Date: Wed, 27 Jun 2001 00:41:26 -0700 From: David Greenman <dg@root.com> To: Mike Tancsa <mike@sentex.net> Cc: freebsd-stable@freebsd.org Subject: Re: More fxp problems (Different ?) Message-ID: <20010627004126.A53263@nexus.root.com> In-Reply-To: <5.1.0.14.0.20010626132049.04048e40@marble.sentex.ca>; from mike@sentex.net on Tue, Jun 26, 2001 at 01:32:51PM -0400 References: <5.1.0.14.0.20010626132049.04048e40@marble.sentex.ca>
next in thread | previous in thread | raw e-mail | index | archive | help
>This is a stange one. It seems like a hardware issue as the same two nics
>on a 815 or 440BX chipset did have these problems.
^ NOT?
...in any case, try this patch and let me know if the problem goes
away:
Index: if_fxp.c
===================================================================
RCS file: /home/ncvs/src/sys/dev/fxp/if_fxp.c,v
retrieving revision 1.110.2.4
diff -c -r1.110.2.4 if_fxp.c
*** if_fxp.c 2001/06/08 20:36:57 1.110.2.4
--- if_fxp.c 2001/06/27 07:48:29
***************
*** 490,501 ****
--- 490,503 ----
* If we are not a 82557 chip, we can enable extended features.
*/
if (sc->chip != FXP_CHIP_82557) {
+ #if 0
/*
* If there is a valid cacheline size (8 or 16 dwords),
* then turn on MWI.
*/
if (pci_read_config(dev, PCIR_CACHELNSZ, 1) != 0)
sc->flags |= FXP_FLAG_MWI_ENABLE;
+ #endif
/* turn on the extended TxCB feature */
sc->flags |= FXP_FLAG_EXT_TXCB;
-DG
David Greenman
Co-founder, The FreeBSD Project - http://www.freebsd.org
President, TeraSolutions, Inc. - http://www.terasolutions.com
Pave the road of life with opportunities.
>MB is an Asus CUV-4x-e Via82ct chipset 686b. Cards are
>1 2940U
>1 3Ware 2 port RAID-1 config (6x000 series)
>2 Intel nics, both show inphy1: <i82555 10/100 media interface> on miib
>inphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
>and both are from the same box. In short, all these parts do not play well
>together. If I pull the two nics and put in cheap old RealTeks the box
>works fine enough. But with the Intels (I have tried manual IRQs, auto,
>changing the slot order etc), I still get the nics locking up on a
>consistent basis. I even tried a card that was made a few years ago, but
>same results.
>
>
>Jun 26 08:52:02 granite /kernel: Timecounter "i8254" frequency 1193182 Hz
>Jun 26 08:52:02 granite /kernel: CPU: Pentium III/Pentium III Xeon/Celeron
>(870.58-MHz 686-class CPU)
>Jun 26 08:52:02 granite /kernel: Origin =
>"GenuineIntel" Id = 0x68a Stepping = 10
>Jun 26 08:52:02 granite /kernel:
>Features=0x383f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,MMX,FX
>SR,SSE>
>un 26 08:52:02 granite /kernel: npx0: INT 16 interface
>Jun 26 08:52:02 granite /kernel: pcib0: <Host to PCI bridge> on motherboard
>Jun 26 08:52:02 granite /kernel: pci0: <PCI bus> on pcib0
>Jun 26 08:52:02 granite /kernel: pcib2: <VIA 82C598MVP (Apollo MVP3)
>PCI-PCI (AGP) bridge> at device 1.0 on pci0
>Jun 26 08:52:02 granite /kernel: pci1: <PCI bus> on pcib2
>Jun 26 08:52:02 granite /kernel: isab0: <VIA 82C686 PCI-ISA bridge> at
>device 4.0 on pci0
>Jun 26 08:52:02 granite /kernel: isa0: <ISA bus> on isab0
>Jun 26 08:52:02 granite /kernel: atapci0: <VIA 82C686 ATA100 controller>
>port 0xd800-0xd80f at device 4.1 on pci0
>Jun 26 08:52:02 granite /kernel: ata0: at 0x1f0 irq 14 on atapci0
>Jun 26 08:52:02 granite /kernel: ata1: at 0x170 irq 15 on atapci0
>Jun 26 08:52:02 granite /kernel: pci0: <VIA 83C572 USB controller> at 4.2
>irq 10
>Jun 26 08:52:02 granite /kernel: pci0: <VIA 83C572 USB controller> at 4.3
>irq 10
>
>I have a very similar mix of parts in my news server and I have never seen
>this happen on the 440BX chipset. Is it something specific to the Via or
>this combo ? The errors were happening at 100 and 10Mb with various duplex
>settings. A few times the NIC totally locked up and I had to reboot the
>machine to recover.
