Re: Python recompile

On 3/10/2025 9:25 PM, Chris M. Thomasson wrote:

On 3/10/2025 8:07 PM, Lawrence D'Oliveiro wrote:

On Mon, 10 Mar 2025 17:56:03 -0700, Chris M. Thomasson wrote:
>

On 3/10/2025 4:19 PM, Lawrence D'Oliveiro wrote:
>

On Mon, 10 Mar 2025 14:40:23 -0700, Chris M. Thomasson wrote:
>

For some damn reason I am remembering that the array that
WaitForMultipleObjects waits on should be "shifted" or even randomized
per call in a server loop that is using EVENTS. IIRC, this was way
back in late 90's and early 2000's.

>
I would hope that the order of entries is not important, as that would
defeat the point of non-determinism.

>
I remember reading a paper on the 50,000 concurrent connections issue
wrt IOCP vs events. Well, it made sure to tell the user to at least
shift the array that WFMO's waits on, per call in the server loop.

>
But why? Surely events relevant to entries in the list should be serviced
in the order they arrive, not in the order in which the entries happen to
occur in the list. There is no other reasonable way to do it.

 Well, it just the way WFMO is implemented. Notice that post from Kaz wrt the index... I remember reading about it is more than one place now, for some damn reason. NT insider comes to mind.

There was an interesting issue with IOCP used with sockets. If one issued multiple overlapped operations, they could complete in any order. So, I would have to do a little sort on them to make sure my program logic got them in order.