From nobody Wed Sep 23 09:36:34 2026 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=quarantine dis=none) header.from=suse.com ARC-Seal: i=1; a=rsa-sha256; t=1785248546; cv=none; d=zohomail.com; s=zohoarc; b=kZlD/zD69bgB0e+0wJ2PRNnl+xBeXoPEz5AsZtueFJ5DxG18Kz1yg2j3pWVKqPg2EvyNi+k3pvHiu3KFB+cU5nOvciL+mLD+LyjzMVUpHI+FQ4Osfh8oCWV8u3jNr+zPbhK44EgJS/FVHLl7F3TIWSLtyScoVSUItQdmRDvYqeg= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1785248546; h=Content-Type: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=YAcoPIBVIk35F4UuhNNjahrwo2SklVXZ2m2wuexNepE=; b=hrGGxtAVWR/vcUKItHiSUeLGFzgoAcGgbp5rg4kn1GvEA51hKBZdXQVmW7hasCnS3tSBNoPXwqz6wqS7BoKFuEhewfJXCx95EcKGQezwLuTzdogmgxRpD+rYRCiAfdVrWZudIqVZ7ACijZe4ljXfZX8JB1nk/8KrM1viz/djEog= 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=quarantine dis=none) Return-Path: Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) by mx.zohomail.com with SMTPS id 1785248546597290.58057033417435; Tue, 28 Jul 2026 07:22:26 -0700 (PDT) Received: from list by lists.xenproject.org with outflank-mailman.1374381.1621539 (Exim 4.92) (envelope-from ) id 1woihD-0002xw-DO; Tue, 28 Jul 2026 14:22:15 +0000 Received: by outflank-mailman (output) from mailman id 1374381.1621539; Tue, 28 Jul 2026 14:22:15 +0000 Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1woihD-0002xn-Ad; Tue, 28 Jul 2026 14:22:15 +0000 Received: by outflank-mailman (input) for mailman id 1374381; Tue, 28 Jul 2026 14:22:14 +0000 Received: from mx.expurgate.net ([194.145.224.10]) by lists.xenproject.org with esmtp (Exim 4.92) (envelope-from ) id 1woihC-0002xG-Fr for xen-devel@lists.xenproject.org; Tue, 28 Jul 2026 14:22:14 +0000 Received: from mx.expurgate.net (helo=localhost) by mx.expurgate.net with esmtp id 1woihB-00HOAI-T1 for xen-devel@lists.xenproject.org; Tue, 28 Jul 2026 16:22:13 +0200 Received: from [10.42.69.11] (helo=localhost) by localhost with ESMTP (eXpurgate MTA 0.9.1) (envelope-from ) id 6a68bafb-5cb7-0a2a0a5109dd-0a2a450bc7ee-44 for ; Tue, 28 Jul 2026 16:22:13 +0200 Received: from [209.85.221.46] (helo=mail-wr1-f46.google.com) by tlsNG-42698a.mxtls.expurgate.net with ESMTPS (eXpurgate 4.57.1) (envelope-from ) id 6a68bb15-b7e8-0a2a450b0019-d155dd2eb983-3 for ; Tue, 28 Jul 2026 16:22:13 +0200 Received: by mail-wr1-f46.google.com with SMTP id ffacd0b85a97d-47f84023916so3909473f8f.3 for ; Tue, 28 Jul 2026 07:22:13 -0700 (PDT) Received: from [10.156.60.236] (ip-037-024-206-209.um08.pools.vodafone-ip.de. [37.24.206.209]) by smtp.gmail.com with ESMTPSA id ffacd0b85a97d-47f85c6f076sm59359286f8f.34.2026.07.28.07.22.12 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Tue, 28 Jul 2026 07:22:12 -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" Authentication-Results: