From patchwork Tue Feb 25 19:17:56 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Tamas K Lengyel X-Patchwork-Id: 11404467 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 671BC14BC for ; Tue, 25 Feb 2020 19:19:55 +0000 (UTC) Received: from lists.xenproject.org (lists.xenproject.org [192.237.175.120]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 4D8D82082F for ; Tue, 25 Feb 2020 19:19:55 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 4D8D82082F Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=intel.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=xen-devel-bounces@lists.xenproject.org Received: from localhost ([127.0.0.1] helo=lists.xenproject.org) by lists.xenproject.org with esmtp (Exim 4.89) (envelope-from ) id 1j6fir-00078z-Do; Tue, 25 Feb 2020 19:18:25 +0000 Received: from us1-rack-iad1.inumbo.com ([172.99.69.81]) by lists.xenproject.org with esmtp (Exim 4.89) (envelope-from ) id 1j6fip-00078m-FX for xen-devel@lists.xenproject.org; Tue, 25 Feb 2020 19:18:23 +0000 X-Inumbo-ID: 914e5e16-5803-11ea-aba8-bc764e2007e4 Received: from mga06.intel.com (unknown [134.134.136.31]) by us1-rack-iad1.inumbo.com (Halon) with ESMTPS id 914e5e16-5803-11ea-aba8-bc764e2007e4; Tue, 25 Feb 2020 19:18:13 +0000 (UTC) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga003.jf.intel.com ([10.7.209.27]) by orsmga104.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 25 Feb 2020 11:18:12 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,485,1574150400"; d="scan'208";a="237776387" Received: from tlengyel-mobl2.amr.corp.intel.com (HELO localhost.localdomain) ([10.254.187.145]) by orsmga003.jf.intel.com with ESMTP; 25 Feb 2020 11:18:11 -0800 From: Tamas K Lengyel To: xen-devel@lists.xenproject.org Date: Tue, 25 Feb 2020 11:17:56 -0800 Message-Id: <628c5cdc73c589e45a19cc0ddb5cf972b00eb3dd.1582658216.git.tamas.lengyel@intel.com> X-Mailer: git-send-email 2.20.1 In-Reply-To: References: MIME-Version: 1.0 Subject: [Xen-devel] [PATCH v10 2/3] x86/mem_sharing: reset a fork X-BeenThere: xen-devel@lists.xenproject.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Cc: Tamas K Lengyel , Tamas K Lengyel , Wei Liu , Konrad Rzeszutek Wilk , Andrew Cooper , Ian Jackson , George Dunlap , Stefano Stabellini , Jan Beulich , Julien Grall , =?utf-8?q?Roger_Pau_Monn=C3=A9?= Errors-To: xen-devel-bounces@lists.xenproject.org Sender: "Xen-devel" Implement hypercall that allows a fork to shed all memory that got allocated for it during its execution and re-load its vCPU context from the parent VM. This allows the forked VM to reset into the same state the parent VM is in a faster way then creating a new fork would be. Measurements show about a 2x speedup during normal fuzzing operations. Performance may vary depending how much memory got allocated for the forked VM. If it has been completely deduplicated from the parent VM then creating a new fork would likely be more performant. Signed-off-by: Tamas K Lengyel --- v10: implemented hypercall continuation similar to the existing range_share op --- xen/arch/x86/mm/mem_sharing.c | 126 +++++++++++++++++++++++++++++++++- xen/include/public/memory.h | 4 ++ 2 files changed, 129 insertions(+), 1 deletion(-) diff --git a/xen/arch/x86/mm/mem_sharing.c b/xen/arch/x86/mm/mem_sharing.c index 8ee37e6943..aa4358aae4 100644 --- a/xen/arch/x86/mm/mem_sharing.c +++ b/xen/arch/x86/mm/mem_sharing.c @@ -1673,7 +1673,6 @@ static int fork(struct domain *d, struct domain *cd) domain_pause(d); cd->parent_paused = true; cd->max_pages = d->max_pages; - cd->max_vcpus = d->max_vcpus; } /* this is preemptible so it's the first to get done */ @@ -1704,6 +1703,91 @@ static int fork(struct domain *d, struct domain *cd) return