Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*simple\s+change\s+to\s+qdisc

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*simple\s+change\s+to\s+qdisc_restart\s*$/: 18 ]

Total 18 documents matching your query.

1. simple change to qdisc_restart() (score: 1): Author: avem@xxxxxxxxxx>; Date: Tue, 20 May 2003 10:22:17 +0200; Any comments regarding the following patch? Thx. -- sch_generic.c.old Tue May 20 09:11:25 2003 +++ sch_generic.c Tue May 20 10:16:11 2003 @@ -77,6 +77,7 @@ int qdisc_restart(struct net_device *dev); /archives/netdev/2003-05/msg00194.html (8,740 bytes)
2. Re: simple change to qdisc_restart() (score: 1): Author: 田充 <mk@xxxxxxxxxxxxxx>; Date: Tue, 20 May 2003 01:28:24 -0700 (PDT); I understand why it is valid, etc., but why do we even want to do this? It is not like this dead-loop detection stuff is a hot-path or anything like that.; /archives/netdev/2003-05/msg00195.html (8,088 bytes)
3. Re: simple change to qdisc_restart() (score: 1): Author: ine <Eric.Lemoine@xxxxxxx>; Date: Tue, 20 May 2003 10:57:25 +0200; I've implemented a prototype that uses per-CPU kernel threads for processing packets coming in from a single interface. The idea is to apply multiple CPUs to a single network interface to be able to; /archives/netdev/2003-05/msg00196.html (9,006 bytes)
4. Re: simple change to qdisc_restart() (score: 1): Author: Miller" <davem@xxxxxxxxxx>; Date: Tue, 20 May 2003 12:36:43 +0200; I think it will make any use of "raw" dev->hard_start_xmit" impossible. Which is what pktgen uses. Sounds like a project at least having packet reordering and cache bouncing in mind. Cheers. --ro; /archives/netdev/2003-05/msg00197.html (9,462 bytes)
5. Re: simple change to qdisc_restart() (score: 1): Author: Eric.Lemoine@xxxxxxx>; Date: Tue, 20 May 2003 13:24:02 +0200; Oops. I forgot to mention that: [*] Currently the NIC simply does ring_idx = IPsrc & (nr_rings-1). -- Eric; /archives/netdev/2003-05/msg00198.html (9,493 bytes)
6. Re: simple change to qdisc_restart() (score: 1): Author: <Robert.Olsson@xxxxxxxxxxx>; Date: Tue, 20 May 2003 13:21:09 +0200; Let me explain a bit more. I developped a kernel module that basically implements per-cpu kernel threads, each being bound to a particular cpu. I also modified the Myrinet NIC driver and firmware so; /archives/netdev/2003-05/msg00199.html (10,345 bytes)
7. Re: simple change to qdisc_restart() (score: 1): Author: Hadi <hadi@xxxxxxxxxxxxxxxx>; Date: Tue, 20 May 2003 14:24:11 +0200; OK! Sounds interesting... So reordering should be guaranteed within "connections" but not per interface. And if you can repeat the trick with per-cpu rings for tx you can eventually eliminate cache b; /archives/netdev/2003-05/msg00204.html (9,816 bytes)
8. Re: simple change to qdisc_restart() (score: 1): Author: adi@xxxxxxxxxxxxxxxx>; Date: Tue, 20 May 2003 08:33:50 -0400 (EDT); His patch should be interesting. I have seen NICs showing up in the market with multiple DMA rings/channels and you can map flows to channels. Locking the device on egress just because one of the rin; /archives/netdev/2003-05/msg00205.html (9,637 bytes)
9. Re: simple change to qdisc_restart() (score: 1): Author: avem@xxxxxxxxxx>; Date: Mon, 26 May 2003 11:15:19 +0200; I'm working on being authorized to give out my code. It may take some time... -- Eric; /archives/netdev/2003-05/msg00321.html (9,139 bytes)
