[$] Lockless algorithms for mere mortals
Time, as some have said, is nature's way of keeping everything from
happening at once. In today's highly concurrent computers, though, time
turns out not to be enough to keep events in order; that task falls to an
extensive set of locking primitives and, below those, the formalized view
of memory known as the Linux kernel memory model. It takes a special kind
of mind to really understand the memory model, though; kernel developers
lacking that particular superpower are likely to make mistakes when working
in areas where the memory model comes into play. Working at that level is
increasingly necessary for performance purposes, though; a recent
conversation points out ways in which the kernel could make that kind of
work easier for ordinary kernel developers.