Re: [PATCH v4 00/10] net: support thread-specific pidfds for send and connect

[Andrew Halaney]

On Mon, Oct 12, 2026 at 05:59:17PM +0200, Paolo Abeni wrote:
> On 10/5/26 17:09, Andrew Halaney wrote:
> > SO_PASSPIDFD/SCM_PIDFD and SO_PEERPIDFD allow to retrieve a pidfd for
> > the thread-group leader. The coredump server using the coredump socket
> > cannot get a handle on the task that took the signal and is writing the
> > coredump easily. Workloads interested in per-thread authentification
> > have similar problems.
> > 
> > Add SO_PASSPIDFD_THREAD/SCM_PIDFD_THREAD and SO_PEERPIDFD_THREAD. We
> > record the sending and the connecting thread in addition to the
> > thread-group leader.
> > 
> > SO_PASSPIDFD_THREAD functions like SO_PASSPIDFD but instead of an
> > SCM_PIDFD message for the thread-group leader, SCM_PIDFD_THREAD sends a
> > pidfd for the specific thread. SO_PASSPIDFD / SO_PASSPIDFD_THREAD are
> > independent, allowing callers to get both styles of pidfd back in their
> > cmsg.
> > 
> > SO_PEERPIDFD_THREAD allows to retrieve a pidfd for the specific thread
> > that called connect(), listen(), or socketpair().
> > 
> > pidfs_coredump() now also stamps the dumping thread's struct pid so a
> > pidfd for that thread reports the coredump like the pidfd of the
> > thread-group leader does.
> > 
> > Signed-off-by: Christian Brauner (Amutable) <brauner@xxxxxxxxxx>
> > Signed-off-by: Andrew Halaney <andrew@xxxxxxxxxxxx>
> > 
> > ---
> > Changes in v4:
> > - Moved get_pid() out of scm_set_cred() into scm_send() (Kuniyuki)
> > - Dropped unnecessary helps (Kuniyuki)
> > - Rebased on v7.3-r6
> > - Collected Kuniyuki's RB tags
> > - Link to v3: https://lore.kernel.org/r/20260928-work-unix-passpidfd-v3-0-532018a91b21@xxxxxxxxxxxx
> Out of sheer ignorance, I don't see the shared, stable tree where we
> can pull from, can you please point that out?

Here ya go!

https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git/log/?h=vfs-7.4.shared.net-next.pidfd

> 
> I think that this late in the cycle, the series could go via Christian's
> tree only and we could get it back after the merge window. WDYT?

This works for me at least, for what its worth.

Thanks,
Andrew