From owner-freebsd-current Thu Dec 23 13: 3:36 1999 Delivered-To: freebsd-current@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by hub.freebsd.org (Postfix) with ESMTP id 82A4714F73 for ; Thu, 23 Dec 1999 13:03:31 -0800 (PST) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id NAA07664; Thu, 23 Dec 1999 13:03:30 -0800 (PST) (envelope-from dillon) Date: Thu, 23 Dec 1999 13:03:30 -0800 (PST) From: Matthew Dillon Message-Id: <199912232103.NAA07664@apollo.backplane.com> To: Bill Paul , julian@whistle.com, scottm@CS.UCLA.EDU, jlemon@americantv.com, brad@shub-internet.org, jabley@patho.gen.nz, phk@critter.freebsd.dk, wollman@khavrinen.lcs.mit.edu, current@FreeBSD.ORG Subject: Re: Woa! May have found something - 'rl' driver and small packets (was Re: Odd TCP glitches in new currents) References: <199912231930.OAA11526@skynet.ctr.columbia.edu> <199912232032.MAA07521@apollo.backplane.com> Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Heh heh. Sorry about this, I believe I have further information on another older problem. Bill, remember those ethernet lockups I was having with the 'xl' driver all those months ago that we could never track down? Well, they happen with the 'dc' driver too. But this time I'm not getting a complete lockup. The network actually continues to work well enough, well, just barely well enough, that I can still use it. slowly. It appears that the 'dc' driver continues to take receive interrupts (see the systat -vm snapshot at the end), but winds up not processing any of the packets. Except when 64 packets accumulate then suddenly all 64 get processed all at once! Then nothing again until the next 64 accumulate. This netstat is on the machine with the 'dc' driver that locked up, when I ping it from another machine. The 'dc' driver still works--- barely. It doesn't processes any packets until 64 have been received, then it processes them all at once. The transmit side appears to work fine and the receive side appears to get interrupts but does not appear to process incoming packets. Yet, obviously, the packets are being accumulated somewhere because I don't have any packet loss, just incredibly long and odd ping times. This occurs when I am running netscape on the same box over a remote X connection (read: Lots of packets going over the network plus lots of local PCI activity talking to the graphics card). Same problem occurs with different graphics adapters but I believe this same problem also occured with the 'xl' driver on the card I had in before I put this card in. dc0: irq 5 at device 9.0 on pci0 dc0: Ethernet address: 00:a0:cc:69:4e:2d dc0@pci0:9:0: class=0x020000 card=0xc00111ad chip=0xc11511ad rev=0x25 hdr=0x00 input (Total) output packets errs bytes packets errs bytes colls 64 0 7188 48 0 4792 0 0 0 0 2 0 332 0 64 0 6962 46 0 4628 0 0 0 0 2 0 348 0 64 0 8268 46 0 4592 0 0 0 0 2 0 348 0 64 0 7704 46 0 4656 0 0 0 0 2 0 332 0 64 0 7228 47 0 4614 0 0 0 0 2 0 332 0 65 0 6972 47 0 4686 0 0 0 0 3 0 522 0 64 0 14472 42 0 4188 0 0 0 0 3 0 422 0 64 0 7724 44 0 4196 0 0 0 0 1 0 134 0 64 0 6768 49 0 4830 0 0 0 0 2 0 332 0 64 0 7440 45 0 4386 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 input (Total) output When I ping the machine faster from another box: input (Total) output packets errs bytes packets errs bytes colls 64 0 6712 50 0 5108 0 64 0 6724 50 0 5132 0 64 0 7948 50 0 5116 0 64 0 6816 48 0 4978 0 64 0 7072 50 0 5208 0 64 0 46144 28 0 3058 0 64 0 37416 31 0 3290 0 64 0 6712 50 0 5108 0 64 0 7004 49 0 4898 0 64 0 6712 46 0 4444 0 64 0 6724 50 0 4724 0 64 0 6432 50 0 4768 0 0 0 0 0 0 0 0 64 0 6432 50 0 4768 0 64 0 6684 50 0 4724 0 64 0 6792 55 0 5554 0 64 0 6876 53 0 5402 0 64 0 6752 52 0 5212 0 64 0 6712 5 1 4622 0 0 0 0 0 0 74 0 0 0 0 0 0 850 0 And watch what happens after I managed to 'ifconfig dc0 media auto', it goes back to normal... suddenly everything is working properly again. input (Total) output packets errs bytes packets errs bytes colls 0 0 0 0 0 170 0 0 0 0 0 0 170 0 32 0 3992 36 0 2494 0 54 0 5374 53 0 5062 0 57 0 7802 58 0 5456 0 54 0 6236 53 0 5062 0 53 0 5314 53 0 5062 0 53 0 5314 53 0 5062 0 59 0 6318 61 0 5642 0 55 0 5510 58 0 5372 0 53 0 5314 53 0 5062 0 53 0 5314 53 0 5062 0 42 0 4180 44 0 3988 0 10 0 1044 13 0 914 0 7 0 742 9 0 566 0 11 users Load 0.00 0.01 0.00 Thu Dec 23 12:50 Mem:KB REAL VIRTUAL VN PAGER SWAP PAGER Tot Share Tot Share Free in out in out Act 19324 3876 31956 5456 30252 count All 63628 4180 2785500 5824 pages Interrupts Proc:r p d s w Csw Trp Sys Int Sof Flt cow 284 total 1 20 54 1 295 284 38 6 13024 wire 40 dc0 irq11 42892 act ata-pci0 i 0.8%Sys 0.8%Intr 0.8%User 0.0%Nice 97.7%Idl 7664 inact fdc0 irq6 | | | | | | | | | | 48 cache atkbd0 irq + 30204 free ppc0 irq7 daefr 16 mse0 irq3 Namei Name-cache Dir-cache prcfr 100 clk irq0 Calls hits % hits % react 128 rtc irq8 1 1 100 pdwake zfod pdpgs Disks acd0 fd0 ofod intrn KB/t 0.00 0.00 %slo-z 6176 buf tps 0 0 tfree 1 dirtybuf MB/s 0.00 0.00 6908 desiredvnodes % busy 0 0 610 numvnodes 25 freevnodes 11 users Load 0.15 0.05 0.01 Thu Dec 23 12:51 Mem:KB REAL VIRTUAL VN PAGER SWAP PAGER Tot Share Tot Share Free in out in out Act 19816 3876 32700 5456 30240 count All 63640 4180 2785500 5824 pages Interrupts Proc:r p d s w Csw Trp Sys Int Sof Flt cow 273 total 22 103 1 667 273 67 6 13024 wire 29 dc0 irq11 42904 act ata-pci0 i 0.0%Sys 0.0%Intr 3.1%User 0.0%Nice 96.9%Idl 7664 inact fdc0 irq6 | | | | | | | | | | 48 cache atkbd0 irq >> 30192 free ppc0 irq7 daefr 16 mse0 irq3 Namei Name-cache Dir-cache prcfr 100 clk irq0 Calls hits % hits % react 128 rtc irq8 pdwake zfod pdpgs Disks acd0 fd0 ofod intrn KB/t 0.00 0.00 %slo-z 6176 buf tps 0 0 tfree 1 dirtybuf MB/s 0.00 0.00 6908 desiredvnodes % busy 0 0 610 numvnodes 25 freevnodes To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message