>
>
>Jun 26 06:25:11 granite /kernel: fxp1: device timeout
>Jun 26 06:25:11 granite /kernel: fxp1: SCB timeout: 0x60, 0x0, 0x0 0x0
>Jun 26 06:25:11 granite /kernel: fxp1: SCB timeout: 0x6, 0x0, 0x0 0x0
>Jun 26 06:25:11 granite /kernel: fxp1: SCB timeout: 0x40, 0x0, 0x0 0x0
>Jun 26 06:25:11 granite /kernel: fxp1: DMA timeout
>Jun 26 06:25:12 granite /kernel: fxp1: SCB timeout: 0x10, 0x0, 0x0 0x0
>Jun 26 06:25:12 granite /kernel: fxp1: DMA timeout
>Jun 26 06:25:12 granite /kernel: fxp1: SCB timeout: 0x10, 0x0, 0x0 0x0
>Jun 26 06:25:12 granite /kernel: fxp1: SCB timeout: 0x10, 0x0, 0x0 0x0
>Jun 26 06:25:17 granite /kernel: fxp1: SCB timeout: 0x1, 0x0, 0x0 0x400
>Jun 26 06:25:19 granite /kernel: fxp1: SCB timeout: 0x81, 0x0, 0x0 0x400
>Jun 26 06:25:31 granite /kernel: fxp1: device timeout
>Jun 26 06:25:31 granite /kernel: fxp1: SCB timeout: 0x60, 0x0, 0x0 0x0
>Jun 26 06:25:31 granite /kernel: fxp1: SCB timeout: 0x6, 0x0, 0x0 0x0
>Jun 26 06:25:31 granite /kernel: fxp1: SCB timeout: 0x40, 0x0, 0x0 0x0
>Jun 26 06:25:31 granite /kernel: fxp1: DMA timeout
>Jun 26 06:25:31 granite /kernel: fxp1: SCB timeout: 0x10, 0x0, 0x0 0x0
>Jun 26 06:25:31 granite /kernel: fxp1: DMA timeout
>Jun 26 06:25:31 granite /kernel: fxp1: SCB timeout: 0x10, 0x0, 0x0 0x0
>Jun 26 06:25:31 granite /kernel: fxp1: SCB timeout: 0x10, 0x0, 0x0 0x0
>Jun 26 08:39:46 granite /kernel: fxp1: Ethernet address 00:d0:b7:92:32:b4
>Jun 26 08:40:09 granite /kernel: fxp1: device timeout
>Jun 26 08:40:24 granite /kernel: arp: 199.212.134.2 is on fxp1 but got
>reply from 00:90:27:b0:35:21 on fxp0
>Jun 26 08:40:32 granite /kernel: fxp1: device timeout
>Jun 26 08:41:04 granite /kernel: fxp1: device timeout
>Jun 26 08:47:04 granite /kernel: fxp0: <Intel Pro 10/100B/100+ Ethernet>
>port 0xb400-0xb43f mem 0xfa000000-0xfa0fffff,0xfa800000-0xfa800fff irq 7 at
>device 9.0 on pci0
>Jun 26 08:47:04 granite /kernel: fxp0: Ethernet address 00:d0:b7:92:32:b4
>Jun 26 08:47:27 granite /kernel: fxp0: device timeout
>Jun 26 08:47:51 granite /kernel: fxp0: device timeout
>Jun 26 08:48:14 granite /kernel: fxp0: device timeout
>
>
>
>To Unsubscribe: send mail to majordomo@FreeBSD.org
>with "unsubscribe freebsd-stable" in the body of the message
>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010627004126.A53263>