From nobody Thu Sep 19 01:23:29 2024 Delivered-To: importer@patchew.org Received-SPF: pass (zohomail.com: domain of lists.xenproject.org designates 192.237.175.120 as permitted sender) client-ip=192.237.175.120; envelope-from=xen-devel-bounces@lists.xenproject.org; helo=lists.xenproject.org; Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of lists.xenproject.org designates 192.237.175.120 as permitted sender) smtp.mailfrom=xen-devel-bounces@lists.xenproject.org; dmarc=pass(p=reject dis=none) header.from=cloud.com ARC-Seal: i=1; a=rsa-sha256; t=1720540370; cv=none; d=zohomail.com; s=zohoarc; b=lA+qlTIXiOmFY95/0/0OTLaJKiZN6rFmufhgBnlqzrkrQcWqdt3TgpUr7Q4xwp2zIeJrinzYLInmLk+32HRT4DcN2iWlG3y+CkMaYykuQv9ZXSgT+Fzs5vGVpPu+8c4cmdYXlEhytXRx7gQwuISetsWvAO1uv0S25v38wq/tfhA= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1720540370; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:Subject:To:To:Message-Id:Reply-To; bh=hNkmBJzEXcz2KBPKDeUXYjkt3AnZqwx9CjUlb8rUPSk=; b=VXm+7nf94oUUYom542HJ4WQkLaaRxslVPWjKfRTY3klUUs5IpmawsUrof4/3k171sU4YONNfpMvqp73P9dmhj17Dlxr/kWJQJxhFmhIg/riUDHq3Zo9P1WuK7Afr1ZU6AP339dAvntI2frTmDVcS2r44kgcPfFA0QeCNERJwD+U= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of lists.xenproject.org designates 192.237.175.120 as permitted sender) smtp.mailfrom=xen-devel-bounces@lists.xenproject.org; dmarc=pass header.from= (p=reject dis=none) Return-Path: Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) by mx.zohomail.com with SMTPS id 1720540370224220.96044163123418; Tue, 9 Jul 2024 08:52:50 -0700 (PDT) Received: from list by lists.xenproject.org with outflank-mailman.756381.1165010 (Exim 4.92) (envelope-from ) id 1sRD8t-000401-Cz; Tue, 09 Jul 2024 15:52:35 +0000 Received: by outflank-mailman (output) from mailman id 756381.1165010; Tue, 09 Jul 2024 15:52:35 +0000 Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1sRD8t-0003zl-89; Tue, 09 Jul 2024 15:52:35 +0000 Received: by outflank-mailman (input) for mailman id 756381; Tue, 09 Jul 2024 15:52:33 +0000 Received: from se1-gles-flk1-in.inumbo.com ([94.247.172.50] helo=se1-gles-flk1.inumbo.com) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1sRD8r-0002zY-Bh for xen-devel@lists.xenproject.org; Tue, 09 Jul 2024 15:52:33 +0000 Received: from mail-ed1-x530.google.com (mail-ed1-x530.google.com [2a00:1450:4864:20::530]) by se1-gles-flk1.inumbo.com (Halon) with ESMTPS id 403eb1a2-3e0b-11ef-8776-851b0ebba9a2; Tue, 09 Jul 2024 17:52:31 +0200 (CEST) Received: by mail-ed1-x530.google.com with SMTP id 4fb4d7f45d1cf-58ef19aa69dso5770798a12.3 for ; Tue, 09 Jul 2024 08:52:31 -0700 (PDT) Received: from EMEAENGAAD19049.citrite.net ([185.25.67.249]) by smtp.gmail.com with ESMTPSA id 4fb4d7f45d1cf-594bd459e1esm1181254a12.73.2024.07.09.08.52.30 