Historical speck list archives
 help / color / mirror / Atom feed

From: Thomas Gleixner <tglx@linutronix.de>
To: speck@linutronix.de
Subject: Re: [PATCH] L1TF KVM ARCH_CAPABILITIES #2
Date: Mon, 16 Jul 2018 22:31:31 +0200 (CEST)	[thread overview]
Message-ID: <alpine.DEB.2.21.1807162225060.1693@nanos.tec.linutronix.de> (raw)
In-Reply-To: <alpine.DEB.2.21.1807162202180.1693@nanos.tec.linutronix.de>

On Mon, 16 Jul 2018, speck for Thomas Gleixner wrote:
> On Sun, 15 Jul 2018, speck for Paolo Bonzini wrote:
> 
> > From: Paolo Bonzini <pbonzini@redhat.com>
> > Subject: [PATCH 2/4] x86: use ARCH_CAPABILITIES to skip L1D flush on vmentry
> > 
> > Bit 3 of ARCH_CAPABILITIES tells a hypervisor that L1D flush on vmentry is
> > not needed.  In that case, KVM can behave as if there is no L1TF bug at all.
> 
> Are you sure? It just says that you can skip the L1D flush, but it does not
> say that this is protected completely also against all the SMP issues

The magic PDF says:

   A nested VMM that finds IA32_FLUSH_CMD is enumerated should check whether
   IA32_ARCH_CAPABILITIES bit 3 9 (SKIP_L1DFL_VMENTRY) is set, which indicates
   that it is not required to flush L1D on VMENTER.
   
   First-level VMMs which perform an L1D flush before VMENTER may set
   SKIP_L1DFL_VMENTRY in the IA32_ARCH_CAPABILITIES value exposed to guests.
   These VMMs should set SKIP_L1DFL_VMENTRY in any case where a nested VMM may
   be present.

I'm not seing how that prevents the SMP crap. And also the information in
the nested VMM should be:

    ..... VMX ... L1D Flush: Host VMM

or such. The mitigation really depends on what the Host VMM does and just
claiming !vulnerable might be horribly misleading.

Thanks,

	tglx

next prev parent reply	other threads:[~2018-07-16 20:31 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-15 13:57 [MODERATED] [PATCH] L1TF KVM ARCH_CAPABILITIES #0 Paolo Bonzini
2018-07-15 13:57 ` [MODERATED] [PATCH] L1TF KVM ARCH_CAPABILITIES #1 Paolo Bonzini
2018-07-16 14:36   ` [MODERATED] " Konrad Rzeszutek Wilk
2018-07-16 20:02   ` Thomas Gleixner
2018-07-17 11:20     ` [MODERATED] " Paolo Bonzini
2018-07-17 11:28       ` Thomas Gleixner
2018-07-17 17:11         ` [MODERATED] " Paolo Bonzini
2018-07-17 19:23           ` Thomas Gleixner
2018-07-15 13:57 ` [MODERATED] [PATCH] L1TF KVM ARCH_CAPABILITIES #2 Paolo Bonzini
2018-07-16 20:04   ` Thomas Gleixner
2018-07-16 20:31     ` Thomas Gleixner [this message]
2018-07-16 20:41       ` [MODERATED] " Luck, Tony
2018-07-16 21:13         ` Thomas Gleixner
2018-07-15 13:57 ` [MODERATED] [PATCH] L1TF KVM ARCH_CAPABILITIES #3 Paolo Bonzini
2018-07-15 13:57 ` [MODERATED] [PATCH] L1TF KVM ARCH_CAPABILITIES #4 Paolo Bonzini
2018-07-16 14:58   ` [MODERATED] " Konrad Rzeszutek Wilk
2018-07-16 20:06   ` Thomas Gleixner
2018-07-17 11:14     ` [MODERATED] " Paolo Bonzini
2018-07-16 20:44 ` [MODERATED] Re: [PATCH] L1TF KVM ARCH_CAPABILITIES #0 Konrad Rzeszutek Wilk
2018-07-17 11:22   ` 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=alpine.DEB.2.21.1807162225060.1693@nanos.tec.linutronix.de \
    --to=tglx@linutronix.de \
    --cc=speck@linutronix.de \
    /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

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 read-only IMAP folder(s) and NNTP newsgroup(s).