boot-utils: Add support for booting Debian images#35
Conversation
'-cpu max' is treated as '-cpu host' when KVM is enabled. Link: https://gitlab.com/qemu-project/qemu/-/commit/bab52d4bba3f22921a690a887b4bd0342f2754cd Signed-off-by: Nathan Chancellor <nathan@kernel.org>
* -machine "virtualiation=true" allows the virtual machine to spawn virtual guests, which can be useful for verifying KVM without an actual arm64 piece of hardware. This only works when KVM is not being used though so it is only enabled when not on an arm64 host. * -machine "virt,gic-version=max" will use the newest Generic Interrupt Controller available. These were taken from Will Deacon's QEMU guide and it was tested on Debian Buster with Debian's QEMU package and a self compiled tip of tree version. Link: https://mirrors.edge.kernel.org/pub/linux/kernel/people/will/docs/qemu/qemu-arm64-howto.html Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Purely aesthetic. Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
There are times where there is a need to have a fuller environment for testing, rather than the simple Busybox based rootfs that is normally used. Introduce a Debian build script, which uses debootstrap and qemu-user-static to generate a bootable rootfs image. boot-qemu.sh is modified accordingly to support this image if the user wants it. Signed-off-by: Nathan Chancellor <nathan@kernel.org>
aa6f659 to
381b188
Compare
|
I added support for Additionally, I fixed an issue where |
| APPEND_STRING+="console=ttyAMA0 " | ||
| # https://lists.nongnu.org/archive/html/qemu-discuss/2018-08/msg00030.html | ||
| # VFS: Cannot open root device "vda" or unknown-block(0,0): error -6 | ||
| ${DEBIAN} && HIGHMEM=,highmem=off |
There was a problem hiding this comment.
should the value assigned to HIGHMEM be a quoted string?
There was a problem hiding this comment.
shellcheck does not complain but I can update that if you feel it is worthwhile.
| @@ -1,2 +1 @@ | |||
| qemu-binaries/ | |||
| rootfs.cpio | |||
There was a problem hiding this comment.
doesn't this mean then that each time we decompress an initrd, that the directory is now dirty?
There was a problem hiding this comment.
No, I just moved it to images/.gitignore.
|
Looks cool! |
Reported-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Suggested-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Requested-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Suggested-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Suggested-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
765951c to
ce8b8a1
Compare
nickdesaulniers
left a comment
There was a problem hiding this comment.
cant wait to try it out! 🆒 🧔
|
Thank you both for the review! |
Each of these commits should be self explanatory.