Tachino Nobuhiro wrote: > > Hello, > > At Fri, 22 Jun 2001 11:52:12 +1000, > [EMAIL PROTECTED] wrote: > > > > [1.] One line summary of the problem: > > > > poll() timeout always takes 10ms too long > > > > [2.] Full description of the problem/report: > > > > Select() timeouts work fine. A timeout between 10n-9 and 10n ms times > > out after 10n ms on average. Poll() timeouts between 10n-9 and 10n ms, > > on the other hand, time out after 10(n+1) ms on average. It's always a > > jiffy too long. This means it's impossible to set a 10ms timeout using > > poll() even though it's possible using select(). The programs and their > > output below [6] demonstrate this. The same behavious occurs with > > linux-2.2 and linux-2.4. > > > I think this is correct behavior. The Single UNIX Specification > describes about the timeout parameter of poll() as follows, > > If none of the defined events have occurred on any selected > file descriptor, poll() waits at least timeout milliseconds > for an event to occur on any of the selected file descriptors. > > On the other hand, select(), > > If the specified condition is false for all of the specified > file descriptors, select() blocks, up to the specified timeout > interval, until the specified condition is true for at least > one of the specified file descriptors. ok, it's a correct behaviour. but having both poll and select timeout at the time specified would also be correct behaviour. better than that, it would be expected behaviour. raf - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