eu.smtp.expurgate.cloud; dkim=pass header.s=google header.d=suse.com header.i="@suse.com" header.h="Content-Transfer-Encoding:Content-Type:In-Reply-To:Autocrypt:Content-Language:References:Cc:To:From:Subject:User-Agent:MIME-Version:Date:Message-ID" DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.com; s=google; t=1785248533; x=1785853333; darn=lists.xenproject.org; h=content-transfer-encoding:content-type:in-reply-to:autocrypt :content-language:references:cc:to:from:subject:user-agent :mime-version:date:message-id:from:to:cc:subject:date:message-id :reply-to:content-type; bh=YAcoPIBVIk35F4UuhNNjahrwo2SklVXZ2m2wuexNepE=; b=NQT3b/l1U/0AHnFoVyVGdUrKKFGRn8j6g5egzmJbO4CCblTXk6M4llqgKZrtjbPtIg uyRnzuNLtjmp7VQYlfi0C0a8w8UQtrAhVlefLM3UizzB8Ue3lGMFifoLK4fhkCl1czDI 1JJHcyVbuwSzwsoPzkCAehhBKDkK+2tPHd74cjZOKGNF5kMhd3p+nAtS+u8et5A9om/x 0lGCXfP5E1LJyfHZZi5TCpCBS4/h62be4ICpHwxi5sZvIXDk76x+esTd8WEPfA3BlXK5 7crs2NTr2ZBtZZwFeJ/rhPYRB0e/5L4E9T9LOCVTRz7Wc2Dxl8BU35IrghmONgCQr1r5 OHjg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1785248533; x=1785853333; h=content-transfer-encoding:content-type:in-reply-to:autocrypt :content-language:references:cc:to:from:subject:user-agent :mime-version:date:message-id:x-gm-gg:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to:content-type; bh=YAcoPIBVIk35F4UuhNNjahrwo2SklVXZ2m2wuexNepE=; b=NO8wQ4YtuZT7DSe/R+npHfP87Q8zGu/gculo4nluzaNUuN0vLLW7QADQ/tAL1L95SD cfwkHbzuC81j3WVSZG9IC7CK9j+xLdK3p0+4MTqnHnbXRTacuBt54Wk+w0xrhIT0V8jD 7xh57B9PEw/L1wPybsKFw/6nSgNqbYpm2aUWTJpHEOmLFjO6HOaPj/TGUelp1ILi6Sp+ 5JdH1L3cy2EFOd2hgzswZGHNiaGNQWINATqnBM+3EPIBNR9P910lPciQMkIiR5ut1T7Q BsAv6x0aiDCGpfTwRLG0k4crfCD2BjOg9dJsiw1ZA7W+miyKZVWCea0amg+KbV/v89kO XN8A== X-Gm-Message-State: AOJu0YwTxgKYLiWaOcCk7fD2JiCf9TvGE6GZMauRthQWQ4kY3kJARUpW gtZH1+9cxRl+4Qi4Ze7IK0s88mf7oAZec+axdpIaX4GIoeceJ2QKPAkm5XHxtv31zySA8O+ETrC ifwHGOQ== X-Gm-Gg: AR+sD12nmcPFH2Nwq84+IJ3jR+z0ggNs8K1Bw6Tpl1wHzUYZxdG+06hrGRaHIzSdVUg Hb8UQ0KmA+SNAdxUFEpaDPYz39S9VyGYN9oodKAf88kMc0fS6KrC+H6HDugs3Z/KNfleWxFzSck q0U29Sgpt2GgUvOIG4JQNc7t8dVZzt2GWlRUse3Oqp/sFO7x3QXoqyVFLfAMrnQmQQpOo6qxGhL XzQvsQ0nSkCO/gi2uNq528SKkzkgddLInwYu83ugb4GqT9uEIOerO+AfjPIuBUkgxSLkFdT95+d /GComwCnD9ue1mzsWlOyXi3/efwTmfyQjpQZgvtoUPLazpLUzMgGeBZYvzXHpqcX77QDpQmzERc SfknWokGuTKerg43ST/3gUJzRGUqt85H9PchsKu6+MCak8ypmA49K4gvPdTafOnXeALBVkmS5wX Q/LxdbB+UJsP3Qq7sVg5ASeudtGZepr5xs3ubQHyrxSx0yqkLtmaqLTQgHqRnqDUtbkjFKf+X+g CHp X-Received: by 2002:a05:6000:22c3:b0:47f:9567:e63b with SMTP id ffacd0b85a97d-47fb1f26115mr3657417f8f.55.1785248533183; Tue, 28 Jul 2026 07:22:13 -0700 (PDT) Message-ID: <274756a3-83da-4594-8909-25a3a9eb94d2@suse.com> Date: Tue, 28 Jul 2026 16:22:11 +0200 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: [PATCH v15 05/10] x86/shadow: make clean-dirty-bitmap post-processing preemptable From: Jan Beulich To: "xen-devel@lists.xenproject.org" Cc: Andrew Cooper , Tim Deegan References: <68c16600-a4bf-4060-a1fc-56c4ae655b03@suse.com> Content-Language: en-US Autocrypt: addr=jbeulich@suse.com; keydata= xsDiBFk3nEQRBADAEaSw6zC/EJkiwGPXbWtPxl2xCdSoeepS07jW8UgcHNurfHvUzogEq5xk hu507c3BarVjyWCJOylMNR98Yd8VqD9UfmX0Hb8/BrA+Hl6/DB/eqGptrf4BSRwcZQM32aZK 7Pj2XbGWIUrZrd70x1eAP9QE3P79Y2oLrsCgbZJfEwCgvz9JjGmQqQkRiTVzlZVCJYcyGGsD /0tbFCzD2h20ahe8rC1gbb3K3qk+LpBtvjBu1RY9drYk0NymiGbJWZgab6t1jM7sk2vuf0Py O9Hf9XBmK0uE9IgMaiCpc32XV9oASz6UJebwkX+zF2jG5I1BfnO9g7KlotcA/v5ClMjgo6Gl MDY4HxoSRu3i1cqqSDtVlt+AOVBJBACrZcnHAUSuCXBPy0jOlBhxPqRWv6ND4c9PH1xjQ3NP nxJuMBS8rnNg22uyfAgmBKNLpLgAGVRMZGaGoJObGf72s6TeIqKJo/LtggAS9qAUiuKVnygo 3wjfkS9A3DRO+SpU7JqWdsveeIQyeyEJ/8PTowmSQLakF+3fote9ybzd880fSmFuIEJldWxp Y2ggPGpiZXVsaWNoQHN1c2UuY29tPsJgBBMRAgAgBQJZN5xEAhsDBgsJCAcDAgQVAggDBBYC AwECHgECF4AACgkQoDSui/t3IH4J+wCfQ5jHdEjCRHj23O/5ttg9r9OIruwAn3103WUITZee e7Sbg12UgcQ5lv7SzsFNBFk3nEQQCACCuTjCjFOUdi5Nm244F+78kLghRcin/awv+IrTcIWF hUpSs1Y91iQQ7KItirz5uwCPlwejSJDQJLIS+QtJHaXDXeV6NI0Uef1hP20+y8qydDiVkv6l IreXjTb7DvksRgJNvCkWtYnlS3mYvQ9NzS9PhyALWbXnH6sIJd2O9lKS1Mrfq+y0IXCP10eS FFGg+Av3IQeFatkJAyju0PPthyTqxSI4lZYuJVPknzgaeuJv/2NccrPvmeDg6Coe7ZIeQ8Yj t0ARxu2xytAkkLCel1Lz1WLmwLstV30g80nkgZf/wr+/BXJW/oIvRlonUkxv+IbBM3dX2OV8 AmRv1ySWPTP7AAMFB/9PQK/VtlNUJvg8GXj9ootzrteGfVZVVT4XBJkfwBcpC/XcPzldjv+3 HYudvpdNK3lLujXeA5fLOH+Z/G9WBc5pFVSMocI71I8bT8lIAzreg0WvkWg5V2WZsUMlnDL9 mpwIGFhlbM3gfDMs7MPMu8YQRFVdUvtSpaAs8OFfGQ0ia3LGZcjA6Ik2+xcqscEJzNH+qh8V m5jjp28yZgaqTaRbg3M/+MTbMpicpZuqF4rnB0AQD12/3BNWDR6bmh+EkYSMcEIpQmBM51qM EKYTQGybRCjpnKHGOxG0rfFY1085mBDZCH5Kx0cl0HVJuQKC+dV2ZY5AqjcKwAxpE75MLFkr wkkEGBECAAkFAlk3nEQCGwwACgkQoDSui/t3IH7nnwCfcJWUDUFKdCsBH/E5d+0ZnMQi+G0A nAuWpQkjM1ASeQwSHEeAWPgskBQL In-Reply-To: <68c16600-a4bf-4060-a1fc-56c4ae655b03@suse.com> Content-Transfer-Encoding: quoted-printable X-purgate-ID: tlsNG-42698a/1785248533-A86CA9EA-898524F5/0/0 X-purgate-type: clean X-purgate-size: 6534 X-ZohoMail-DKIM: pass (identity @suse.com) X-ZM-MESSAGEID: 1785248548728158500 Content-Type: text/plain; charset="utf-8" What this currently amounts to is blowing the domain's tables. To make that call preemptable and to record state in a race-free manner, it needs to be moved inside the paging-locked region. Since the corresponding HAP function wants to acquire the P2M lock, the hook invocation itself cannot be moved. Call the shadow function directly, and prevent the hook from being invoked in that case. To prevent "extra" accounting from being done more than once in paging_log_dirty_op(), the mapping of the