Date: Sat, 10 Jul 2010 16:24:43 +0100 From: "Robert N. M. Watson" <rwatson@FreeBSD.org> To: Gabor PALI <pgj@FreeBSD.org> Cc: svn-src-head@freebsd.org Subject: Re: svn commit: r209119 - head/sys/sys Message-ID: <4A28A601-C87F-47C6-8CBE-5F1BF866CA4A@FreeBSD.org> In-Reply-To: <4C37713D.5060202@FreeBSD.org> References: <4C376B0E.9050505@FreeBSD.org> <alpine.BSF.2.00.1007091949170.94277@fledge.watson.org> <4C37713D.5060202@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 9 Jul 2010, at 19:58, Gabor PALI wrote: >> I assume there are reasonable alternatives that work around the >> potential race with a small probability of a missed or extra update, >> or similar, which would be fine. >=20 > In a few words: As far as I know, 64-bit atomic counters could be > implemented by using cmpxchg8b or by a plain uint64_t variable = protected > by some kind of locking(9). If we can do it in one atomic in the common case, and two atomics in an = edge case, that sounds fine. I think any use of locking(9) would be = sufficiently costly as to not be worth the improvements in consistency, = given the frequency of statistics operations. Robert=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4A28A601-C87F-47C6-8CBE-5F1BF866CA4A>