On Tue, 24 Oct 2006, Patrick McHardy wrote:
Russell Stuart wrote:
On Mon, 2006-10-23 at 14:39 +0200, Patrick McHardy wrote:
The implementation may be different, but the intention and the
result is the same. I probably would mind less if it wouldn't
affect userspace compatibility, but we need to carry this stuff
for ever even if we add another implementation that covers all
qdiscs.
So where is the overlap? Your patch will make qdiscs
that use packet length work with ATM. Ours makes the
rest, ie those that use Alexy's RTAB, work with ATM.
They would probably both apply with minimal conflicts.
You have previously said you have no intention of
changing the current RTAB implantation with your STAB
patch.
No, my patch works for qdiscs with and without RTABs, this
is where they overlap.
Patrick, your patch does not solve the alignment issue of the RTABs.
(The kernel part of) our patch (Me and Russells), simply gives the ability
to align the RTAB. We will still need this ability when your STAB patch
is in place.
<cut>
Sure, I'm just mentioning it. Is seems to be quite deeply codified
in userspace though.
Yes, the userspace patch is where we do the "ugly" complicated stuff.
<cut>
I'm not sure I understand what you're saying here. The transmission
time gets _smaller_ by transmitting over ATM? Does this have anything
to do with the off-by-one during rate table calculation you or
Jesper noticed?
There is nothing I would describe as an "off-by-one
error" in the RTAB calculation, so I can't be sure
what you are referring to.
Either you or Jesper pointed to this code in iproute:
It was me, Jesper.
for (i=0; i<256; i++) {
unsigned sz = (i<<cell_log);
...
rtab[i] = tc_core_usec2tick(1000000*((double)sz/bps));
which tends to underestimate the transmission time by using
the smallest possible size for each cell.
Correct.
My original patch only corrected this "off-by-one error" (and parsed on an
overhead argument to the kernel).
Russell, took the step further and made it possible to adjust the
alignment from userspace.
<cut>
Cheers,
Jesper Brouer
--
-------------------------------------------------------------------
MSc. Master of Computer Science
Dept. of Computer Science, University of Copenhagen
Author of http://www.adsl-optimizer.dk
-------------------------------------------------------------------
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
