From: Demi Marie Obenour <demiobenour@gmail.com>
To: Alyssa Ross <hi@alyssa.is>
Cc: Spectrum OS Development <devel@spectrum-os.org>
Subject: Re: [PATCH 1/2] Create Nix derivation for building verity images
Date: Fri, 7 Nov 2025 14:24:03 -0500 [thread overview]
Message-ID: <e2ee3782-d72d-4cd2-a271-9e8d7df2e3dd@gmail.com> (raw)
In-Reply-To: <875xbnxu5c.fsf@alyssa.is>
[-- Attachment #1.1.1: Type: text/plain, Size: 6269 bytes --]
On 11/6/25 06:44, Alyssa Ross wrote:
> Demi Marie Obenour <demiobenour@gmail.com> writes:
>
>> On 11/6/25 05:20, Alyssa Ross wrote:
>>> Demi Marie Obenour <demiobenour@gmail.com> writes:
>>>
>>>> This gets rid of a lot of duplicated code and allows building the verity
>>>> roothash and superblock only when needed. It also removes a hack used
>>>> to work around make limitations. Furthermore,
>>>> 'veritysetup --root-hash-file' is used to avoid an awk script.
>>>>
>>>> Signed-off-by: Demi Marie Obenour <demiobenour@gmail.com>
>>>> ---
>>>> nix-shell --pure --run 'make run' in host/initramfs fails. This is a
>>>> preexisting bug and I will send a separate patch for it.
>>>> ---
>>>> host/initramfs/Makefile | 25 +++++--------------------
>>>> host/initramfs/shell.nix | 4 +++-
>>>> host/rootfs/Makefile | 24 +++++-------------------
>>>> host/rootfs/shell.nix | 3 +++
>>>> host/verity.nix | 19 +++++++++++++++++++
>>>> lib/common.mk | 1 -
>>>> pkgs/default.nix | 1 +
>>>> release/live/Makefile | 26 +++++---------------------
>>>> release/live/default.nix | 4 +++-
>>>> 9 files changed, 44 insertions(+), 63 deletions(-)
>>>
>>>> diff --git a/host/rootfs/Makefile b/host/rootfs/Makefile
>>>> index 00d125774bb7b98736d0928c69cb307740cee034..bb602e2745fb5873204f453b35fc529c5c96f64a 100644
>>>> --- a/host/rootfs/Makefile
>>>> +++ b/host/rootfs/Makefile
>>>> @@ -82,25 +82,11 @@ clean:
>>>> rm -rf build
>>>> .PHONY: clean
>>>>
>>>> -# veritysetup format produces two files, but Make only (portably)
>>>> -# supports one output per rule, so we combine the two outputs then
>>>> -# define two more rules to separate them again.
>>>> -build/rootfs.verity: $(dest)
>>>> - $(VERITYSETUP) format $(dest) build/rootfs.verity.superblock.tmp \
>>>> - | awk -F ':[[:blank:]]*' '$$1 == "Root hash" {print $$2; exit}' \
>>>> - > build/rootfs.verity.roothash.tmp
>>>> - cat build/rootfs.verity.roothash.tmp build/rootfs.verity.superblock.tmp \
>>>> - > $@
>>>> - rm build/rootfs.verity.roothash.tmp build/rootfs.verity.superblock.tmp
>>>> -build/rootfs.verity.roothash: build/rootfs.verity
>>>> - head -n 1 build/rootfs.verity > $@
>>>> -build/rootfs.verity.superblock: build/rootfs.verity
>>>> - tail -n +2 build/rootfs.verity > $@
>>>>
>>>> -build/live.img: ../../scripts/format-uuid.sh ../../scripts/make-gpt.sh ../../scripts/sfdisk-field.awk build/rootfs.verity.superblock build/rootfs.verity.roothash $(dest)
>>>> +build/live.img: ../../scripts/format-uuid.sh ../../scripts/make-gpt.sh ../../scripts/sfdisk-field.awk $(ROOT_FS_VERITY) $(ROOT_FS_VERITY_ROOTHASH) $(dest)
>>>> ../../scripts/make-gpt.sh $@.tmp \
>>>> - build/rootfs.verity.superblock:verity:$$(../../scripts/format-uuid.sh "$$(dd if=build/rootfs.verity.roothash bs=32 skip=1 count=1 status=none)") \
