On Apr 11, 2008, at 1:12 AM, Justin C. Walker wrote:
On Apr 10, 2008, at 10:35 , Michael Cashwell wrote:
On Apr 10, 2008, at 11:44 AM, Justin C. Walker wrote:
You are correct that an ethernet driver does not have permission
to drop one kind of packet over another, I don't think you are
correct in your implication that it cares about network-layer
Did I imply it does? I didn't intend to.
You did say that the driver should not have "more permission" to
drop UDP over TCP packets. I don't know of any driver that pays
attention to IP frame type (in particular because they can't always
tell; as you say, fragmentation is fun).
OK, that was only in reference to an implication that I took from
William's post that the loss was OK simply because the data were UDP
even if it was happening in the driver. We are pretty clearly in
violent agreement here.
I don't think TCP/UDP or QoS is involved in rimas' problem. That
really does look like a driver issue.
I don't think so, although I suppose it's worth looking at. The
driver can only do so much; it's part of a bucket brigade, and if it
doesn't have buffers, that's it for the outgoing packets. The
driver has no clue where the packets come from; it's just all bits
I'm thinking in the area of the driver not servicing the hardware
often enough and the loss actually occurs in the Ethernet controller
hardware. But I'd see that as a driver issue since servicing the
hardware is its job. It may just be a buffer reservation problem though.
Assuming that someone's data is less important just because they
used UDP is flawed.
I don't think there is any assumption regarding that except in the
UDP code, where it's assumed that if there's no space for the
packet, it's OK to drop it.
In terms of any code at issue for rimas' problem, yes.
I'd just run across the "UDP == throw-away" mindset in William's post
in other contexts (mainly sysadmins configuring routers) and wanted to
point out that it's not really correct in all cases.
Do not post admin requests to the list. They will be ignored.
Darwin-kernel mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden