Notices by Christian Brauner ?? (brauner@mastodon.social)

I've done a series that adds support for AF_UNIX sockets in coredumps. Userspace provides an AF_UNIX socket path via core_pattern and the kernel connects to it, shuts down the read side and writes the coredump to the socket.

This means no more super privileged usermode helper upcalls and makes for a very nice API experience. I captured coredumps simply via socat:

https://lore.kernel.org/20250430-work-coredump-socket-v1-0-2faf027dbb47@kernel.org

The receiver can use SO_PEERPIDFD to get a stable handle on the crashed process.

#kernel #linux #coredump

1/n

Oh yeah, I already mentioned this a while ago but I also added support for pidfds in Linux coredumps.

/proc/sys/kernel/core_pattern has been extended so userspace can instruct the kernel to install a pidfd for the
crashing process into the usermode helper process, e.g., systemd-coredump.

Replaced sudo with run0 and I'm very happy to have one less setuid binary to worry about.

#systemd #Linux

Picked back up the work for VFS {g,u}id squashing. IOW, mapping all {g,u}ids down to a single {g,u}id. Any process that doesn't have that {g,u}id but is still privileged otherwise will write to disk as the squashed {g,u}id. I just finished a draft and selftests that miraculously work.

https://web.git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git/log/?h=work.idmapped.squash

Probably "bugs galore" at this point. Needs more thinking.

#linux #kernel #vfs #filesystems

I think multi-threaded exec behavior needs to be illegal:
https://web.git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git/commit/?h=vfs-6.15.pidfs&id=3ae8576617e88a2b3cdbad6d8087be3454a4b326

de_thread() is an insult to humanity.

https://lwn.net/Articles/1010667

I want this for struct file allocations.

We've removed #reiserfs and now we're removing #sysv https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git/log/?h=vfs-6.15.sysv

The code removal continues.

#kernel #linux #fs

@osandov 🤡

@axboe @osandov This was actually the first time I ever read Phoronix comments. Immediately regretted it.

@osandov you already made it to Phoronix.
https://www.phoronix.com/news/Linux-6.14-Faster-kcore-Reads

util-linux has merged support for statmount() and listmount() to list mounts:
https://github.com/util-linux/util-linux/pull/3092#event-15865684400

This should be a nice performance improvement.

See https://brauner.io/2024/12/16/list-all-mounts.html for some details on listmount() and statmount().

#linux #kernel

I've written a post that shows how to list all mounts in all mount namespaces (all mounts on the system) using new apis we added to the #vfs this year.

https://brauner.io/2024/12/16/list-all-mounts.html

#kernel #linux #vfs

@corbet fwiw, I tend to not use it unless the message really isn't getting across.

@corbet Any specific example that triggered this?

I can probably mange to free up another 8 bytes in struct file by killing f_version and moving it into the few filesystems that care about it.

That would give us back 8 bytes for future expansions.

It's a rough draft and lacking commit messages but it should work (famous last words)...

https://github.com/brauner/linux/commits/vfs.file.f_version

#linux #kernel

I just added nine new EINVALs to the move_mount() manpage...

That commit message

kernel test robot noticed a 9.1% improvement of stress-ng.sockfd.ops_per_sec on:

commit: d9f21b3613337b55cc9d4a6ead484dca68475143 ("af_unix: Try to run GC async.")
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d9f21b3613337b55cc9d4a6ead484dca68475143

is very scary to people who remember the "pleasantness" we've already gotten out of this codepath.

I'm done writing groff.
https://github.com/brauner/man-pages-md

fyi https://fosdem.org/2024/schedule/track/kernel
happening rn.

I hadn't read that commit message until now. 😅