This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: ppc64 floating point usage [was Re: PPC64 Compiler bug !!]


>>>>> Eric Christopher writes:

Eric> Right, I can understand all of that, I worry that when you're choosing
Eric> register preferences it'll take an integer register in the unlikely case
Eric> that you have all of your floating point registers occupied giving you
Eric> an insn that can't match it's constraints. I thought * was normally used
Eric> when you have multiple alternatives and want to ignore one/multiple of
Eric> them for choosing register preferences. I just don't see a valid
Eric> constraint as a possible choice for register preference.

Eric> If I'm wrong here I apologize for the furor :)

	Quoting the GCC Internals Manual, Machine Descriptions, Operand
Constraints, Constraint Modifier Characters:

"* 
      Says that the following character should be ignored when choosing
      register preferences. * has no effect on the meaning of the
      constraint as a constraint, and no effect on reloading."

Reload will ensure that the instruction matches its constraints.

David


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]