On Wednesday, February 20, 2019 at 12:59:28 AM UTC-8, Dmitry Vyukov wrote: > > Hi, > > FYI, if you haven't seen this yet: > > https://github.com/abseil/abseil-cpp/blob/master/absl/synchronization/mutex.h > > https://github.com/abseil/abseil-cpp/blob/master/absl/synchronization/mutex.cc > > > Some minor improvements are possible, but overall it's a piece of art. > Originally developed by Mike Burrows > (https://en.wikipedia.org/wiki/Michael_Burrows). If you like learning > synchronization primitives, this one will be a good time investment. > The main design goal I guess is handling all possible usage scenarios > gracefully (i.e. not just readers, or writers, or light contention, or > heavy contention, or few waiters, or thousands of waiters, etc) which > is expected of a primitive used in literally millions of place across > thousands of systems. It also has some interesting features like > contention profiling and deadlock detection. FWIW it also fits into a > single intptr (i.e. 4 bytes on 32-bit systems). > > There is also a bit simplified version of the same algo in nsync library: > https://github.com/google/nsync/blob/master/README > https://github.com/google/nsync/blob/master/internal/mu.c > It's C and does not have contention profiling and deadlock detection, > so may be easier for leaning of the mutex algorithm itself. >
Interesting! Will take a look. Thank you for the heads up Dmitry. :^) -- --- You received this message because you are subscribed to the Google Groups "Scalable Synchronization Algorithms" group. To unsubscribe from this group and stop receiving emails from it, send an email to lock-free+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/lock-free/ccb8a2a8-5d39-4381-b848-2a5e67df0f17%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
