This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: Your Jan 8th, 2001 rtlanal.c:note_stores change
- From: kenner at vlsi1 dot ultra dot nyu dot edu (Richard Kenner)
- To: rth at redhat dot com
- Cc: davem at redhat dot com, gcc-bugs at gcc dot gnu dot org, jakub at redhat dot com
- Date: Mon, 15 Apr 02 16:24:26 EDT
- Subject: Re: Your Jan 8th, 2001 rtlanal.c:note_stores change
> But we don't have a way to represent *a* call value, which is the point.
> The only thing we can represent is *the* call value.
Why do you think what we have at present is anything but "a" value?
I guess I meant "the whole value". When you have a SET, we are setting
the source to the value (the *whole* value) of the destination.
What do you think we buy by assuming *anything* about a call value?
We don't. However, we always find that lying in this sort of way
comes back to haunt us. I can't find any record of what bug I was
fixing, but I know I got an ICE due to having multiple registers in
different modes thought to be the same value.