On Wed, 13 Nov 2024 16:20:40 +0100, Christoph Hellwig wrote:
> currently blk-mq reorders requests when adding them to the plug because
> the request list can't do efficient tail appends. When the plug is
> directly issued using ->queue_rqs that means reordered requests are
> passed to the driver, which can lead to very bad I/O patterns when
> not corrected, especially on rotational devices (e.g. NVMe HDD) or
> when using zone append.
>
> [...]
Applied, thanks!
[1/6] nvme-pci: reverse request order in nvme_queue_rqs
commit: beadf0088501d9dcf2454b05d90d5d31ea3ba55f
[2/6] virtio_blk: reverse request order in virtio_queue_rqs
commit: a982a5637e91fd4491da4bb8eb79cde3fed2300e
[3/6] block: remove rq_list_move
commit: dd53d238d4ca7ff1a0e10eb0205e842560f1394d
[4/6] block: add a rq_list type
commit: dc56e2a78ba20c472c419c6f2ed8fb7d0c95ad90
[5/6] block: don't reorder requests in blk_add_rq_to_plug
commit: 05634083a304eb3d4b5733fe8d6e2fd0c1041ca8
[6/6] block: don't reorder requests in blk_mq_add_to_batch
commit: 65550dea43815d01f45d7ece38e458855d5e2962
Best regards,
--
Jens Axboe
