All of lore.kernel.org
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Nadav Amit <namit@cs.technion.ac.il>
Cc: kvm@vger.kernel.org, nadav.amit@gmail.com
Subject: Re: [PATCH 07/21] KVM: x86: Emulator considers imm as memory operand
Date: Wed, 05 Nov 2014 12:36:52 +0100	[thread overview]
Message-ID: <545A0BD4.3070300@redhat.com> (raw)
In-Reply-To: <1414922101-17626-8-git-send-email-namit@cs.technion.ac.il>

On 02/11/2014 10:54, Nadav Amit wrote:
> The emulator mistakenly considers some of the immediate operands as memory
> operands, performs memory read and uses the wrong data.  By default, every
> operand is marked as OP_MEM, so if it is not changed, memory read may be
> wrongly emulated and the wrong value would be used.  Consider for instance the
> ROR instruction - src2 (the number of times) would be read from memory instead
> of being used as immediate.
> 
> Mark every immediate operand as such to avoid this problem.
> 
> Signed-off-by: Nadav Amit <namit@cs.technion.ac.il>
> ---
>  arch/x86/kvm/emulate.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
> index f456783..e624d62 100644
> --- a/arch/x86/kvm/emulate.c
> +++ b/arch/x86/kvm/emulate.c
> @@ -4269,6 +4269,7 @@ static int decode_operand(struct x86_emulate_ctxt *ctxt, struct operand *op,
>  		fetch_register_operand(op);
>  		break;
>  	case OpCL:
> +		op->type = OP_IMM;
>  		op->bytes = 1;
>  		op->val = reg_read(ctxt, VCPU_REGS_RCX) & 0xff;
>  		break;
> @@ -4276,6 +4277,7 @@ static int decode_operand(struct x86_emulate_ctxt *ctxt, struct operand *op,
>  		rc = decode_imm(ctxt, op, 1, true);
>  		break;
>  	case OpOne:
> +		op->type = OP_IMM;
>  		op->bytes = 1;
>  		op->val = 1;
>  		break;
> @@ -4334,21 +4336,27 @@ static int decode_operand(struct x86_emulate_ctxt *ctxt, struct operand *op,
>  		ctxt->memop.bytes = ctxt->op_bytes + 2;
>  		goto mem_common;
>  	case OpES:
> +		op->type = OP_IMM;
>  		op->val = VCPU_SREG_ES;
>  		break;
>  	case OpCS:
> +		op->type = OP_IMM;
>  		op->val = VCPU_SREG_CS;
>  		break;
>  	case OpSS:
> +		op->type = OP_IMM;
>  		op->val = VCPU_SREG_SS;
>  		break;
>  	case OpDS:
> +		op->type = OP_IMM;
>  		op->val = VCPU_SREG_DS;
>  		break;
>  	case OpFS:
> +		op->type = OP_IMM;
>  		op->val = VCPU_SREG_FS;
>  		break;
>  	case OpGS:
> +		op->type = OP_IMM;
>  		op->val = VCPU_SREG_GS;
>  		break;
>  	case OpImplicit:
> 

I'm including this for stable@ because, until commit c44b4c6ab80e (KVM:
emulate: clean up initializations in init_decode_cache, 2014-04-16),
this would be harmless.  The type would be set to OP_REG instead of
OP_IMM, but it would not matter because no writeback is done for Src2
operands.

Now, the uninitialized op->type may be an OP_MEM from a previous
instruction.

