From: Hannes Reinecke <hare@suse.de>
To: Bart Van Assche <bvanassche@acm.org>,
"Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>,
Doug Gilbert <dgilbert@interlog.com>,
Daniel Wagner <daniel.wagner@suse.com>,
James Bottomley <james.bottomley@hansenpartnership.com>,
linux-scsi@vger.kernel.org
Subject: Re: [RFC PATCH 0/4] scsi: use xarray for devices and targets
Date: Wed, 27 May 2020 18:59:23 +0200 [thread overview]
Message-ID: <6bfb4925-b714-80da-e8f5-9e2093fc14e5@suse.de> (raw)
In-Reply-To: <f1cb4faf-816e-9f71-aa74-ddf023b197f4@acm.org>
On 5/27/20 6:36 PM, Bart Van Assche wrote:
> On 2020-05-27 07:13, Hannes Reinecke wrote:
>> Hi all,
>>
>> based on the ideas from Doug Gilbert here's now my take on using
>> xarrays for devices and targets.
>> It revolves around two ideas:
>> - 'channel' and 'id' are never ever used to the full 32 bit range;
>> 'channels' are well below 10, and no driver is using more than
>> 16 bits for the id. So we can reduce the type of 'channel' and
>> 'id' to 16 bits, and use the 32 bit value 'channel << 16 | id'
>> as the index into the target xarray.
>> - Most SCSI LUNs are below 256 (to ensure compability with older
>> systems). So there we can use the LUN number as the index into
>> the xarray; for larger LUN numbers we'll allocate a separate
>> index.
>>
>> With these change we can implement an efficient lookup mechanism,
>> devolving into direct lookup for most cases.
>> And iteration should be as efficient as the current, list-based,
>> approach.
>>
>> This is compile-tested only, to give you an impression of the
>> overall idea and to get the discussion rolling.
>
> Hi Hannes,
>
> My understanding of the xarray concept is that it provides two
> advantages over using linked lists:
> - Faster lookups.
> - Requires less memory.
>
> Will we benefit from any of these advantages in the SCSI code? Hadn't
> James Bottomley already brought up that lookup by (channel, target, lun)
> only happens from some LLDs and from the procfs code?
>
It's not only lookup, it's iteration in general.
Which affects scanning and device removal; especially the latter is
_very_ error prone (just look at scsi_target_reap etc), so any reduction
in complexity is a good thing in general methinks.
> Are there any use cases where the number of SCSI devices is large enough
> to benefit from the memory reduction?
>
I would assume that we're seeing benefits as soon as we're in the range
of tens to hundreds of devices; then list lookup will be eating up more
time and space as xarrays.
And the big benefit of using xarrays is that we will be alerted if an
element with the same indices is being added; we've already had issues
in the past here which are notoriously difficult to track down.
Cheers,
Hannes
--
Dr. Hannes Reinecke Teamlead Storage & Networking
hare@suse.de +49 911 74053 688
SUSE Software Solutions GmbH, Maxfeldstr. 5, 90409 Nürnberg
HRB 36809 (AG Nürnberg), Geschäftsführer: Felix Imendörffer
next prev parent reply other threads:[~2020-05-27 16:59 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-27 14:13 [RFC PATCH 0/4] scsi: use xarray for devices and targets Hannes Reinecke
2020-05-27 14:13 ` [PATCH 1/4] scsi: convert target lookup to xarray Hannes Reinecke
2020-05-27 14:57 ` Johannes Thumshirn
2020-05-27 15:06 ` Hannes Reinecke
2020-05-28 7:24 ` Daniel Wagner
2020-05-28 7:52 ` Hannes Reinecke
2020-05-28 16:28 ` Bart Van Assche
2020-05-29 5:01 ` kbuild test robot
2020-05-31 9:10 ` Dan Carpenter
2020-05-27 14:13 ` [PATCH 2/4] target_core_pscsi: use __scsi_device_lookup() Hannes Reinecke
2020-05-27 15:05 ` Johannes Thumshirn
2020-05-27 14:13 ` [PATCH 3/4] scsi: move target device list to xarray Hannes Reinecke
2020-05-27 15:34 ` Johannes Thumshirn
2020-05-27 20:13 ` kbuild test robot
2020-05-30 2:47 ` kbuild test robot
2020-05-27 14:14 ` [PATCH 4/4] scsi: remove direct device lookup per host Hannes Reinecke
2020-05-28 8:00 ` kbuild test robot
2020-05-27 16:36 ` [RFC PATCH 0/4] scsi: use xarray for devices and targets Bart Van Assche
2020-05-27 16:59 ` Hannes Reinecke [this message]
2020-05-28 3:59 ` Douglas Gilbert
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=6bfb4925-b714-80da-e8f5-9e2093fc14e5@suse.de \
--to=hare@suse.de \
--cc=bvanassche@acm.org \
--cc=daniel.wagner@suse.com \
--cc=dgilbert@interlog.com \
--cc=hch@lst.de \
--cc=james.bottomley@hansenpartnership.com \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
/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