Thread (11 messages) read the whole thread 11 messages, 3 authors, 2026-07-01

Re: [PATCH v4 bpf-next 2/3] bpf: add bpf_init_inode_xattr kfunc for atomic inode labeling

From: David Windsor <hidden>
Date: 2026-06-30 18:46:51
Also in: bpf, linux-fsdevel, linux-integrity, linux-kselftest, lkml, selinux

On Thu, Jun 25, 2026 at 7:23 AM Christian Brauner [off-list ref] wrote:
<snip>
We expose a bunch of VFS heavy operations for various security modules
and this is really not different. For xattrs we have it all centralized
in the VFS and in general all VFS related bpf kfuncs should continue
living there and be registered there. Anything that's just bpf infra
specific can go to security/bpf/kfuncs.c instead. But anyway, it's a bpf
specific helper so it's the bpf maintainer's call.
After Alexei's requested changes removing the attach-time checks,
there's really not much left to go in an LSM-specific kfuncs file. The
bpf infra plumbing for registering the kfunc and bpf_xattrs_used()
seem to be the only LSM-specific bits aside from the kfunc.

I am willing to put this code anywhere. I've tried to CC all involved
in all 3 patches, even though there's some split in concerns.
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help