq always.
Got it ! I was looking only at mapped_device() in dm.c. But for request based
DMs, the queue is prepared differently in dm_mq_init_request_queue(), using
blk_mq_init_allocated_queue() and blk_register_queue() afterward in
dm_setup_md_queue().
Sending a V5 to fix that.
Thanks.
--
Damien Le Moal
Western Digital Research
--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel
