RFR : fix Assembler::prefix_and_encode for x86

[Doug Simon]

Ok, I’ll integrate your change - thanks!

> On Jan 13, 2015, at 11:09 AM, Igor Ignatyev <igor.ignatyev at oracle.com> wrote:
> 
> Hi Doug,
> 
> 03eda0a202e9[1] was integrated into Graal only, so this fix isn't needed for upstream.
> 
> -Igor
> 
> [1]
>> changeset:   16064:03eda0a202e9
>> user:        Christian Wimmer <christian.wimmer at oracle.com>
>> date:        Fri Jun 06 11:06:42 2014 -0700
>> files:       graal/com.oracle.graal.asm.amd64/src/com/oracle/graal/asm/amd64/AMD64Assembler.java src/cpu/x86/vm/assembler_x86.cpp src/cpu/x86/vm/assembler_x86.hpp
>> description:
>> Better fix for prefix byte for AMD64 byte instruction
> 
> On 01/13/2015 01:03 PM, Doug Simon wrote:
>> Are you sure this fix should be made in the Graal code base? Seems like an upstream patch to me.
>> 
>>> On Jan 13, 2015, at 10:58 AM, Igor Ignatyev <igor.ignatyev at oracle.com> wrote:
>>> 
>>> Hi all,
>>> 
>>> please review the fix for Assembler::prefix_and_encode method for 32bit:
>>> 16064:03eda0a202e9 changed signature of this method and implementation for 64bit (in assembler_x86.cpp), however 32bit inline version wasn't changed.
>>> 
>>> http://cr.openjdk.java.net/~iignatyev/graal/i586_prefix_and_encode/webrev.00/
>>> 
>>> --
>>> Thanks,
>>> Igor
>>