From: "Roger Pau Monné" <roger.pau@citrix.com>
To: Jens Axboe <axboe@kernel.dk>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
"martin.peterson@oracle.com" <martin.peterson@oracle.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>
Subject: Re: [PATCH v1 7/7] xen-block: implement indirect descriptors
Date: Thu, 18 Apr 2013 17:14:30 +0200 [thread overview]
Message-ID: <51700DD6.1020603@citrix.com> (raw)
In-Reply-To: <20130418142642.GY4816@kernel.dk>
On 18/04/13 16:26, Jens Axboe wrote:
>>>>>>> I've just set that to something that brings a performance benefit
>>>>>>> without having to map an insane number of persistent grants in blkback.
>>>>>>>
>>>>>>> Yes, the values are correct, but the device request queue (rq) is only
>>>>>>> able to provide read requests with 64 segments or write requests with
>>>>>>> 128 segments. I haven't been able to get larger requests, even when
>>>>>>> setting this to 512 or higer.
>>>>>>
>>>>>> What are you using to drive the requests? 'fio'?
>>>>>
>>>>> Yes, I've tried fio with several "bs=" values, but it doesn't seem to
>>>>> change the size of the underlying requests. Have you been able to get
>>>>> bigger requests?
>>>>
>>>> Martin, Jens,
>>>> Any way to drive more than 128 segments?
>>>
>>> If the driver is bio based, then there's a natural size constraint on
>>> the number of vecs in the bio. So to get truly large requests, the
>>> driver would need to merge incoming sequential IOs (similar to how it's
>>> done for rq based drivers).
>>
>> When you say rq based drivers, you mean drivers with a request queue?
>>
>> We are already using a request queue in blkfront, and I'm setting the
>> maximum number of segments per request using:
>>
>> blk_queue_max_segments(<rq>, <segments>);
>>
>> But even when setting <segments> to 256 or 512, I only get read requests
>> with 64 segments and write requests with 128 segments from the queue.
>
> What kernel are you testing? The plugging is usually what will trigger a
> run of the queue, for rq based drivers. What does your fio job look
> like?
>
I'm currently testing on top of Konrad for-jens-3.9 branch, which is
3.8.0-rc7. This is how my fio job looks like:
[read]
rw=read
size=900m
bs=4k
directory=/root/fio
loops=100000
I've tried several bs sizes; 4k, 16k, 128k, 1m, 10m, and as far as I can
see requests from the queue don't have more than 64 segments.
Also tried a simple dd with several bs sizes and got the same result.
next prev parent reply other threads:[~2013-04-18 15:14 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-27 11:10 [PATCH v1 0/7] xen-block: indirect descriptors Roger Pau Monne
2013-03-27 11:10 ` [PATCH v1 1/7] xen-blkback: print stats about persistent grants Roger Pau Monne
2013-04-09 14:47 ` Konrad Rzeszutek Wilk
2013-03-27 11:10 ` [PATCH v1 2/7] xen-blkback: use balloon pages for all mappings Roger Pau Monne
2013-04-09 14:47 ` Konrad Rzeszutek Wilk
2013-04-15 8:05 ` Roger Pau Monné
2013-04-15 8:12 ` [Xen-devel] " Roger Pau Monné
2013-04-17 14:00 ` Konrad Rzeszutek Wilk
2013-04-15 9:14 ` Roger Pau Monné
2013-04-17 14:05 ` Konrad Rzeszutek Wilk
2013-04-09 15:46 ` Konrad Rzeszutek Wilk
2013-04-15 8:21 ` Roger Pau Monné
2013-03-27 11:10 ` [PATCH v1 3/7] xen-blkback: implement LRU mechanism for persistent grants Roger Pau Monne
2013-04-09 15:42 ` Konrad Rzeszutek Wilk
2013-04-15 11:19 ` Roger Pau Monné
2013-04-17 14:15 ` Konrad Rzeszutek Wilk
2013-03-27 11:10 ` [PATCH v1 4/7] xen-blkback: move pending handles list from blkbk to pending_req Roger Pau Monne
2013-03-27 11:10 ` [PATCH v1 5/7] xen-blkback: make the queue of free requests per backend Roger Pau Monne
2013-04-09 16:13 ` Konrad Rzeszutek Wilk
2013-04-15 13:50 ` Roger Pau Monné
2013-04-17 14:16 ` Konrad Rzeszutek Wilk
2013-03-27 11:10 ` [PATCH v1 6/7] xen-blkback: expand map/unmap functions Roger Pau Monne
2013-03-27 11:10 ` [PATCH v1 7/7] xen-block: implement indirect descriptors Roger Pau Monne
2013-04-09 18:49 ` Konrad Rzeszutek Wilk
2013-04-15 17:01 ` Roger Pau Monné
2013-04-17 14:25 ` Konrad Rzeszutek Wilk
2013-04-17 17:04 ` Roger Pau Monné
2013-04-17 17:27 ` Konrad Rzeszutek Wilk
2013-04-18 12:43 ` Jens Axboe
2013-04-18 14:16 ` Roger Pau Monné
2013-04-18 14:26 ` Jens Axboe
2013-04-18 15:14 ` Roger Pau Monné [this message]
2013-04-18 15:58 ` 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 \
--in-reply-to=51700DD6.1020603@citrix.com \
--to=roger.pau@citrix.com \
--cc=axboe@kernel.dk \
--cc=konrad.wilk@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=martin.peterson@oracle.com \
--cc=xen-devel@lists.xen.org \
/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;
as well as URLs for NNTP newsgroup(s).