From patchwork Mon May 8 15:44:54 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mathias Krause X-Patchwork-Id: 13234674 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 726A5C77B75 for ; Mon, 8 May 2023 15:45:36 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S234563AbjEHPpe (ORCPT ); Mon, 8 May 2023 11:45:34 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:55430 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233932AbjEHPpY (ORCPT ); Mon, 8 May 2023 11:45:24 -0400 Received: from mail-ej1-x62d.google.com (mail-ej1-x62d.google.com [IPv6:2a00:1450:4864:20::62d]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 20DAC9003 for ; Mon, 8 May 2023 08:45:14 -0700 (PDT) Received: by mail-ej1-x62d.google.com with SMTP id a640c23a62f3a-965e93f915aso667683266b.2 for ; Mon, 08 May 2023 08:45:14 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=grsecurity.net; s=grsec; t=1683560712; x=1686152712; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=zv5jxo4A4M09xK+CgdnY7TNGrFXWGeW2rvoeJckh5pQ=; b=O0OUEazRF08OvMNX4AUXuCLff5KJlzxTgl8NTWFPa/f/5mNevEUlbvovv89G1iChlV vczdRlan42b9SoIBSM48ydK91aQaDPzp9JBnykSbaflRDlplXaK8endjzKwRr965kuwI z3gGtS11yan8s2spcpqvqwSxv2bxmx2xFxxEYL6J1Lf77FgQbfDQQFRqibtz6PXkKlrw /nZdcg1MdkPDYGoS2OOp2IrV6L9wl4fRzUOrL4TRHeYBrEf51y6hAEZ+VKPTFtesyXKk itq3gQIIGVwvCFjBTrodlvnQt5E2KBunVXnkhbafzW9mJMQn3pcQyViRUCSFMxnhz/7D sETA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1683560712; x=1686152712; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=zv5jxo4A4M09xK+CgdnY7TNGrFXWGeW2rvoeJckh5pQ=; b=Dak6UNwtKCSq/RG+vomSkk6GTZzR1S0uKymteEne62QBQEDhIUvPEPUEfZxjPbwGpz JS/eWStdqXSd7l4hcaGiCDheiytoukakN8PNQpWeXOvR/RfJOY83gDULvZ7N2bX1lGDl rsrfnBANDG9028XRxja8ngKkEh6fakml7H2ZnQDvUBiWtzOhwK3wurC3QwKcoCePqK61 Z/QB9531CwXc44rjYw0b+8tIzKF6/pNqFwhw+9oBSHawfnaxEG0dJ9Njg2MGtIpu8Spe OLbIWu8SXqKaamHdD+Xrjs3+rj9mJd/HdfL5hOoD3w81dCkseRsetiPAV1SS40UqrD+k FZag== X-Gm-Message-State: AC+VfDx4c9o75Yq9YBopYuYy97vrEssoAiqQMGtz12jIYJHLrGCFn1j/ hToq7AGs+F6Gd3sE5lT0Mwoou6t8mZ7sscuYG5FSRw== X-Google-Smtp-Source: ACHHUZ4fDZtM3R1LfVlXHrpCv6tS1WN5fBuNPEcR8H9Es36SKvCWWCHDqKQ+17j4+nQe+D8YWM7KKw== X-Received: by 2002:a17:907:3e1b:b0:967:3963:dab8 with SMTP id hp27-20020a1709073e1b00b009673963dab8mr2687816ejc.7.1683560712549; Mon, 08 May 2023 08:45:12 -0700 (PDT) Received: from localhost.localdomain (p549211c7.dip0.t-ipconnect.de. [84.146.17.199]) by smtp.gmail.com with ESMTPSA id kw3-20020a170907770300b0096621c999c6sm121758ejc.79.2023.05.08.08.45.11 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 08 May 2023 08:45:11 -0700 (PDT) From: Mathias Krause To: stable@vger.kernel.org Cc: Paolo Bonzini , Sean Christopherson , kvm@vger.kernel.org, Mathias Krause Subject: [PATCH 6.2 2/5] KVM: x86: Do not unload MMU roots when only toggling CR0.WP with TDP enabled Date: Mon, 8 May 2023 17:44:54 +0200 Message-Id: <20230508154457.29956-3-minipli@grsecurity.net> X-Mailer: git-send-email 2.39.2 In-Reply-To: <20230508154457.29956-1-minipli@grsecurity.net> References: <20230508154457.29956-1-minipli@grsecurity.net> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org [ Upstream commit 01b31714bd90be2784f7145bf93b7f78f3d081e1 ] There is no need to unload the MMU roots with TDP enabled when only CR0.WP has changed -- the paging structures are still valid, only the permission bitmap needs to be updated. One heavy user of toggling CR0.WP is grsecurity's KERNEXEC feature to implement kernel W^X. The optimization brings a huge performance gain for this case as the following micro-benchmark running 'ssdd 10 50000' from rt-tests[1] on a grsecurity L1 VM shows (runtime in seconds, lower is better): legacy TDP shadow kvm-x86/next@d8708b 8.43s 9.45s 70.3s +patch 5.39s 5.63s 70.2s For legacy MMU this is ~36% faster, for TDP MMU even ~40% faster. Also TDP and legacy MMU now both have a similar runtime which vanishes the need to disable TDP MMU for grsecurity. Shadow MMU sees no measurable difference and is still slow, as expected. [1] https://git.kernel.org/pub/scm/utils/rt-tests/rt-tests.git Signed-off-by: Mathias Krause Link: https://lore.kernel.org/r/20230322013731.102955-3-minipli@grsecurity.net Co-developed-by: Sean Christopherson Signed-off-by: Sean Christopherson Signed-off-by: Mathias Krause --- arch/x86/kvm/x86.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 0ec7a0cb5da8..d8d679ee5327 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -904,6 +904,18 @@ EXPORT_SYMBOL_GPL(load_pdptrs); void kvm_post_set_cr0(struct kvm_vcpu *vcpu, unsigned long old_cr0, unsigned long cr0) { + /* + * CR0.WP is incorporated into the MMU role, but only for non-nested, + * indirect shadow MMUs. If TDP is enabled, the MMU's metadata needs + * to be updated, e.g. so that emulating guest translations does the + * right thing, but there's no need to unload the root as CR0.WP + * doesn't affect SPTEs. + */ + if (tdp_enabled && (cr0 ^ old_cr0) == X86_CR0_WP) { + kvm_init_mmu(vcpu); + return; + } + if ((cr0 ^ old_cr0) & X86_CR0_PG) { kvm_clear_async_pf_completion_queue(vcpu); kvm_async_pf_hash_reset(vcpu);