>>>> - $(dest):root:$$(../../scripts/format-uuid.sh "$$(head -c 32 build/rootfs.verity.roothash)")
>>>> + "$$ROOT_FS_VERITY:verity:$$(../../scripts/format-uuid.sh "$$(dd "if=$$ROOT_FS_VERITY_ROOTHASH" bs=32 skip=1 count=1 status=none)")" \
>>>> + $(dest):root:$$(../../scripts/format-uuid.sh "$$(head -c 32 "$$ROOT_FS_VERITY_ROOTHASH")")
>>>> mv $@.tmp $@
>>>>
>>>> debug:
>>>> @@ -110,7 +96,7 @@ debug:
>>>> $(VMLINUX)
>>>> .PHONY: debug
>>>>
>>>> -run: build/live.img $(EXT_FS) build/rootfs.verity.roothash
>>>> +run: build/live.img $(EXT_FS) $(ROOT_FS_VERITY_ROOTHASH)
>>>> @set -x && \
>>>> ext="$$(mktemp build/spectrum-rootfs-extfs.XXXXXXXXXX.img)" && \
>>>> truncate -s 10G "$$ext" && \
>>>> @@ -131,7 +117,7 @@ run: build/live.img $(EXT_FS) build/rootfs.verity.roothash
>>>> -device virtconsole,chardev=virtiocon0 \
>>>> -drive file=build/live.img,if=virtio,format=raw,readonly=on \
>>>> -drive file=/proc/self/fd/3,if=virtio,format=raw \
>>>> - -append "earlycon console=hvc0 roothash=$$(< build/rootfs.verity.roothash) intel_iommu=on nokaslr" \
>>>> + -append "earlycon console=hvc0 roothash=$$(< "$$ROOT_FS_VERITY_ROOTHASH") intel_iommu=on nokaslr" \
>>>> -device virtio-keyboard \
>>>> -device virtio-mouse \
>>>> -device virtio-gpu \
>>>> diff --git a/host/rootfs/shell.nix b/host/rootfs/shell.nix
>>>> index 1bf61bebf418333624e799cc8ca231f5783206f4..f16e4905adfbc8faebde19d0a1364ad9df90219b 100644
>>>> --- a/host/rootfs/shell.nix
>>>> +++ b/host/rootfs/shell.nix
>>>> @@ -5,6 +5,7 @@
>>>> import ../../lib/call-package.nix (
>>>> { callSpectrumPackage, rootfs, pkgsStatic, srcOnly, stdenv
>>>> , btrfs-progs, cryptsetup, jq, netcat, qemu_kvm, reuse, util-linux
>>>> +, verity
>>>> }:
>>>>
>>>> rootfs.overrideAttrs (
>>>> @@ -20,5 +21,7 @@ rootfs.overrideAttrs (
>>>> KERNEL = "${passthru.kernel}/${stdenv.hostPlatform.linux-kernel.target}";
>>>> LINUX_SRC = srcOnly passthru.kernel.configfile;
>>>> VMLINUX = "${passthru.kernel.dev}/vmlinux";
>>>> + ROOT_FS_VERITY = "${verity}/rootfs.verity.superblock";
>>>> + ROOT_FS_VERITY_ROOTHASH = "${verity}/rootfs.verity.roothash";
>>>> };
>>>> })) (_: {})
>>>
>>> Surely this would break interactive development of the rootfs?
>>> If I'm in a Nix shell, and make a change to any part of the rootfs, the
>>> verity data in the environment will be out of date. I'd have to leave
>>> and re-enter the Nix shell after /any/ change, waiting for an evaluation
>>> each time, as opposed to the current situation where that's only
>>> necessary when modifying Nix code or other Spectrum components.
>>
>> It would. Are there alternatives you can recommend? I don't want the
>> updater and the installer to have to use two different copies.
>
> Have the host/rootfs derivation install the verity files alongside the
> rootfs image. Then host/rootfs/Makefile is the single place we generate
> the verity images, and it will still be regenerated by make when in a
> Nix shell.
Is it okay to instead remove dm-verity protection for the verity images?
Given that we discussed using virtiofs for live development, I don't think
the verity protection is necessary. It also slows down live development.
