Subject: Re: [boost] [spinlock] Spin on volatile read, NUMA fairness?
From: Niall Douglas (s_sourceforge_at_[hidden])
Date: 2014-12-05 08:29:54
On 5 Dec 2014 at 11:41, Gavin Lambert wrote:
> On 5/12/2014 02:32, Niall Douglas wrote:
> > On Intel the least strong read you can do is an acquire. It makes me
> > very wary of ever using atomics with relaxed because you simply can't
> > test them on Intel. It was a big reason I invested in that ARM board.
> I use Relacy Race Detector whenever I want to test anything to do with
> atomics, particularly the weaker orderings. It does require extracting
> the code in question to a little self-contained example along with a bit
> of extra decoration but it works quite well.
Never used that one myself.
I generally use valgrind drd which requires annotation plus clang
threadsanitiser which usually does not. My biggest problem is with
windows only code paths, I have been known to run valgrind on wine :(
-- ned Productions Limited Consulting http://www.nedproductions.biz/ http://ie.linkedin.com/in/nialldouglas/
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk