From: Henning Schild <henning.schild@siemens.com>
To: Claudius Heine <claudius.heine.ext@siemens.com>
Cc: <isar-users@googlegroups.com>, Claudius Heine <ch@denx.de>
Subject: Re: [PATCH 2/2] integrate ubifs image type
Date: Tue, 29 Jan 2019 10:10:50 +0100 [thread overview]
Message-ID: <20190129101050.7681afdc@md1za8fc.ad001.siemens.net> (raw)
In-Reply-To: <5fab2f59-9f15-713f-f2c8-15886800a3dc@siemens.com>
On Tue, 29 Jan 2019 09:16:02 +0100
Claudius Heine <claudius.heine.ext@siemens.com> wrote:
> Hi Henning,
>
> On 28/01/2019 18.22, Henning Schild wrote:
> > Am Mon, 28 Jan 2019 14:40:36 +0100
> > schrieb Claudius Heine <claudius.heine.ext@siemens.com>:
> >
> >> Hi Henning,
> >>
> >> On 28/01/2019 14.03, Henning Schild wrote:
> >>> Am Mon, 28 Jan 2019 13:28:21 +0100
> >>> schrieb "[ext] claudius.heine.ext@siemens.com"
> >>> <claudius.heine.ext@siemens.com>:
> >>>
> >>>> From: Claudius Heine <ch@denx.de>
> >>>>
> >>>> Signed-off-by: Claudius Heine <ch@denx.de>
> >>>> ---
> >>>> doc/user_manual.md | 1 +
> >>>> meta-isar/conf/local.conf.sample | 1 +
> >>>> .../multiconfig/qemuamd64-buster-ubifs.conf | 16 +++++++
> >>>> meta/classes/ubifs-img.bbclass | 42
> >>>> +++++++++++++++++++
> >>>> scripts/ci_build.sh | 1 + 5 files
> >>>> changed, 61 insertions(+) create mode 100644
> >>>> meta-isar/conf/multiconfig/qemuamd64-buster-ubifs.conf create
> >>>> mode 100644 meta/classes/ubifs-img.bbclass
> >>>>
> >>>> diff --git a/doc/user_manual.md b/doc/user_manual.md
> >>>> index c4fe42a..c9924ad 100644
> >>>> --- a/doc/user_manual.md
> >>>> +++ b/doc/user_manual.md
> >>>> @@ -476,6 +476,7 @@ Isar contains additional image type classes
> >>>> that can be used as reference:
> >>>> - `ext4-img`
> >>>> - `rpi-sdimg`
> >>>> - `targz-img`
> >>>> + - `ubifs-img`
> >>>> - `wic-img`
> >>>>
> >>>> ---
> >>>> diff --git a/meta-isar/conf/local.conf.sample
> >>>> b/meta-isar/conf/local.conf.sample index a671b20..9ea366c 100644
> >>>> --- a/meta-isar/conf/local.conf.sample
> >>>> +++ b/meta-isar/conf/local.conf.sample
> >>>> @@ -53,6 +53,7 @@ BBMULTICONFIG = " \
> >>>> hikey-stretch \
> >>>> qemuamd64-buster \
> >>>> qemuamd64-buster-tgz \
> >>>> + qemuamd64-buster-ubifs \
> >>>> rpi-jessie \
> >>>> "
> >>>>
> >>>> diff --git
> >>>> a/meta-isar/conf/multiconfig/qemuamd64-buster-ubifs.conf
> >>>> b/meta-isar/conf/multiconfig/qemuamd64-buster-ubifs.conf new
> >>>> file mode 100644 index 0000000..7c638b9 --- /dev/null
> >>>> +++ b/meta-isar/conf/multiconfig/qemuamd64-buster-ubifs.conf
> >>>> @@ -0,0 +1,16 @@
> >>>> +# This software is a part of ISAR.
