From: Avi Kivity <avi@redhat.com>
To: Gleb Natapov <gleb@redhat.com>
Cc: Francis Moreau <francis.moro@gmail.com>, kvm@vger.kernel.org
Subject: Re: 2.6.38-rc6: general protection error inside KVM 64 bits guest
Date: Sun, 06 Mar 2011 14:55:17 +0200 [thread overview]
Message-ID: <4D738435.1030906@redhat.com> (raw)
In-Reply-To: <20110306124238.GC25565@redhat.com>
On 03/06/2011 02:42 PM, Gleb Natapov wrote:
> >
> IO permission checking for 64-bit guest in KVM is wrong. The patch bellow
> should fix it.
>
>
> diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h
> index 50ebc32..7ef5b86 100644
> --- a/arch/x86/include/asm/kvm_emulate.h
> +++ b/arch/x86/include/asm/kvm_emulate.h
> @@ -142,10 +142,8 @@ struct x86_emulate_ops {
> int (*pio_out_emulated)(int size, unsigned short port, const void *val,
> unsigned int count, struct kvm_vcpu *vcpu);
>
> - bool (*get_cached_descriptor)(struct desc_struct *desc,
> - int seg, struct kvm_vcpu *vcpu);
> - void (*set_cached_descriptor)(struct desc_struct *desc,
> - int seg, struct kvm_vcpu *vcpu);
> + bool (*get_cached_descriptor)(void *p, int seg, struct kvm_vcpu *vcpu);
> + void (*set_cached_descriptor)(void *p, int seg, struct kvm_vcpu *vcpu);
Ugh, void *.
Add a u64 *highword parameter, or something.
> @@ -1764,25 +1764,35 @@ static bool emulator_io_port_access_allowed(struct x86_emulate_ctxt *ctxt,
> struct x86_emulate_ops *ops,
> u16 port, u16 len)
> {
> - struct desc_struct tr_seg;
> + union {
> + struct desc_struct tss32;
> +#ifdef CONFIG_X86_64
> + struct ldttss_desc64 tss64;
> +#endif
> + } tr_seg;
> int r;
> u16 io_bitmap_ptr;
> u8 perm, bit_idx = port& 0x7;
> unsigned mask = (1<< len) - 1;
> + unsigned long base;
>
> ops->get_cached_descriptor(&tr_seg, VCPU_SREG_TR, ctxt->vcpu);
> - if (!tr_seg.p)
> + if (!tr_seg.tss32.p)
> return false;
> - if (desc_limit_scaled(&tr_seg)< 103)
> + if (desc_limit_scaled(&tr_seg.tss32)< 103)
> return false;
> - r = ops->read_std(get_desc_base(&tr_seg) + 102,&io_bitmap_ptr, 2,
> - ctxt->vcpu, NULL);
> + base = get_desc_base(&tr_seg.tss32);
> +#ifdef CONFIG_X86_64
> + if (ctxt->mode == X86EMUL_MODE_PROT64)
> + base |= ((u64)tr_seg.tss64.base3)<< 32;
> +#endif
> + r = ops->read_std(base + 102,&io_bitmap_ptr, 2, ctxt->vcpu, NULL);
> if (r != X86EMUL_CONTINUE)
> return false;
Note: if we get a fault here, we ought to propagate it. Only happens if
there's a race, since the cpu checks for these exceptions.
--
error compiling committee.c: too many arguments to function
next prev parent reply other threads:[~2011-03-06 12:55 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-03-05 9:40 2.6.38-rc6: general protection error inside KVM 64 bits guest Francis Moreau
2011-03-06 12:42 ` Gleb Natapov
2011-03-06 12:55 ` Avi Kivity [this message]
2011-03-06 14:35 ` Gleb Natapov
2011-03-06 15:02 ` Avi Kivity
2011-03-06 15:08 ` Avi Kivity
2011-03-09 7:05 ` Francis Moreau
2011-03-09 9:25 ` Gleb Natapov
2011-03-09 9:26 ` Avi Kivity
2011-03-09 9:28 ` Francis Moreau
2011-03-09 9:29 ` Avi Kivity
2011-03-09 9:30 ` Francis Moreau
2011-03-09 9:32 ` Gleb Natapov
2011-03-09 10:03 ` Avi Kivity
2011-03-09 10:07 ` Francis Moreau
2011-05-27 10:22 ` Francis Moreau
2011-05-27 12:48 ` Gleb Natapov
2011-05-27 14:30 ` Francis Moreau
2011-03-06 19:36 ` Francis Moreau
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=4D738435.1030906@redhat.com \
--to=avi@redhat.com \
--cc=francis.moro@gmail.com \
--cc=gleb@redhat.com \
--cc=kvm@vger.kernel.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