core: Add OSTREE_COMMIT_META_KEY_ARCH #2121
Conversation
There was a problem hiding this comment.
Hmm yeah, I think that makes sense. We can then use that in Zincati instead of the more "implementation-specific" coreos-assembler.basearch. Might be worth linking to coreos/coreos-assembler@e02ef26 in the commit message. LGTM otherwise!
Re.
I'll probably end up changing rpm-ostree at least to use
the RPM architecture for this key and fail if it doesn't match
the booted value.
doesn't that theoretically break going from a 32-bit to 64-bit kernel? I'm not sure if 32-bit kernels would still show x86_64 on a 64-bit capable CPU even if it was compiled for 32-bit. (Or I guess this applies more generally going between compatible "sub-architectures").
src/libostree/ostree-core.h
Outdated
| * | ||
| * Since: 2020.4 | ||
| */ | ||
| #define OSTREE_COMMIT_META_KEY_ARCH "ostree.arch" |
There was a problem hiding this comment.
Small bikeshed to just spell this out as OSTREE_COMMIT_META_KEY_ARCHITECTURE "ostree.architecture", though good as is too.
There was a problem hiding this comment.
doesn't that theoretically break going from a 32-bit to 64-bit kernel?
Right, so far we haven't hit that problem for any system we care about that I know. e.g. no one sane runs i686 kernels anymore, and for ARM in Fedora it's just aarch64. But I guess if we did do this in rpm-ostree we'd probably want to use the libdnf architecture mapping for compatible types to do it "properly".
There was a problem hiding this comment.
I'm not sure if 32-bit kernels would still show x86_64 on a 64-bit capable CPU even if it was compiled for 32-bit.
No, 32 bit kernels report as i686.
Add a standard key for this. We actually had a case in OpenShift builds recently where a `ppc64le` image was pushed over an `x86_64` one and this started failing at runtime with a not immediately obvious error. I'll probably end up changing rpm-ostree at least to use the RPM architecture for this key and fail if it doesn't match the booted value. Possibly that should live in ostree but it would involve adding architecture schema here, which gets into a big mess. Let's just standardize the key. xref coreos/coreos-assembler@e02ef26
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: cgwalters, jlebon The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
Add a standard key for this. We actually had a case in OpenShift
builds recently where a ppc64le image was pushed over an
x86_64one and this started failing at runtime with a not immediately
obvious error.
I'll probably end up changing rpm-ostree at least to use
the RPM architecture for this key and fail if it doesn't match
the booted value.
Possibly that should live in ostree but it would involve adding
architecture schema here, which gets into a big mess. Let's
just standardize the key.