All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Nadav Har'El" <nyh@math.technion.ac.il>
To: Avi Kivity <avi@redhat.com>
Cc: kvm@vger.kernel.org, Sheng Yang <sheng@linux.intel.com>
Subject: Re: [PATCH 18/24] Exiting from L2 to L1
Date: Sun, 12 Sep 2010 21:51:17 +0200	[thread overview]
Message-ID: <20100912195117.GA11002@fermat.math.technion.ac.il> (raw)
In-Reply-To: <4C8D0C19.4050003@redhat.com>

On Sun, Sep 12, 2010, Avi Kivity wrote about "Re: [PATCH 18/24] Exiting from L2 to L1":
> >>>>>+	vmcs12->vm_entry_intr_info_field =
> >>>>>+		vmcs_read32(VM_ENTRY_INTR_INFO_FIELD);
>...
> prepare_vmcs12() is called in response for a 2->1 vmexit which is first 
> trapped by 0, yes?  Because it's called immediately after a vmexit, 
> VM_ENTRY_INTR_INFO_FIELD is guaranteed to have been cleared by the 
> processor.

Indeed - it is cleared in the real vmcs, i.e., vmcs02, but we also need
to clear it in the emulated vmcs that L1 sees for L2, i.e., vmcs12.

The original code (quoted above) just copied the vmcs02 value to vmcs12,
which was (mostly) fine because the vmcs02 value has a correctly-cleared
bit - but you asked to avoid the vmread.

So the second option is to just explicitly remove the valid bit from
vmcs12->vm_entry_intro_info_field, which I do now.
vmcs12->vm_entry_intro_info_field was set by L1 before it entered L2, and now
that L2 is exiting back to L1, we need to clear the valid bit.

The more I think about it, the more I become convinced that the second
option is indeed better than the first option (the original code in the
patch).

> There are two cases where VM_ENTRY_INTR_INFO_FIELD can potentially not 
> be cleared by hardware:
>...
> If neither of these are valid, the code can be removed.  If only the 
> second, we might make it conditional.

Again, unless I'm misunderstanding what you mean, the hardware only
modified vmcs02 (the hardware vmcs), not vmcs12. We need to modify vmcs12
as well, to remove the "valid" bit. If we don't, when L1 enters into the same
L2 again, the same old value will be copied again from vmcs12 to vmcs02,
and cause an injection of the same interrupt again.

And by the way, I haven't said this enough, but thanks for your continued
reviews and all your very useful corrections for these patches!

Nadav.

-- 
Nadav Har'El                        |       Sunday, Sep 12 2010, 5 Tishri 5771
nyh@math.technion.ac.il             |-----------------------------------------
Phone +972-523-790466, ICQ 13349191 |I before E except after C. We live in a
http://nadav.harel.org.il           |weird society!

  reply	other threads:[~2010-09-12 19:51 UTC|newest]

