Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751438AbdGYJuW (ORCPT ); Tue, 25 Jul 2017 05:50:22 -0400 Received: from mx2.suse.de ([195.135.220.15]:34240 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750855AbdGYJuU (ORCPT ); Tue, 25 Jul 2017 05:50:20 -0400 From: Juergen Gross To: linux-kernel@vger.kernel.org, xen-devel@lists.xenproject.org, x86@kernel.org Cc: boris.ostrovsky@oracle.com, hpa@zytor.com, tglx@linutronix.de, mingo@redhat.com, Juergen Gross Subject: [PATCH 3/3] xen: fix hvm guest with kaslr enabled Date: Tue, 25 Jul 2017 11:50:13 +0200 Message-Id: <20170725095013.2939-4-jgross@suse.com> X-Mailer: git-send-email 2.12.3 In-Reply-To: <20170725095013.2939-1-jgross@suse.com> References: <20170725095013.2939-1-jgross@suse.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2174 Lines: 70 A Xen HVM guest running with KASLR enabled will die rather soon today due to the shared info page mapping is using va() too early. This was introduced by commit a5d5f328b0e2baa5ee7c119fd66324eb79eeeb66 ("xen: allocate page for shared info page from low memory"). In order to fix this use early_memremap() to get a temporary virtual address for shared info until va() can be used safely. Signed-off-by: Juergen Gross --- arch/x86/xen/enlighten_hvm.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) diff --git a/arch/x86/xen/enlighten_hvm.c b/arch/x86/xen/enlighten_hvm.c index d23531f5f17e..de503c225ae1 100644 --- a/arch/x86/xen/enlighten_hvm.c +++ b/arch/x86/xen/enlighten_hvm.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -21,6 +22,8 @@ #include "mmu.h" #include "smp.h" +static unsigned long shared_info_pfn; + void xen_hvm_init_shared_info(void) { struct xen_add_to_physmap xatp; @@ -28,7 +31,7 @@ void xen_hvm_init_shared_info(void) xatp.domid = DOMID_SELF; xatp.idx = 0; xatp.space = XENMAPSPACE_shared_info; - xatp.gpfn = virt_to_pfn(HYPERVISOR_shared_info); + xatp.gpfn = shared_info_pfn; if (HYPERVISOR_memory_op(XENMEM_add_to_physmap, &xatp)) BUG(); } @@ -51,8 +54,16 @@ static void __init reserve_shared_info(void) pa += PAGE_SIZE) ; + shared_info_pfn = PHYS_PFN(pa); + memblock_reserve(pa, PAGE_SIZE); - HYPERVISOR_shared_info = __va(pa); + HYPERVISOR_shared_info = early_memremap(pa, PAGE_SIZE); +} + +static void __init xen_hvm_init_mem_mapping(void) +{ + early_memunmap(HYPERVISOR_shared_info, PAGE_SIZE); + HYPERVISOR_shared_info = __va(PFN_PHYS(shared_info_pfn)); } static void __init init_hvm_pv_info(void) @@ -221,5 +232,6 @@ const struct hypervisor_x86 x86_hyper_xen_hvm = { .init_platform = xen_hvm_guest_init, .pin_vcpu = xen_pin_vcpu, .x2apic_available = xen_x2apic_para_available, + .init_mem_mapping = xen_hvm_init_mem_mapping, }; EXPORT_SYMBOL(x86_hyper_xen_hvm); -- 2.12.3