It turned out that the `--memory-size` given at install time is
not persistent. All later anita runs use an arch-dependent default
(32M for i386, so this might explain why it was never stable).
Assuming most machines have >1GiB RAM per processor core available,
we can also increase the runtime size further (better to waste some
RAM than to wait very long because it starts swapping things out).
We choose 512MiB for 64-bit targets and 256MiB for 32-bit ones.
However, we don't need that much for the initial installation step
and it also decides the size of the swap partition. So we use a
smaller size initially that's just enough to get us through the
installation quickly enough (192MiB & 128MiB).
Change-Id: I255c41aeb92cda29ed23a236017472982e839530
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/52484
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Add a set of Dockerfiles for build testing. If you have an x86 machine
and ~20GiB free disk space, run `make register` and `make -jxx` in
util/manibuilder and go eat some pizza. The former runs a privileged
docker container to set binfmt_misc up for qemu (read the code, don't
trust it).
Regarding the build targets, this is the original state of Manibuilder
as it was used to build-test `flashrom-1.0`. Some fixes to the frame-
work were applied, but fixups for the targets will be done in separate
patches to maintain their original state for reference.
Change-Id: I60863a5c7d70dde71486fccb66cb59b30ba4d982
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/c/flashrom/+/23005
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
