From: Sean Christopherson <seanjc@google.com>
To: Yosry Ahmed <yosry@kernel.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
Jim Mattson <jmattson@google.com>,
kvm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v4 2/9] KVM: SVM: Refactor SVM instruction handling on #GP intercept
Date: Fri, 3 Apr 2026 11:18:55 -0700 [thread overview]
Message-ID: <adAEjw56f_4cuUNR@google.com> (raw)
In-Reply-To: <20260316202732.3164936-3-yosry@kernel.org>
On Mon, Mar 16, 2026, Yosry Ahmed wrote:
> +/* Return 0 if not SVM instr, otherwise return associated exit_code */
> +static u64 svm_instr_exit_code(struct kvm_vcpu *vcpu)
To make it very clear what this is doing how about:
static u64 svm_get_decoded_instr_exit_code(struct kvm_vcpu *vcpu)
> {
> struct x86_emulate_ctxt *ctxt = vcpu->arch.emulate_ctxt;
>
> if (ctxt->b != 0x1 || ctxt->opcode_len != 2)
> - return NONE_SVM_INSTR;
> + return 0;
This should assert to ensure there's no collision with '0', i.e.
BUILD_BUG_ON(!SVM_EXIT_VMRUN || !SVM_EXIT_VMLOAD || !SVM_EXIT_VMSAVE);
> + if (!is_guest_mode(vcpu))
> + return kvm_emulate_instruction(vcpu,
> + EMULTYPE_VMWARE_GP | EMULTYPE_NO_DECODE);
Since you're moving this anyways:
if (!is_guest_mode(vcpu))
return kvm_emulate_instruction(vcpu, EMULTYPE_VMWARE_GP |
EMULTYPE_NO_DECODE);
Actually! Better idea, for this code and for the page_address_valid() checks:
invert the checks to reduce indentation, i.e. end up with:
/* FIXME: Handle SVM instructions through the emulator */
svm_exit_code = svm_get_decoded_instr_exit_code(vcpu);
if (svm_exit_code) {
if (!is_guest_mode(vcpu))
return svm_invoke_exit_handler(vcpu, svm_exit_code);
if (nested_svm_check_permissions(vcpu))
return 1;
if (!page_address_valid(vcpu, kvm_register_read(vcpu, VCPU_REGS_RAX)))
goto reinject;
/*
* FIXME: Only synthesize a #VMEXIT if L1 sets the intercept,
* but only after the VMLOAD/VMSAVE exit handlers can properly
* handle VMLOAD/VMSAVE from L2 with VLS enabled in L1 (i.e.
* RAX is an L2 GPA that needs translation through L1's NPT).
*/
nested_svm_simple_vmexit(svm, svm_exit_code);
return 1;
}
/*
* VMware backdoor emulation on #GP interception only handles
* IN{S}, OUT{S}, and RDPMC, and only for L1.
*/
if (!enable_vmware_backdoor || is_guest_mode(vcpu))
goto reinject;
return kvm_emulate_instruction(vcpu, EMULTYPE_VMWARE_GP | EMULTYPE_NO_DECODE);
> +
> reinject:
> kvm_queue_exception_e(vcpu, GP_VECTOR, error_code);
> return 1;
> --
> 2.53.0.851.ga537e3e6e9-goog
>
next prev parent reply other threads:[~2026-04-03 18:18 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-16 20:27 [PATCH v4 0/9] KVM: SVM: Fixes for VMCB12 checks and mapping Yosry Ahmed
2026-03-16 20:27 ` [PATCH v4 1/9] KVM: SVM: Properly check RAX in the emulator for SVM instructions Yosry Ahmed
2026-03-16 20:56 ` Yosry Ahmed
2026-03-16 20:27 ` [PATCH v4 2/9] KVM: SVM: Refactor SVM instruction handling on #GP intercept Yosry Ahmed
2026-04-03 18:18 ` Sean Christopherson [this message]
2026-04-03 21:45 ` Yosry Ahmed
2026-03-16 20:27 ` [PATCH v4 3/9] KVM: SVM: Properly check RAX on #GP intercept of SVM instructions Yosry Ahmed
2026-04-03 17:39 ` Sean Christopherson
2026-04-03 19:00 ` Sean Christopherson
2026-04-03 21:43 ` Yosry Ahmed
2026-04-03 22:16 ` Sean Christopherson
2026-03-16 20:27 ` [PATCH v4 4/9] KVM: SVM: Move RAX legality check to SVM insn interception handlers Yosry Ahmed
2026-03-16 20:27 ` [PATCH v4 5/9] KVM: SVM: Check EFER.SVME and CPL on #GP intercept of SVM instructions Yosry Ahmed
2026-03-16 20:27 ` [PATCH v4 6/9] KVM: SVM: Treat mapping failures equally in VMLOAD/VMSAVE emulation Yosry Ahmed
2026-03-16 20:27 ` [PATCH v4 7/9] KVM: nSVM: Fail emulation of VMRUN/VMLOAD/VMSAVE if mapping vmcb12 fails Yosry Ahmed
2026-03-16 20:27 ` [PATCH v4 8/9] KVM: selftests: Rework svm_nested_invalid_vmcb12_gpa Yosry Ahmed
2026-03-16 20:27 ` [PATCH v4 9/9] KVM: selftests: Drop 'invalid' from svm_nested_invalid_vmcb12_gpa's name Yosry Ahmed
2026-04-03 19:05 ` [PATCH v4 0/9] KVM: SVM: Fixes for VMCB12 checks and mapping Sean Christopherson
2026-04-03 21:45 ` Yosry Ahmed
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=adAEjw56f_4cuUNR@google.com \
--to=seanjc@google.com \
--cc=jmattson@google.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=yosry@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