From: Gleb Natapov <gleb@redhat.com>
To: Arthur Chunqi Li <yzt356@gmail.com>
Cc: kvm@vger.kernel.org, pbonzini@redhat.com, jan.kiszka@web.de
Subject: Re: [PATCH v3] KVM: nVMX: Set segment infomation of L1 when L2 exits
Date: Mon, 15 Jul 2013 10:09:24 +0300 [thread overview]
Message-ID: <20130715070924.GI11772@redhat.com> (raw)
In-Reply-To: <1373870041-15154-1-git-send-email-yzt356@gmail.com>
On Mon, Jul 15, 2013 at 02:34:01PM +0800, Arthur Chunqi Li wrote:
> When L2 exits to L1, segment infomations of L1 are not set correctly.
> According to Intel SDM 27.5.2(Loading Host Segment and Descriptor
> Table Registers), segment base/limit/access right of L1 should be
> set to some designed value when L2 exits to L1. This patch fixes
> this.
>
> Signed-off-by: Arthur Chunqi Li <yzt356@gmail.com>
> ---
> arch/x86/kvm/vmx.c | 58 +++++++++++++++++++++++++++++++++++++++++++---------
> 1 file changed, 48 insertions(+), 10 deletions(-)
>
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index 064d0be..0ea9614 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -7948,6 +7948,8 @@ static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
> static void load_vmcs12_host_state(struct kvm_vcpu *vcpu,
> struct vmcs12 *vmcs12)
> {
> + struct kvm_segment seg_cs, seg_ds, seg_tr;
> +
You do not need three of them. Just use one and reinit whatever is
different.
> if (vmcs12->vm_exit_controls & VM_EXIT_LOAD_IA32_EFER)
> vcpu->arch.efer = vmcs12->host_ia32_efer;
> else if (vmcs12->vm_exit_controls & VM_EXIT_HOST_ADDR_SPACE_SIZE)
> @@ -8001,16 +8003,6 @@ static void load_vmcs12_host_state(struct kvm_vcpu *vcpu,
> vmcs_writel(GUEST_SYSENTER_EIP, vmcs12->host_ia32_sysenter_eip);
> vmcs_writel(GUEST_IDTR_BASE, vmcs12->host_idtr_base);
> vmcs_writel(GUEST_GDTR_BASE, vmcs12->host_gdtr_base);
> - vmcs_writel(GUEST_TR_BASE, vmcs12->host_tr_base);
> - vmcs_writel(GUEST_GS_BASE, vmcs12->host_gs_base);
> - vmcs_writel(GUEST_FS_BASE, vmcs12->host_fs_base);
> - vmcs_write16(GUEST_ES_SELECTOR, vmcs12->host_es_selector);
> - vmcs_write16(GUEST_CS_SELECTOR, vmcs12->host_cs_selector);
> - vmcs_write16(GUEST_SS_SELECTOR, vmcs12->host_ss_selector);
> - vmcs_write16(GUEST_DS_SELECTOR, vmcs12->host_ds_selector);
> - vmcs_write16(GUEST_FS_SELECTOR, vmcs12->host_fs_selector);
> - vmcs_write16(GUEST_GS_SELECTOR, vmcs12->host_gs_selector);
> - vmcs_write16(GUEST_TR_SELECTOR, vmcs12->host_tr_selector);
>
> if (vmcs12->vm_exit_controls & VM_EXIT_LOAD_IA32_PAT)
> vmcs_write64(GUEST_IA32_PAT, vmcs12->host_ia32_pat);
> @@ -8018,6 +8010,52 @@ static void load_vmcs12_host_state(struct kvm_vcpu *vcpu,
> vmcs_write64(GUEST_IA32_PERF_GLOBAL_CTRL,
> vmcs12->host_ia32_perf_global_ctrl);
>
> + /* Set L1 segment info according to Intel SDM
> + 27.5.2 Loading Host Segment and Descriptor-Table Registers */
> + memset(&seg_cs, 0, sizeof(struct kvm_segment));
> + memset(&seg_ds, 0, sizeof(struct kvm_segment));
> + memset(&seg_tr, 0, sizeof(struct kvm_segment));
> + seg_cs.base = 0;
> + seg_cs.limit = 0xFFFFFFFF;
> + seg_cs.selector = vmcs12->host_cs_selector;
> + seg_cs.type = AR_TYPE_CODE_MASK | AR_TYPE_READABLE_MASK
> + | AR_TYPE_ACCESSES_MASK;
Those defines are for AR filed of vmcs, not for type field of segment
register, they only accidentally work here, but I don't like to mix
things like that. SDM says what the numerical value the field should
have so just use that, it will be easy to see that code does what spec
says.
> + seg_cs.present = 1;
> + if (vmcs12->vm_exit_controls & VM_EXIT_HOST_ADDR_SPACE_SIZE)
> + seg_cs.l = 1;
> + else
> + seg_cs.db = 1;
> + seg_cs.s = 1;
> + seg_cs.g = 1;
> + vmx_set_segment(vcpu, &seg_cs, VCPU_SREG_CS);
> + seg_ds.base = 0;
> + seg_ds.limit = 0xFFFFFFFF;
> + seg_ds.type = AR_TYPE_READABLE_MASK | AR_TYPE_ACCESSES_MASK;
> + seg_ds.present = 1;
> + seg_ds.s = 1;
> + seg_ds.db = 1;
> + seg_ds.g = 1;
> + seg_ds.selector = vmcs12->host_ds_selector;
> + vmx_set_segment(vcpu, &seg_ds, VCPU_SREG_DS);
> + seg_ds.selector = vmcs12->host_es_selector;
> + vmx_set_segment(vcpu, &seg_ds, VCPU_SREG_ES);
> + seg_ds.selector = vmcs12->host_ss_selector;
> + vmx_set_segment(vcpu, &seg_ds, VCPU_SREG_SS);
> + seg_ds.selector = vmcs12->host_fs_selector;
> + seg_ds.base = vmcs12->host_fs_base;
> + vmx_set_segment(vcpu, &seg_ds, VCPU_SREG_FS);
> + seg_ds.selector = vmcs12->host_gs_selector;
> + seg_ds.base = vmcs12->host_gs_base;
> + vmx_set_segment(vcpu, &seg_ds, VCPU_SREG_GS);
> + seg_tr.base = vmcs12->host_tr_base;
> + seg_tr.limit = 0x67;
> + seg_tr.selector = vmcs12->host_tr_selector;
> + seg_tr.type = AR_TYPE_CODE_MASK | AR_TYPE_READABLE_MASK
> + | AR_TYPE_ACCESSES_MASK;
> + seg_tr.present = 1;
> + vmx_set_segment(vcpu, &seg_tr, VCPU_SREG_TR);
> +
> +
> kvm_set_dr(vcpu, 7, 0x400);
> vmcs_write64(GUEST_IA32_DEBUGCTL, 0);
> }
> --
> 1.7.9.5
--
Gleb.
prev parent reply other threads:[~2013-07-15 7:09 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-15 6:34 [PATCH v3] KVM: nVMX: Set segment infomation of L1 when L2 exits Arthur Chunqi Li
2013-07-15 7:09 ` Gleb Natapov [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=20130715070924.GI11772@redhat.com \
--to=gleb@redhat.com \
--cc=jan.kiszka@web.de \
--cc=kvm@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=yzt356@gmail.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox