Date: Thu, 5 Jan 2006 00:02:24 -0600 From: Vulpes Velox <v.velox@vvelox.net> To: Heinz Suez <malagana15@yahoo.es> Cc: freebsd-hardware@freebsd.org Subject: Re: SiI 3112 time out problems and suggestions for a good sata controller. Message-ID: <20060105000224.2b274960@vixen42.vulpes> In-Reply-To: <20060105050138.16343.qmail@web26411.mail.ukl.yahoo.com> References: <20060104215937.3ba2c636@vixen42.vulpes> <20060105050138.16343.qmail@web26411.mail.ukl.yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
My plans are to get a new one. Unfurnately can't sell it, because it is built onto the board. I am currently working on picking a good one. Don't need RAID though, I plan to implement that using GEOM later on. Any opinions on if I should worry about corruption with those occasional messages till then? On Thu, 5 Jan 2006 06:01:38 +0100 (CET) Heinz Suez <malagana15@yahoo.es> wrote: > Hi. I had the same problem a month ago with the same controller. > The problem was that i bought this pci sata controller with the > sil3112 for an old computer (dual p3 550, 512mb ram pc133, pci > 2.1), so the transfer rate & buffer was higher than the mother > board was able to deliver...in other words, the sata controller > (the sil3121 chipset i mean) was not 100% backwards compatible with > pci 2.1 when it should be. I tried this controller with a brand new > computer and worked just fine with lots of hdd's. Now, you have 2 > solutions: Sell the controller if it's a pci card or buy a more > expensive card with a different chipset. There're a lot of sata > chipsets that support pci 2.1 or older computers, i have seen lots > of users running sata raid at full speed on p2 machines. Good luck. > Vulpes Velox <v.velox@vvelox.net> escribi=F3: I am getting entries > like this in dmesg. ad10: TIMEOUT - WRITE_DMA retrying (2 retries > left) LBA=3D355277839 ad10: TIMEOUT - WRITE_DMA retrying (2 retries > left) LBA=3D358666287 >=20 > smartctl -t long /dev/ad10 shows it to be good though. The results > from the times I have run it show it completing successfully every > time. I believe I have no reason to suspect a bad disk and it just a > crappy chip. From what I have read in the mailing list archives, I > believe this to be true. Any one have any opinions otherwise on this > or the like? >=20 >=20 > The machine this is running on has releng_5 on it. >=20 > I was wondering if any one had any suggestions on a good for a good > sata pci card that preferably has 4 connectors on it. >=20 >=20 > SiI 3112 sysctl entry if any one is curious. > dev.atapci.1.%desc: SiI 3112 SATA150 controller > dev.atapci.1.%driver: atapci > dev.atapci.1.%location: slot=3D11 function=3D0 > dev.atapci.1.%pnpinfo: vendor=3D0x1095 device=3D0x3112 subvendor=3D0x1095 > subdevice=3D0x6112 class=3D0x010400 dev.atapci.1.%parent: pci1
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060105000224.2b274960>