Repository navigation
Description
The weekly audit found 6 Edera commit(s) on edera/mainline (a85e73be200d) that edera/6.18-lts (492d0ef30003) does not have. Audit run.
| Mainline commit | Subject | Backport status |
|---|---|---|
3f7ca89e3d70 |
xen/gntdev: don't copy to user space under priv->lock (#79) | none |
1dc86dd35ab6 |
x86/xen: clear the user pgd pointer when freeing it | #82 |
e79780b9d1b6 |
xen/gntdev: reset the copy batch's writeable flag (#81) | none |
8f315d39b605 |
xen/balloon: let the OOM notifier grow an empty domU balloon (#84) | none |
ec5588ab8435 |
x86: fault: emulate trampolines that start with endbr (#87) | none |
7d91f39740ed |
xen-pciback: free a device's config fields before its dev_data (#88) | none |
Each one is resolved by landing its backport on edera/6.18-lts, or, if it must never go there, by labelling its mainline pull request no-backport or listing its subject in .automation/backport-skip on edera/6.18-lts. A commit that was ported under another subject or in another form also goes in the skip file. This issue is updated weekly and closes itself once nothing is missing.
Claude's triage
All six listed commits are really missing from edera/6.18-lts at 492d0ef30003. None of them has been ported in another form, is already in linux-6.18.y, or is inapplicable to 6.18, so none should be skipped. Five are in flight as open backport pull requests (#82, #83, #85, #91, #92). One, e79780b9d1b6 (mainline #81), has no LTS pull request yet and its backport is marked backport-waiting, so it needs action.
3f7ca89e3d70 xen/gntdev: don't copy to user space under priv->lock (#79)
In flight (#83)
gntdev_ioctl_map_grant_ref() on the LTS tip still calls copy_to_user() with priv->lock held. The only LTS commit on drivers/xen/gntdev.c is f673bdd7025d (NUMA placeholder pages), and it is unrelated. Let #83 land.
1dc86dd35ab6 x86/xen: clear the user pgd pointer when freeing it
In flight (#82)
arch/x86/xen/mmu_pv.c on the LTS tip does not clear page->private after the user pgd is freed. Let #82 land.
e79780b9d1b6 xen/gntdev: reset the copy batch's writeable flag (#81)
Missing
gntdev_ioctl_grant_copy() on the LTS tip (drivers/xen/gntdev.c:1111) resets nr_ops and nr_pages but not writeable, so a batch fresh from kmalloc() can reach gntdev_put_pages() with that flag uninitialised. The fix is the same one-line addition and the context lines are identical on LTS, so it should apply as is with a line offset. It does not depend on 3f7ca89e3d70 (#79), and no later commit in the list depends on it. Backport it: mainline pull request #81 is in STATUS as backport-waiting.
8f315d39b605 xen/balloon: let the OOM notifier grow an empty domU balloon (#84)
In flight (#85)
xen_try_alloc_unpopulated_blocks() does not exist on the LTS tip. The prerequisites it builds on are already on LTS: the OOM deflate notifier and the 2 MiB unpopulated-allocator blocks. Let #85 land.
ec5588ab8435 x86: fault: emulate trampolines that start with endbr (#87)
In flight (#91)
The EMUTRAMP matchers in arch/x86/mm/fault.c on the LTS tip contain no endbr handling. Let #91 land.
7d91f39740ed xen-pciback: free a device's config fields before its dev_data (#88)
In flight (#92)
pcistub_device_release() on the LTS tip still frees dev_data and clears drvdata before freeing the config fields, so the leak is present. Let #92 land.
Activity