From owner-freebsd-stable Tue Aug 14 10: 0:54 2001 Delivered-To: freebsd-stable@freebsd.org Received: from smtp1.sentex.ca (smtp1.sentex.ca [199.212.134.4]) by hub.freebsd.org (Postfix) with ESMTP id B82FC37B403 for ; Tue, 14 Aug 2001 10:00:33 -0700 (PDT) (envelope-from mike@sentex.net) Received: from simoeon.sentex.net (pyroxene.sentex.ca [199.212.134.18]) by smtp1.sentex.ca (8.11.1/8.11.1) with ESMTP id f7EGxuv37083; Tue, 14 Aug 2001 12:59:56 -0400 (EDT) (envelope-from mike@sentex.net) Message-Id: <5.1.0.14.0.20010814122135.01abc0a0@marble.sentex.ca> X-Sender: mdtpop@marble.sentex.ca X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Tue, 14 Aug 2001 12:54:26 -0400 To: "Michael M." , freebsd-stable@FreeBSD.ORG From: Mike Tancsa Subject: Re: fxp SCB timeout problems, anyone have a solution? In-Reply-To: <5.1.0.14.2.20010814120333.04649230@127.0.0.1> References: <5.1.0.14.0.20010814081137.0593b758@192.168.0.12> <5.1.0.14.2.20010813181107.04174140@127.0.0.1> <20010810100632.D18533@nexus.root.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1"; format=flowed Content-Transfer-Encoding: quoted-printable Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Try /usr/ports/benchmarks/netperf in 100baseTX. I forget which test, but=20 one of them should be able to do it as I remember reading results on a pair= =20 of OC-3 cards that pushed close to 120Mb/s. Or, check to see if the ping=20 flood is generating or trying to generate data up near the medium limit.= e.g. netstat -ni;sleep 10;netstat -ni;sleep 10;netstat -ni to see if you are near 100Mb/s also, see if the problem happens at 10Mb full-duplex. Thats what I run=20 here on the ET version of the chip and I dont run into the problems. Also,= =20 I have never had the machine panic on me, even with the EM version which I= =20 have problems with period. ---Mike At 12:13 PM 8/14/01 -0400, Michael M. wrote: >Good thing you asked.. this one slipped by me. It turns out I can= duplicate >the problem on both the A *AND* the A2 board.. when I was testing the >machine with the A board, I didn't realize that it was plugged in at >100baseTX (f/d), while the A2 board was in running at "10baseT/UTP". >I did some shuffling.. tried the A board in the 10baseT hub.. was able to= =20 >cause >kernel panic with just a few ping -fs1500's ... BUT, when I plugged either >machine into the 100baseTX switch, not only could I not crash them, but the >fxp SCB timeout messages disappeared. (Perhaps I just couldn't generate >enough data??). > >So.. to sum it up, both the A and A2 board will both have fxp SCB timeout >messages, as well as kernel panics if running at 10baseT. The timeout >messages and kernel panics go away if they're zipping along at 100baseTX. >(Unless I'm not sending enough data to crash them at this higher speed...) > >Hope this helps! > >Mike > > >At 08:16 AM 8/14/2001 -0400, Mike Tancsa wrote: > >>Very strange. I have been using the ET version of the board without the=20 >>timeout problems. the EM however, I can reproduce them in very short=20 >>order. I have a very similar board as yours (AOpen) >> >>fxp1: port 0xc400-0xc43f mem=20 >>0xd5101000-0xd5101fff irq 11 at device 8.0 on pci1 >>fxp1: Ethernet address 00:01:80:05:d2:67 >>inphy1: on miibus1 >>inphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto >> >>But I dont see these problems. Are you running in 10baseT ? 100BaseTX ?= =20 >>Half or full duplex ? >> >> ---Mike >> >>At 07:07 PM 8/13/2001 -0400, Michael M. wrote: >>>Greetings.. >>> >>>I've been experiencing problems with fxp SCB timeouts too.. I've= actually >>>had the problem for almost a month now.. starting with a build from >>>7/17/2001, 06:42:09 EDT. I installed FreeBSD on this system about >>>a month ago, running a -STABLE build dated 7/17/2001. I've cvsup'd a >>>couple of times since then, through my recent build today: >>> >>>FreeBSD 4.4-PRERELEASE (CORP2) #4: Mon Aug 13 14:34:10 EDT 2001 >>> >>>I'm still experiencing the same problem. For the past month, this= machine >>>would hard lock every few days, when it was sitting idle (not in=20 >>>production), >>>seemingly without reason. After reading some of the new posts about the >>>fxp messages on the list, I decided to do some testing... If I set up a= =20 >>>ping -f >>>to hammer away at the machine in question from just 2 other machines on >>>the LAN, I can make corp2 (the machine with the fxp problem) kernel= panic. >>> >>>At first it was surprising to me that this machine was experiencing a=20 >>>problem, >>>as it is running the same board as another machine that was just built= that >>>has been running -STABLE fine for at least 2 