From: Avi Kivity <avi@redhat.com>
To: Joerg Roedel <joerg.roedel@amd.com>
Cc: Alexander Graf <agraf@suse.de>,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 08/12] kvm/svm: cache nested intercepts
Date: Wed, 29 Jul 2009 16:13:35 +0300 [thread overview]
Message-ID: <4A704AFF.8070702@redhat.com> (raw)
In-Reply-To: <1248872192-30881-9-git-send-email-joerg.roedel@amd.com>
On 07/29/2009 03:56 PM, Joerg Roedel wrote:
> Signed-off-by: Joerg Roedel<joerg.roedel@amd.com>
> ---
> arch/x86/kvm/svm.c | 30 +++++++++++++++++++++++-------
> 1 files changed, 23 insertions(+), 7 deletions(-)
>
> diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
> index 31467b1..9192c9a 100644
> --- a/arch/x86/kvm/svm.c
> +++ b/arch/x86/kvm/svm.c
> @@ -86,6 +86,15 @@ struct nested_state {
>
> /* gpa pointers to the real vectors */
> u64 vmcb_msrpm;
> +
> + /* cache for intercepts of the guest */
> + u16 intercept_cr_read;
> + u16 intercept_cr_write;
> + u16 intercept_dr_read;
> + u16 intercept_dr_write;
> + u32 intercept_exceptions;
> + u64 intercept;
> +
> };
>
> struct vcpu_svm {
> @@ -1459,7 +1468,6 @@ static int nested_svm_exit_handled_real(struct vcpu_svm *svm,
> void *arg2,
> void *opaque)
> {
> - struct vmcb *nested_vmcb = (struct vmcb *)arg1;
> bool kvm_overrides = *(bool *)opaque;
> u32 exit_code = svm->vmcb->control.exit_code;
>
> @@ -1486,38 +1494,38 @@ static int nested_svm_exit_handled_real(struct vcpu_svm *svm,
> switch (exit_code) {
> case SVM_EXIT_READ_CR0 ... SVM_EXIT_READ_CR8: {
> u32 cr_bits = 1<< (exit_code - SVM_EXIT_READ_CR0);
> - if (nested_vmcb->control.intercept_cr_read& cr_bits)
> + if (svm->nested.intercept_cr_read& cr_bits)
> return 1;
> break;
> }
> case SVM_EXIT_WRITE_CR0 ... SVM_EXIT_WRITE_CR8: {
> u32 cr_bits = 1<< (exit_code - SVM_EXIT_WRITE_CR0);
> - if (nested_vmcb->control.intercept_cr_write& cr_bits)
> + if (svm->nested.intercept_cr_write& cr_bits)
> return 1;
> break;
> }
> case SVM_EXIT_READ_DR0 ... SVM_EXIT_READ_DR7: {
> u32 dr_bits = 1<< (exit_code - SVM_EXIT_READ_DR0);
> - if (nested_vmcb->control.intercept_dr_read& dr_bits)
> + if (svm->nested.intercept_dr_read& dr_bits)
> return 1;
> break;
> }
> case SVM_EXIT_WRITE_DR0 ... SVM_EXIT_WRITE_DR7: {
> u32 dr_bits = 1<< (exit_code - SVM_EXIT_WRITE_DR0);
> - if (nested_vmcb->control.intercept_dr_write& dr_bits)
> + if (svm->nested.intercept_dr_write& dr_bits)
> return 1;
> break;
> }
> case SVM_EXIT_EXCP_BASE ... SVM_EXIT_EXCP_BASE + 0x1f: {
> u32 excp_bits = 1<< (exit_code - SVM_EXIT_EXCP_BASE);
> - if (nested_vmcb->control.intercept_exceptions& excp_bits)
> + if (svm->nested.intercept_exceptions& excp_bits)
> return 1;
> break;
> }
> default: {
> u64 exit_bits = 1ULL<< (exit_code - SVM_EXIT_INTR);
> nsvm_printk("exit code: 0x%x\n", exit_code);
> - if (nested_vmcb->control.intercept& exit_bits)
> + if (svm->nested.intercept& exit_bits)
> return 1;
> }
> }
> @@ -1808,6 +1816,14 @@ static int nested_svm_vmrun(struct vcpu_svm *svm, void *arg1,
>
> svm->nested.vmcb_msrpm = nested_vmcb->control.msrpm_base_pa;
>
> + /* cache intercepts */
> + svm->nested.intercept_cr_read = nested_vmcb->control.intercept_cr_read;
> + svm->nested.intercept_cr_write = nested_vmcb->control.intercept_cr_write;
> + svm->nested.intercept_dr_read = nested_vmcb->control.intercept_dr_read;
> + svm->nested.intercept_dr_write = nested_vmcb->control.intercept_dr_write;
> + svm->nested.intercept_exceptions = nested_vmcb->control.intercept_exceptions;
> + svm->nested.intercept = nested_vmcb->control.intercept;
> +
> force_new_asid(&svm->vcpu);
> svm->vmcb->control.exit_int_info = nested_vmcb->control.exit_int_info;
> svm->vmcb->control.exit_int_info_err = nested_vmcb->control.exit_int_info_err;
>
I don't see the benefit of this patch. Accessing the cache is just as
expensive as accessing the real vmcb.
