All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ben Gardon <bgardon@google.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: Sean Christopherson <seanjc@google.com>,
	LKML <linux-kernel@vger.kernel.org>, kvm <kvm@vger.kernel.org>,
	Peter Xu <peterx@redhat.com>, Peter Shier <pshier@google.com>,
	Peter Feiner <pfeiner@google.com>,
	Junaid Shahid <junaids@google.com>,
	Jim Mattson <jmattson@google.com>,
	Yulei Zhang <yulei.kernel@gmail.com>,
	Wanpeng Li <kernellwp@gmail.com>,
	Vitaly Kuznetsov <vkuznets@redhat.com>,
	Xiao Guangrong <xiaoguangrong.eric@gmail.com>
Subject: Re: [PATCH 15/24] kvm: mmu: Wrap mmu_lock cond_resched and needbreak
Date: Wed, 27 Jan 2021 12:08:19 -0800	[thread overview]
Message-ID: <CANgfPd9FaPhQiEkJ=VHKiVWZ_5S3k2uWHU+ViCi4nEF=GU4qsw@mail.gmail.com> (raw)
In-Reply-To: <f1ef3118-2a8e-4bf2-b3b0-60ac4947e106@redhat.com>

On Tue, Jan 26, 2021 at 12:48 PM Paolo Bonzini <pbonzini@redhat.com> wrote:
>
> On 26/01/21 19:11, Ben Gardon wrote:
> > When I did a strict replacement I found ~10% worse memory population
> > performance.
> > Running dirty_log_perf_test -v 96 -b 3g -i 5 with the TDP MMU
> > disabled, I got 119 sec to populate memory as the baseline and 134 sec
> > with an earlier version of this series which just replaced the
> > spinlock with an rwlock. I believe this difference is statistically
> > significant, but didn't run multiple trials.
> > I didn't take notes when profiling, but I'm pretty sure the rwlock
> > slowpath showed up a lot. This was a very high contention scenario, so
> > it's probably not indicative of real-world performance.
> > In the slow path, the rwlock is certainly slower than a spin lock.
> >
> > If the real impact doesn't seem too large, I'd be very happy to just
> > replace the spinlock.
>
> Ok, so let's use the union idea and add a "#define KVM_HAVE_MMU_RWLOCK"
> to x86.  The virt/kvm/kvm_main.c MMU notifiers functions can use the
> #define to pick between write_lock and spin_lock.

I'm not entirely sure I understand this suggestion. Are you suggesting
we'd have the spinlock and rwlock in a union in struct kvm but then
use a static define to choose which one is used by other functions? It
seems like if we're using static defines the union doesn't add value.
If we do use the union, I think the advantages offered by __weak
wrapper functions, overridden on a per-arch basis, are worthwhile.

>
> For x86 I want to switch to tdp_mmu=1 by default as soon as parallel
> page faults are in, so we can use the rwlock unconditionally and drop
> the wrappers, except possibly for some kind of kvm_mmu_lock/unlock_root
> that choose between read_lock for TDP MMU and write_lock for shadow MMU.
>
> Thanks!
>
> Paolo
>

  reply	other threads:[~2021-01-27 20:09 UTC|newest]