Thread overview: 147+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-06-13 12:22 [PATCH 0/24] Nested VMX, v5 Nadav Har'El
2010-06-13 12:23 ` [PATCH 1/24] Move nested option from svm.c to x86.c Nadav Har'El
2010-06-14  8:11   ` Avi Kivity
2010-06-15 14:27     ` Nadav Har'El
2010-06-13 12:23 ` [PATCH 2/24] Add VMX and SVM to list of supported cpuid features Nadav Har'El
2010-06-14  8:13   ` Avi Kivity
2010-06-15 14:31     ` Nadav Har'El
2010-06-13 12:24 ` [PATCH 3/24] Implement VMXON and VMXOFF Nadav Har'El
2010-06-14  8:21   ` Avi Kivity
2010-06-16 11:14     ` Nadav Har'El
2010-06-16 11:26       ` Avi Kivity
2010-06-15 20:18   ` Marcelo Tosatti
2010-06-16  7:50     ` Nadav Har'El
2010-06-13 12:24 ` [PATCH 4/24] Allow setting the VMXE bit in CR4 Nadav Har'El
2010-06-15 11:09   ` Gleb Natapov
2010-06-15 14:44     ` Nadav Har'El
2010-06-13 12:25 ` [PATCH 5/24] Introduce vmcs12: a VMCS structure for L1 Nadav Har'El
2010-06-14  8:33   ` Avi Kivity
2010-06-14  8:49     ` Nadav Har'El
2010-06-14 12:35       ` Avi Kivity
2010-06-16 12:24     ` Nadav Har'El
2010-06-16 13:10       ` Avi Kivity
2010-06-22 14:54     ` Nadav Har'El
2010-06-22 16:53       ` Nadav Har'El
2010-06-23  8:07         ` Avi Kivity
2010-08-08 15:09           ` Nadav Har'El
2010-08-10  3:24             ` Avi Kivity
2010-06-23  7:57       ` Avi Kivity
2010-06-23  9:15         ` Alexander Graf
2010-06-23  9:24           ` Avi Kivity
2010-06-23 12:07         ` Nadav Har'El
2010-06-23 12:13           ` Avi Kivity
2010-06-13 12:25 ` [PATCH 6/24] Implement reading and writing of VMX MSRs Nadav Har'El
2010-06-14  8:42   ` Avi Kivity
2010-06-23  8:13     ` Nadav Har'El
2010-06-23  8:24       ` Avi Kivity
2010-06-13 12:26 ` [PATCH 7/24] Understanding guest pointers to vmcs12 structures Nadav Har'El
2010-06-14  8:48   ` Avi Kivity
2010-08-02 12:25     ` Nadav Har'El
2010-08-02 13:38       ` Avi Kivity
2010-06-15 12:14   ` Gleb Natapov
2010-08-01 15:16     ` Nadav Har'El
2010-08-01 15:25       ` Gleb Natapov
2010-08-02  8:57         ` Nadav Har'El
2010-06-13 12:26 ` [PATCH 8/24] Hold a vmcs02 for each vmcs12 Nadav Har'El
2010-06-14  8:57   ` Avi Kivity
2010-07-06  9:50   ` Dong, Eddie
2010-08-02 13:38     ` Nadav Har'El
2010-06-13 12:27 ` [PATCH 9/24] Implement VMCLEAR Nadav Har'El
2010-06-14  9:03   ` Avi Kivity
2010-06-15 13:47   ` Gleb Natapov
2010-06-15 13:50     ` Avi Kivity
2010-06-15 13:54       ` Gleb Natapov
2010-08-05 11:50         ` Nadav Har'El
2010-08-05 11:53           ` Gleb Natapov
2010-08-05 12:01             ` Nadav Har'El
2010-08-05 12:05               ` Avi Kivity
2010-08-05 12:10                 ` Nadav Har'El
2010-08-05 12:13                   ` Avi Kivity
2010-08-05 12:29                     ` Nadav Har'El
2010-08-05 12:03           ` Avi Kivity
2010-07-06  2:56   ` Dong, Eddie
2010-08-03 12:12     ` Nadav Har'El
2010-06-13 12:27 ` [PATCH 10/24] Implement VMPTRLD Nadav Har'El
2010-06-14  9:07   ` Avi Kivity
2010-08-05 11:13     ` Nadav Har'El
2010-06-16 13:36   ` Gleb Natapov
2010-07-06  3:09   ` Dong, Eddie
2010-08-05 11:35     ` Nadav Har'El
2010-06-13 12:28 ` [PATCH 11/24] Implement VMPTRST Nadav Har'El
2010-06-14  9:15   ` Avi Kivity
2010-06-16 13:53     ` Gleb Natapov
2010-06-16 15:33       ` Nadav Har'El
2010-06-13 12:28 ` [PATCH 12/24] Add VMCS fields to the vmcs12 Nadav Har'El
2010-06-14  9:24   ` Avi Kivity
2010-06-16 14:18   ` Gleb Natapov
2010-06-13 12:29 ` [PATCH 13/24] Implement VMREAD and VMWRITE Nadav Har'El
