From mboxrd@z Thu Jan  1 00:00:00 1970
From: Weidong Han <weidong.han@intel.com>
Subject: Re: Xen 4.0.0-rc7 problem/hang with vt-d DMAR	 parsing
Date: Thu, 25 Mar 2010 17:05:57 +0800
Message-ID: <4BAB2775.5010208@intel.com>
References: <20100323193748.GW1878@reaktio.net>
	<C7CECCF9.E36E%keir.fraser@eu.citrix.com>
	<20100323200515.GZ1878@reaktio.net>
	<ED3036A092A28F4C91B0B4360DD128EABE03D6BC@shzsmsx502.ccr.corp.intel.com>
	<4BA9DA400200007800036ABB@vpn.id2.novell.com>
	<4BA9D512.9090902@intel.com>
	<4BA9ED8D0200007800036B3F@vpn.id2.novell.com>
	<4BA9F0C1.7080809@intel.com>
	<4BAA01560200007800036B9B@vpn.id2.novell.com>
	<4BAAB488.3020007@intel.com>
	<4BAB304A0200007800036D8B@vpn.id2.novell.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Return-path: <xen-devel-bounces@lists.xensource.com>
In-Reply-To: <4BAB304A0200007800036D8B@vpn.id2.novell.com>
List-Unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xensource.com>
List-Help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-Subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
Sender: xen-devel-bounces@lists.xensource.com
Errors-To: xen-devel-bounces@lists.xensource.com
To: Jan Beulich <JBeulich@novell.com>
Cc: "xen-devel@lists.xensource.com" <xen-devel@lists.xensource.com>, Keir Fraser <keir.fraser@eu.citrix.com>, "Cui,
	Dexuan" <dexuan.cui@intel.com>
List-Id: xen-devel@lists.xenproject.org

Jan Beulich wrote:
>>>> Weidong Han <weidong.han@intel.com> 25.03.10 01:55 >>>
>>>>         
>> Do you mean to know which case fails on length checking? How about below 
>> patch?
>>     
>
> Yes, something like this. Although I'd prefer to have a general
> sizeof(struct acpi_dmar_entry_header) check before the switch
> statement (to avoid even looking at out of range header fields),
> and "break"s instead of "goto disable"s.
>
>   
Ok. Updated the patch according to your suggestion. Thanks.

diff -r a4eac162dcb9 xen/drivers/passthrough/vtd/dmar.c
--- a/xen/drivers/passthrough/vtd/dmar.c    Thu Mar 25 01:05:03 2010 +0800
+++ b/xen/drivers/passthrough/vtd/dmar.c    Fri Mar 26 01:59:55 2010 +0800
@@ -659,26 +659,71 @@ static int __init acpi_parse_dmar(struct
     while ( ((unsigned long)entry_header) <
             (((unsigned long)dmar) + table->length) )
     {
+        if ( entry_header->length < sizeof(struct acpi_dmar_entry_header) )
+        {
+            dprintk(XENLOG_ERR VTDPREFIX,
+                    "Invalid ACPI DMAR entry length: 0x%x\n",
+                    entry_header->length);
+            ret = -EINVAL;
+            break;
+        }
+
         switch ( entry_header->type )
         {
         case ACPI_DMAR_DRHD:
             if ( iommu_verbose )
                 dprintk(VTDPREFIX, "found ACPI_DMAR_DRHD:\n");
+
+            if ( entry_header->length < sizeof(struct acpi_table_drhd) )
+            {
+                dprintk(XENLOG_ERR VTDPREFIX,
+                        "  Invalid length: 0x%x\n", entry_header->length);
+                ret = -EINVAL;
+                break;
+            }
+
             ret = acpi_parse_one_drhd(entry_header);
             break;
         case ACPI_DMAR_RMRR:
             if ( iommu_verbose )
                 dprintk(VTDPREFIX, "found ACPI_DMAR_RMRR:\n");
+
+            if ( entry_header->length < sizeof(struct acpi_table_rmrr) )
+            {
+                dprintk(XENLOG_ERR VTDPREFIX,
+                        "  Invalid length: 0x%x\n", entry_header->length);
+                ret = -EINVAL;
+                break;
+            }
+
             ret = acpi_parse_one_rmrr(entry_header);
             break;
         case ACPI_DMAR_ATSR:
             if ( iommu_verbose )
                 dprintk(VTDPREFIX, "found ACPI_DMAR_ATSR:\n");
+
+            if ( entry_header->length < sizeof(struct acpi_table_atsr) )
+            {
+                dprintk(XENLOG_ERR VTDPREFIX,
+                        "  Invalid length: 0x%x\n", entry_header->length);
+                ret = -EINVAL;
+                break;
+            }
+
             ret = acpi_parse_one_atsr(entry_header);
             break;
         default: