This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Use of register 17 on x86
- To: rth at redhat dot com
- Subject: Re: Use of register 17 on x86
- From: kenner at vlsi1 dot ultra dot nyu dot edu (Richard Kenner)
- Date: Sun, 18 Feb 01 18:26:08 EST
- Cc: gcc at gcc dot gnu dot org
> Also, why isn't x86 using CC0 anymore? That would seem to be the most
> efficient, both in terms of generated code and compilation speed.
It's is impossible to represent a comparison followed by multiple
cmove instructions.
True.
Given the improvements in code generation with the new backend, I
think that gives lie to CC0 being more efficient in terms of generated
code.
Maybe, but the downside is that you have the explicit clobber of the register
in many arithmetic insns, making it harder to optimize them because you
can't just insert one without knowing if reg 17 is dead there.