months now (it's current= build >>>date is 7/24/2001). After further investigation, I found corp2 was= actually >>>running the Intel "D815EEA2" board - a slightly different model than the >>>"D815EEA" board the other (working) machine is running. >>> >>>A snippet from Intel's website: >>> >>>What is the difference between Intel Desktop Boards D815EEA and >>>D815EPEA and the Intel Desktop Boards D815EEA2 and D815EPEA2? >>>The main differences are Intel=AE Desktop Boards D815EEA2 and D815EPEA2 >>>now support two additional USB ports out the back panel (a total of=20 >>>four). Also >>>the game port on the back panel has been removed. >>>Note: The D815EEA2 and D815EPEA2 require a different I/O shield and >>>software image than the one used on D815EEA and D815EPEA. >>> >>>Possible bug in just the A2? >>> >>>Anyway.. here is the dmesg, and stack trace from the crash: >>>(notice the fxp0 / SCB timeouts too..) >>> >>>Fatal trap 10: trace trap while in kernel mode >>>instruction pointer =3D 0x8:0xc0212d86 >>>stack pointer =3D 0x10:0xc025085c >>>frame pointer =3D 0x10:0x0 >>>code segment =3D base 0x0, limit 0xfffff, type 0x1b >>> =3D DPL 0, pres 1, def32 1, gran 1 >>>processor eflags =3D interrupt enabled, IOPL =3D 0 >>>current process =3D Idle >>>interrupt mask =3D none >>>trap number =3D 10 >>>panic: trace trap >>> >>>syncing disks... 4 4 4 4 4 4 4 fxp0: SCB timeout: 0x70, 0x0, 0x50 0x400 >>>4 4 4 4 4 fxp0: SCB timeout: 0x80, 0x0, 0x50 0x400 >>>fxp0: SCB timeout: 0x80, 0x0, 0x50 0x400 >>>4 4 4 4 4 4 fxp0: SCB timeout: 0x80, 0x0, 0x50 0x400 >>>fxp0: SCB timeout: 0x80, 0x0, 0x50 0x400 >>>4 4 >>>giving up on 4 buffers >>>ad0: WRITE command timeout tag=3D0 serv=3D0 - resetting >>>ata0: resetting devices .. done >>>fxp0: SCB timeout: 0x80, 0x0, 0x50 0x400 >>>fxp0: SCB timeout: 0x80, 0x0, 0x50 0x400 >>>Uptime: 11m10s >>> >>>dumping to dev #ad/0x30001, offset 24787 >>>dump ata0: resetting devices .. done >>>254 253 252 251 250 249 248 247 246 245 244 243 242 241 240 239 238 237= =20 >>>236 235 234 233 232 231 230 229 228 227 226 225 224 223 222 221 220 219= =20 >>>218 217 216 215 214 213 212 211 210 209 208 207 206 205 204 203 202 201= =20 >>>200 199 198 197 196 195 194 193 192 191 190 189 188 187 186 185 184 183= =20 >>>182 181 180 179 178 177 176 175 174 173 172 171 170 169 168 167 166 165= =20 >>>164 163 162 161 160 159 158 157 156 155 154 153 152 151 150 149 148 147= =20 >>>146 145 144 143 142 141 140 139 138 137 136 135 134 133 132 131 130 129= =20 >>>128 127 126 125 124 123 122 121 120 119 118 117 116 115 114 113 112 111= =20 >>>110 109 108 107 106 105 104 103 102 101 100 99 98 97 96 95 94 93 92 91=20 >>>90 89 88 87 86 85 84 83 82 81 80 79 78 77 76 75 74 73 72 71 70 69 68 67= =20 >>>66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 48 47 46 45 44 43= =20 >>>42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19= =20 >>>18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 succeeded >>>Automatic reboot in 15 seconds - press a key on the console to abort >>>Rebooting... >>>Copyright (c) 1992-2001 The FreeBSD Project. >>>Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994 >>> The Regents of the University of California. All rights= reserved. >>>FreeBSD 4.4-PRERELEASE #4: Mon Aug 13 14:34:10 EDT 2001 >>> root@corp2.:/usr/obj/usr/src/sys/CORP2 >>>Timecounter "i8254" frequency 1193182 Hz >>>Timecounter "TSC" frequency 996769942 Hz >>>CPU: Pentium III/Pentium III Xeon/Celeron (996.77-MHz 686-class CPU) >>> Origin =3D "GenuineIntel" Id =3D 0x68a Stepping =3D 10 >>>Features=3D0x383f9ff >>>real memory =3D 267124736 (260864K bytes) >>>config> di sn0 >>>config> di lnc0 >>>config> di ie0 >>>config> di cs0 >>>config> q >>>avail memory =3D 257245184 (251216K bytes) >>>Preloaded elf kernel "kernel" at 0xc02d5000. >>>Preloaded userconfig_script "/boot/kernel.conf" at 0xc02d509c. >>>Pentium Pro MTRR support enabled >>>md0: Malloc disk >>>npx0: on motherboard >>>npx0: INT 16 interface >>>pcib0: on motherboard >>>pci0: on pcib0 >>>pci0: at 2.0 irq 11 >>>pcib1: at device 30.0 on= pci0 >>>pci1: on pcib1 >>>fxp0: port 0xdf00-0xdf3f mem=20 >>>0xff8ff000-0xff8fffff irq 11 at device 8.0 on pci1 >>>fxp0: Ethernet address 00:03:47:xx:xx:xx >>>inphy0: on miibus0 >>>inphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto >>>isab0: at device 31.0 on= pci0 >>>isa0: on isab0 >>>atapci0: port 0xffa0-0xffaf at device=20 >>>31.1 on pci0 >>>ata0: at 0x1f0 irq 14 on atapci0 >>>ata1: at 0x170 irq 15 on atapci0 >>>pci0: at 31.2 irq 11 >>>pci0: (vendor=3D0x8086, dev=3D0x2443) at 31.3 irq 9 >>>pci0: at 31.4 irq 10 >>>pci0: (vendor=3D0x8086, dev=3D0x2445) at 31.5 irq 9 >>>orm0: