Re: [boost] Boost.Threads, N2178, N2184, et al

From: Roland Schwarz (roland.schwarz_at_[hidden])
Date: 2007-03-26 16:38:47

Stefan Seefeld wrote:
> The point really is that we want a means to interrupt an otherwise
> non-interruptible (i.e. blocking) call, so this requires support
> from the system layer.

True.
But we will need this in either case. Cancellation also is not
feasible if we can't get out of the kernel.
But: a (std) library implementor can emulate this by making use of async
IO at the OS level.

Roland