Paolo

  reply	other threads:[~2014-11-05 11:37 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-02  9:54 [PATCH 00/21] Fixes for various KVM bugs Nadav Amit
2014-11-02  9:54 ` [PATCH 01/21] KVM: x86: decode_modrm does not regard modrm correctly Nadav Amit
2014-11-05 11:14   ` Paolo Bonzini
2014-11-02  9:54 ` [PATCH 02/21] KVM: x86: No error-code on real-mode exceptions Nadav Amit
2014-11-02  9:54 ` [PATCH 03/21] KVM: x86: Emulator should set DR6 upon GD like real CPU Nadav Amit
2014-11-02  9:54 ` [PATCH 04/21] KVM: x86: Clear DR6[0:3] on #DB during handle_dr Nadav Amit
2014-11-02  9:54 ` [PATCH 05/21] KVM: x86: Breakpoints do not consider CS.base Nadav Amit
2014-11-02  9:54 ` [PATCH 06/21] KVM: x86: Emulator MOV-sreg uses incorrect size Nadav Amit
2014-11-05 11:28   ` Paolo Bonzini
2014-11-02  9:54 ` [PATCH 07/21] KVM: x86: Emulator considers imm as memory operand Nadav Amit
2014-11-05 11:36   ` Paolo Bonzini [this message]
2014-11-02  9:54 ` [PATCH 08/21] KVM: x86: Reset FPU state during reset Nadav Amit
2014-11-05 12:04   ` Paolo Bonzini
2014-11-05 13:20     ` Nadav Amit
2014-11-05 14:55       ` Paolo Bonzini
2014-11-05 20:31         ` Nadav Amit
2014-11-06  8:58           ` Paolo Bonzini
2014-11-06  9:13             ` Nadav Amit
2014-11-06  9:44               ` Paolo Bonzini
2014-11-06  9:56                 ` Nadav Amit
2014-11-06 10:44                   ` Paolo Bonzini
2014-11-06 17:38                 ` Radim Krčmář
2014-11-02  9:54 ` [PATCH 09/21] KVM: x86: SYSCALL cannot clear eflags[1] Nadav Amit
2014-11-02  9:54 ` [PATCH 10/21] KVM: x86: Wrong flags on CMPS and SCAS emulation Nadav Amit
2014-11-02  9:54 ` [PATCH 11/21] KVM: x86: Emulate push sreg as done in Core Nadav Amit
2014-11-02  9:54 ` [PATCH 12/21] KVM: x86: MOV to CR3 can set bit 63 Nadav Amit
2015-02-10 16:15   ` Jan Kiszka
2015-02-10 16:18     ` Paolo Bonzini
2015-02-10 16:34       ` Jan Kiszka
2015-02-10 16:42         ` Paolo Bonzini
2014-11-02  9:54 ` [PATCH 13/21] KVM: x86: Do not update EFLAGS on faulting emulation Nadav Amit
2014-11-02  9:54 ` [PATCH 14/21] KVM: x86: Software disabled APIC should still deliver NMIs Nadav Amit
2014-11-05 12:30   ` Paolo Bonzini
2014-11-05 20:45     ` Nadav Amit
2014-11-06  9:34       ` Paolo Bonzini
2014-11-06 16:45         ` Radim Krčmář
2014-11-10 17:35           ` Paolo Bonzini
2014-11-10 18:06             ` Radim Krčmář
2014-11-14 15:00           ` Paolo Bonzini
2014-11-26 17:01             ` Nadav Amit
2014-11-26 18:00               ` Paolo Bonzini
2014-11-27 13:39               ` Radim Krčmář
2014-11-27 21:45                 ` Nadav Amit
2014-11-27 22:26                   ` Radim Krčmář
2014-12-01 16:30                     ` Paolo Bonzini
2014-12-01 17:49                       ` Radim Krčmář
2014-11-02  9:54 ` [PATCH 15/21] KVM: x86: Combine the lgdt and lidt emulation logic Nadav Amit
2014-11-02  9:54 ` [PATCH 16/21] KVM: x86: Inject #GP when loading system segments with non-canonical base Nadav Amit
2014-11-02  9:54 ` [PATCH 17/21] KVM: x86: Remove redundant and incorrect cpl check on task-switch Nadav Amit
2014-11-02  9:54 ` [PATCH 18/21] KVM: x86: Emulator mis-decodes VEX instructions on real-mode Nadav Amit
2014-11-08  7:25   ` Paolo Bonzini
2014-11-02  9:54 ` [PATCH 19/21] KVM: x86: Warn on APIC base relocation Nadav Amit
2014-11-02  9:55 ` [PATCH 20/21] KVM: x86: MOVNTI emulation min opsize is not respected Nadav Amit
2014-11-05 12:18   ` Paolo Bonzini
2014-11-05 19:58     ` Nadav Amit
2014-11-05 19:58     ` Nadav Amit
2014-11-06  9:23   ` Paolo Bonzini
2014-11-02  9:55 ` [PATCH 21/21] KVM: x86: Return UNHANDLABLE on unsupported SYSENTER Nadav Amit
2014-11-05 12:31 ` [PATCH 00/21] Fixes for various KVM bugs 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=545A0BD4.3070300@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=nadav.amit@gmail.com \
    --cc=namit@cs.technion.ac.il \
    /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.