10. simple change to qdisc_restart() (score: 1): Author: Eric Lemoine <Eric.Lemoine@xxxxxxx>; Date: Tue, 20 May 2003 10:22:17 +0200; Hi, Any comments regarding the following patch? Thx. -- sch_generic.c.old Tue May 20 09:11:25 2003 +++ sch_generic.c Tue May 20 10:16:11 2003 @@ -77,6 +77,7 @@ int qdisc_restart(struct net_device *de; /archives/netdev/2003-05/msg00596.html (8,740 bytes)
11. Re: simple change to qdisc_restart() (score: 1): Author: "David S. Miller" <davem@xxxxxxxxxx>; Date: Tue, 20 May 2003 01:28:24 -0700 (PDT); I understand why it is valid, etc., but why do we even want to do this? It is not like this dead-loop detection stuff is a hot-path or anything like that.; /archives/netdev/2003-05/msg00597.html (8,200 bytes)
12. Re: simple change to qdisc_restart() (score: 1): Author: Eric Lemoine <Eric.Lemoine@xxxxxxx>; Date: Tue, 20 May 2003 10:57:25 +0200; I've implemented a prototype that uses per-CPU kernel threads for processing packets coming in from a single interface. The idea is to apply multiple CPUs to a single network interface to be able to; /archives/netdev/2003-05/msg00598.html (9,138 bytes)
13. Re: simple change to qdisc_restart() (score: 1): Author: Robert Olsson <Robert.Olsson@xxxxxxxxxxx>; Date: Tue, 20 May 2003 12:36:43 +0200; I think it will make any use of "raw" dev->hard_start_xmit" impossible. Which is what pktgen uses. Sounds like a project at least having packet reordering and cache bouncing in mind. Cheers. --ro; /archives/netdev/2003-05/msg00599.html (9,668 bytes)
14. Re: simple change to qdisc_restart() (score: 1): Author: Eric Lemoine <Eric.Lemoine@xxxxxxx>; Date: Tue, 20 May 2003 13:24:02 +0200; Oops. I forgot to mention that: [*] Currently the NIC simply does ring_idx = IPsrc & (nr_rings-1). -- Eric; /archives/netdev/2003-05/msg00600.html (9,785 bytes)
15. Re: simple change to qdisc_restart() (score: 1): Author: Eric Lemoine <Eric.Lemoine@xxxxxxx>; Date: Tue, 20 May 2003 13:21:09 +0200; Let me explain a bit more. I developped a kernel module that basically implements per-cpu kernel threads, each being bound to a particular cpu. I also modified the Myrinet NIC driver and firmware so; /archives/netdev/2003-05/msg00601.html (10,555 bytes)
16. Re: simple change to qdisc_restart() (score: 1): Author: Robert Olsson <Robert.Olsson@xxxxxxxxxxx>; Date: Tue, 20 May 2003 14:24:11 +0200; OK! Sounds interesting... So reordering should be guaranteed within "connections" but not per interface. And if you can repeat the trick with per-cpu rings for tx you can eventually eliminate cache b; /archives/netdev/2003-05/msg00606.html (10,108 bytes)
17. Re: simple change to qdisc_restart() (score: 1): Author: Jamal Hadi <hadi@xxxxxxxxxxxxxxxx>; Date: Tue, 20 May 2003 08:33:50 -0400 (EDT); His patch should be interesting. I have seen NICs showing up in the market with multiple DMA rings/channels and you can map flows to channels. Locking the device on egress just because one of the rin; /archives/netdev/2003-05/msg00607.html (9,935 bytes)
18. Re: simple change to qdisc_restart() (score: 1): Author: Eric Lemoine <Eric.Lemoine@xxxxxxx>; Date: Mon, 26 May 2003 11:15:19 +0200; I'm working on being authorized to give out my code. It may take some time... -- Eric; /archives/netdev/2003-05/msg00723.html (9,464 bytes)

Current List: 1 - 18
Page: [1]