--
Sincerely,
Demi Marie Obenour (she/her/hers)
[-- Attachment #1.1.2: OpenPGP public key --]
[-- Type: application/pgp-keys, Size: 7253 bytes --]
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2025-11-07 19:24 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-05 22:33 [PATCH 0/2] Move verity and EFI creation to separate Nix derivations Demi Marie Obenour
2025-11-05 22:33 ` [PATCH 1/2] Create Nix derivation for building verity images Demi Marie Obenour
2025-11-06 10:20 ` Alyssa Ross
2025-11-06 10:55 ` Demi Marie Obenour
2025-11-06 11:44 ` Alyssa Ross
2025-11-07 19:24 ` Demi Marie Obenour [this message]
2025-11-13 11:32 ` Alyssa Ross
2025-11-05 22:33 ` [PATCH 2/2] Move UKI creation to a separate derivation Demi Marie Obenour
2025-11-08 4:47 ` [PATCH v2 0/2] Move verity and EFI creation to separate Nix derivations Demi Marie Obenour
2025-11-08 4:47 ` [PATCH v2 1/2] Build verity images in rootfs Nix derivation Demi Marie Obenour
2025-11-08 4:47 ` [PATCH v2 2/2] Move UKI creation to a separate derivation Demi Marie Obenour
2025-11-12 0:59 ` [PATCH v3 0/2] Move verity and EFI creation to separate Nix derivations Demi Marie Obenour
2025-11-12 0:59 ` [PATCH v3 1/2] Build verity images in rootfs Nix derivation Demi Marie Obenour
2025-11-13 11:46 ` Alyssa Ross
2025-11-13 22:33 ` Demi Marie Obenour
2025-11-14 11:53 ` Alyssa Ross
2025-11-12 0:59 ` [PATCH v3 2/2] Move UKI creation to a separate derivation Demi Marie Obenour
2025-11-13 11:57 ` Alyssa Ross
2025-11-13 22:42 ` Demi Marie Obenour
2025-11-14 11:58 ` Alyssa Ross
2025-11-19 8:15 ` [PATCH v4 0/2] Move verity and EFI creation to separate Nix derivations Demi Marie Obenour
2025-11-19 8:15 ` [PATCH v4 1/2] Build verity images in rootfs Nix derivation Demi Marie Obenour
2025-11-25 12:27 ` Alyssa Ross
2025-11-25 12:31 ` Alyssa Ross
2025-11-19 8:15 ` [PATCH v4 2/2] Move UKI creation to a separate derivation Demi Marie Obenour
2025-11-22 1:21 ` [PATCH v5 0/2] Move verity and EFI creation to separate Nix derivations Demi Marie Obenour
2025-11-22 1:21 ` [PATCH v5 1/2] Build verity images in rootfs Nix derivation Demi Marie Obenour
2025-11-25 12:34 ` Alyssa Ross
2025-11-22 1:21 ` [PATCH v5 2/2] Move UKI creation to a separate derivation Demi Marie Obenour
2025-11-25 12:41 ` Alyssa Ross
2025-11-26 19:10 ` [PATCH v6 0/2] Move verity and EFI creation to separate Nix derivations Demi Marie Obenour
2025-11-26 19:10 ` [PATCH v6 1/2] Build verity images in rootfs Nix derivation Demi Marie Obenour
2025-11-27 19:23 ` Alyssa Ross
2025-11-26 19:10 ` [PATCH v6 2/2] Move UKI creation to a separate derivation Demi Marie Obenour
2025-11-28 10:47 ` Alyssa Ross
2025-11-28 19:27 ` Demi Marie Obenour
2025-11-28 11:02 ` Alyssa Ross
2025-11-28 19:25 ` Demi Marie Obenour
2025-11-28 20:12 ` Alyssa Ross
2025-11-26 18:58 ` [PATCH v5 0/2] Move verity and EFI creation to separate Nix derivations Demi Marie Obenour
2025-11-26 18:58 ` [PATCH v5 1/2] Build verity images in rootfs Nix derivation Demi Marie Obenour
2025-11-26 18:58 ` [PATCH v5 2/2] Move UKI creation to a separate derivation Demi Marie Obenour
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=e2ee3782-d72d-4cd2-a271-9e8d7df2e3dd@gmail.com \
--to=demiobenour@gmail.com \
--cc=devel@spectrum-os.org \
--cc=hi@alyssa.is \
/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.
Code repositories for project(s) associated with this public inbox
https://spectrum-os.org/git/crosvm
https://spectrum-os.org/git/doc
https://spectrum-os.org/git/mktuntap
https://spectrum-os.org/git/nixpkgs
https://spectrum-os.org/git/spectrum
https://spectrum-os.org/git/ucspi-vsock
https://spectrum-os.org/git/www
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).