general high-level discussion about spectrum
 help / color / mirror / Atom feed
From: Alyssa Ross <hi@alyssa.is>
To: discuss@spectrum-os.org
Cc: Demi Marie Obenour <demiobenour@gmail.com>
Subject: This Week in Spectrum, 2025-W28
Date: Mon, 14 Jul 2025 13:25:51 +0200	[thread overview]
Message-ID: <87tt3f6mkg.fsf@alyssa.is> (raw)

[-- Attachment #1: Type: text/plain, Size: 2388 bytes --]

This week, I finally finished the networking fixes I've kept mentioning
for the last few weeks.  Now, if the VMM for the network provider VM
exits for whatever reason, it's possible to recover without rebooting
the whole system.  I've also fixed running AppImages, which had recently
regressed, and added a test for that functionality to ensure it doesn't
happen again.  Along the way came various improvements to Spectrum's
integration test framework.  As I work on tests and the test framework,
each new test is easier than the last, and it's a great feeling!

Now, I'm looking into a Cloud Hypervisor bug that causes Spectrum's File
Chooser portal implementation to intermittently not work.  When that's
fixed, I of course already have a Spectrum regression test ready for
that too.  In this case, because it only doesn't work some of the time,
having a test that can (usually) reproduce it has made debugging so much
easier than trying to reproduce by hand.

I investigated a couple more newly introduced Nixpkgs regressions that
affected Spectrum.  One was a crosvm segfault on startup, caused by
minijail not taking into account that kernel headers could include
features not available on the running kernel.  They already had a fix,
and tagged a new release including it when I asked, but sadly it seems
that the regression will be reintroduced next time a new kernel version
is added[1]. :(

[1]: https://groups.google.com/a/chromium.org/g/minijail/c/7YR25MbzGG0

Demi continued working towards sound in Spectrum.  She sent another
revision of her patch to run PipeWire in guests[2], and sent a large
batch of security fixes upstream to PipeWire[3], some of which have
already been accepted[4].  She also continued her work fixing
unsoundness in Cloud Hyprevisor with a PR[5] to the rust-vmm vfio crate,
and started a discussion in the review thread for the specification of
the virtio-media device about its security properties[6].

[2]: https://spectrum-os.org/lists/archives/spectrum-devel/638beeaa-2351-4f51-81a6-bc58883930c2@gmail.com/
[3]: https://gitlab.freedesktop.org/pipewire/pipewire/-/merge_requests/2423
[4]: https://gitlab.freedesktop.org/pipewire/pipewire/-/merge_requests/2423#note_2993653
[5]: https://github.com/rust-vmm/vfio/pull/103
[6]: https://spectrum-os.org/lists/archives/spectrum-devel/67378956-e1dd-4d7f-9b66-97887b625936@gmail.com/

See you next week!

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 227 bytes --]

                 reply	other threads:[~2025-07-14 11:26 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=87tt3f6mkg.fsf@alyssa.is \
    --to=hi@alyssa.is \
    --cc=demiobenour@gmail.com \
    --cc=discuss@spectrum-os.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).