Date: Tue, 15 Jun 2021 15:51:22 -0400 From: Kurt Hackenberg <kh@panix.com> To: freebsd-questions@freebsd.org Subject: Re: Is a successful call to write(2) atomic? Message-ID: <034b6d41-218a-0782-052e-c1c74a007897@panix.com> In-Reply-To: <CADqw_gKYTV-tRKxPFD1q78_gnpR2DWwmkTxLZJFmo4CUGqGqsg@mail.gmail.com> References: <22440.1623740785@segfault.tristatelogic.com> <44e15917-0c92-08f2-462e-a1b3705f9afb@panix.com> <CADqw_gKYTV-tRKxPFD1q78_gnpR2DWwmkTxLZJFmo4CUGqGqsg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2021/06/15 15:16, Michael Schuster wrote: > Let's take a step back: an atomic write() either writes everything or > nothing - and that's all. There's nothing in that claim that says > "everything must be in a contiguous block", nor, that all the data must be > written in a single "operation" by the underlying system. > > So after consideration I don't think the observed behaviour is violating > the claim that write() is atomic - I welcome correction, of course :-) You're just talking about the exact meaning of that word "atomic". Ronald used it to mean indivisible, as atoms were once thought to be; you advocate a slightly different meaning. That's a digression. Ronald's problem is clear, no matter what word we use. Paul and I know the problem and have suggested two well-known solutions.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?034b6d41-218a-0782-052e-c1c74a007897>