3.2.101-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: Dan Williams <dan.j.willi...@intel.com>

commit 085331dfc6bbe3501fb936e657331ca943827600 upstream.

Commit 75f139aaf896 "KVM: x86: Add memory barrier on vmcs field lookup"
added a raw 'asm("lfence");' to prevent a bounds check bypass of
'vmcs_field_to_offset_table'.

The lfence can be avoided in this path by using the array_index_nospec()
helper designed for these types of fixes.

Signed-off-by: Dan Williams <dan.j.willi...@intel.com>
Signed-off-by: Thomas Gleixner <t...@linutronix.de>
Acked-by: Paolo Bonzini <pbonz...@redhat.com>
Cc: Andrew Honig <aho...@google.com>
Cc: k...@vger.kernel.org
Cc: Jim Mattson <jmatt...@google.com>
Link: 
https://lkml.kernel.org/r/151744959670.6342.3001723920950249067.st...@dwillia2-desk3.amr.corp.intel.com
[bwh: Backported to 3.2:
 - Replace max_vmcs_field with the local size variable
 - Adjust context]
Signed-off-by: Ben Hutchings <b...@decadent.org.uk>
---
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -29,6 +29,7 @@
 #include <linux/ftrace_event.h>
 #include <linux/slab.h>
 #include <linux/tboot.h>
+#include <linux/nospec.h>
 #include "kvm_cache_regs.h"
 #include "x86.h"
 
@@ -570,23 +571,21 @@ static unsigned short vmcs_field_to_offs
        FIELD(HOST_RSP, host_rsp),
        FIELD(HOST_RIP, host_rip),
 };
-static const int max_vmcs_field = ARRAY_SIZE(vmcs_field_to_offset_table);
 
 static inline short vmcs_field_to_offset(unsigned long field)
 {
-       if (field >= max_vmcs_field)
-               return -1;
+       const size_t size = ARRAY_SIZE(vmcs_field_to_offset_table);
+       unsigned short offset;
 
-       /*
-        * FIXME: Mitigation for CVE-2017-5753.  To be replaced with a
-        * generic mechanism.
-        */
-       asm("lfence");
-
-       if (vmcs_field_to_offset_table[field] == 0)
+       BUILD_BUG_ON(size > SHRT_MAX);
+       if (field >= size)
                return -1;
 
-       return vmcs_field_to_offset_table[field];
+       field = array_index_nospec(field, size);
+       offset = vmcs_field_to_offset_table[field];
+       if (offset == 0)
+               return -1;
+       return offset;
 }
 
 static inline struct vmcs12 *get_vmcs12(struct kvm_vcpu *vcpu)

Reply via email to