top level table also becomes conditional (upon the subsequent loop doing nothing anyway). Further note that this now also avoids sh{,adow}_clean_dirty_bitmap() needlessly being called when no paging mode was enabled at all for the domain. (The function continues to be called even when log-dirty mode isn't enabled.) Signed-off-by: Jan Beulich --- The hook functions were/are slightly mis-named, as they don't clean anything. However, to keep names similar (and hence possible to grep for) the new directly called function is still named in the same manner. Of course we could purge the hook altogether, calling the HAP function directly as well. If so, question would be whether to do so right here or in a public follow-up (perhaps the latter). In the course of doing this work I've noticed that libxenguest uses XEN_DOMCTL_SHADOW_OP_CLEAN even on the last iteration. Wouldn't it suffice to invoke the non-cleaning (peek) operation there instead, reducing the overhead of shadow's post-processing? --- v15: Fix !SHADOW_PAGING build. v14: New. --- a/xen/arch/x86/include/asm/domain.h +++ b/xen/arch/x86/include/asm/domain.h @@ -234,7 +234,7 @@ struct paging_domain { union { struct { unsigned long done:PADDR_BITS - PAGE_SHIFT; - unsigned long i4:PAGETABLE_ORDER; + unsigned long i4:PAGETABLE_ORDER + 1; unsigned long i3:PAGETABLE_ORDER; } log_dirty; }; --- a/xen/arch/x86/include/asm/shadow.h +++ b/xen/arch/x86/include/asm/shadow.h @@ -56,6 +56,9 @@ int shadow_domain_init(struct domain *d) * paging_vcpu_init() in paging.c */ void shadow_vcpu_init(struct vcpu *v); =20 +/* Post-processing necessary after the log-dirty bitmap was cleaned. */ +bool shadow_clean_dirty_bitmap(struct domain *d); + #ifdef CONFIG_SHADOW_PAGING =20 /* Enable an arbitrary shadow mode. Call once at domain creation. */ --- a/xen/arch/x86/mm/paging.c +++ b/xen/arch/x86/mm/paging.c @@ -473,8 +473,9 @@ static int paging_log_dirty_op(struct do goto out; } =20 - l4 =3D paging_map_log_dirty_bitmap(d); i4 =3D d->arch.paging.preempt.log_dirty.i4; + if ( i4 < LOGDIRTY_NODE_ENTRIES ) + l4 =3D paging_map_log_dirty_bitmap(d); i3 =3D d->arch.paging.preempt.log_dirty.i3; pages =3D d->arch.paging.preempt.log_dirty.done; =20 @@ -542,6 +543,19 @@ static int paging_log_dirty_op(struct do if ( l4 ) unmap_domain_page(l4); =20 + /* + * In the shadow case post-processing may take long and hence needs to= be + * preemptable. Since continuation state is protected by the paging lo= ck, + * we need to call its function (and record state) before dropping that + * lock. The hook invocation further down therefore will be skipped. + */ + if ( !rv && clean && paging_mode_shadow(d) && + !shadow_clean_dirty_bitmap(d) ) + { + d->arch.paging.preempt.log_dirty.i4 =3D LOGDIRTY_NODE_ENTRIES; + rv =3D -ERESTART; + } + if ( !rv ) { d->arch.paging.preempt.dom =3D NULL; @@ -569,10 +583,12 @@ static int paging_log_dirty_op(struct do =20 if ( pages < sc->pages ) sc->pages =3D pages; - if ( clean ) + if ( clean && paging_mode_hap(d) ) { - /* We need to further call clean_dirty_bitmap() functions of speci= fic - * paging modes (shadow or hap). Safe because the domain is pause= d. */ + /* + * We need to further call HAP's clean_dirty_bitmap() function (sh= adow + * was dealt with above). Safe because the domain is paused. + */ d->arch.paging.log_dirty.ops->clean(d); } domain_unpause(d); --- a/xen/arch/x86/mm/shadow/common.c +++ b/xen/arch/x86/mm/shadow/common.c @@ -71,7 +71,6 @@ DEFINE_PER_CPU(uint32_t,trace_shadow_pat =20 static int cf_check sh_enable_log_dirty(struct domain *d); static int cf_check sh_disable_log_dirty(struct domain *d); -static void cf_check sh_clean_dirty_bitmap(struct domain *d); =20 static void cf_check shadow_update_paging_modes(struct vcpu *v); =20 @@ -82,7 +81,7 @@ int shadow_domain_init(struct domain *d) static const struct log_dirty_ops sh_ops =3D { .enable =3D sh_enable_log_dirty, .disable =3D sh_disable_log_dirty, - .clean =3D sh_clean_dirty_bitmap, + .clean =3D ZERO_BLOCK_PTR, }; =20 INIT_PAGE_LIST_HEAD(&d->arch.paging.shadow.pinned_shadows); @@ -2606,14 +2605,18 @@ static int cf_check sh_disable_log_dirty /* This function is called when we CLEAN log dirty bitmap. See * paging_log_dirty_op() for details. */ -static void cf_check sh_clean_dirty_bitmap(struct domain *d) +bool shadow_clean_dirty_bitmap(struct domain *d) { - paging_lock(d); + bool preempted =3D false; + + ASSERT(paging_locked_by_me(d)); + /* Need to revoke write access to the domain's pages again. * In future, we'll have a less heavy-handed approach to this, * but for now, we just unshadow everything except Xen. */ - shadow_blow_tables(d, NULL); - paging_unlock(d); + shadow_blow_tables(d, &preempted); + + return !preempted; } =20 /*************************************************************************= */ --- a/xen/arch/x86/mm/shadow/none.c +++ b/xen/arch/x86/mm/shadow/none.c @@ -7,11 +7,6 @@ static int cf_check _toggle_log_dirty(st return -EOPNOTSUPP; } =20 -static void cf_check _clean_dirty_bitmap(struct domain *d) -{ - ASSERT(is_pv_domain(d)); -} - static void cf_check _update_paging_modes(struct vcpu *v) { ASSERT_UNREACHABLE(); @@ -23,7 +18,7 @@ int shadow_domain_init(struct domain *d) static const struct log_dirty_ops sh_none_ops =3D { .enable =3D _toggle_log_dirty, .disable =3D _toggle_log_dirty, - .clean =3D _clean_dirty_bitmap, + .clean =3D ZERO_BLOCK_PTR, }; =20 paging_log_dirty_init(d, &sh_none_ops);