Søren Schmidt wrote: > Travis Mikalson wrote: >> Søren Schmidt wrote: >>> Anyhow, enough ramblings, the attached patch makes 48bit access work >>> on the Tyan, so that problem should be fixored, can you all please >>> test that and let me know how it goes, thanks! >>> >>> Patch is against RELENG_7, but should fit on RELENG_6 and -current as >>> well. >>> >>> -Søren >> >> That's certainly a good thing to have found and fixed, but >> unfortunately I cannot easily test this specific aspect that you've >> fixed :) >> >> I patched and booted with a disk plugged into my on-board SATA, the >> memory corruption issue is still there so all I can really see is init >> and sh sig11'ing and zfs spitting out checksum errors, sorry. > Hmm, I havn't booted on this disk yet, but I can mount it and run a > complete buildworld on it without a hickup, more tomorrow when I have a > little more time, now I need to get some shuteye :) I booted from a HD connected to a PCI SATA controller and plugged my second disk (of a ZFS mirror) into the on-board SATA. zpool scrub still counted up about 50 checksum errors (for 1.2GB of data) on the disk that was connected to the on-board SATA controller. No change from before, but I don't think I had ever yet put enough data on my disk at any point to make it past 137G anyway. If you're not booting from the disk, in order to really get a good handle on the subtle data corruption you may want to setup ZFS so you have some failing checksums to look at. Or just boot from your SATA disk and watch the explosion :) -- TerraNovaNet Internet Services - Key Largo, FL Voice: (305)453-4011 x101 Fax: (305)451-5991 http://www.terranova.net/ ---------------------------------------------- Life's not fair, but the root password helps.Received on Sat Dec 08 2007 - 01:44:06 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:23 UTC