Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 5 Jun 2015 21:22:27 +0000 (UTC)
From:      Neel Natu <neel@FreeBSD.org>
To:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org
Subject:   svn commit: r284046 - head/sys/amd64/vmm
Message-ID:  <201506052122.t55LMR7u056330@svn.freebsd.org>

next in thread | raw e-mail | index | archive | help
Author: neel
Date: Fri Jun  5 21:22:26 2015
New Revision: 284046
URL: https://svnweb.freebsd.org/changeset/base/284046

Log:
  The 'verify_gla()' function is used to ensure that the effective address
  after decoding the instruction matches the one provided by hardware.
  
  Prior to r283293 'vie->num_valid' used to contain the actual length of
  the instruction whereas now it contains the maximum instruction length
  possible. This introduced a bug when calculating a RIP-relative base address.
  
  Fix this by using 'vie->num_processed' rather than 'vie->num_valid' as the
  length of the emulated instruction.
  
  Reported and tested by:	tychon
  MFC after:	1 week

Modified:
  head/sys/amd64/vmm/vmm_instruction_emul.c

Modified: head/sys/amd64/vmm/vmm_instruction_emul.c
==============================================================================
--- head/sys/amd64/vmm/vmm_instruction_emul.c	Fri Jun  5 20:38:22 2015	(r284045)
+++ head/sys/amd64/vmm/vmm_instruction_emul.c	Fri Jun  5 21:22:26 2015	(r284046)
@@ -2342,7 +2342,7 @@ verify_gla(struct vm *vm, int cpuid, uin
 		 * instruction
 		 */
 		if (vie->base_register == VM_REG_GUEST_RIP)
-			base += vie->num_valid;
+			base += vie->num_processed;
 	}
 
 	idx = 0;



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201506052122.t55LMR7u056330>