This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: SLOW_UNALIGNED_ACCESS granularity
- To: kenner at vlsi1 dot ultra dot nyu dot edu (Richard Kenner)
- Subject: Re: SLOW_UNALIGNED_ACCESS granularity
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Wed, 29 Dec 1999 13:19:18 -0700
- cc: dje at watson dot ibm dot com, gcc-patches at gcc dot gnu dot org
- Reply-To: law at cygnus dot com
In message <9912292015.AA16454@vlsi1.ultra.nyu.edu>you write:
> I tried looking at adding an ALIGN parameter, but it is not always
> clear what value to use and I did not have a good instinct about what
> to use as a default.
>
> I don't follow. The default would simply *ignore* the ALIGN parameter.
>
> Or perhaps we mean different things:
I believe you and David where talking about two different things. I believe
David was concerned about what value to pass to the macro invocation,
particularly when there is no obvious value lying around.
> is the highest alignment it's known to have. If the caller knows
> nothing about the alignment, it will call it with BITS_PER_UNIT, since
> that's the highest alignment it's known to have.
Agreed. At least that's safe until we have to support a bit addressable
machine :-)
jeff