Date: Wed, 7 Apr 1999 23:23:22 +0200 From: Stefan Esser <se@mi.uni-koeln.de> To: Michel TALON <talon@lpthe.jussieu.fr> Cc: freebsd-stable@freebsd.org, Stefan Esser <se@freebsd.org> Subject: Re: Strange PCI lookup at boot Message-ID: <19990407232322.B1574@dialup124.mi.uni-koeln.de> In-Reply-To: <199904071123.NAA20523@niobe.lpthe.jussieu.fr>; from Michel TALON on Wed, Apr 07, 1999 at 01:23:04PM %2B0200 References: <199904071123.NAA20523@niobe.lpthe.jussieu.fr>
next in thread | previous in thread | raw e-mail | index | archive | help
On 1999-04-07 13:23 +0200, Michel TALON <talon@lpthe.jussieu.fr> wrote: > now that i have two machines running FreeBSD-3.1-Stable, i > can see a strange behaviour of the code looking for PCI > devices at boot on one of them. here is dmesg: Two machines with different behaviour. Hmmm, what's the difference between those two machines (main-board, BIOS, ...) ? > Copyright (c) 1992-1999 FreeBSD Inc. > Copyright (c) 1982, 1986, 1989, 1991, 1993 > The Regents of the University of California. All rights reserved. > FreeBSD 3.1-STABLE #0: Tue Apr 6 15:56:51 CEST 1999 > talon@aglae.lpthe.jussieu.fr:/usr/src/sys/compile/AGLAE > Timecounter "i8254" frequency 1193182 Hz > Timecounter "TSC" frequency 200004435 Hz > CPU: Pentium Pro (200.00-MHz 686-class CPU) > Origin = "GenuineIntel" Id = 0x616 Stepping=6 > Features=0xf9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR,PGE,MCA,CMOV> > real memory = 134217728 (131072K bytes) > avail memory = 127905792 (124908K bytes) > Preloaded elf kernel "kernel" at 0xf028c000. > Preloaded userconfig_script "/kernel.config" at 0xf028c09c. > Preloaded splash_image_data "/boot/splash.bmp" at 0xf028c0e8. > Preloaded elf module "splash_bmp.ko" at 0xf028c138. > Probing for devices on PCI bus 0: > chip0: <Intel 82378IB PCI to ISA bridge> rev 0x43 on pci0.2.0 > xl0: <3Com 3c900-COMBO Etherlink XL> rev 0x00 int a irq 11 on pci0.4.0 > xl0: Ethernet address: 00:60:97:1f:75:ca > xl0: selecting 10baseT transceiver, half duplex > vga0: <Matrox MGA 2064W graphics accelerator> rev 0x01 int a irq 255 on pci0.5 > ide_pci0: <PCI IDE controller (busmaster capable)> rev 0x01 int a irq 14 on pc > .8.0 > chip1: <Intel 82453KX/GX (Orion) PCI memory controller> rev 0x04 on pci0.20.0 > chip2: <Intel 82454KX/GX (Orion) host to PCI bridge> rev 0x04 on pci0.25.0 > Probing for devices on PCI bus 1: The Orion mainboards I heard about (never have seen one ;-) had some on-board components on PCI bus 0 and most available PCI slots connected to a second Host Bridge chip as bus 1. The PCI code does the right thing in such a case (AFAIK), but it looks like your system only has one Host Bridge, and a silly number of buses is probed ... > Probing for devices on PCI bus 127: The maximum number of buses assumed to exist in a system is derived from some config space register in the chip-set. Please send me the output of the following command: pciconf -r pci0:25:0 0x48 The two leftmost byte values printed will be the secondary and subordinate bus numbers programmed into the chip-set by the PCI BIOS. > changing root device to wd0s1a > WARNING: / was not properly dismounted > ffs_mountfs: superblock updated for soft updates It appears the system is operational and you are just annoyed by those useless probe message lines ? > Of course on the other one there is probe for PCI bus 0 and 1 > only. I wonder what could cause such strange lookup. The assumption, that there might be devices on a higher numbered bus. > I have already answered this same question, but got no > answer, so i waited to see if this problem would survive an > upgrade to recent Stable, which is the case. Sorry, didn't see your earlier question. I may be able to improve the probe code based on the value you report (the pciconf -r result). Regards, STefan 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?19990407232322.B1574>