Subject: Re: pointer atomicity
To: None <tech-kern@NetBSD.org>
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
List: tech-kern
Date: 01/11/2006 16:24:16
> Is there any port where the simple read or simple write of a single
> pointer is nonatomic?
I think the VAX may count, if the pointer crosses a page boundary - if
the latter-read part incurs a page fault that demands a read from swap,
the process may be blocked for a more or less arbitrarily long time in
between reading the two parts. (The only way to avoid this would be to
reread both parts if either part faults, and that is just asking for
livelock if the earlier-read part is pushed out of core by reading the
latter-read part in, though I suppose DEC could have done it.)
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse@rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B