kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Marcelo Tosatti <mtosatti@redhat.com>
To: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Cc: Takuya Yoshikawa <takuya.yoshikawa@gmail.com>,
	avi@redhat.com, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 3/4 changelog-v2] KVM: Switch to srcu-less get_dirty_log()
Date: Thu, 8 Mar 2012 21:08:35 -0300	[thread overview]
Message-ID: <20120309000835.GA25838@amt.cnet> (raw)
In-Reply-To: <20120308103545.24b81093.yoshikawa.takuya@oss.ntt.co.jp>

On Thu, Mar 08, 2012 at 10:35:45AM +0900, Takuya Yoshikawa wrote:
> Marcelo Tosatti <mtosatti@redhat.com> wrote:
> 
> > What is worrying are large memory cases: think of the 50GB slot case.
> > 100ms hold time is pretty bad (and reacquiring the lock is relatively
> > simple).
> > 
> 
> OK, I agree basically.
> 
> But let me explain one thing before deciding what I should do next.
> 
> With my method, even when we use a 50GB slot, the hold time will be under
> 10ms -- not 100ms -- if the memory actually updated from the last time is
> 1GB (256K dirty pages).
> 
> > >   8747274.0   10973.3       33.3      -31%    -3%    256K
>   Note that this unit-test was done on my tiny core-i3 32-bit host.
>   On servers which can install more than 50GB memory, this will become
>   much faster: actually my live migration tests done on Xeon saw much
>   better numbers.  Unit-test has been tuned for the worst case.
> 
> I admit that if the dirty memory size is more than 10GB, we may see over
> 100ms hold time you are worrying about.
> 
>   For that I was proposing introducing a new GET_DIRTY_LOG API which can
>   restrict the number of dirty pages we get the log - but this is a long
>   term goal.
> 
> 
> So, I am OK to try to introduce cond_resched_lock_cb() as you suggested.
> But, as I explained above, my current implementation does not introduce
> any real regression concerning to mmu_lock hold time:
> 
>   Before we could see 10ms hold time in real workloads:
>   > funcgraph_entry:      ! 9783.060 us |  kvm_mmu_slot_remove_write_access();
> 
>   I have never seen ms hold time with my method in the same workloads.
> 
> So, it would be helpful if you can apply the patch series and I can work
> on top of that: although I cannot use servers with 100GB memory now,
> migrating a guest with 16GB memory or so may be possible later: I need
> to reserve servers for that.

Makes sense.

It looks good to me, Avi can you review & ack please?

> I also want to know "mmu_lock -- TLB flush"-decoupling plan.  We will not
> need to introduce cond_resched_lock_cb() in sched.h if that is possible.
> 
> Thanks,
> 	Takuya

  reply	other threads:[~2012-03-09  0:08 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-03-01 10:30 [PATCH 0/4] KVM: srcu-less dirty logging -v2 Takuya Yoshikawa
2012-03-01 10:31 ` [PATCH 1/4] KVM: MMU: Split the main body of rmap_write_protect() off from others Takuya Yoshikawa
2012-03-12  7:39   ` Takuya Yoshikawa
2012-03-12  7:52     ` Takuya Yoshikawa
2012-03-01 10:32 ` [PATCH 2/4] KVM: Avoid checking huge page mappings in get_dirty_log() Takuya Yoshikawa
2012-03-02  2:56   ` Takuya Yoshikawa
2012-03-02  5:11     ` Takuya Yoshikawa
2012-03-12 18:04   ` Avi Kivity
2012-03-13  9:20     ` Takuya Yoshikawa
2012-03-13 10:12       ` Avi Kivity
2012-03-13 23:04   ` Marcelo Tosatti
2012-03-14  1:04     ` Takuya Yoshikawa
2012-03-14  5:34     ` Takuya Yoshikawa
2012-03-14 10:58       ` Marcelo Tosatti
2012-03-01 10:33 ` [PATCH 3/4] KVM: Switch to srcu-less get_dirty_log() Takuya Yoshikawa
2012-03-03  5:21   ` [PATCH 3/4 changelog-v2] " Takuya Yoshikawa
2012-03-06 11:15     ` Marcelo Tosatti
2012-03-06 14:43       ` Takuya Yoshikawa
2012-03-06 15:01         ` Marcelo Tosatti
2012-03-06 15:23           ` Takuya Yoshikawa
2012-03-06 15:28             ` Marcelo Tosatti
2012-03-07  8:07               ` Takuya Yoshikawa
2012-03-07 23:25                 ` Marcelo Tosatti
2012-03-08  1:35                   ` Takuya Yoshikawa
2012-03-09  0:08                     ` Marcelo Tosatti [this message]
2012-03-12 12:05                       ` Avi Kivity
2012-03-07  8:18               ` Takuya Yoshikawa
2012-03-07 23:20                 ` Marcelo Tosatti
2012-03-16  5:03   ` [PATCH 3/4] " Xiao Guangrong
2012-03-16  6:55     ` Takuya Yoshikawa
2012-03-16  7:30       ` Xiao Guangrong
2012-03-16  7:55         ` Takuya Yoshikawa
2012-03-16  8:28           ` Xiao Guangrong
2012-03-16  9:44             ` Takuya Yoshikawa
2012-03-19  9:34               ` Xiao Guangrong
2012-03-19 10:15                 ` Takuya Yoshikawa
2012-03-01 10:34 ` [PATCH 4/4] KVM: Remove unused dirty_bitmap_head and nr_dirty_pages Takuya Yoshikawa
2012-03-03  5:12 ` [PATCH 0/4] KVM: srcu-less dirty logging -v2 Takuya Yoshikawa
2012-03-20 14:43 ` Avi Kivity

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=20120309000835.GA25838@amt.cnet \
    --to=mtosatti@redhat.com \
    --cc=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=takuya.yoshikawa@gmail.com \
    --cc=yoshikawa.takuya@oss.ntt.co.jp \
    /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).