| To: | Harald Welte <laforge@xxxxxxxxxxxx> |
|---|---|
| Subject: | Re: [PATCH + RFC] neighbour/ARP cache scalability |
| From: | Tim Gardner <timg@xxxxxxx> |
| Date: | Tue, 21 Sep 2004 14:58:37 -0600 |
| Cc: | Andi Kleen <ak@xxxxxxx>, YOSHIFUJI Hideaki / ???????????? <yoshfuji@xxxxxxxxxxxxxx>, pekkas@xxxxxxxxxx, netdev@xxxxxxxxxxx |
| In-reply-to: | <20040921203404.GA3236@xxxxxxxxxxxxxxxxxxxxxxx> |
| Organization: | TriplePoint, Inc. |
| References: | <20040922.001448.73843048.yoshfuji@xxxxxxxxxxxxxx> <Pine.LNX.4.44.0409211856260.9906-100000@xxxxxxxxxx> <20040922.010428.104988024.yoshfuji@xxxxxxxxxxxxxx> <1095784761.3934.52.camel@xxxxxxxxxxx> <20040921173134.GC12132@xxxxxxxxxxxxx> <1095789507.3934.69.camel@xxxxxxxxxxx> <20040921181525.GB18938@xxxxxxxxxxxxx> <20040921203404.GA3236@xxxxxxxxxxxxxxxxxxxxxxx> |
| Reply-to: | timg@xxxxxxx |
| Sender: | netdev-bounce@xxxxxxxxxxx |
> My personal (simplistic) favourite is still a simple threshold (absolute > value / percentage) for incomplete neighbour entries. This way we make > sure that we cannot starve 'real' (fully resolved) entries at the cost > of incomplete ones. > > > -Andi It's not like NUD doesn't already have an overflow policy. neigh_forced_gc() performs a cleanup on NUD_INCOMPLETE entries when gc_thresh3 is exceeded. rtg -- timg@xxxxxxx http://www.tpi.com 406-443-5357(MT) 503-601-0234(OR) |
| Previous by Date: | Re: [PATCH + RFC] neighbour/ARP cache scalability, Harald Welte |
|---|---|
| Next by Date: | Re: PPP-over-L2TP kernel support, new patch for review, Benjamin LaHaise |
| Previous by Thread: | Re: [PATCH + RFC] neighbour/ARP cache scalability, Harald Welte |
| Next by Thread: | Re: [PATCH + RFC] neighbour/ARP cache scalability, Harald Welte |
| Indexes: | [Date] [Thread] [Top] [All Lists] |