public inbox for [email protected]
 help / color / mirror / Atom feed
From: Jens Axboe <[email protected]>
To: Kanchan Joshi <[email protected]>,
	[email protected], [email protected], [email protected]
Cc: [email protected], [email protected],
	[email protected], [email protected]
Subject: Re: [PATCH for-next v3 0/4] fixed-buffer for uring-cmd/passthrough
Date: Fri, 2 Sep 2022 10:32:16 -0600	[thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>

On 9/2/22 10:06 AM, Jens Axboe wrote:
> On 9/2/22 9:16 AM, Kanchan Joshi wrote:
>> Hi,
>>
>> Currently uring-cmd lacks the ability to leverage the pre-registered
>> buffers. This series adds the support in uring-cmd, and plumbs
>> nvme passthrough to work with it.
>>
>> Using registered-buffers showed peak-perf hike from 1.85M to 2.17M IOPS
>> in my setup.
>>
>> Without fixedbufs
>> *****************
>> # taskset -c 0 t/io_uring -b512 -d128 -c32 -s32 -p0 -F1 -B0 -O0 -n1 -u1 /dev/ng0n1
>> submitter=0, tid=5256, file=/dev/ng0n1, node=-1
>> polled=0, fixedbufs=0/0, register_files=1, buffered=1, QD=128
>> Engine=io_uring, sq_ring=128, cq_ring=128
>> IOPS=1.85M, BW=904MiB/s, IOS/call=32/31
>> IOPS=1.85M, BW=903MiB/s, IOS/call=32/32
>> IOPS=1.85M, BW=902MiB/s, IOS/call=32/32
>> ^CExiting on signal
>> Maximum IOPS=1.85M
> 
> With the poll support queued up, I ran this one as well. tldr is:
> 
> bdev (non pt)	122M IOPS
> irq driven	51-52M IOPS
> polled		71M IOPS
> polled+fixed	78M IOPS
> 
> Looking at profiles, it looks like the bio is still being allocated
> and freed and not dipping into the alloc cache, which is using a
> substantial amount of CPU. I'll poke a bit and see what's going on...

It's using the fs_bio_set, and that doesn't have the PERCPU alloc cache
enabled. With the below, we then do:

polled+fixed	82M

I suspect the remainder is due to the lack of batching on the request
freeing side, at least some of it. Haven't really looked deeper yet.

One issue I saw - try and use passthrough polling without having any
poll queues defined and it'll stall just spinning on completions. You
need to ensure that these are processed as well - look at how the
non-passthrough io_uring poll path handles it.


diff --git a/block/bio.c b/block/bio.c
index 3d3a2678fea2..cba6b1c02eb8 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -1754,7 +1754,7 @@ static int __init init_bio(void)
 	cpuhp_setup_state_multi(CPUHP_BIO_DEAD, "block/bio:dead", NULL,
 					bio_cpu_dead);
 
-	if (bioset_init(&fs_bio_set, BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS))
+	if (bioset_init(&fs_bio_set, BIO_POOL_SIZE, 0, BIOSET_NEED_BVECS | BIOSET_PERCPU_CACHE))
 		panic("bio: can't allocate bios\n");
 
 	if (bioset_integrity_create(&fs_bio_set, BIO_POOL_SIZE))

-- 
Jens Axboe

  reply	other threads:[~2022-09-02 16:32 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20220902152701epcas5p1d4aca8eebc90fb96ac7ed5a8270816cf@epcas5p1.samsung.com>
2022-09-02 15:16 ` [PATCH for-next v3 0/4] fixed-buffer for uring-cmd/passthrough Kanchan Joshi
     [not found]   ` <CGME20220902152706epcas5p11f1a11710a6ba985f4679a5cfe131bf8@epcas5p1.samsung.com>
2022-09-02 15:16     ` [PATCH for-next v3 1/4] io_uring: introduce io_uring_cmd_import_fixed Kanchan Joshi
     [not found]   ` <CGME20220902152709epcas5p1a1bd433cac6040c492e347edae484ca5@epcas5p1.samsung.com>
2022-09-02 15:16     ` [PATCH for-next v3 2/4] io_uring: introduce fixed buffer support for io_uring_cmd Kanchan Joshi
2022-09-02 23:13       ` Jens Axboe
     [not found]   ` <CGME20220902152712epcas5p2622e861ac4a5ae9820a9af9442d556b4@epcas5p2.samsung.com>
2022-09-02 15:16     ` [PATCH for-next v3 3/4] block: add helper to map bvec iterator for passthrough Kanchan Joshi
2022-09-02 23:14       ` Jens Axboe
     [not found]   ` <CGME20220902152717epcas5p26905ce7cb48e9d278976a301d73c297f@epcas5p2.samsung.com>
2022-09-02 15:16     ` [PATCH for-next v3 4/4] nvme: wire up fixed buffer support for nvme passthrough Kanchan Joshi
2022-09-02 16:06   ` [PATCH for-next v3 0/4] fixed-buffer for uring-cmd/passthrough Jens Axboe
2022-09-02 16:32     ` Jens Axboe [this message]
2022-09-02 18:46       ` Kanchan Joshi
2022-09-02 19:32         ` Jens Axboe
2022-09-02 21:25           ` Jens Axboe
2022-09-03  9:34             ` Kanchan Joshi
2022-09-03 17:00             ` Jens Axboe
2022-09-04 17:01               ` Kanchan Joshi
2022-09-04 20:17                 ` Jens Axboe
2022-09-05  5:52                   ` Kanchan Joshi
2022-09-05 17:48                     ` Jens Axboe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    [email protected] \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox