Kernel-level Work
-
Syscall latency... and some uses of speculative execution
Moving from UEK5 to UEK6 brought about an unwelcome surprise: an increase in syscall latency on some x86 systems. The root cause, as we will see, was slightly slower evaluation of audit rules, which, given that they are evaluated for every syscall, is not great.
In this post we start off by exploring the root cause which turns out to not be UEK specific, it also impacts upstream kernels as well. Then we detail the fixes and how they take advantage of the speculative out-of-order nature of the CPU pipeline.
The changes, even though they target low-level optimizations, are quite straight-forward, almost trivial.
-
Stéphane Graber: Two months in
It’s now been two months since I left my position at Canonical and went freelance! A lot of things have now all fallen into place to the point where it almost feels like having a normal work routine again
Kernel and ZFS builds
As mentioned in an earlier post, after over a year of rolling my own kernels and manually installing them on all my systems, I’ve decided to spend a bit of time automating the whole process and putting in place a proper build and publishing pipeline.