From: Ian Lance Taylor <ian@airs.com>
To: neroden@twcny.rr.com
Cc: gcc@gcc.gnu.org, gdb@sources.redhat.com, binutils@sources.redhat.com
Subject: Re: Complicated config puzzle (m68k-apollo-bsd, v810, aix4.3+ maintainers please reply)
Date: Sat, 09 Nov 2002 18:57:00 -0000 [thread overview]
Message-ID: <m33cqaqgh6.fsf@gossamer.airs.com> (raw)
In-Reply-To: <20021109234729.GA15157@doctormoo>
Nathanael Nerode <neroden@twcny.rr.com> writes:
> I'm trying to untangle the maze that is toplevel configuration. I need
> some advice as to desired behavior.
>
> For the *host* m68k-apollo-bsd, currently
> host=build: sets BISON=yacc
> host!=build: sets BISON=bison
>
> BISON represents a build-side tool, correct?
> So we should set BISON=yacc whenever this is the *build* machine, right?
Yes.
(But does anybody really care about m68k-apollo-bsd any more?)
> For the *target* aix >= 4.3, currently
> host=build: sets AR_FOR_TARGET=ar -X32_64
> sets NM_FOR_TARGET=nm -B -X32_64
> host!=build: sets AR_FOR_TARGET=${target_alias}-ar
> sets NM_FOR_TARGET=${target_alias}-nm
>
> This just doesn't seem right at all. Under what circumstances do we
> want these flags? It looks like they're always valid for GNU ar, GNU
> nm, AIX ar, and AIX nm; I'm guessing we want them even in the Canadian
> case.
>
> I'm further guessing that we want ${target_alias}-ar whenever
> build!=target, with the flags added.
>
> I can do this. :-)
I believe the -X32_64 option only matters when using the native tools,
not when using the GNU tools. When host != build, we can probably
assume we are using the GNU tools. So we need the flags when host =
build = target = aix >= 4.3. In other cases, we don't need the flags.
It probably doesn't hurt to add them when target = aix >= 4.3.
> For the *target* v810*, currently
> host=build: sets CC_FOR_TARGET=ca732 -ansi
> sets AS_FOR_TARGET=as732
> sets AR_FOR_TARGET=ar732
> host!=build: sets CC_FOR_TARGET = ${target_alias}-gcc
> sets AS_FOR_TARGET = ${target_alias}-as
> sets AR_FOR_TARGET = ${target_alias}-ar
>
> This also looks wrong (very, very wrong), but I'm not sure what to do
> with it. I'll hazard a guess that the '732' compilers should be used
> when build=target=v810* and not otherwise? But I could easily be wrong.
I don't know for sure , but I'll bet this is another case where it
only matters when host = target = build and we are not using the GNU
tools.
Ian
prev parent reply other threads:[~2002-11-10 2:57 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-11-09 15:47 Nathanael Nerode
2002-11-09 18:57 ` Ian Lance Taylor [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=m33cqaqgh6.fsf@gossamer.airs.com \
--to=ian@airs.com \
--cc=binutils@sources.redhat.com \
--cc=gcc@gcc.gnu.org \
--cc=gdb@sources.redhat.com \
--cc=neroden@twcny.rr.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox