Re: [PATCHSET 0/5] Enable alloc caching and batched freeing for passthrough

[PATCHSET 0/5] Enable alloc caching and batched freeing for passthrough · Jens Axboe <axboe@kernel.dk> · 2022-09-22
[PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Jens Axboe <axboe@kernel.dk> · 2022-09-22
Re: [PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Pankaj Raghav <hidden> · 2022-09-23
Re: [PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Pankaj Raghav <hidden> · 2022-09-23
Re: [PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Jens Axboe <axboe@kernel.dk> · 2022-09-23
Re: [PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Damien Le Moal <hidden> · 2022-09-24
Re: [PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Jens Axboe <axboe@kernel.dk> · 2022-09-24
Re: [PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Damien Le Moal <hidden> · 2022-09-24
Re: [PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Pankaj Raghav <hidden> · 2022-09-24
Re: [PATCH 1/5] block: enable batched allocation for blk_mq_alloc_request() · Jens Axboe <axboe@kernel.dk> · 2022-09-24
[PATCH 2/5] block: change request end_io handler to pass back a return value · Jens Axboe <axboe@kernel.dk> · 2022-09-22
[PATCH 5/5] nvme: enable batched completions of passthrough IO · Jens Axboe <axboe@kernel.dk> · 2022-09-22
[PATCH 3/5] block: allow end_io based requests in the completion batch handling · Jens Axboe <axboe@kernel.dk> · 2022-09-22
[PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Jens Axboe <axboe@kernel.dk> · 2022-09-22
Re: [PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Christoph Hellwig <hch@infradead.org> · 2022-09-23
Re: [PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Jens Axboe <axboe@kernel.dk> · 2022-09-23
Re: [PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Christoph Hellwig <hch@infradead.org> · 2022-09-26
Re: [PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Jens Axboe <axboe@kernel.dk> · 2022-09-26
Re: [PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Christoph Hellwig <hch@infradead.org> · 2022-09-26
Re: [PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Jens Axboe <axboe@kernel.dk> · 2022-09-26
Re: [PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Christoph Hellwig <hch@infradead.org> · 2022-09-26
Re: [PATCH 4/5] nvme: split out metadata vs non metadata end_io uring_cmd completions · Jens Axboe <axboe@kernel.dk> · 2022-09-26
Re: [PATCHSET 0/5] Enable alloc caching and batched freeing for passthrough · Christoph Hellwig <hch@infradead.org> · 2022-09-23
Re: [PATCHSET 0/5] Enable alloc caching and batched freeing for passthrough · Jens Axboe <axboe@kernel.dk> · 2022-09-23
Re: [PATCHSET 0/5] Enable alloc caching and batched freeing for passthrough · Christoph Hellwig <hch@infradead.org> · 2022-09-23
Re: [PATCHSET 0/5] Enable alloc caching and batched freeing for passthrough · Jens Axboe <axboe@kernel.dk> · 2022-09-23

From: Christoph Hellwig <hch@infradead.org>
Date: 2022-09-23 15:16:49
Also in: linux-nvme, linux-scsi

On Thu, Sep 22, 2022 at 12:28:00PM -0600, Jens Axboe wrote:

This is good for a 10% improvement for passthrough performance. For
a non-drive limited test case, passthrough IO is now more efficient
than the regular bdev O_DIRECT path.

How so?  If it ends up faster we are doing something wrong in the
normal path as there should be no fundamental difference in the work
that is being done.

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help