rc; } +/* + * The fork reset operation is intended to be used on short-lived forks only. + */ +static int fork_reset(struct domain *d, struct domain *cd, + struct mem_sharing_op_fork_reset *fr) +{ + int rc = 0; + struct p2m_domain* p2m = p2m_get_hostp2m(cd); + struct page_info *page, *tmp; + unsigned long list_position = 0, preempt_count = 0, start = fr->opaque; + + domain_pause(cd); + + page_list_for_each_safe(page, tmp, &cd->page_list) + { + p2m_type_t p2mt; + p2m_access_t p2ma; + gfn_t gfn; + mfn_t mfn; + bool shared = false; + + list_position++; + + /* Resume were we left of before preemption */ + if ( start && list_position < start ) + continue; + + mfn = page_to_mfn(page); + if ( mfn_valid(mfn) ) + { + + gfn = mfn_to_gfn(cd, mfn); + mfn = __get_gfn_type_access(p2m, gfn_x(gfn), &p2mt, &p2ma, + 0, NULL, false); + + if ( p2m_is_ram(p2mt) && !p2m_is_shared(p2mt) ) + { + /* take an extra reference, must work for a shared page */ + if( !get_page(page, cd) ) + { + ASSERT_UNREACHABLE(); + return -EINVAL; + } + + shared = true; + preempt_count += 0x10; + + /* + * Must succeed, it's a shared page that exists and + * thus its size is guaranteed to be 4k so we are not splitting + * large pages. + */ + rc = p2m->set_entry(p2m, gfn, INVALID_MFN, PAGE_ORDER_4K, + p2m_invalid, p2m_access_rwx, -1); + ASSERT(!rc); + + put_page_alloc_ref(page); + put_page(page); + } + } + + if ( !shared ) + preempt_count++; + + /* Preempt every 2MiB (shared) or 32MiB (unshared) - arbitrary. */ + if ( preempt_count >= 0x2000 ) + { + if ( hypercall_preempt_check() ) + { + rc = -ERESTART; + break; + } + preempt_count = 0; + } + } + + if ( rc ) + fr->opaque = list_position; + else if ( !(rc = hvm_copy_context_and_params(cd, d)) ) + fork_tsc(cd, d); + + domain_unpause(cd); + return rc; +} + int mem_sharing_memop(XEN_GUEST_HANDLE_PARAM(xen_mem_sharing_op_t) arg) { int rc; @@ -1973,7 +2057,17 @@ int mem_sharing_memop(XEN_GUEST_HANDLE_PARAM(xen_mem_sharing_op_t) arg) goto out; if ( !mem_sharing_enabled(pd) && (rc = mem_sharing_control(pd, true)) ) + { + rcu_unlock_domain(pd); goto out; + } + + rc = -EINVAL; + if ( pd->max_vcpus != d->max_vcpus ) + { + rcu_unlock_domain(pd); + goto out; + } rc = fork(pd, d); @@ -1985,6 +2079,36 @@ int mem_sharing_memop(XEN_GUEST_HANDLE_PARAM(xen_mem_sharing_op_t) arg) break; } + case XENMEM_sharing_op_fork_reset: + { + struct domain *pd; + + rc = -ENOSYS; + if ( !mem_sharing_is_fork(d) ) + goto out; + + rc = rcu_lock_live_remote_domain_by_id(d->parent->domain_id, &pd); + if ( rc ) + goto out; + + rc = fork_reset(pd, d, &mso.u.fork_reset); + + rcu_unlock_domain(pd); + + if ( rc > 0 ) + { + if ( __copy_to_guest(arg, &mso, 1) ) + rc = -EFAULT; + else + rc = hypercall_create_continuation(__HYPERVISOR_memory_op, + "lh", XENMEM_sharing_op, + arg); + } + else + mso.u.fork_reset.opaque = 0; + break; + } + default: rc = -ENOSYS; break; diff --git a/xen/include/public/memory.h b/xen/include/public/memory.h index c1dbad060e..7ca07c01dd 100644 --- a/xen/include/public/memory.h +++ b/xen/include/public/memory.h @@ -483,6 +483,7 @@ DEFINE_XEN_GUEST_HANDLE(xen_mem_access_op_t); #define XENMEM_sharing_op_audit 7 #define XENMEM_sharing_op_range_share 8 #define XENMEM_sharing_op_fork 9 +#define XENMEM_sharing_op_fork_reset 10 #define XENMEM_SHARING_OP_S_HANDLE_INVALID (-10) #define XENMEM_SHARING_OP_C_HANDLE_INVALID (-9) @@ -537,6 +538,9 @@ struct xen_mem_sharing_op { domid_t parent_domain; /* IN: parent's domain id */ uint16_t _pad[3]; /* Must be set to 0 */ } fork; + struct mem_sharing_op_fork_reset { /* OP_FORK_RESET */ + uint64_aligned_t opaque; /* Must be set to 0 */ + } fork_reset; } u; }; typedef struct xen_mem_sharing_op xen_mem_sharing_op_t;