From: Jens Axboe Date: Thu, 22 Sep 2022 17:41:51 +0000 (-0600) Subject: nvme: enable batched completions of passthrough IO X-Git-Url: http://git.maquefel.me/?a=commitdiff_plain;h=851eb780decb7180bcf09fad0035cba9aae669df;p=linux.git nvme: enable batched completions of passthrough IO Now that the normal passthrough end_io path doesn't need the request anymore, we can kill the explicit blk_mq_free_request() and just pass back RQ_END_IO_FREE instead. This enables the batched completion from freeing batches of requests at the time. This brings passthrough IO performance at least on par with bdev based O_DIRECT with io_uring. With this and batche allocations, peak performance goes from 110M IOPS to 122M IOPS. For IRQ based, passthrough is now also about 10% faster than previously, going from ~61M to ~67M IOPS. Reviewed-by: Anuj Gupta Reviewed-by: Sagi Grimberg Reviewed-by: Keith Busch Co-developed-by: Stefan Roesch Signed-off-by: Jens Axboe --- diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c index f9d1f7e4d6d1b..914b142b6f2b0 100644 --- a/drivers/nvme/host/ioctl.c +++ b/drivers/nvme/host/ioctl.c @@ -430,8 +430,7 @@ static enum rq_end_io_ret nvme_uring_cmd_end_io(struct request *req, else io_uring_cmd_complete_in_task(ioucmd, nvme_uring_task_cb); - blk_mq_free_request(req); - return RQ_END_IO_NONE; + return RQ_END_IO_FREE; } static enum rq_end_io_ret nvme_uring_cmd_end_io_meta(struct request *req,