> >>>> +# Copyright (c) Siemens AG, 2018
> >>>> +#
> >>>> +# SPDX-License-Identifier: MIT
> >>>> +
> >>>> +MACHINE ?= "qemuamd64"
> >>>> +
> >>>> +DISTRO ?= "debian-buster"
> >>>> +DISTRO_ARCH ?= "amd64"
> >>>> +
> >>>> +KERNEL_NAME ?= "amd64"
> >>>> +
> >>>> +MKUBIFS_ARGS ?= "-m 0x1000 -e 0x3e000 -c 1500"
> >>>> +IMAGE_TYPE ?= "ubifs-img"
> >>>> +
> >>>> +IMAGE_INSTALL += "sshd-regen-keys"
> >>>> diff --git a/meta/classes/ubifs-img.bbclass
> >>>> b/meta/classes/ubifs-img.bbclass new file mode 100644
> >>>> index 0000000..f5e17d3
> >>>> --- /dev/null
> >>>> +++ b/meta/classes/ubifs-img.bbclass
> >>>> @@ -0,0 +1,42 @@
> >>>> +# This software is a part of ISAR.
> >>>> +# Copyright (c) Siemens AG, 2018
> >>>> +
> >>>> +python() {
> >>>> + if not d.getVar("MKUBIFS_ARGS"):
> >>>> + bb.fatal("MKUBIFS_ARGS must be set")
> >>>> +}
> >>>> +
> >>>> +inherit image
> >>>
> >>> I dislike all this variable setting and additional mounting.
> >>>
> >>>> +UBIFS_IMAGE_FILE ?= "${IMAGE_FULLNAME}.ubifs.img"
> >>>> +
> >>>> +IMAGER_INSTALL += "mtd-utils"
> >>>> +
> >>>> +PP = "/home/builder/${PN}"
> >>>
> >>> That is already defined in another file in Isar. We should rather
> >>> find a common location and refactor.
> >>>
> >>>> +PP_DEPLOY = "${PP}/deploy"
> >>>> +PP_ROOTFS = "${PP}/rootfs"
> >>>
> >>> Similar comments might apply here.
> >>
> >> Do you think that refactoring should be part of this patchset?
> >
> > I am not sure, just wanted to write it down. On the one hand it
> > would be nice to solve all drive-by issues as we go. On the other
> > hand that will slow down development and impose on contributors.
> > So feel free to ignore my comments, the inconsistencies are not your
> > fault and asking you to fix them would be too much.
>
> Maybe we should open issues for all those inconsistencies and
> annoyances, so that they can be fixed or at least be documented.
We do not use issues, so we would need another way to track not fully
resolved discussions from the list. But i agree, it would be nice to
keep track of things in a structured way to fix them eventually.
Maybe a TODO.md?
> At some point making a complete code review (not just patches) on
> isar would be great, just to find and fix those.
I thought the same, maybe an Isar hackathon.
> >>>> +BUILDROOT = "${BUILDCHROOT_DIR}${PP}"
> >>>> +BUILDROOT_DEPLOY = "${BUILDCHROOT_DIR}${PP_DEPLOY}"
> >>>> +BUILDROOT_ROOTFS = "${BUILDCHROOT_DIR}${PP_ROOTFS}"
> >>>
> >>> Same here, and "buildroot" to me is something else ;).
> >>
> >> That is part of my plot to slowly merge bitbake, isar and
> >> buildroot ;)
> >
> > Good first step. We actually have buildroot isar recipes in
> > jailhouse-images, so this name can potentially get confusing for
> > real.
>
> I took this variable name from:
>
> https://github.com/ilbers/isar/blob/22f2387b3791a1c6c88b8c4424285d7788c3743c/meta/classes/dpkg-base.bbclass#L49
>
> So fixing this should be done as well then?
I guess in that case the name is OK.
Henning
> >
> >>>
> >>>> +do_ubifs_image[stamp-extra-info] = "${DISTRO}-${MACHINE}"
> >>>> +
> >>>> +# Generate ubifs filesystem image
> >>>> +do_ubifs_image() {
> >>>> + rm -f ${DEPLOY_DIR_IMAGE}/${UBIFS_IMAGE_FILE}
> >>>> +
> >>>> + buildchroot_do_mounts
> >>>> +
> >>>> + sudo flock ${MOUNT_LOCKFILE} -c ' \
> >>>> + mkdir -p ${BUILDROOT_DEPLOY} ${BUILDROOT_ROOTFS}
> >>>> + mount --bind ${DEPLOY_DIR_IMAGE} ${BUILDROOT_DEPLOY}
> >>>> + mount --bind ${IMAGE_ROOTFS} ${BUILDROOT_ROOTFS}
> >>>> + '
> >>>
> >>> I think this mounting should also be required by a proper ext4
> >>> class. On the other hand the wic one moves the image into the
> >>> final deploy folder.
> >>> Not sure what is best but maybe a good idea to go for "always
> >>> move" or "always mount".
> >>
> >> Hmm.. In OE this task would not to touch the deploy directory and
> >> *copy* all artifacts in a dedicated task AFAIK. But OE has a
> >> completely different image creation pipeline and infrastructure
> >> than isar, so who knows how things should be done here.
> >
> > We currently generate and deploy in one task. The fun thing is that
> > we need multiple commands and can probably end up in weird states
> > if we get interrupted or fail.
> >
> > Changing that general pattern is a discussion of its own, since it
> > would change the API.
> >
> > Staying in the current API i think the mount and generate directly
> > into deploy ... no more chmod/chown mv ... whatever. Would be the
> > cleanest way. We already have a reliable umount mechanism to clean
> > up if things go wrong.
> >
> > If you agree: Please include that deploydir/rootfs mount into the
> > common mounts and update your imager. I will do the same for wic
> > and maybe ext4.
>
> Ok. Will do that.
>
> >
> >> I am open to discussion about implementing some rules and
> >> documentation about how things should be done in isar.
> >
> > I assume that consistent references are equally efficient as
> > documentation, maybe more.
>
> Right!
>
> Cheers,
> Claudius
>
> >
> > Henning
> >
> >> Claudius
> >>
> >>>
> >>> Henning
> >>>
> >>>> + # Create ubifs image using buildchroot tools
> >>>> + sudo chroot ${BUILDCHROOT_DIR} /usr/sbin/mkfs.ubifs
> >>>> ${MKUBIFS_ARGS} \
> >>>> + -r "${PP_ROOTFS}"
> >>>> "${PP_DEPLOY}/${UBIFS_IMAGE_FILE}" +}
> >>>> +
> >>>> +addtask ubifs_image before do_build after do_copy_boot_files
> >>>> do_install_imager_deps diff --git a/scripts/ci_build.sh
> >>>> b/scripts/ci_build.sh index f3523e8..dcde0b4 100755
> >>>> --- a/scripts/ci_build.sh
> >>>> +++ b/scripts/ci_build.sh
> >>>> @@ -115,6 +115,7 @@ else
> >>>> multiconfig:qemuamd64-stretch:isar-image-base \
> >>>> multiconfig:qemuamd64-buster:isar-image-base \
> >>>> multiconfig:qemuamd64-buster-tgz:isar-image-base \
> >>>> + multiconfig:qemuamd64-buster-ubifs:isar-image-base \
> >>>> multiconfig:rpi-jessie:isar-image-base
> >>>> # qemu-user-static of <= buster too old to build that
> >>>> #multiconfig:qemuarm64-buster:isar-image-base
> >>>
> >>
> >
>
next prev parent reply other threads:[~2019-01-29 9:10 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-28 12:28 [PATCH 0/2] Integrate UBIFS support (+doc fixes) claudius.heine.ext
2019-01-28 12:28 ` [PATCH 1/2] doc/user: added targz-img to image type list claudius.heine.ext
2019-01-28 12:28 ` [PATCH 2/2] integrate ubifs image type claudius.heine.ext
2019-01-28 12:36 ` Jan Kiszka
2019-01-28 12:41 ` Claudius Heine
2019-01-28 13:03 ` Henning Schild
2019-01-28 13:40 ` Claudius Heine
2019-01-28 17:22 ` Henning Schild
2019-01-29 8:16 ` Claudius Heine
2019-01-29 8:24 ` Claudius Heine
2019-01-29 9:14 ` Henning Schild
2019-01-29 9:10 ` Henning Schild [this message]
2019-01-29 9:26 ` Claudius Heine
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=20190129101050.7681afdc@md1za8fc.ad001.siemens.net \
--to=henning.schild@siemens.com \
--cc=ch@denx.de \
--cc=claudius.heine.ext@siemens.com \
--cc=isar-users@googlegroups.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