sysroot: Add ostree_sysroot_write_deployments_with_options()#745
sysroot: Add ostree_sysroot_write_deployments_with_options()#745cgwalters wants to merge 2 commits intoostreedev:masterfrom
Conversation
|
☔ The latest upstream changes (presumably 5d413df) made this pull request unmergeable. Please resolve the merge conflicts. |
fb2f00b to
b000868
Compare
|
☔ The latest upstream changes (presumably e665e51) made this pull request unmergeable. Please resolve the merge conflicts. |
b000868 to
cc990d6
Compare
|
🛫 Rebased 🛬 |
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
|
☔ The latest upstream changes (presumably 7b2370d) made this pull request unmergeable. Please resolve the merge conflicts. |
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
More sophisticated users of libostree like rpm-ostree need control over things like the system repository. Previously we introduced a "no cleanup" flag to `ostree_sysroot_simple_write_deployment()`, but that's a high level API that does filtering on its own. Since rpm-ostree needs more control, let's expose the bare essentials of the "sysroot commit" operation with an extensible options structure, where one of the options is whether or not to do post-transaction repository operations.
cc990d6 to
92be53b
Compare
|
🛫 Rebased 🛬 |
| * | ||
| * Assuming @new_deployments have already been deployed in place on | ||
| * disk, atomically update bootloader configuration. | ||
| * Older verison of ostree_sysroot_write_deployments() - will perform |
There was a problem hiding this comment.
s/ostree_sysroot_write_deployments/ostree_sysroot_write_deployments_with_options/ ?
| * | ||
| * Assuming @new_deployments have already been deployed in place on disk via | ||
| * ostree_sysroot_deploy_tree(), atomically update bootloader configuration. By | ||
| * default, no-post-transaction cleanup will be performed. You should invoke |
There was a problem hiding this comment.
Adding a comment here that I've addressed your comments on the code comments ⬇️
More sophisticated users of libostree like rpm-ostree need control over things like the system repository. Previously we introduced a "no cleanup" flag to `ostree_sysroot_simple_write_deployment()`, but that's a high level API that does filtering on its own. Since rpm-ostree needs more control, let's expose the bare essentials of the "sysroot commit" operation with an extensible options structure, where one of the options is whether or not to do post-transaction repository operations. Closes: #745 Approved by: jlebon
|
💔 Test failed - status-atomicjenkins |
|
@rh-atomic-bot retry |
|
☀️ Test successful - status-atomicjenkins |
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much.
This is prep for livefs. We need to tweak the logic from what core libostree has in `ostree_sysroot_simple_write_deployment()`, and while we could land improved logic there, I think it makes sense to carry this here until we're confident enough in the logic to make it ABI. This does depend on a [new libostree API](ostreedev/ostree#745) that allows writing deployments without doing cleanup. The `bump_mtime()` bit is also prep for livefs, carrying in this patch to avoid splitting things too much. Closes: #709 Approved by: jlebon
Depends: #744
More sophisticated users of libostree like rpm-ostree need control over things
like the system repository. Previously we introduced a "no cleanup" flag to
ostree_sysroot_simple_write_deployment(), but that's a high level API thatdoes filtering on its own.
Since rpm-ostree needs more control, let's expose the bare essentials of the
"sysroot commit" operation with an extensible options structure, where one of
the options is whether or not to do post-transaction repository operations.