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 09 Jul 2024 08:52:30 -0700 (PDT) X-Outflank-Mailman: Message body and most headers restored to incoming version X-BeenThere: xen-devel@lists.xenproject.org List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Errors-To: xen-devel-bounces@lists.xenproject.org Precedence: list Sender: "Xen-devel" X-Inumbo-ID: 403eb1a2-3e0b-11ef-8776-851b0ebba9a2 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cloud.com; s=cloud; t=1720540351; x=1721145151; darn=lists.xenproject.org; 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=hNkmBJzEXcz2KBPKDeUXYjkt3AnZqwx9CjUlb8rUPSk=; b=cNRC02mPsgCxTMSIftyD7WmM3tRiO/JjJvtuazvKd+V89U64qUKalgXJxPWcsKW12v MqAejNNS2uVTsPlEJ+zPhDont9Xadayg+U644LD82fsykXMFbplBqdBiJYC/ZRW0ROJR yZUhk3HOUbWuBSEzT5tohI8s5hXNkT/iN8Uj8= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1720540351; x=1721145151; 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=hNkmBJzEXcz2KBPKDeUXYjkt3AnZqwx9CjUlb8rUPSk=; b=Wm42QZQKYS/98zEVpTUxAMQmf+9CPifGxjbCoLLFgVifyVfd/24ZUMKdhwPua/v31T 7gFhDT4qd/ukdt+NcdyHwtr/IFJmi0lt4RJ/y8/En0AGixzELDfzUcK079LBs8H5oMc5 wYKETca1wQfc7z+m7XWzhaH5chmpQdOHZFD0y40R6DhySJM3NLNcLw5Nuv8Ha/1pKvKB 2DWz1Setz5eT3Rdr4HMuPczJMXd73e+R1hYeZs4+8GuWaLPBgSivuZHCkJIa2EvkvM4P oiDOco5a4vxUZUQRLSFEgzNFyEcKDQnl22FPesqe7VVoaraGN4pZZsa0fjM3RJ2DwfyR Eweg== X-Gm-Message-State: AOJu0YyfTKNzljICbWLvGjF26xg+T7Q7s7aC1ekdkFvWembHHJDUneSy qVQGgmaQH1IoD9fnCZrEOqBxYWnrJw27cVHh1pl0SmwW06w/7rh7+7CwAJF6hNioBMLVKpjF25P A X-Google-Smtp-Source: AGHT+IExzK2GEXIrMlan9uaEtW4BhXbWTYYP9wHO4l3cjDaACOrct0xZK0SpUV1xuDHRRGbXmAK44g== X-Received: by 2002:a50:ee85:0:b0:57d:619:7721 with SMTP id 4fb4d7f45d1cf-594bb5807c7mr1792406a12.21.1720540350701; Tue, 09 Jul 2024 08:52:30 -0700 (PDT) From: Alejandro Vallejo To: Xen-devel Cc: Alejandro Vallejo , Jan Beulich , Andrew Cooper , =?UTF-8?q?Roger=20Pau=20Monn=C3=A9?= Subject: [PATCH for-4.20 4/4] x86/fpu: Split fpu_setup_fpu() in two Date: Tue, 9 Jul 2024 16:52:23 +0100 Message-Id: X-Mailer: git-send-email 2.34.1 In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-ZohoMail-DKIM: pass (identity @cloud.com) X-ZM-MESSAGEID: 1720540370530116600 Content-Type: text/plain; charset="utf-8" It's doing too many things at once and there's no clear way of defining what it's meant to do. This patch splits the function in two. 1. A reset function, parameterized by the FCW value. FCW_RESET means to r= eset the state to power-on reset values, while FCW_DEFAULT means to reset t= o the default values present during vCPU creation. 