cbloom rants

7/18/2011

07-18-11 - cblib Relacy

›
Announce : cblib now has its own version of "Relacy". This is no replacement for the original - go get Dmitry's Relacy Ra...
5 comments:

07-18-11 - MCS list-based lock

›
We did CLH before, but MCS really is better, so let's go through it. MCS is another list/node based lock from the ancient days. The ...
7/17/2011

07-17-11 - Per-thread event mutexes

›
Another class of mutex implementations that we haven't talked about yet are those based on a list of waiting threads. Again this is a ge...

07-17-11 - CLH list-based lock

›
The multi-way ticket lock we just did is very similar to some classic spin locks. I found this nice page : scalable synchronization pseudo...

07-17-11 - Atman's Multi-way Ticket Lock

›
But first - Atman sent me a note about the ticket lock which is related and worth sharing. Imagine we're in a "high performanc...
7/16/2011

07-16-11 - Ticket FIFO Mutex

›
The Linux kernel internally uses a FIFO spinlock that they call "ticket lock". A ticket or "bakery" algorithm is quite ...
2 comments:
7/15/2011

07-15-11 - Review of many Mutex implementations

›
This is gonna be a long one. The point of this is not that you should go off and implement your own mutex (don't). The point is that it...
5 comments:
7/14/2011

07-14-11 - Some obscure threading APIs

›
Futex : Futex from a Win32 programmer's perspective is an enhanced version of the windows Event. It has absolutely nothing to do wit...
4 comments:

07-14-11 - compare_exchange_strong vs compare_exchange_weak

›
The C++0x standard was revised a while ago to split compare_exchange (aka CAS) into two ops. A quick note on the difference : bool compa...
5 comments:

07-14-11 - ARM Atomics

›
Some quick notes for reference in the future. In general when I'm porting atomics to a new platform, what I would really like is a do...
2 comments:
‹
›
Home
View web version
Powered by Blogger.