| Message ID | 20181117214354.822-4-axboe@kernel.dk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/7] block: avoid ordered task state change for polled IO | expand |
On Sat, Nov 17, 2018 at 02:43:50PM -0700, Jens Axboe wrote: > This relies on the fc taget ops setting ->poll_queue, which > nobody does. Otherwise it just checks if something has > completed, which isn't very useful. Please also remove the unused ->poll_queue method. Otherwise looks fine: Reviewed-by: Christoph Hellwig <hch@lst.de>
On 11/19/18 12:59 AM, Christoph Hellwig wrote: > On Sat, Nov 17, 2018 at 02:43:50PM -0700, Jens Axboe wrote: >> This relies on the fc taget ops setting ->poll_queue, which >> nobody does. Otherwise it just checks if something has >> completed, which isn't very useful. > > Please also remove the unused ->poll_queue method. > > Otherwise looks fine: > > Reviewed-by: Christoph Hellwig <hch@lst.de> Sure, will do.
On 11/19/2018 7:19 AM, Jens Axboe wrote: > On 11/19/18 12:59 AM, Christoph Hellwig wrote: >> On Sat, Nov 17, 2018 at 02:43:50PM -0700, Jens Axboe wrote: >>> This relies on the fc taget ops setting ->poll_queue, which >>> nobody does. Otherwise it just checks if something has >>> completed, which isn't very useful. >> Please also remove the unused ->poll_queue method. >> >> Otherwise looks fine: >> >> Reviewed-by: Christoph Hellwig <hch@lst.de> > Sure, will do. > Looks fine on my end Reviewed-by: James Smart <jsmart2021@gmail.com>
diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c index 98c3c77f48f6..de797c641265 100644 --- a/drivers/nvme/host/fc.c +++ b/drivers/nvme/host/fc.c @@ -2302,38 +2302,6 @@ nvme_fc_queue_rq(struct blk_mq_hw_ctx *hctx, return nvme_fc_start_fcp_op(ctrl, queue, op, data_len, io_dir); } -static struct blk_mq_tags * -nvme_fc_tagset(struct nvme_fc_queue *queue) -{ - if (queue->qnum == 0) - return queue->ctrl->admin_tag_set.tags[queue->qnum]; - - return queue->ctrl->tag_set.tags[queue->qnum - 1]; -} - -static int -nvme_fc_poll(struct blk_mq_hw_ctx *hctx, unsigned int tag) - -{ - struct nvme_fc_queue *queue = hctx->driver_data; - struct nvme_fc_ctrl *ctrl = queue->ctrl; - struct request *req; - struct nvme_fc_fcp_op *op; - - req = blk_mq_tag_to_rq(nvme_fc_tagset(queue), tag); - if (!req) - return 0; - - op = blk_mq_rq_to_pdu(req); - - if ((atomic_read(&op->state) == FCPOP_STATE_ACTIVE) && - (ctrl->lport->ops->poll_queue)) - ctrl->lport->ops->poll_queue(&ctrl->lport->localport, - queue->lldd_handle); - - return ((atomic_read(&op->state) != FCPOP_STATE_ACTIVE)); -} - static void nvme_fc_submit_async_event(struct nvme_ctrl *arg) { @@ -2404,7 +2372,6 @@ static const struct blk_mq_ops nvme_fc_mq_ops = { .init_request = nvme_fc_init_request, .exit_request = nvme_fc_exit_request, .init_hctx = nvme_fc_init_hctx, - .poll = nvme_fc_poll, .timeout = nvme_fc_timeout, };
This relies on the fc taget ops setting ->poll_queue, which nobody does. Otherwise it just checks if something has completed, which isn't very useful. Signed-off-by: Jens Axboe <axboe@kernel.dk> --- drivers/nvme/host/fc.c | 33 --------------------------------- 1 file changed, 33 deletions(-)