public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoffer Dall <christoffer.dall@linaro.org>
To: Marc Zyngier <marc.zyngier@arm.com>
Cc: kvm@vger.kernel.org, kvmarm@lists.cs.columbia.edu,
	Steve Capper <steve.capper@linaro.org>
Subject: Re: [PATCH 4/4] arm/arm64: KVM: use kernel mapping to perform invalidation on page fault
Date: Fri, 9 Jan 2015 13:51:13 +0100	[thread overview]
Message-ID: <20150109125113.GT21092@cbox> (raw)
In-Reply-To: <1420718349-24152-5-git-send-email-marc.zyngier@arm.com>

On Thu, Jan 08, 2015 at 11:59:09AM +0000, Marc Zyngier wrote:
> When handling a fault in stage-2, we need to resync I$ and D$, just
> to be sure we don't leave any old cache line behind.
> 
> That's very good, except that we do so using the *user* address.
> Under heavy load (swapping like crazy), we may end up in a situation
> where the page gets mapped in stage-2 while being unmapped from
> userspace by another CPU.
> 
> At that point, the DC/IC instructions can generate a fault, which
> we handle with kvm->mmu_lock held. The box quickly deadlocks, user
> is unhappy.
> 
> Instead, perform this invalidation through the kernel mapping,
> which is guaranteed to be present. The box is much happier, and so
> am I.
> 
> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>

This looks good to me!

Thanks,
-Christoffer

      parent reply	other threads:[~2015-01-09 12:50 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-08 11:59 [PATCH 0/4] arm/arm64: KVM: Random selection of MM related fixes Marc Zyngier
2015-01-08 11:59 ` [PATCH 1/4] mm: Correct ordering of *_clear_flush_young_notify Marc Zyngier
2015-01-08 13:12   ` Paolo Bonzini
2015-01-08 19:00   ` Andrea Arcangeli
2015-01-12 10:15     ` Steve Capper
2015-01-08 11:59 ` [PATCH 2/4] arm/arm64: KVM: Use set/way op trapping to track the state of the caches Marc Zyngier
2015-01-09 11:19   ` Christoffer Dall
2015-01-09 11:38     ` Marc Zyngier
2015-01-09 12:12       ` Christoffer Dall
2015-01-08 11:59 ` [PATCH 3/4] arm/arm64: KVM: Flush caches to memory on unmap Marc Zyngier
2015-01-09 12:30   ` Christoffer Dall
2015-01-09 14:35     ` Marc Zyngier
2015-01-11 12:30       ` Christoffer Dall
2015-01-12 11:15         ` Marc Zyngier
2015-01-12 20:13           ` Christoffer Dall
2015-01-13 13:47             ` Christoffer Dall
2015-01-13 13:57               ` Marc Zyngier
2015-01-08 11:59 ` [PATCH 4/4] arm/arm64: KVM: use kernel mapping to perform invalidation on page fault Marc Zyngier
2015-01-08 12:30   ` Peter Maydell
2015-01-08 13:07     ` Marc Zyngier
2015-01-08 13:16       ` Peter Maydell
2015-01-08 15:06         ` Marc Zyngier
2015-01-08 15:21           ` Peter Maydell
2015-01-09 12:50             ` Christoffer Dall
2015-01-09 13:03               ` Peter Maydell
2015-01-09 14:16                 ` Marc Zyngier
2015-01-09 15:28                   ` Peter Maydell
2015-01-09 17:18                     ` Marc Zyngier
2015-01-11 12:33                     ` Christoffer Dall
2015-01-11 17:37                       ` Peter Maydell
2015-01-11 17:58                         ` Christoffer Dall
2015-01-11 18:27                           ` Peter Maydell
2015-01-11 18:38                             ` Christoffer Dall
2015-01-12  9:58                               ` Marc Zyngier
2015-01-12 20:10                                 ` Christoffer Dall
2015-01-13 11:38                                   ` Marc Zyngier
2015-01-13 12:04                                     ` Christoffer Dall
2015-01-13 12:12                                       ` Peter Maydell
2015-01-13 13:35                                         ` Christoffer Dall
2015-01-13 13:41                                           ` Peter Maydell
2015-01-13 13:49                                             ` Christoffer Dall
2015-01-15 12:00                                           ` Mark Rutland
2015-01-15 13:00                                             ` Christoffer Dall
2015-01-15 15:47                                               ` Mark Rutland
2015-01-09 12:51   ` Christoffer Dall [this message]

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=20150109125113.GT21092@cbox \
    --to=christoffer.dall@linaro.org \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.cs.columbia.edu \
    --cc=marc.zyngier@arm.com \
    --cc=steve.capper@linaro.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