2. A x87/SSE state loader (equivalent to the old function when it took a = data pointer). Signed-off-by: Alejandro Vallejo --- I'm still not sure what the old function tries to do. The state we start vC= PUs in is _similar_ to the after-finit, but it's not quite (`ftw` is not -1). I= went for the "let's not deviate too much from previous behaviour", but maybe we = did intend for vCPUs to start as if `finit` had just been executed? --- xen/arch/x86/domain.c | 7 +++-- xen/arch/x86/hvm/hvm.c | 19 ++++++++----- xen/arch/x86/i387.c | 50 +++++++++++---------------------- xen/arch/x86/include/asm/i387.h | 27 +++++++++++++++--- 4 files changed, 56 insertions(+), 47 deletions(-) diff --git a/xen/arch/x86/domain.c b/xen/arch/x86/domain.c index ccadfe0c9e70..245899cc792f 100644 --- a/xen/arch/x86/domain.c +++ b/xen/arch/x86/domain.c @@ -1198,9 +1198,10 @@ int arch_set_info_guest( is_pv_64bit_domain(d) ) v->arch.flags &=3D ~TF_kernel_mode; =20 - vcpu_setup_fpu(v, v->arch.xsave_area, - flags & VGCF_I387_VALID ? &c.nat->fpu_ctxt : NULL, - FCW_DEFAULT); + if ( flags & VGCF_I387_VALID ) + vcpu_setup_fpu(v, &c.nat->fpu_ctxt); + else + vcpu_reset_fpu(v, FCW_DEFAULT); =20 if ( !compat ) { diff --git a/xen/arch/x86/hvm/hvm.c b/xen/arch/x86/hvm/hvm.c index 09b1426ee314..bedbd2a0b888 100644 --- a/xen/arch/x86/hvm/hvm.c +++ b/xen/arch/x86/hvm/hvm.c @@ -1162,10 +1162,17 @@ static int cf_check hvm_load_cpu_ctxt(struct domain= *d, hvm_domain_context_t *h) seg.attr =3D ctxt.ldtr_arbytes; hvm_set_segment_register(v, x86_seg_ldtr, &seg); =20 - /* Cover xsave-absent save file restoration on xsave-capable host. */ - vcpu_setup_fpu(v, xsave_enabled(v) ? NULL : v->arch.xsave_area, - ctxt.flags & XEN_X86_FPU_INITIALISED ? ctxt.fpu_regs : = NULL, - FCW_RESET); + /* + * On Xen 4.1 and later the FPU state is restored on a later HVM conte= xt, so + * what we're doing here is initialising the FPU state for guests from= even + * older versions of Xen. In general such guests only use legacy x87/S= SE + * state, and if they did use XSAVE then our best-effort strategy is t= o make + * an XSAVE header for x87 and SSE hoping that's good enough. + */ + if ( ctxt.flags & XEN_X86_FPU_INITIALISED ) + vcpu_setup_fpu(v, &ctxt.fpu_regs); + else + vcpu_reset_fpu(v, FCW_RESET); =20 v->arch.user_regs.rax =3D ctxt.rax; v->arch.user_regs.rbx =3D ctxt.rbx; @@ -4005,9 +4012,7 @@ void hvm_vcpu_reset_state(struct vcpu *v, uint16_t cs= , uint16_t ip) v->arch.guest_table =3D pagetable_null(); } =20 - if ( v->arch.xsave_area ) - v->arch.xsave_area->xsave_hdr.xstate_bv =3D 0; - vcpu_setup_fpu(v, v->arch.xsave_area, NULL, FCW_RESET); + vcpu_reset_fpu(v, FCW_RESET); =20 arch_vcpu_regs_init(v); v->arch.user_regs.rip =3D ip; diff --git a/xen/arch/x86/i387.c b/xen/arch/x86/i387.c index a964b84757ec..7851f1b3f6e4 100644 --- a/xen/arch/x86/i387.c +++ b/xen/arch/x86/i387.c @@ -310,41 +310,25 @@ int vcpu_init_fpu(struct vcpu *v) return xstate_alloc_save_area(v); } =20 -void vcpu_setup_fpu(struct vcpu *v, struct xsave_struct *xsave_area, - const void *data, unsigned int fcw_default) +void vcpu_reset_fpu(struct vcpu *v, uint16_t fcw) { - fpusse_t *fpu_sse =3D &v->arch.xsave_area->fpu_sse; - - ASSERT(!xsave_area || xsave_area =3D=3D v->arch.xsave_area); - - v->fpu_initialised =3D !!data; - - if ( data ) - { - memcpy(fpu_sse, data, sizeof(*fpu_sse)); - if ( xsave_area ) - xsave_area->xsave_hdr.xstate_bv =3D XSTATE_FP_SSE; - } - else if ( xsave_area && fcw_default =3D=3D FCW_DEFAULT ) - { - xsave_area->xsave_hdr.xstate_bv =3D 0; - fpu_sse->mxcsr =3D MXCSR_DEFAULT; - } - else - { - memset(fpu_sse, 0, sizeof(*fpu_sse)); - fpu_sse->fcw =3D fcw_default; - fpu_sse->mxcsr =3D MXCSR_DEFAULT; - if ( v->arch.xsave_area ) - { - v->arch.xsave_area->xsave_hdr.xstate_bv &=3D ~XSTATE_FP_SSE; - if ( fcw_default !=3D FCW_DEFAULT ) - v->arch.xsave_area->xsave_hdr.xstate_bv |=3D X86_XCR0_X87; - } - } + v->fpu_initialised =3D false; + *v->arch.xsave_area =3D (struct xsave_struct) { + .fpu_sse =3D { + .mxcsr =3D MXCSR_DEFAULT, + .fcw =3D fcw, + }, + .xsave_hdr.xstate_bv =3D fcw =3D=3D FCW_RESET ? X86_XCR0_X87 : 0, + }; +} =20 - if ( xsave_area ) - xsave_area->xsave_hdr.xcomp_bv =3D 0; +void vcpu_setup_fpu(struct vcpu *v, const void *data) +{ + v->fpu_initialised =3D true; + *v->arch.xsave_area =3D (struct xsave_struct) { + .fpu_sse =3D *(fpusse_t*)data, + .xsave_hdr.xstate_bv =3D XSTATE_FP_SSE, + }; } =20 /* Free FPU's context save area */ diff --git a/xen/arch/x86/include/asm/i387.h b/xen/arch/x86/include/asm/i38= 7.h index a783549db991..ce699fc66663 100644 --- a/xen/arch/x86/include/asm/i387.h +++ b/xen/arch/x86/include/asm/i387.h @@ -31,10 +31,29 @@ void vcpu_restore_fpu_nonlazy(struct vcpu *v, bool need= _stts); void vcpu_restore_fpu_lazy(struct vcpu *v); void vcpu_save_fpu(struct vcpu *v); void save_fpu_enable(void); - int vcpu_init_fpu(struct vcpu *v); -struct xsave_struct; -void vcpu_setup_fpu(struct vcpu *v, struct xsave_struct *xsave_area, - const void *data, unsigned int fcw_default); void vcpu_destroy_fpu(struct vcpu *v); + +/* + * Restore `v`'s FPU to known values + * + * If fcw =3D=3D FCW_RESET, then the reset state is power-on RESET. + * + * Otherwise `mxcsr` is set to `MXCSR_DEFAULT`, `fcw` is overriden with the + * `fcw` argument and everything else is zeroed out. + * + * @param v vCPU containing the FPU + * @param fcw Intended FPU Control Word + */ +void vcpu_reset_fpu(struct vcpu *v, uint16_t fcw); + +/* + * Load x87/SSE state into `v`'s FPU + * + * Overrides the XSAVE header to set the state components to be x87 and SS= E. + * + * @param v vCPU containing the FPU + * @param data 512-octet blob for x87/SSE state + */ +void vcpu_setup_fpu(struct vcpu *v, const void *data); #endif /* __ASM_I386_I387_H */ --=20 2.34.1