Age | Commit message (Collapse) | Author |
|
Fixes: 1c98a5b3b36f ("bin/debvm-run: qemu (>> 1:8.0) provides symlinks for qemu-system-${debarch} as well as qemu-system-any")
|
|
as well as qemu-system-any
|
|
qemu makes heavy use of fd passing, so we better avoid user-passed fds.
|
|
We need the intermediate shell process to clean the temporary files with
the kernel and the initrd - unless we delete them before running qemu.
This method should help with killing a qemu e.g. using a timeout.
|
|
|
|
I really should have tested this part, but CI did.
Fixes: 954ba600ffb7 ("debvm-run: massively speed up tcg emulation of arm cpus")
|
|
pauth emulation is very intensive on the CPU and thus there is a
non-cryptographic alternative that provides a speedup of 3 to 4.
https://qemu-project.gitlab.io/qemu/system/arm/cpu-features.html#tcg-vcpu-features
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Reported-by: Emanuele Rocca <ema@debian.org>
Tested-by: Emanuele Rocca <ema@debian.org>
Closes: #1033643
|
|
|
|
We now extend /etc/profile to invoke setterm -resize when detecting an
interactive login on a serial console. This should fix 90% of wrongly
sized serial consoles.
|
|
Reported-by: Fabian Gruenbichler
|
|
These clearly benefit the amd64 case as well. Experience with the
virtio gpu is not as clear cut as it may crash qemu.
|
|
Reported-by: Arnd Bergmann <arnd@arndb.de>
|
|
|
|
Reported-by: Gioele Barabucci <gioele@svario.it>
Closes: #1030255
|
|
|
|
install and use multiarch kernels for sibling architectures
See merge request helmutg/debvm!28
|
|
We already pass it in the kvm case where it should become host and all
should be fine. However in the non-kvm case, the default gic can only
support 8 cores. If we want to go beyond that, we need a higher
gic-version. We couldn't measure a performance difference between max
and unset, so we'll just always pass max.
Thanks: Arnd Bergmann <arnd@arndb.de>
Thanks: Johannes Schauer Marin Rodrigues <josch@debian.org>
|
|
Reported-by: Arnd Bergmann <arnd@arndb.de>
|
|
In a multiarch setting, VMARCH shall describe the contained userland and
KERNELARCH shall describe the architecture of the contained kernel.
|
|
Fixes: 66b9374cc19c ("debvm-run: extract kernel image before inspecting it")
|
|
elf-arch may report armel and armhf as arm.
|
|
Reported-by: Arnd Bergmann <arnd@arndb.de>
|
|
We completely treat all 32bit arm as armhf now. Booting the rpi kernel
installed by debvm-create for armel will not work at all. The only sane
way to boot armel is to enable a multiarch kernel.
Refine the VMARCH guess from /bin/true using the file output for the
kernel image for the common cases of i386+amd64 and armel/armhf+arm64.
|
|
|
|
|
|
This was never tried, but autopkgtests indicate that smp limits also
apply to the native case.
|
|
|
|
It's not as simple as disabling the rng when pci is unavailable. For
m68k, we can select -machine virt and then we can have a
virtio-rng-device. Thus make that part configurable.
|
|
While the host version only works on kvm, max means the same and also
works on tcg.
Signed-off-by: Johannes Schauer Marin Rodrigues <josch@debian.org>
Signed-off-by: Helmut Grohne <helmut@subdivi.de>
|
|
Fixes: aee1eb3d1c28 ("debvm-run: no need for true randomness -- use /dev/urandom to be faster")
|
|
|
|
Now that we pass accel=kvm:tcg, there is no difference anymore.
|
|
-enable-kvm is equivalent to -M accel=kvm, but we allow a fallback to
tcg in case kvm is not available thus shifting this test into qemu.
Reported-by: Bastian Blank <waldi@debian.org>
Reported-by: Johannes Schauer Marin Rodrigues <josch@debian.org>
|
|
Fixes: bfaa3d124fab ("debvm-run: on arm64 there is no default machine -- set one")
|
|
When running on armhf natively, there also is no default machine, so we
should also pass -machine virt there. That likely affects all arms and
riscv64 as well.
|
|
|
|
|
|
|
|
Give an example of how to use it with ports and add a few
architecture-specific cases. Note that m68k and sparc64 do not actually
work, because they lack PCI.
|
|
Reported-by: Jochen Sprickerhof <git@jochen.sprickerhof.de>
|
|
Reported-by: Jochen Sprickerhof <git@jochen.sprickerhof.de>
|
|
|
|
Reported-by: Helmut Grohne <helmut@subdivi.de>
|
|
Reported-by: Helmut Grohne <helmut@subdivi.de>
|
|
|
|
Jochen Sprickerhof found that systemd interprets the TERM variable
according to our intended meaning and applies it to pid 1. It doesn't
apply it to serial-getty, though so we keep that code for now.
|
|
Resolve conflicts in debvm-create arising from adding --skip and moving
the functionality to be skipped to share/customize-*.sh.
|
|
* debvm-create will now parse a new kernel cmdline debvm.term and if
present will pass its value to agetty as TERM.
* debvm-run will now detect whether it is running in a terminal in
non-graphic mode and pass its environment TERM variable as debvm.term
to the kernel cmdline thus closing the loop.
|
|
The purpose of this change is adding support files to be referenced and
called from these tools. Those support files shall be located in
../share and this way of locating them shall work both in-source and
when installed.
|