17 Nov
2006
17 Nov
'06
12:13 p.m.
On 11/17/06, Christopher Kohlhoff <chris@kohlhoff.com> wrote: [...]
I spent some time investigating this today and it smells a lot like a bug in the way Windows implements I/O completion ports. I can go into more detail if anybody's actually interested, but in any case I will be constructing a small test case in order to report the problem.
I'm curious and interested. You can reply on private mail if you want. Giovanni P. Deretta