From patchwork Thu Oct 4 11:56:01 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Paul Mackerras X-Patchwork-Id: 10625997 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 4DAE515A6 for ; Thu, 4 Oct 2018 11:56:38 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 3EFDE28E47 for ; Thu, 4 Oct 2018 11:56:38 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 3338B28E58; Thu, 4 Oct 2018 11:56:38 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-8.0 required=2.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,MAILING_LIST_MULTI,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id B2EEE28E75 for ; Thu, 4 Oct 2018 11:56:37 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727646AbeJDStc (ORCPT ); Thu, 4 Oct 2018 14:49:32 -0400 Received: from ozlabs.org ([203.11.71.1]:57839 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727627AbeJDStb (ORCPT ); Thu, 4 Oct 2018 14:49:31 -0400 Received: from authenticated.ozlabs.org (localhost [127.0.0.1]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ozlabs.org (Postfix) with ESMTPSA id 42Qrvj0sTWzB4NF; Thu, 4 Oct 2018 21:56:33 +1000 (AEST) Authentication-Results: ozlabs.org; dmarc=none (p=none dis=none) header.from=ozlabs.org DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=ozlabs.org; s=201707; t=1538654193; bh=yHS904xXThXhIel/gAmPE0yHe35LIiWoYponRErAQOQ=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=kR9q+X6i6Bf3njtyu+XSPeLSTZ3sxTtde2U2gtY9ZNxia6va8LVGFYR5O9vSSXjh4 ch4Xg5hADgzvy0rfHpMCVCoiR0MwXjmRhll4JsCBV1Wn4HbSvIUz3hIEafwpvPBe46 3NunjfXuLiFiuNzSGPrlCTbg/4NdUFyd0VJxbQYkthXfB8mjZBoRRjmOWB70DLn73H 8NBIe35goxPhaTP90QsiCDJIk/qobOBzmreigE3iAynTPvRROUwqXaKE7ISbNeWduO VmeCIDU7ynU1OH0v51078oMYl7qCWKwS2ezEbldCv7ASpolz2cextCm87yC0oWQ7aG memWKOg46F+kQ== From: Paul Mackerras To: kvm@vger.kernel.org, kvm-ppc@vger.kernel.org Cc: linuxppc-dev@ozlabs.org, David Gibson Subject: [PATCH v4 24/32] KVM: PPC: Book3S HV: Use hypercalls for TLB invalidation when nested Date: Thu, 4 Oct 2018 21:56:01 +1000 Message-Id: <1538654169-15602-25-git-send-email-paulus@ozlabs.org> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1538654169-15602-1-git-send-email-paulus@ozlabs.org> References: <1538654169-15602-1-git-send-email-paulus@ozlabs.org> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP This adds code to call the H_TLB_INVALIDATE hypercall when running as a guest, in the cases where we need to invalidate TLBs (or other MMU caches) as part of managing the mappings for a nested guest. Calling H_TLB_INVALIDATE lets the nested hypervisor inform the parent hypervisor about changes to partition-scoped page tables or the partition table without needing to do hypervisor-privileged tlbie instructions. Signed-off-by: Paul Mackerras Reviewed-by: David Gibson --- arch/powerpc/include/asm/kvm_book3s_64.h | 5 +++++ arch/powerpc/kvm/book3s_64_mmu_radix.c | 30 ++++++++++++++++++++++++++++-- arch/powerpc/kvm/book3s_hv_nested.c | 30 ++++++++++++++++++++++++------ 3 files changed, 57 insertions(+), 8 deletions(-) diff --git a/arch/powerpc/include/asm/kvm_book3s_64.h b/arch/powerpc/include/asm/kvm_book3s_64.h index a02f0b3..aa5bf85 100644 --- a/arch/powerpc/include/asm/kvm_book3s_64.h +++ b/arch/powerpc/include/asm/kvm_book3s_64.h @@ -24,6 +24,7 @@ #include #include #include +#include #ifdef CONFIG_PPC_PSERIES static inline bool kvmhv_on_pseries(void) @@ -117,6 +118,10 @@ struct kvm_nested_guest *kvmhv_get_nested(struct kvm *kvm, int l1_lpid, bool create); void