There is an inconformity between qemu and kernel of memory alignment
check in memory hotplug. Both of qemu and kernel will do the start
address alignment check in memory hotplug. But it's 2M in qemu
while 128M in kernel. It leads to an issue when memory hotplug.
Currently, the kata image is a nvdimm device, which will plug into the VM as
a dimm. If another dimm is pluged, it will reside on top of that nvdimm.
So, the start address of the second dimm may not pass the alginment
check in kernel if the nvdimm size doesn't align with 128M.
There are 3 ways to address this issue I think:
1. fix the alignment size in kernel according to qemu. I think people
in linux kernel community will not accept it.
2. do alignment check in qemu and force the start address of hotplug
in alignment with 128M, which means there maybe holes between memory blocks.
3. obey the rule in user end, which means fix it in kata.
I think the second one is the best, but I can't do that for some reason.
Thus, the last one is the choice here.
Fixes: #1769
Signed-off-by: Jianyong Wu <jianyong.wu@arm.com>
While we introduced IMAGE_REGISTRY, we didn't actually update the
corresponding Dockerfiles to utilize it. Let's add
Fixes: #1622
Signed-off-by: Eric Ernst <eric.g.ernst@gmail.com>
This fixes the QAT driver version and provides a check early in the
building process to make sure the driver exists. It also provides
hints to users on how to fix themselves if the driver changes again.
Fixes: #1618
Signed-off-by: Eric Adams <eric.adams@intel.com>
Instead of having different VERSION files spread accross the project,
let's always use the one in the topsrcdir and remove all the others,
keeping only a synlink to the topsrcdir one.
Fixes: #1579
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Currently, musl toolchain installation on arm64 is just downloading from
a website. It's unsafe in case the website corrupts. So build musl
toolchain from source if it can't be downloaded.
Fixes: #1481
Signed-off-by: Jianyong Wu <jianyong.wu@arm.com>
Update the Intel QAT Dockerfile to work with the 2.0 repos, fix some
bugs with building Debian/Ubuntu rootfs, and update the latest QAT
driver. Updated copyright.
Fixes: #1419
Signed-off-by: Adams, Eric <eric.adams@intel.com>
Update the Intel QAT Dockerfile to work with the 2.0 repos, fix some
bugs with building Debian/Ubuntu rootfs, and update the latest QAT
driver. Updated copyright.
Fixes: #1419
Signed-off-by: Adams, Eric <eric.adams@intel.com>
For building rootfs with docker, glibc based rust target should be installed on ppc64le.
Additionally, protobuf-compiler would be required on ppc64le as it is not present by default.
Fixes: #1417
Signed-off-by: Amulya Meka <amulmek1@in.ibm.com>
Give the user chance to specify their own registry in event the default
provided are not accessible, desirable.
Fixes: #1393
Signed-off-by: Eric Ernst <eric.g.ernst@gmail.com>
- Install the required protobuf-compiler on Ubuntu
- Install correct libc Rust target (glibc on s390x)
- Do not skip Rust installation on s390x
Fixes: #1266
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.com>
Since there is no Rust target for musl on s390x, builds on s390x should use
glibc. This commit removes the "Cannot build Rust agent on s390x" restriction
in rootfs.sh and only installs musl in the build environment when it is
required.
Fixes: #1262
Signed-off-by: Jakob Naucke <jakob.naucke@ibm.com>
Include on the terms glossary some words about the agent init as
later on README its explain how the agent can be switched between
systemd and init.
Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
Changed the user-visible urls to point to the right Kata Containers
files/repositories.
Fixes#234
Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
To avoid hitting quota limit on docker.io, this changes the Fedora
and CentOS dockerfiles to pull the image from the projects registries.
Fixes#1324
Signed-off-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
The osbuilder scripts include a bunch of code for installing cmake, even
though cmake is never invoked. versions.yaml claims it's needed to build
grpc-rs, but that doesn't appear to be in our dependency graph.
Presumably, we used to need this, but don't any more. So, remove all cmake
references.
Fixes#1309
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This is not needed for Fedora, RHEL, and CentOS, but it is required when
using any other host OS. Having --security-opt apparmor=unconfined used
unconditionally is a no go as it'd break podman.
The reason this was only added when building for SUSE (as target distro)
was because debian and ubuntu condition would fall-through the switch to
the suse case (which makes me think that the fall-through was not
accidental).
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Falling through the switch cases in docker_extra_args() looks like a
typo and causes issues when building with podman, as `--security-opt
apparmor=unconfinded" shouldn't be passed if Apparmor is no enable on
the system.
Fixes: #1241
Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com>
Run static checks prior to building the agent.Checks
fail if run after since the compilation process
produces new rust code.
Signed-off-by: Archana Shinde <archana.m.shinde@intel.com>
Improve Kata Containers memory footprint by reducing debian
image size.
Without this change:
Debian image -> 256MB
With this change:
Debian image -> 128MB
Note: this change *will not* impact ubuntu image.
fixes#1188
Signed-off-by: Julio Montes <julio.montes@intel.com>
Define `before_starting_container` and `after_stopping_container`
functions, these functions run before and after the container that
builds the rootfs respectively.
Signed-off-by: Julio Montes <julio.montes@intel.com>
When building with AGENT_SOURCE_BIN pointing to an already built
kata-agent binary, the target directory needs to be created in the
rootfs tree.
Fixes#873
Signed-off-by: Ralf Haferkamp <rhafer@suse.com>
Since we always build musl kata-agent, there is no need to build
it inside a musl container. We can just build on the host and then
copy the binary to the target rootfs.
There are still a lot to clean up and it should be made so for ALL
target distros instead of just alpine. But this is at least working
for alpine first.
Signed-off-by: Peng Tao <bergwolf@hyper.sh>
Disable reflink when using DAX. Reflink is a xfs feature that cannot be
used together with DAX.
fixes kata-containers/osbuilder#456
fixes#577
Signed-off-by: Julio Montes <julio.montes@intel.com>
For building rust agent on ppc64le, the rust toolchain is built using
the LIBC implementation - gnu instead of musl.
Fixes: #481
Signed-off-by: Amulya Meka <amulmek1@in.ibm.com>
Don't modify /sbin/init on the build host when using command `AGENT_INIT="yes" ./rootfs.sh centos` to build rootfs.
Fixes: #430
Signed-off-by: liangxianlong <liang.xianlong@zte.com.cn>
tools/osbuilder/scripts/lib.sh was making references to agent and
runtime repositories to get golang, rust, cmake and musl versions.
Since runtime and agent repos are consolidating, we only need to
make reference to our versions.yaml in this (kata-containers) repo.
Fixes: #234.
Signed-off-by: Salvador Fuentes <salvador.fuentes@intel.com>
- Removes `CODE_OF_CONDUCT.md` and `CONTRIBUTING.md` from osbuilder
directory.
- Fixes a reference from `image-builder/README.md` to
`rootfs-builder/README.md`
- Updates the main `README.md` making a reference to the local
`tools/osbuilder/README.md`
Signed-off-by: Salvador Fuentes <salvador.fuentes@intel.com>
move all osbuilder files into `tools` directory to be able
to merge this into kata-containers repo.
Signed-off-by: Salvador Fuentes <salvador.fuentes@intel.com>