Thread overview: 75+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-12 18:10 [PATCH 00/24] Allow parallel page faults with TDP MMU Ben Gardon
2021-01-12 18:10 ` [PATCH 01/24] locking/rwlocks: Add contention detection for rwlocks Ben Gardon
2021-01-12 21:20   ` kernel test robot
2021-01-12 18:10 ` [PATCH 02/24] sched: Add needbreak " Ben Gardon
2021-01-12 18:10 ` [PATCH 03/24] sched: Add cond_resched_rwlock Ben Gardon
2021-01-12 18:10 ` [PATCH 04/24] kvm: x86/mmu: change TDP MMU yield function returns to match cond_resched Ben Gardon
2021-01-20 18:38   ` Sean Christopherson
2021-01-21 20:22     ` Paolo Bonzini
2021-01-26 14:11     ` Paolo Bonzini
2021-01-12 18:10 ` [PATCH 05/24] kvm: x86/mmu: Fix yielding in TDP MMU Ben Gardon
2021-01-20 19:28   ` Sean Christopherson
2021-01-22  1:06     ` Ben Gardon
2021-01-12 18:10 ` [PATCH 06/24] kvm: x86/mmu: Skip no-op changes in TDP MMU functions Ben Gardon
2021-01-20 19:51   ` Sean Christopherson
2021-01-25 23:51     ` Ben Gardon
2021-01-12 18:10 ` [PATCH 07/24] kvm: x86/mmu: Add comment on __tdp_mmu_set_spte Ben Gardon
2021-01-26 14:13   ` Paolo Bonzini
2021-01-12 18:10 ` [PATCH 08/24] kvm: x86/mmu: Add lockdep when setting a TDP MMU SPTE Ben Gardon
2021-01-20 19:58   ` Sean Christopherson
2021-01-26 14:13   ` Paolo Bonzini
2021-01-12 18:10 ` [PATCH 09/24] kvm: x86/mmu: Don't redundantly clear TDP MMU pt memory Ben Gardon
2021-01-20 20:06   ` Sean Christopherson
2021-01-26 14:14   ` Paolo Bonzini
2021-01-12 18:10 ` [PATCH 10/24] kvm: x86/mmu: Factor out handle disconnected pt Ben Gardon
2021-01-20 20:30   ` Sean Christopherson
2021-01-26 14:14   ` Paolo Bonzini
2021-01-12 18:10 ` [PATCH 11/24] kvm: x86/mmu: Put TDP MMU PT walks in RCU read-critical section Ben Gardon
2021-01-20 22:19   ` Sean Christopherson
2021-01-12 18:10 ` [PATCH 12/24] kvm: x86/kvm: RCU dereference tdp mmu page table links Ben Gardon
2021-01-22 18:32   ` Sean Christopherson
2021-01-26 18:17     ` Ben Gardon
2021-01-12 18:10 ` [PATCH 13/24] kvm: x86/mmu: Only free tdp_mmu pages after a grace period Ben Gardon
2021-01-12 18:10 ` [PATCH 14/24] kvm: mmu: Wrap mmu_lock lock / unlock in a function Ben Gardon
2021-01-13  2:35   ` kernel test robot
2021-01-13  2:35     ` kernel test robot
2021-01-12 18:10 ` [PATCH 15/24] kvm: mmu: Wrap mmu_lock cond_resched and needbreak Ben Gardon
2021-01-21  0:19   ` Sean Christopherson
2021-01-21 20:17     ` Paolo Bonzini
2021-01-26 14:38     ` Paolo Bonzini
2021-01-26 17:47       ` Ben Gardon
2021-01-26 17:55         ` Paolo Bonzini
2021-01-26 18:11           ` Ben Gardon
2021-01-26 20:47             ` Paolo Bonzini
2021-01-27 20:08               ` Ben Gardon [this message]
2021-01-27 20:55                 ` Paolo Bonzini
2021-01-27 21:20                   ` Ben Gardon
2021-01-28  8:18                     ` Paolo Bonzini
2021-01-12 18:10 ` [PATCH 16/24] kvm: mmu: Wrap mmu_lock assertions Ben Gardon
2021-01-26 14:29   ` Paolo Bonzini
2021-01-12 18:10 ` [PATCH 17/24] kvm: mmu: Move mmu_lock to struct kvm_arch Ben Gardon
2021-01-12 18:10 ` [PATCH 18/24] kvm: x86/mmu: Use an rwlock for the x86 TDP MMU Ben Gardon
2021-01-21  0:45   ` Sean Christopherson
2021-01-12 18:10 ` [PATCH 19/24] kvm: x86/mmu: Protect tdp_mmu_pages with a lock Ben Gardon
2021-01-21 19:22   ` Sean Christopherson
2021-01-21 21:32     ` Sean Christopherson
2021-01-26 14:27       ` Paolo Bonzini
2021-01-26 21:47         ` Ben Gardon
2021-01-26 22:02         ` Sean Christopherson
2021-01-26 22:09           ` Sean Christopherson
2021-01-27 12:40           ` Paolo Bonzini
2021-01-26 13:37   ` Paolo Bonzini
2021-01-26 21:07     ` Ben Gardon
2021-01-12 18:10 ` [PATCH 20/24] kvm: x86/mmu: Add atomic option for setting SPTEs Ben Gardon
2021-01-13  0:05   ` kernel test robot
2021-01-13  0:05     ` kernel test robot
2021-01-26 14:21   ` Paolo Bonzini
2021-01-12 18:10 ` [PATCH 21/24] kvm: x86/mmu: Use atomic ops to set SPTEs in TDP MMU map Ben Gardon
2021-01-12 18:10 ` [PATCH 22/24] kvm: x86/mmu: Flush TLBs after zap in TDP MMU PF handler Ben Gardon
2021-01-21  0:05   ` Sean Christopherson
2021-01-12 18:10 ` [PATCH 23/24] kvm: x86/mmu: Freeze SPTEs in disconnected pages Ben Gardon
2021-01-12 18:10 ` [PATCH 24/24] kvm: x86/mmu: Allow parallel page faults for the TDP MMU Ben Gardon
2021-01-21  0:55   ` Sean Christopherson
2021-01-26 21:57     ` Ben Gardon
2021-01-27 17:14       ` Sean Christopherson
2021-01-26 13:37   ` Paolo Bonzini

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='CANgfPd9FaPhQiEkJ=VHKiVWZ_5S3k2uWHU+ViCi4nEF=GU4qsw@mail.gmail.com' \
    --to=bgardon@google.com \
    --cc=jmattson@google.com \
    --cc=junaids@google.com \
    --cc=kernellwp@gmail.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=peterx@redhat.com \
    --cc=pfeiner@google.com \
    --cc=pshier@google.com \
    --cc=seanjc@google.com \
    --cc=vkuznets@redhat.com \
    --cc=xiaoguangrong.eric@gmail.com \
    --cc=yulei.kernel@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.