Clang and AArch64 backend patches to support arm v8 Neon

Jiangning Liu liujiangning1 at
Wed Jul 31 02:09:45 PDT 2013

Hi Ana,

The patch looks good for me.

The MCHammer test result shows encoding/decoding and assemble/disassemble
can all pass for the following three instruction classes,

* Vector Arithmetic
* Vector Immediate
* Vectore Pairwise Arithmetic

There is a typo for the last one in your commit log. Remove 'e' from

BTW, it would be better to add tag [PATCH] in your subject, because all of
patches in this mail list are with this tag.


2013/7/31 Ana Pazos <apazos at>

> Hello Tim and LLVM reviewers,
> I have updated the Clang and aarch64 backend patches to support arm v8 Neon
> to merge cleanly with the latest Clang/LLVM code.
> You have reviewed these patches before, but they have not been merged yet.
> We would like to have them merged now.
> Tim (and anyone else interested), can you do the review of the rebased
> patches and merge them?
> Thanks,
> Ana.
> -----Original Message-----
> From: Tim Northover [mailto:t.p.northover at]
> Sent: Thursday, January 31, 2013 10:52 AM
> To: Ana Pazos
> Cc: cfe-commits at; llvm-commits; anapazos at
> Subject: Re: clang and aarch64 backend changes to support Neon
> Hi Ana,
> Thanks for the patches.
> > Here are the initial clang and aarch64 backend changes to support Neon.
> Just a note for others that I believe these are the changes that will bring
> the AArch64 backend as committed in line with what has already been
> reviewed
> so far. I don't think they need separate technical review (though obviously
> if anyone *does* have comments...).
> > If everyone is ok with these changes, can you Tim please go ahead and
> > commit them?
> As soon as I'm allowed, I will.
> Tim.
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the cfe-commits mailing list