| To: | David Stevens <dlstevens@xxxxxxxxxx> |
|---|---|
| Subject: | Re: TCP rto estimation patch |
| From: | Wesley Eddy <weddy@xxxxxxxxxxxx> |
| Date: | Fri, 23 Apr 2004 19:19:14 -0400 |
| Cc: | linux-kernel@xxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx |
| In-reply-to: | <OFE28EC49F.335C9D5D-ON88256E7F.0073BC2E-88256E7F.0073F700@us.ibm.com> |
| References: | <20040423142445.GC501@grc.nasa.gov> <OFE28EC49F.335C9D5D-ON88256E7F.0073BC2E-88256E7F.0073F700@us.ibm.com> |
| Reply-to: | weddy@xxxxxxxxxxxx |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mutt/1.5.5.1i |
On Fri, Apr 23, 2004 at 02:06:39PM -0700, David Stevens wrote: > I believe "2" and "3" are the scale factors for the fixed-point > representation > of the data, not the "alpha" and "beta" I remember from the estimator > paper. > The math is done with bit shifts instead of floating point multiplications. It's the same thing though. The 2 means beta of 1/4 and the 3 means alpha of 1/8. -Wes
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: kernel BUG at page_alloc.c:98 -- compiling with distcc, Carson Gaspar |
|---|---|
| Next by Date: | Re: [PATCH] Report vegas connection info, David S. Miller |
| Previous by Thread: | Re: TCP rto estimation patch, David S. Miller |
| Next by Thread: | [PATCH] pktgen dependency (was: Re: Linux 2.4.27-pre1), Geert Uytterhoeven |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