--
error compiling committee.c: too many arguments to function
next prev parent reply other threads:[~2009-07-29 13:08 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-07-29 12:56 [PATCH 0/12] Nested SVM cleanups Joerg Roedel
2009-07-29 12:56 ` [PATCH 01/12] kvm/svm: make nested svm debugging runtime configurable Joerg Roedel
2009-07-29 13:22 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 02/12] kvm/svm: add helper functions for global interrupt flag Joerg Roedel
2009-07-29 13:15 ` Alexander Graf
2009-07-29 13:44 ` Joerg Roedel
2009-07-29 12:56 ` [PATCH 03/12] kvm/svm: optimize nested #vmexit Joerg Roedel
2009-07-29 13:23 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 04/12] kvm/svm: optimize nested vmrun Joerg Roedel
2009-07-29 13:27 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 05/12] kvm/svm: copy only necessary parts of the control area on vmrun/vmexit Joerg Roedel
2009-07-29 13:30 ` Alexander Graf
2009-07-29 13:47 ` Joerg Roedel
2009-07-29 13:48 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 06/12] kvm/svm: complete interrupts after handling nested exits Joerg Roedel
2009-07-29 13:46 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 07/12] kvm/svm: move nested svm state into seperate struct Joerg Roedel
2009-07-29 13:49 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 08/12] kvm/svm: cache nested intercepts Joerg Roedel
2009-07-29 13:13 ` Avi Kivity [this message]
2009-07-29 13:13 ` Joerg Roedel
2009-07-29 13:26 ` Avi Kivity
2009-07-29 13:47 ` Joerg Roedel
2009-07-29 13:50 ` Alexander Graf
2009-07-29 13:52 ` Joerg Roedel
2009-07-29 12:56 ` [PATCH 09/12] kvm/svm: consolidate nested_svm_exit_handled Joerg Roedel
2009-07-29 13:56 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 10/12] kvm/svm: do nested vmexit in nested_svm_exit_handled Joerg Roedel
2009-07-29 14:15 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 11/12] kvm/svm: handle #pf intercepts in nested_svm_exit_handled directly Joerg Roedel
2009-07-29 14:20 ` Alexander Graf
2009-07-29 14:44 ` Avi Kivity
2009-07-29 14:41 ` Alexander Graf
2009-07-29 14:49 ` Avi Kivity
2009-07-29 14:46 ` Alexander Graf
2009-07-29 12:56 ` [PATCH 12/12] kvm/svm: simplify nested_svm_check_exception Joerg Roedel
2009-07-29 14:21 ` Alexander Graf
2009-07-29 13:05 ` [PATCH 0/12] Nested SVM cleanups Avi Kivity
2009-07-29 13:09 ` Joerg Roedel
2009-07-29 13:31 ` Avi Kivity
2009-07-29 13:32 ` Avi Kivity
2009-07-29 13:29 ` Alexander Graf
2009-07-29 13:38 ` Joerg Roedel
2009-07-29 13:48 ` Avi Kivity
2009-07-29 13:48 ` Joerg Roedel
2009-07-29 13:59 ` Avi Kivity
2009-07-29 13:58 ` Joerg Roedel
2009-07-29 14:07 ` Avi Kivity
2009-07-29 13:06 ` Avi Kivity
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=4A704AFF.8070702@redhat.com \
--to=avi@redhat.com \
--cc=agraf@suse.de \
--cc=joerg.roedel@amd.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@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;
as well as URLs for NNTP newsgroup(s).