Date: Sat, 09 Dec 2006 12:27:28 +0100 From: "J. Martin Petersen" <jmp.lists@alvorlig.dk> To: freebsd-current@FreeBSD.ORG Cc: Peter_Losher@isc.org Subject: Re: FreeBSD 6 (or 7) on a HP DL140 G3 SATA Message-ID: <457A9DA0.1020601@alvorlig.dk> In-Reply-To: <4578E1AD.3090505@isc.org> References: <4578E1AD.3090505@isc.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Peter Losher wrote: > Just got my hands on a HP DL140 G3 (generation 3) server, which has a > LSI Logic SAS controller. I suppose you mean the one with the hot-swappable disks? > You can run the live CD or install FreeBSD > (6.2-RC1 or 7-CURRENT) onto the HD's, but when you boot from the disks, > the boot loader barfs: > > -=- > int=0000000d err=0000001a efl=00030287 eip=0000291d > eax=1400000a ebx=00000b3e ecx=00000000 edx=0000c900 > esi=00000d1c edi=00000001 ebp=00000206 esp=00000200 > cs=c900 ds=9a00 es=9a00 fs=0000 gs=0000 ss=9a00 > cs:eip=cc 68 32 06 ff 34 e8 db-fc 83 c4 04 89 46 fe 5f > 5e c9 c3 55 8b ec 1e 33-c0 8e d8 a0 75 04 3c 00 > ss:esp=1c 0d 02 00 00 00 34 02-d8 45 1c 0d 00 00 00 14 > 00 00 00 00 01 00 00 00-1c 0d 02 00 00 00 00 00 > BTX halted > -=- > > Here is the relevant BIOS output from the controller: > > -=- > LSI Logic Corp. MPT SAS BIOS > MPTBIOS-6.06.08.00 (2006.05.16) > Copyright 2000-2006 LSI Logic Corp. > > > SLOT ID LUN VENDOR PRODUCT REVISION CAPACITY > ---- --- --- -------- ---------------- ---------- ---------- > 3 LSILogic SAS1068-IR 1.10.01.00 > -=- > > Has anyone gotten one of these to run FreeBSD from the HD's successfully? Yes; we got similar results until we upgraded the system BIOS and the controller firmware (both are available somewhere on hp.com). We're now having a different problem; once -CURRENT (or 6.2-RC1) is installed and running, the disk performance is abysmal. I/O-wise it is maxing out around 6MB/sec, which seems far from impressive. We have tried both with and without RAID configured through the controller (i.e., with RAID-1 and with just two seperate disks), and results are identical. There does not seem to be any spurious interrupts, and there is no load on the system when doing disk I/O. This is for -CURRENT as of yesterday evening (after the latest commits to dev/mpt). Martin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?457A9DA0.1020601>