As Christian points out [1], even though it's privileged, this interface has a lot of footguns. There are better options these days (e.g. eBPF), so it would be good to start discouraging its use and mark it as deprecated. [1]: https://lore.kernel.org/linux-fsdevel/20250212-giert-spannend-8893f1eaba7d@brauner/ Signed-off-by: Jeff Layton --- I meant to send this months ago, but never got around to it. Let's at least mark this as deprecated and see who complains. We could also consider adding a pr_warn_once() that fires the first time someone calls acct(2) if we want to kill this in a more near-term timeframe. --- init/Kconfig | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/init/Kconfig b/init/Kconfig index fa79feb8fe57bb01d8ce8f35e33535709b57d452..160c1c4ef253593d62650cd5a53f3421bc9372d3 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -624,8 +624,9 @@ config SCHED_HW_PRESSURE arch_update_hw_pressure() and arch_scale_thermal_pressure(). config BSD_PROCESS_ACCT - bool "BSD Process Accounting" + bool "BSD Process Accounting (DEPRECATED)" depends on MULTIUSER + default n help If you say Y here, a user level program will be able to instruct the kernel (via a special system call) to write process accounting @@ -635,7 +636,9 @@ config BSD_PROCESS_ACCT command name, memory usage, controlling terminal etc. (the complete list is in the struct acct in ). It is up to the user level program to do useful things with this - information. This is generally a good idea, so say Y. + information. This mechanism is antiquated and has significant + scalability issues. You probably want to use eBPF instead. Say + N unless you really need this. config BSD_PROCESS_ACCT_V3 bool "BSD Process Accounting version 3 file format" --- base-commit: 7f98ab9da046865d57c102fd3ca9669a29845f67 change-id: 20260106-bsd-acct-c60be8e6ae62 Best regards, -- Jeff Layton