From: Mike McDonald <mikemac@mikemac.com>
To: Jun Sun <jsun@mvista.com>
Cc: linux-mips@oss.sgi.com
Subject: Re: Cross compiling RPMs
Date: Fri, 26 Jan 2001 13:11:35 -0800 [thread overview]
Message-ID: <200101262111.NAA13006@saturn.mikemac.com> (raw)
In-Reply-To: Your message of "Fri, 26 Jan 2001 12:51:31 PST." <20010126125131.G9325@mvista.com>
>Date: Fri, 26 Jan 2001 12:51:31 -0800
>From: Jun Sun <jsun@mvista.com>
>To: Florian Lohoff <flo@rfc822.org>
>Subject: Re: Cross compiling RPMs
>The "noarch" means the installed target is arch-independent. The
>standard setup in mvista CDK is to let target boot from NFS root fs,
>where NFS host can be linux/i386, Linux/ppc and Sun/Sparc (perhaps
>Win/i386 as well, I am not sure). Those packages are meant to be
>installed to all those hosts, and therefore "noarch" :-0.
Hmm, I would have thought they should be designated for the type of
system they were instead to run on. The fast you're installing them
into an NFS root on some other machine shouldn't change that. Can't
any ole rpm be forced to install on some random NFS server? Then by
your reasoning, all rpms would be noarch, wouldn't they?
>Native compiling is easy. Cross-compiling is cool. :-)
>
>Well, not exactly. When you are dealing with head-less, disk-less
>memory-scarce embedded devices with ad hoc run-time environments,
>cross-compiling is your only choice.
>
>Jun
Precisely! In our case, we get drops from various contractors who
are doing developement/porting to a wide variety of platforms. (So far
we have i386, mipsel, arm, and sh3. No alpha or sparc yet.) We'll get
multiple drops from the contractors over time. We need to be able to
1) rebuild the binaries from the supplied sources (some vendors have
delivered binaries that did NOT come from the sources they claimed!),
2) build a test suite for that drop and 3) build an initial ramdisk,
bootable CD, or NFS root dir to test the drop. Building the test
environment will include some subset (usually a real small subset) of
the whole drop but we still need to be able to rebuild everything.
Most of these systems we're dealing with have no native compiling
capability, so cross compiling is the only choice.
And then sometimes we get tarballs instead of rpms, but that's a
different can of worms.
Mike McDonald
mikemac@mikemac.com
next prev parent reply other threads:[~2001-01-26 21:11 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-01-26 18:15 Cross compiling RPMs Mike McDonald
2001-01-26 18:37 ` Pete Popov
2001-01-26 19:27 ` Mike McDonald
2001-01-26 19:39 ` Pete Popov
2001-01-26 19:47 ` Mike McDonald
2001-01-26 20:23 ` Florian Lohoff
2001-01-26 20:51 ` Jun Sun
2001-01-26 21:11 ` Mike McDonald [this message]
2001-01-27 0:31 ` Pete Popov
2001-01-26 21:14 ` Mike McDonald
2001-01-27 10:52 ` Karel van Houten
2001-01-27 10:52 ` Karel van Houten
2001-01-27 22:57 ` Mike McDonald
2001-01-28 12:10 ` Ralf Baechle
2001-01-28 17:45 ` Mike McDonald
2001-01-29 0:05 ` Ralf Baechle
2001-01-29 15:23 ` Maciej W. Rozycki
2001-01-27 7:42 ` Maciej W. Rozycki
2001-01-27 18:50 ` Ralf Baechle
2001-01-28 18:30 ` Geert Uytterhoeven
2001-01-28 19:46 ` Pim van Riezen
2001-01-28 19:46 ` Pim van Riezen
2001-01-29 0:02 ` Ralf Baechle
2001-01-29 8:44 ` Geert Uytterhoeven
2001-01-29 21:54 ` Ralf Baechle
2001-01-29 15:57 ` Maciej W. Rozycki
2001-01-30 0:12 ` Mike McDonald
2001-01-30 9:46 ` Maciej W. Rozycki
2001-01-28 18:27 ` Geert Uytterhoeven
2001-01-26 20:28 ` Jun Sun
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=200101262111.NAA13006@saturn.mikemac.com \
--to=mikemac@mikemac.com \
--cc=jsun@mvista.com \
--cc=linux-mips@oss.sgi.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.