kvmhv_put_nested(struct kvm_nested_guest *gp); +/* Encoding of first parameter for H_TLB_INVALIDATE */ +#define H_TLBIE_P1_ENC(ric, prs, r) (___PPC_RIC(ric) | ___PPC_PRS(prs) | \ + ___PPC_R(r)) + /* Power architecture requires HPT is at least 256kiB, at most 64TiB */ #define PPC_MIN_HPT_ORDER 18 #define PPC_MAX_HPT_ORDER 46 diff --git a/arch/powerpc/kvm/book3s_64_mmu_radix.c b/arch/powerpc/kvm/book3s_64_mmu_radix.c index 4c1eccb..ae0e3ed 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_radix.c +++ b/arch/powerpc/kvm/book3s_64_mmu_radix.c @@ -201,17 +201,43 @@ static void kvmppc_radix_tlbie_page(struct kvm *kvm, unsigned long addr, unsigned int pshift, unsigned int lpid) { unsigned long psize = PAGE_SIZE; + int psi; + long rc; + unsigned long rb; if (pshift) psize = 1UL << pshift; + else + pshift = PAGE_SHIFT; addr &= ~(psize - 1); - radix__flush_tlb_lpid_page(lpid, addr, psize); + + if (!kvmhv_on_pseries()) { + radix__flush_tlb_lpid_page(lpid, addr, psize); + return; + } + + psi = shift_to_mmu_psize(pshift); + rb = addr | (mmu_get_ap(psi) << PPC_BITLSHIFT(58)); + rc = plpar_hcall_norets(H_TLB_INVALIDATE, H_TLBIE_P1_ENC(0, 0, 1), + lpid, rb); + if (rc) + pr_err("KVM: TLB page invalidation hcall failed, rc=%ld\n", rc); } static void kvmppc_radix_flush_pwc(struct kvm *kvm, unsigned int lpid) { - radix__flush_pwc_lpid(lpid); + long rc; + + if (!kvmhv_on_pseries()) { + radix__flush_pwc_lpid(lpid); + return; + } + + rc = plpar_hcall_norets(H_TLB_INVALIDATE, H_TLBIE_P1_ENC(1, 0, 1), + lpid, TLBIEL_INVAL_SET_LPID); + if (rc) + pr_err("KVM: TLB PWC invalidation hcall failed, rc=%ld\n", rc); } static unsigned long kvmppc_radix_update_pte(struct kvm *kvm, pte_t *ptep, diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c index 26151e8..35f8111 100644 --- a/arch/powerpc/kvm/book3s_hv_nested.c +++ b/arch/powerpc/kvm/book3s_hv_nested.c @@ -298,14 +298,32 @@ void kvmhv_nested_exit(void) } } +static void kvmhv_flush_lpid(unsigned int lpid) +{ + long rc; + + if (!kvmhv_on_pseries()) { + radix__flush_tlb_lpid(lpid); + return; + } + + rc = plpar_hcall_norets(H_TLB_INVALIDATE, H_TLBIE_P1_ENC(2, 0, 1), + lpid, TLBIEL_INVAL_SET_LPID); + if (rc) + pr_err("KVM: TLB LPID invalidation hcall failed, rc=%ld\n", rc); +} + void kvmhv_set_ptbl_entry(unsigned int lpid, u64 dw0, u64 dw1) { - if (cpu_has_feature(CPU_FTR_HVMODE)) { + if (!kvmhv_on_pseries()) { mmu_partition_table_set_entry(lpid, dw0, dw1); - } else { - pseries_partition_tb[lpid].patb0 = cpu_to_be64(dw0); - pseries_partition_tb[lpid].patb1 = cpu_to_be64(dw1); + return; } + + pseries_partition_tb[lpid].patb0 = cpu_to_be64(dw0); + pseries_partition_tb[lpid].patb1 = cpu_to_be64(dw1); + /* L0 will do the necessary barriers */ + kvmhv_flush_lpid(lpid); } static void kvmhv_set_nested_ptbl(struct kvm_nested_guest *gp) @@ -482,7 +500,7 @@ static void kvmhv_flush_nested(struct kvm_nested_guest *gp) spin_lock(&kvm->mmu_lock); kvmppc_free_pgtable_radix(kvm, gp->shadow_pgtable, gp->shadow_lpid); spin_unlock(&kvm->mmu_lock); - radix__flush_tlb_lpid(gp->shadow_lpid); + kvmhv_flush_lpid(gp->shadow_lpid); kvmhv_update_ptbl_cache(gp); if (gp->l1_gr_to_hr == 0) kvmhv_remove_nested(gp); @@ -766,7 +784,7 @@ static void kvmhv_emulate_tlbie_lpid(struct kvm_vcpu *vcpu, spin_lock(&kvm->mmu_lock); kvmppc_free_pgtable_radix(kvm, gp->shadow_pgtable, gp->shadow_lpid); - radix__flush_tlb_lpid(gp->shadow_lpid); + kvmhv_flush_lpid(gp->shadow_lpid); spin_unlock(&kvm->mmu_lock); break; case 1: