| To: | Alexey Kuznetsov <kuznet@xxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: 2.6.10 TCP troubles -- suggested patch |
| From: | "David S. Miller" <davem@xxxxxxxxxxxxx> |
| Date: | Tue, 15 Feb 2005 15:26:51 -0800 |
| Cc: | kuznet@xxxxxxxxxxxxx, shemminger@xxxxxxxx, hubert.tonneau@xxxxxxxxxxxxxx, romieu@xxxxxxxxxxxxx, niv@xxxxxxxxxx, rick.jones2@xxxxxx, netdev@xxxxxxxxxxx |
| In-reply-to: | <20050212200318.GB28895@xxxxxxxxxxxxxxx> |
| References: | <0525M9211@xxxxxxxxxxxxxxxxxxxxx> <20050211150420.74737b2e@xxxxxxxxxxxxxxxxx> <20050211170740.2608419b.davem@xxxxxxxxxxxxx> <20050212141641.GA27456@xxxxxxxxxxxxxxx> <20050212114132.5f7b7ffe.davem@xxxxxxxxxxxxx> <20050212200318.GB28895@xxxxxxxxxxxxxxx> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
On Sat, 12 Feb 2005 23:03:18 +0300 Alexey Kuznetsov <kuznet@xxxxxxxxxxxxx> wrote: > Actually, that anti-MacOS never worked well. If segment with forced PSH > was not transmitted in time, even forced PSHs could be deleted. > Your patch with setting PSH right before (or in) tcp_transmit_skb() must > work. Unless these segments are not tso. Yes, it never did work well. But now we understand more deeply the nature of this beast, we can probably refine it. In short, for properly working TCP stream with no drops and no reordering, Darwin delays ACKs until delack timer fires or PSH is seen :-) |
| Previous by Date: | Re: 2.6.10 TCP troubles -- suggested patch, David S. Miller |
|---|---|
| Next by Date: | Re: [PATCH net-drivers-2.6 2/10] e1000: use netif_poll_{enable|disable}, Francois Romieu |
| Previous by Thread: | Re: 2.6.10 TCP troubles -- suggested patch, Alexey Kuznetsov |
| Next by Thread: | Re: 2.6.10 TCP troubles -- suggested patch, Rick Jones |
| Indexes: | [Date] [Thread] [Top] [All Lists] |