This is the mail archive of the gcc@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: rs6000.md/altivec.md problem in setting of vector registers

From: David Edelsohn <dje at watson dot ibm dot com>
To: Dorit Naishlos <DORIT at il dot ibm dot com>, Dale Johannesen <dalej at apple dot com>
Cc: gcc at gcc dot gnu dot org
Date: Tue, 09 Mar 2004 13:46:43 -0500
Subject: Re: rs6000.md/altivec.md problem in setting of vector registers
References: <OF45AA3A9B.10F22048-ONC2256E4D.005E56AD-C2256E4E.0000D3E5@il.ibm.com> <086DC735-6E3B-11D8-8C0E-000A95D7CD40@apple.com>

> There are several cost-computation functions in rs6000.c, especially
> rs6000_register_move_cost looks promising.  Doesn't the RA call into there
> at some point?  Seems like it should.  That would be the right place, if
> it does.

	REGISTER_MOVE_COST only is provided the mode and register classes,
not the actual instruction, so it has no knowledge about the operands.
The costs is not wrong, but the RA is inquiring about the wrong mode.
Either RA needs to ask about the correct mode or insv needs to generate
patterns for vector types.

David

References:
- Re: rs6000.md/altivec.md problem in setting of vector registers
  - From: Dorit Naishlos
- Re: rs6000.md/altivec.md problem in setting of vector registers
  - From: Dale Johannesen

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