This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH 2/N, MPX, x86_64] Intel MPX support in glibc for x86_64


On Wed, Apr 9, 2014 at 4:22 PM, Richard Henderson <rth@twiddle.net> wrote:
> On 04/09/2014 03:56 PM, H.J. Lu wrote:
>> +#  else
>> + byte 0x66,0x0f,0x1b,0x84,0x24,0xc0,0x02,0x00,0x00
>> + byte 0x66,0x0f,0x1b,0x8c,0x24,0xd0,0x02,0x00,0x00
>> + byte 0x66,0x0f,0x1b,0x84,0x24,0xe0,0x02,0x00,0x00
>> + byte 0x66,0x0f,0x1b,0x8c,0x24,0xf0,0x02,0x00,0x00
>> ^^^^^^^^^  This should be .byte.
>> +#  endif
>> +# endif
>> 3.
>> byte 0x66,0x0f,0x1b,0x84,0x24,0xc0,0x02,0x00,0x00
>>                                               ^^^^^^^^^^^^^^^^^^^^^^^ This
>> hardcods the offset.  Please add
>>
>> #if LR_BND_OFFSET != xxxx
>> #  error LR_BND_OFFSET is not xxx
>> #endif
>
> Wouldn't it be just as easy to do
>
>   .byte 0x66,0x0f,0x1b,0x84,0x24; .long LR_BND_OFFSET
>
> after all, you know how and where that offset is going to be encoded.

Sure.  We can do that after verifying LR_BND_OFFSET >= 0x80.


H.J.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]