Re: Macnetworkprog Digest, Vol 9, Issue 98
Re: Macnetworkprog Digest, Vol 9, Issue 98
- Subject: Re: Macnetworkprog Digest, Vol 9, Issue 98
- From: Tim Hewett <email@hidden>
- Date: Mon, 03 Dec 2012 22:36:23 +0000
I tried Terry's setsockopt option and related mapping to WMM levels and it has solved the problem. Previously initiating a file copy while sending real-time packets would virtually stop the latter completely. Now both coexist without problem.
Whether this approach works in the reverse scenario where the initiator is on an ethernet and recipient on WiFi, currently untested and reliant on the base station reacting to the flags in the ethernet frames for WMM, is uncertain but looks likely to be inherent in the implementation of the standards.
Thanks for the suggestions.
Tim.
On 3 Dec 2012, at 20:00, email@hidden wrote:
> On 1 Dec 2012, at 15:44, Tim Hewett <email@hidden> wrote:
>
>> Is there an API for setting priority packets for WMM on a WiFi interface?
>
> Have you tried kCFStreamNetworkServiceType?
>
> Last I checked this translated to the SOL_SOCKET / SO_TRAFFIC_CLASS socket option, which is in Darwin but not in the public SDK, but the CFSocketStream values are public API.
>
> Of course, for these to be effective the Wi-Fi driver much honour them. I don't know which, if any, of our Mac OS X Wi-Fi drivers do. Which brings us back to Terry's suggestion of using a trace to verify what's happening on the wire.
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Macnetworkprog mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden