On Fri, 12 Oct 2007 12:02:41 +0100 Andy Green <andy@xxxxxxxxxxx> wrote: > Somebody in the thread at some point said: > > Andy Green wrote, > >> Why don't you try removing the O_NONBLOCK since you test with > >> select... > > > > Never, ever, do that ... select only returns a hint that an IO operation > > might succeed, not a guarantee (google for "spurious wakeup"), hence a > > subsequent read/write might block if the descriptor isn't set to > > O_NONBLOCK, and that's almost always not what's wanted or expected. > > Wah, man select says it too > > Under Linux, select() may report a socket file descriptor > as "ready for reading", while nevertheless a subsequent read > blocks. This could for example happen when data has arrived but > upon examination has wrong checksum and is discarded. There > may be other circumstances in which a file descriptor is > spuriously reported as ready. Thus it may be safer to use O_NON- > BLOCK on sockets that should not block. > > Pretty freakish, thanks for the tip. Its the documented behaviour of select. select tells you "at the moment in time I checked the next whatever operation would not block". It makes no further guarantees, thus select should almost always be used with non blocking I/O. Alan
