This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: find_reloads_address_1 question
- To: weigand at immd1 dot informatik dot uni-erlangen dot de (Ulrich Weigand)
- Subject: Re: find_reloads_address_1 question
- From: Joern Rennecke <amylaar at onetel dot net dot uk>
- Date: Mon, 8 Oct 2001 00:04:18 +0100 (BST)
- Cc: amylaar at onetel dot net dot uk (Joern Rennecke), weigand at immd1 dot informatik dot uni-erlangen dot de (Ulrich Weigand), gcc at gcc dot gnu dot org, uweigand at de dot ibm dot com, hpenner at de dot ibm dot com
> However, if there is a canonical form for addresses that reload
> depends upon (across all architectures), then I guess combine
> would be at fault here for not properly canonifying addresses
> after substing into them ... I'll reexamine my test case and
> try to find out what's going on here.
Yes. Constants should be combined, and come last in a plus.
So in this case, combine should have generated an address like:
(plus (plus (reg) (reg)) (const))