From patchwork Thu Sep 15 00:04:44 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vishal Annapurve X-Patchwork-Id: 12976694 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 04B82C6FA82 for ; Thu, 15 Sep 2022 00:05:17 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229967AbiIOAFO (ORCPT ); Wed, 14 Sep 2022 20:05:14 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:57704 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229890AbiIOAFH (ORCPT ); Wed, 14 Sep 2022 20:05:07 -0400 Received: from mail-pj1-x1049.google.com (mail-pj1-x1049.google.com [IPv6:2607:f8b0:4864:20::1049]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id E222813E90 for ; Wed, 14 Sep 2022 17:05:01 -0700 (PDT) Received: by mail-pj1-x1049.google.com with SMTP id a24-20020a17090a8c1800b0020266349974so12812885pjo.8 for ; Wed, 14 Sep 2022 17:05:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date; bh=rhI1CHpHnoLTF9sMnY1velsatE7KzqpaHm2Zg/NlpWU=; b=csQ+wHtL/4fYR6KUnDudrTc0f7rZcbdEOOHzeTVhZIgqP16I2UHMCa4nKJvwkDmIa1 04EiDNsGGYsa/ZoenTCo4pyTo5zKHaBi91efe9OaAlNaRgO+snXtratc/rVMkxsvUMAp 31VMO3yfzfh6Cm4pENS/+J5g+8gNzyc0ln9enq47YOyqB9UBPrN61cTtiZFrII4Zm9dK uveTLe0u+t73xwiUy8LdyI3z8CSL2y+7uHVWryjbrvAmMzs+ImrIt3aSe3KOP9StVKdb rd0EjF6Ab8sNMS2PV4Ir+kXJLtvty5YMcWkjdEe9vCJ0JX+Qgw9g7PeYKlpbOg1nAy/X 0K0w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date; bh=rhI1CHpHnoLTF9sMnY1velsatE7KzqpaHm2Zg/NlpWU=; b=kRhmD0K2cfRUUlITj2VuIDp/Q2OqDXCj44yhyPjfh1Qu5bQ+xan/Y6bzEpa49qa3rw TL/cl2jyHorxPi8k4ID9w6deJoZOE7zGcnSUTmW9yN4hitqWr5YUcwGEv/W5InWSvhBn hS5IXWSjxCS3icCXH7RsF40SBo1BEvPHIqdi1GyT8BfffSJfvHtdVaYgzlUI0rrTx832 pkyM5SDrR3LjQztq+SQmXqnVmOChCiCqXjZMCdoOUoNhIVXT1c25exuCxA2bebs+ILsR XJercOG5vooUdu+5LB8bBOTNbJ6+r2h4qKwZtMgafN3zq4XMv43D3ePaKOxVBpW2F3EF +thg== X-Gm-Message-State: ACgBeo33AO1tlFQ+fyuEmFr2N65gsUPSsxcsrg0V+yMR7+7VPeRz8itI 5IqeIVMehbbGlmwz5ysO311kxzdhsD3NqXIk X-Google-Smtp-Source: AA6agR4J+7XJTDvy3acXZ7Xp5xXcYKJyhmTX7ZLEn97ViM70m/5htdtIvuFapGYGrwQDbuzUWFwXekmcH4/ukG+F X-Received: from vannapurve2.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:41f8]) (user=vannapurve job=sendgmr) by 2002:a62:170b:0:b0:53b:93dc:966b with SMTP id 11-20020a62170b000000b0053b93dc966bmr40331680pfx.29.1663200301099; Wed, 14 Sep 2022 17:05:01 -0700 (PDT) Date: Thu, 15 Sep 2022 00:04:44 +0000 In-Reply-To: <20220915000448.1674802-1-vannapurve@google.com> Mime-Version: 1.0 References: <20220915000448.1674802-1-vannapurve@google.com> X-Mailer: git-send-email 2.37.2.789.g6183377224-goog Message-ID: <20220915000448.1674802-5-vannapurve@google.com> Subject: [V2 PATCH 4/8] KVM: selftests: x86: Precompute the result for is_{intel,amd}_cpu() From: Vishal Annapurve To: x86@kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org Cc: pbonzini@redhat.com, shuah@kernel.org, bgardon@google.com, seanjc@google.com, oupton@google.com, peterx@redhat.com, vkuznets@redhat.com, dmatlack@google.com, Vishal Annapurve Precedence: bulk List-ID: X-Mailing-List: linux-kselftest@vger.kernel.org Cache the vendor CPU type in a global variable so that multiple calls to is_intel_cpu() do not need to re-execute CPUID. Add cpu vendor check in kvm_hypercall() so that it executes correct vmcall/vmmcall instruction when running on Intel/AMD hosts. This avoids exit to KVM which anyway tries to patch the instruction according to the cpu type. As part of this change, sync the global variable is_cpu_amd into the guest so the guest can determine which hypercall instruction to use without needing to re-execute CPUID for every hypercall. Suggested-by: Sean Christopherson Signed-off-by: Vishal Annapurve --- .../testing/selftests/kvm/lib/x86_64/processor.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/kvm/lib/x86_64/processor.c b/tools/testing/selftests/kvm/lib/x86_64/processor.c index 25ae972f5c71..c0ae938772f6 100644 --- a/tools/testing/selftests/kvm/lib/x86_64/processor.c +++ b/tools/testing/selftests/kvm/lib/x86_64/processor.c @@ -19,6 +19,7 @@ #define MAX_NR_CPUID_ENTRIES 100 vm_vaddr_t exception_handlers; +static bool is_cpu_amd; static void regs_dump(FILE *stream, struct kvm_regs *regs, uint8_t indent) { @@ -1019,7 +1020,7 @@ static bool cpu_vendor_string_is(const char *vendor) bool is_intel_cpu(void) { - return cpu_vendor_string_is("GenuineIntel"); + return (is_cpu_amd == false); } /* @@ -1027,7 +1028,7 @@ bool is_intel_cpu(void) */ bool is_amd_cpu(void) { - return cpu_vendor_string_is("AuthenticAMD"); + return (is_cpu_amd == true); } void kvm_get_cpu_address_width(unsigned int *pa_bits, unsigned int *va_bits) @@ -1182,9 +1183,15 @@ uint64_t kvm_hypercall(uint64_t nr, uint64_t a0, uint64_t a1, uint64_t a2, { uint64_t r; - asm volatile("vmcall" + if (is_amd_cpu()) + asm volatile("vmmcall" : "=a"(r) : "a"(nr), "b"(a0), "c"(a1), "d"(a2), "S"(a3)); + else + asm volatile("vmcall" + : "=a"(r) + : "a"(nr), "b"(a0), "c"(a1), "d"(a2), "S"(a3)); + return r; } @@ -1314,8 +1321,10 @@ bool vm_is_unrestricted_guest(struct kvm_vm *vm) void kvm_selftest_arch_init(void) { + is_cpu_amd = cpu_vendor_string_is("AuthenticAMD"); } void kvm_arch_post_vm_elf_load(struct kvm_vm *vm) { + sync_global_to_guest(vm, is_cpu_amd); }