Re: [PATCH 15/16][fold-const.c] Fix bigendian HFmode in native_interpret_real

[Jeff Law <law@redhat.com>]

On 07/08/2015 03:43 AM, Richard Biener wrote:
> On Wed, Jul 8, 2015 at 12:07 AM, Jeff Law <law@redhat.com> wrote:
>> On 07/07/2015 06:37 AM, Alan Lawrence wrote:
>>>
>>> As per https://gcc.gnu.org/ml/gcc-patches/2015-04/msg01346.html. Fixes
>>> FAIL of advsimd-intrinsics vcreate.c on aarch64_be-none-elf from
>>> previous patch.
>>>
>>> 15_native_interpret_real.patch
>>>
>>>
>>> commit e2e7ca148960a82fc88128820f17e7cbd14173cb
>>> Author: Alan Lawrence<alan.lawrence@arm.com>
>>> Date:   Thu Apr 9 10:54:40 2015 +0100
>>>
>>>       Fix native_interpret_real for HFmode floats on Bigendian with
>>> UNITS_PER_WORD>=4
>>>
>>>       (with missing space)
>>
>> OK with ChangeLog in proper form.
>
> Err - but now offset can become negative?  Shouldn't it rather error out
> before as it requires at least 4 bytes for big-endian?
I managed to convince myself the value wouldn't be negative when reviewing.

>
> That said - the whole thing looks it doesn't expect GET_MODE_SIZE < 4
> and your "fix" is just very obfuscated (if it really is a fix).
While I couldn't convince myself the function as a whole was prepared 
for smaller objects, I don't think Alan's patch made things worse.  One 
could argue the whole bloody thing ought to be rewritten though.

I'd also managed to convince myself the other instances of "3" weren't 
problematical.

jeff