James Greenhalgh <james.greenha...@arm.com> writes:
> On Wed, Aug 30, 2017 at 04:34:40PM +0100, Richard Sandiford wrote:
>> Ping for a few AArch64 patches:
>> 
>>     [AArch64] Remove use of wider vector modes
>>     https://gcc.gnu.org/ml/gcc-patches/2017-08/msg01249.html
>>
>>     [AArch64] Rename cmp_result iterator
>>     https://gcc.gnu.org/ml/gcc-patches/2017-08/msg01250.html
>>
>>     [AArch64] Tighten address register subreg checks
>>     https://gcc.gnu.org/ml/gcc-patches/2017-08/msg01252.html
>
> I've reviewed these three, which all looked OK to me.
>
>>     [AArch64] Tweak aarch64_classify_address interface
>>     https://gcc.gnu.org/ml/gcc-patches/2017-08/msg01251.html
>
> I've reviewed this one with a comment on avoiding the duplication
> of the new enum values across the back end.

Thanks.

>>     [61/77] Use scalar_int_mode in the AArch64 port
>>     https://gcc.gnu.org/ml/gcc-patches/2017-07/msg00701.html
>> 
>>     [75/77] Use scalar_mode in the AArch64 port
>>     https://gcc.gnu.org/ml/gcc-patches/2017-07/msg00715.html
>
> Do you have a link to a short primer on what these changes do? I don't really
> have any objections to them if that's the direction the rest of the compiler
> is going in, but I also didn't really understand what was happening here.

https://gcc.gnu.org/ml/gcc-patches/2017-07/msg00639.html describes
the overall rationale for the changes.  When SVE is added, AArch64
will be the one backend for which the size of a normal machine_mode
is a variable rather than a constant.  But scalar modes will continue
to be constant-sized, so when dealing with them, it's better to use
the new scalar mode classes where possible.

Thanks,
Richard

Reply via email to