2010-06-14  9:36   ` Avi Kivity
2010-06-16 14:48     ` Gleb Natapov
2010-08-04 13:42       ` Nadav Har'El
2010-08-04 16:09     ` Nadav Har'El
2010-08-04 16:41       ` Avi Kivity
2010-06-16 15:03   ` Gleb Natapov
2010-08-04 11:46     ` Nadav Har'El
2010-06-13 12:29 ` [PATCH 14/24] Prepare vmcs02 from vmcs01 and vmcs12 Nadav Har'El
2010-06-14 11:11   ` Avi Kivity
2010-06-17  8:50   ` Gleb Natapov
2010-07-06  6:25   ` Dong, Eddie
2010-06-13 12:30 ` [PATCH 15/24] Move register-syncing to a function Nadav Har'El
2010-06-13 12:30 ` [PATCH 16/24] Implement VMLAUNCH and VMRESUME Nadav Har'El
2010-06-14 11:41   ` Avi Kivity
2010-09-26 11:14     ` Nadav Har'El
2010-09-26 12:56       ` Avi Kivity
2010-09-26 13:06         ` Nadav Har'El
2010-09-26 13:51           ` Avi Kivity
2010-06-17 10:59   ` Gleb Natapov
2010-09-16 16:06     ` Nadav Har'El
2010-06-13 12:31 ` [PATCH 17/24] No need for handle_vmx_insn function any more Nadav Har'El
2010-06-13 12:31 ` [PATCH 18/24] Exiting from L2 to L1 Nadav Har'El
2010-06-14 12:04   ` Avi Kivity
2010-09-12 14:05     ` Nadav Har'El
2010-09-12 14:29       ` Avi Kivity
2010-09-12 17:05         ` Nadav Har'El
2010-09-12 17:21           ` Avi Kivity
2010-09-12 19:51             ` Nadav Har'El [this message]
2010-09-13  8:48               ` Avi Kivity
2010-09-13  5:53             ` Sheng Yang
2010-09-13  8:52               ` Avi Kivity
2010-09-13  9:01                 ` Nadav Har'El
2010-09-13  9:34                   ` Avi Kivity
2010-09-14 13:07     ` Nadav Har'El
2010-06-13 12:32 ` [PATCH 19/24] Deciding if L0 or L1 should handle an L2 exit Nadav Har'El
2010-06-14 12:24   ` Avi Kivity
2010-09-16 14:42     ` Nadav Har'El
2010-06-13 12:32 ` [PATCH 20/24] Correct handling of interrupt injection Nadav Har'El
2010-06-14 12:29   ` Avi Kivity
2010-06-14 12:48     ` Avi Kivity
2010-09-16 15:25     ` Nadav Har'El
2010-06-13 12:33 ` [PATCH 21/24] Correct handling of exception injection Nadav Har'El
2010-06-13 12:33 ` [PATCH 22/24] Correct handling of idt vectoring info Nadav Har'El
2010-06-17 11:58   ` Gleb Natapov
2010-09-20  6:37     ` Nadav Har'El
2010-09-20  9:34       ` Gleb Natapov
2010-09-20 10:03         ` Nadav Har'El
2010-09-20 10:11           ` Avi Kivity
2010-09-22 23:15             ` Nadav Har'El
2010-09-26 15:14               ` Avi Kivity
2010-09-26 15:18                 ` Gleb Natapov
2010-09-20 10:20           ` Gleb Natapov
2010-06-13 12:34 ` [PATCH 23/24] Handling of CR0.TS and #NM for Lazy FPU loading Nadav Har'El
2010-06-13 12:34 ` [PATCH 24/24] Miscellenous small corrections Nadav Har'El
2010-06-14 12:34 ` [PATCH 0/24] Nested VMX, v5 Avi Kivity
2010-06-14 13:03   ` Nadav Har'El
2010-06-15 10:00     ` Avi Kivity
2010-10-17 12:03       ` Nadav Har'El
2010-10-17 12:10         ` Avi Kivity
2010-10-17 12:39           ` Nadav Har'El
2010-10-17 13:35             ` Avi Kivity
2010-07-09  8:59 ` Dong, Eddie
2010-07-11  8:27   ` Nadav Har'El
2010-07-11 11:05     ` Alexander Graf
2010-07-11 12:49       ` Nadav Har'El
2010-07-11 13:12         ` Avi Kivity
2010-07-11 15:39           ` Nadav Har'El
2010-07-11 15:45             ` Avi Kivity
2010-07-11 13:20     ` Avi Kivity
2010-07-15  3:27 ` Sheng Yang

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=20100912195117.GA11002@fermat.math.technion.ac.il \
    --to=nyh@math.technion.ac.il \
    --cc=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=sheng@linux.intel.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.