Date: Mon, 15 Mar 1999 10:18:37 -0800 (PST) From: Matthew Jacob <mjacob@feral.com> To: Andrew Gallatin <gallatin@cs.duke.edu> Cc: alpha@FreeBSD.ORG Subject: revisiting...old business...Alpha I/O vs. Mem Mapping Message-ID: <Pine.LNX.4.04.9903151016380.7456-100000@feral-gw> In-Reply-To: <14043.64310.216520.299580@grasshopper.cs.duke.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
> Gary Palmer writes: > > found-> vendor=0x1077, dev=0x1020, revid=0x02 > > class=01-00-00, hdrtype=0x00, mfdev=0 > > subordinatebus=0 secondarybus=0 > > intpin=a, irq=17 > > map[0]: type 4, range 32, base 00000000, size 8 > > map[1]: type 1, range 32, base 82221000, size 12 > > Qlogic ISP Driver, FreeBSD CAM Version 0.98, Core Version 1.6 > > isp0: <Qlogic ISP 10X0 PCI SCSI Adapter> rev 0x02 int a irq 17 on pci1.1.0 > > isp0: using I/O space register mapping > > If I remember right, the 20152 in an AS600 doesn't support I/O space > mapping of devices behind it. That address of 00000000 sure looks > bogus to me anyway.. For devices behind a ppb, we should be checking > that PCIM_CMD_PORTEN is set on the bridge its behind.. I don't really > know how to architect that though.. > > At any rate, try setting SCSI_ISP_PREFER_MEM_MAP to 1 in isp_pci.c & > see if that helps. > > Drew Drew- you wanted to make MEM map the default for all alpha platforms, but I've had a reluctance to do so. If I/O mapping is broken here, perhaps the best place to fix this is in the PCI code? What do you think? -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.04.9903151016380.7456-100000>