[llvm-dev] Wide load/store optimization question

upcfrost via llvm-dev llvm-dev at lists.llvm.org
Fri Jun 16 13:36:13 PDT 2017

Same here, my backend only has 64bit load/store. But i still use 64bit virt regs and expand/declare missing instructions by myself. 
I'll try looking into sparc backend, thanks. Also, only after writing this post I found a bunch of built-in transforms. Still trying to understand how to use those.
By the way, constraint-wise (alignment), is there any difference between virt regclass and regtuple?
Best regards,Petr

Отправлено со смартфона Samsung Galaxy.
-------- Исходное сообщение --------От: 陳韋任 <chenwj.cs97g at g2.nctu.edu.tw> Дата: 16.06.17  22:03  (GMT+02:00) Кому: tstellar at redhat.com Копия: Peter Bel <upcfrost at gmail.com>, LLVM Developers Mailing List <llvm-dev at lists.llvm.org> Тема: Re: [llvm-dev] Wide load/store optimization question 
One thing you can do is define a register class that is made up of register

tuples e.g. r0r1, r2r3, etc., and use that register class for the 64-bit

load/store instructions.  This will allow you to do the load/store

merging before register allocation without the register constraints.
​Our backend only support load/store for i64 type, hence i64 is not legal for us.I guess Peter's Epiphany arch has similar situation.​
IIRC, LLVM expand load i64 to two load i32. Right now, we have to customlowering load i64 to load v2i32, then map v2i32 to the tuple register (similarto Sparc backend). How can we use the tuple register for those two i32? ​Any existing example?
Wei-Ren Chen (陳韋任)
Homepage: https://people.cs.nctu.edu.tw/~chenwj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20170616/fef6253a/attachment.html>

More information about the llvm-dev mailing list