17 Oct
2008
17 Oct
'08
8:31 p.m.
Fons Adriaensen wrote:
On Fri, Oct 17, 2008 at 09:26:45PM +0200, Joern
Nettingsmeier wrote:
ah, of course. one reader, one writer only. sorry for the noise.
--
jörn nettingsmeier
home://germany/45128 essen/lortzingstr. 11/
http://spunk.dnsalias.org
phone://+49/201/491621
Kurt is up in Heaven now.
i see. but what about the other scenario:
thread A reads read_ptr.
thread B runs and increments it.
thread A increases the old (wrong) read_ptr value by one and writes a
wrong value.
isn't that a problem here?
This can't happen. For both read_ptr and write_ptr
only *one* of the two sides will ever modify it.
That is really the basic idea behind all this.
The other side just reads it, and will never read
a value that could trigger an error.