Coredump Socket Support Merged For Linux 6.16
([Linux Kernel] 6 Hours Ago
Coredump Over Sockets)
- Reference: 0001550123
- News link: https://www.phoronix.com/news/Linux-6.16-Coredump-Sockets
- Source link:
As an alternative to Coredumps dumping to a file or a pipe connected to a user-mode helper process, Linux 6.16 is introducing the ability to send Coredumps over an AF_UNIX socket.
As the latest improvement to help in debugging Linux systems, Linux 6.16 introduces the ability to send Coredumps via sockets. Microsoft engineer Christian brauner who authored this Coredump socket support explained the situation and how the Linux world benefits from this work:
"The new coredump socket will allow userspace to not have to rely on usermode helpers for processing coredumps and provides a saf way to handle them instead of relying on super privileged coredumping helpers
This will also be significantly more lightweight since the kernel doens't have to do a fork()+exec() for each crashing process to spawn a usermodehelper. Instead the kernel just connects to the AF_UNIX socket and userspace can process it concurrently however it sees fit. Support for userspace is incoming starting with systemd-coredump
There's more work coming in that direction next cycle."
More details on the Coredump over sockets support via [1]this merge earlier in the week to the mainline kernel.
[1] https://lore.kernel.org/lkml/20250529180352.1935517-1-dave.hansen@linux.intel.com/
As the latest improvement to help in debugging Linux systems, Linux 6.16 introduces the ability to send Coredumps via sockets. Microsoft engineer Christian brauner who authored this Coredump socket support explained the situation and how the Linux world benefits from this work:
"The new coredump socket will allow userspace to not have to rely on usermode helpers for processing coredumps and provides a saf way to handle them instead of relying on super privileged coredumping helpers
This will also be significantly more lightweight since the kernel doens't have to do a fork()+exec() for each crashing process to spawn a usermodehelper. Instead the kernel just connects to the AF_UNIX socket and userspace can process it concurrently however it sees fit. Support for userspace is incoming starting with systemd-coredump
There's more work coming in that direction next cycle."
More details on the Coredump over sockets support via [1]this merge earlier in the week to the mainline kernel.
[1] https://lore.kernel.org/lkml/20250529180352.1935517-1-dave.hansen@linux.intel.com/
aviallon