From: "Daniel P. Berrange" <berrange@redhat.com>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] RFC: changing ROM bundling in tar dists for releases
Date: Thu, 31 Aug 2017 13:29:25 +0100 [thread overview]
Message-ID: <20170831122925.GJ17315@redhat.com> (raw)
A while back now I posted an RFC patch that changes qemu-X.Y.Z.tar.bz2
archive to *not* have any ROMs or 3rd party libs bundled, and create a
new dist qemu-bundled-X.Y.Z.tar.bz2 for the fully bundled dist:
http://lists.gnu.org/archive/html/qemu-devel/2017-04/msg03335.html
With 2.10 out of the way, and KVM Forum approaching, I figure now is
a reasonable time to revive the idea to see if it has support
The core issues I'm aiming to solve are:
- Distro vendors don't want the bundled ROMs / libs. They want to
fully build everything from source to ensure they are distributing
clean soure & builds in compliance with the licenses. Currently they
strip bundled bits from the build tree, but would prefer if the source
dist did not have them either.
- The qemu release dists get ever larger as we add more ROMS. Adding
EFI ROM builds for i386, x86_64, and aarch64 will make the dists
larger still.
There are the following options I see
1. Keep existing dist, and add a new minimal one
qemu-X.Y.Z.tar.bz2 - qemu source + bundled ROMS + libs
qemu-minimal-X.Y.Z.tar.bz2 - qemu source only
Least impact for current non-distro users, distros just switch.
2. Change existing dist, and add a new one with everything
qemu-X.Y.Z.tar.bz2 - qemu source only
qemu-full-X.Y.Z.tar.bz2 - qemu source + bundled ROMS + libs
Non-distro users need to download a different dist from what they
have known previously, but otherwise unchanged build process.
3. Change existing dist, and add a new one with bundled bits
qemu-X.Y.Z.tar.bz2 - qemu source only
qemu-addons-X.Y.Z.tar.bz2 - bundled ROMS + libs
Non-distro users need to manually download & install an
extra piece compared to now.
4. Change existing dist, never distribute bundled ROMs + libs at all
qemu-X.Y.Z.tar.bz2 - qemu source only
Only here for completeness, not a serious suggestion
5. Change nothing
Continue ignoring the problem I aim to solve.
My patch did option 2, but I really open to any of 1/2/3 without a
strong preference, as long as we can get some agreement on one. I
guess option 1 might be easiest to get agreement on given its minimal
impact.
Regards,
Daniel
--
|: https://berrange.com -o- https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org -o- https://fstop138.berrange.com :|
|: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
next reply other threads:[~2017-08-31 12:29 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-31 12:29 Daniel P. Berrange [this message]
2017-08-31 12:49 ` [Qemu-devel] RFC: changing ROM bundling in tar dists for releases Peter Maydell
2017-09-01 8:31 ` Gerd Hoffmann
2017-09-01 9:24 ` Daniel P. Berrange
2017-09-01 9:49 ` Gerd Hoffmann
2017-09-01 11:00 ` Peter Maydell
2017-09-01 14:11 ` Gerd Hoffmann
2017-09-01 15:13 ` Daniel P. Berrange
2017-09-01 17:45 ` Gerd Hoffmann
2017-08-31 15:47 ` Philippe Mathieu-Daudé
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=20170831122925.GJ17315@redhat.com \
--to=berrange@redhat.com \
--cc=qemu-devel@nongnu.org \
/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;
as well as URLs for NNTP newsgroup(s).