linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Robin Murphy <robin.murphy@arm.com>
To: Cong Wang <xiyou.wangcong@gmail.com>
Cc: iommu@lists.linux-foundation.org,
	LKML <linux-kernel@vger.kernel.org>,
	Joerg Roedel <joro@8bytes.org>,
	John Garry <john.garry@huawei.com>
Subject: Re: [Patch v3 2/3] iommu: optimize iova_magazine_free_pfns()
Date: Wed, 22 Jan 2020 17:34:31 +0000	[thread overview]
Message-ID: <e8789016-858a-b354-aa98-637e1d453fc3@arm.com> (raw)
In-Reply-To: <CAM_iQpXbjf8MuL17kZhxawXYBJm6t5-ho77F_VWR30L-9FS4Kg@mail.gmail.com>

On 21/01/2020 5:29 pm, Cong Wang wrote:
> On Tue, Jan 21, 2020 at 1:52 AM Robin Murphy <robin.murphy@arm.com> wrote:
>>
>> On 18/12/2019 4:39 am, Cong Wang wrote:
>>> If the magazine is empty, iova_magazine_free_pfns() should
>>> be a nop, however it misses the case of mag->size==0. So we
>>> should just call iova_magazine_empty().
>>>
>>> This should reduce the contention on iovad->iova_rbtree_lock
>>> a little bit, not much at all.
>>
>> Have you measured that in any way? AFAICS the only time this can get
>> called with a non-full magazine is in the CPU hotplug callback, where
>> the impact of taking the rbtree lock and immediately releasing it seems
>> unlikely to be significant on top of everything else involved in that
>> operation.
> 
> This patchset is only tested as a whole, it is not easy to deploy
> each to production and test it separately.
> 
> Is there anything wrong to optimize a CPU hotplug path? :) And,
> it is called in alloc_iova_fast() too when, for example, over-cached.

And if the IOVA space is consumed to the point that we've fallen back to 
that desperate last resort, what do you think the chances are that a 
significant number of percpu magazines will be *empty*? Also bear in 
mind that in that case we've already walked the rbtree once, so any 
notion of still being fast is long, long gone.

As for CPU hotplug, it's a comparatively rare event involving all manner 
of system-wide synchronisation, and the "optimisation" of shaving a few 
dozen CPU cycles off at one point *if* things happen to line up 
correctly is taking a cup of water out of a lake. If the domain is busy 
at the time, then once again chances are the magazines aren't empty and 
having an extra check redundant with the loop condition simply adds 
(trivial, but nonzero) overhead to every call. And if the domain isn't 
busy, then the lock is unlikely to be contended anyway.

Sorry, but without convincing evidence, this change just looks like 
churn for the sake of it.

Robin.

  reply	other threads:[~2020-01-22 17:34 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-18  4:39 [Patch v3 0/3] iommu: reduce spinlock contention on fast path Cong Wang
2019-12-18  4:39 ` [Patch v3 1/3] iommu: avoid unnecessary magazine allocations Cong Wang
2020-01-21 11:11   ` Robin Murphy
2020-01-21 17:21     ` Cong Wang
2020-01-22 17:07       ` Robin Murphy
2020-01-22 17:54         ` Cong Wang
2019-12-18  4:39 ` [Patch v3 2/3] iommu: optimize iova_magazine_free_pfns() Cong Wang
2020-01-21  9:52   ` Robin Murphy
2020-01-21 17:29     ` Cong Wang
2020-01-22 17:34       ` Robin Murphy [this message]
2020-01-22 17:45         ` Cong Wang
2019-12-18  4:39 ` [Patch v3 3/3] iommu: avoid taking iova_rbtree_lock twice Cong Wang
2019-12-19  9:51   ` John Garry
2020-01-21  9:56   ` Robin Murphy
2020-03-03 11:33     ` John Garry
2020-01-20 23:10 ` [Patch v3 0/3] iommu: reduce spinlock contention on fast path Cong Wang
  -- strict thread matches above, loose matches on Subject: below --
2019-12-06 21:38 Cong Wang
2019-12-06 21:38 ` [Patch v3 2/3] iommu: optimize iova_magazine_free_pfns() Cong Wang

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=e8789016-858a-b354-aa98-637e1d453fc3@arm.com \
    --to=robin.murphy@arm.com \
    --cc=iommu@lists.linux-foundation.org \
    --cc=john.garry@huawei.com \
    --cc=joro@8bytes.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=xiyou.wangcong@gmail.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;
as well as URLs for NNTP newsgroup(s).