WIP: switch composefs-boot to use bootc-kernel-cmdline#180
WIP: switch composefs-boot to use bootc-kernel-cmdline#180jeckersb wants to merge 1 commit intocomposefs:mainfrom
Conversation
|
requires bootc parts - bootc-dev/bootc#1611 |
cgwalters
left a comment
cgwalters
left a comment
There was a problem hiding this comment.
Looks sane to me overall, though I am a bit uncertain if bootc will continue to use these portions of composefs-boot.
crates/composefs-boot/Cargo.toml
Outdated
|
|
||
| [dependencies] | ||
| anyhow = { version = "1.0.87", default-features = false } | ||
| bootc-kernel-cmdline = { git = "https://github.com/jeckersb/bootc", branch = "cmdline" } |
There was a problem hiding this comment.
This is a git-repository-level circular dependency - that's what I was trying to get at earlier. It works fine at the Rust level of course because we're only compiling an individual crate from that git repo.
OTOH doing this will make it easier to share private dependencies between these two projects, so I'm overall in favor.
In theory of course in the future we could actually publish the cmdline crate. I know there's at least one other probable user in e.g. https://github.com/coreos/afterburn/blob/5191ebd8d9d8ead57e8c2abae74028d223028750/src/util/cmdline.rs#L66
There was a problem hiding this comment.
Yeah I think publishing it would probably be best, it's generally useful. Plus right now the CI here is failing because it's unpublished so it would fix that instead of requiring us to disable it or hack around it.
There was a problem hiding this comment.
The simplest thing honestly seems to me to not touch the karg parsing in composefs-boot right now and move on to higher priority things.
But yeah we can definitely publish the crate, I'd like to improve our automation for doing so for the other crates in the bootc-dev org.
Signed-off-by: John Eckersberg <jeckersb@redhat.com>
jeckersb
force-pushed
the
bootc-kernel-cmdline
branch
from
3172b7a to
db72642
Compare
|
Updated Cargo.toml to at least point to bootc where bootc-dev/bootc#1611 was merged in, rather than my fork |
|
Moving this back to draft until we can get bootc-dev/infra#20 sorted out |
2 participants
Signed-off-by: John Eckersberg jeckersb@redhat.com