Date: Mon, 18 Jun 2001 23:47:17 +0900 From: Makoto MATSUSHITA <matusita@jp.FreeBSD.org> To: current@FreeBSD.ORG Subject: Problems with ata probing twice. Message-ID: <20010618234717K.matusita@jp.FreeBSD.org> In-Reply-To: <20010618091710.A72197@peitho.fxp.org> References: <20010618141337.L728@tao.org.uk> <20010618141337.L728@tao.org.uk> <20010618091710.A72197@peitho.fxp.org>
next in thread | previous in thread | raw e-mail | index | archive | help
jedgar> Commenting hints.psm.0.* and hint.atkbd.0.* from /boot/device.hints jedgar> (http://docs.freebsd.org/cgi/getmsg.cgi?fetch=84052+0+current/freebsd-current) jedgar> works here. I have similar problem; 'ata' bus is detected twice. We jp.FreeBSD.org have donated TrustGuard iSV, which is 13 PCs in 4U (<URL:http://www.trustguard.co.jp/products/server/isv.htm>). FreeBSD 4-stable works fine, but 5-current (as of mid-March/2001; early-June/2001 kernel causes unwanted panic but it's another story) kernel detects ata bus twice, even if there is only one ata bus (and one HDD on the bus). (from dmesg output) atapci0: <Intel PIIX4 ATA33 controller> port 0xf000-0xf00f at device 7.1 on pci0 ata0: at 0x1f0 irq 14 on atapci0 ata1: at 0x170 irq 15 on atapci0 I've tried to figure out what's going on with atacontrol(8), and got: ushi % sudo atacontrol info ata0 Master slot: ad0: <IBM-DJSA-220/JS4OAC3A> ATA/ATAPI rev 5 Slave slot: no device present ushi % sudo atacontrol info ata1 Master slot: ad0: <IBM-DJSA-220/JS4OAC3A> ATA/ATAPI rev 5 Slave slot: no device present Apparantly, the kernel attaches same bus/HDD _twice_. Detecting atkbd/psm twice can be disabled with device.hints, but how to do with ata bus? I've commented out all hw.ata.* line in device.hints but no lacks... This causes another problem. In fact, this iSV has two fxp ethernet device, and one of them uses IRQ 15. Since ata1 is detected before fxp, I've got: (from dmesg output) fxp0: <Intel Pro 10/100B/100+ Ethernet> port 0xe400-0xe43f mem 0xd2420000-0xd243ffff,0xd2440000-0xd2440fff irq 15 at device 8.0 on pci0 fxp0: could not map interrupt device_probe_and_attach: fxp0 attach returned 6 fxp0: <Intel Pro 10/100B/100+ Ethernet> port 0xe800-0xe83f mem 0xd2400000-0xd241ffff,0xd2441000-0xd2441fff irq 11 at device 9.0 on pci0 fxp0: Ethernet address 00:10:f3:03:52:32 inphy0: <i82555 10/100 media interface> on miibus0 inphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto Kernel refuses to attach first fxp device. *** Is there any way to stop detecting ata1 in this case? Why kernel detects ata1 which doesn't exist? Since this machine will be 5-current buildbox for our SNAPSHOTs service, there are no options that using 4-stable kernel instead. -- - Makoto `MAR' MATSUSHITA To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010618234717K.matusita>