Note, the bugfix for the memory leak on a missing disk on Hyper-V
has been incorporated into 4.9 and 4.10 so has been removed from
our patch queue.
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
this removes all test targets from the top level makefile with the
exception of `make test` that now calls `$(MAKE) -C test`.
all tests now use `moby run` by instead of the older `./scripts`.
this removes the need for dedicated qemu/hyperkit test targets.
Signed-off-by: Dave Tucker <dt@docker.com>
This builds the dhcpcd package from a fixed Alpine 3.5 base and adds
the dhcpcd daemon from the Alpine base image. This should make the
build reproducible and allows us to use the git tree hash.
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
This makes it easier for CI to grab artifacts from the ephemeral build
VMs and provides a stable location so they can be used in subsequent
tests. For example, running the LTP tests.
Signed-off-by: Dave Tucker <dt@docker.com>
This also adds the cherry-picked commit f1c635b439a5c017 ("scsi: storvsc: Workaround
for virtual DVD SCSI version") from 4.11 for the 4.9 and 4.10 kernels. This commit
fixes a crash/memory leak on Hyper-V when no disk drives are attached and if one
boots of a CD-ROM drive.
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
The test is currently not hooked up, but it pulls the latest 4.9.x
kernel image, builds a simple kernel module against it and then creates
a test package containing the kernel module and a shell script.
The shell script prints out the 'modinfo' and then tries to 'insmod'
the module. It checks the output of 'dmesg' of the load succeeded.
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
This means that we limit to one place where we use network access
in building, and in future all other package builds can be deterministic.
Signed-off-by: Justin Cormack <justin.cormack@docker.com>
Drop `hyperkit` from the `linuxkit run` invocation, thus causing the linuxkit
tool to pick the platform's default backend (which is qemu on my Linux system,
which works better than hyperkit in this environment).
Signed-off-by: Ian Campbell <ian.campbell@docker.com>
- Use 'disk-size' for disk size
- Use 'disk' for the path to the disk image
- Use 'fw' for the path to the EFI firmware
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
Bring it into line with official images: remove `jq`, `sfdisk` which
are not needed; we used them for disk mounting but this is down standalone.
Add the tools for other filesystems, which are needed for `dm`, `btrfs`.
Signed-off-by: Justin Cormack <justin.cormack@docker.com>
Building debug kernels (with additional run time checks and debugging)
was broken a few commits back. This adds back support for building debug
kernels.
In addition, it builds and uploads debug kernels for selected kernel
series (4.9.x LTS and latest stable). The tag for these kernels has
a "_dbg" suffix.
Update documentation.
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
This fixes an issue introduced in the split of push and run.
Before this commit, an image could be pushed to a custom location from
the presence of the `CLOUDSDK_IMAGE_NAME` variable. The subsquent run
would fail as the name `test-ltp` or `test` had been hard coded in the
Make target. We now use a target-specific variable which is only set if
there is not an existing variable in the environment
Signed-off-by: Dave Tucker <dt@docker.com>
This fixes a regression where `linuxkit push gcp` no longer allows for a
custom destination to be provided via the `-img-name` flag or the
`CLOUDSDK_IMAGE_NAME` environment variable
Signed-off-by: Dave Tucker <dt@docker.com>
- Use a RUN command per artefact created
- Use WORKDIR to avoid "cd /linux" on every RUN command
- Copy all relevant build artefacts to /out
- Only create one additional layer in final stage
- Add System.map to output image
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@docker.com>
