On Fri, Oct 12, 2001 at 08:39:26PM +0200, Neal H Walfield wrote: > When you say that you are running it as a normal user, does the normal > user have access to the device or are you using a file store?
Th user had read access to the device /dev/hd0. > I am > also interested in seeing exactly what commands you executed and what > the geometry of the drive is. "parted" and then "check 1". The geometry as seen under GNU/Hurd on one disk is: Disk geometry for /dev/hd0: 0.000-4887.421 megabytes Not very helpful. Anyway, the geometry of the disks as detected by the kernel is: 662/240/63 UDMA The other disk has also a geometry, but I have to reboot to find out how GNU Mach sees it. Linux reports it as a 5605/255/63 (physical 89355/16/63) > Also, have you tried Parted under > GNU/Linux? If so, what kind of results are you getting. The check itself, as far as it runs, seems to work ok. I get the same warnings about free space at the end of the partition etc as in GNU/Linux. In GNU/Linux, nothing else happens (no crash, no error). > Finally, > could you give a more detailed explanation of what crashing GNU Mach > means: are you brought to ddb, are any errors sent to the console? The error is different each time, the crash itself spurious. I am quite certain that random data in the kernel address space is corrupted. The symptoms are page faults, assertion failures and hangs in appearingly random places (schedular, virtual memory handler, etc). > Parted only uses the store interface; nothing terribly special. Seems that there is an access pattern that makes GNU Mach run wild. I forgot if I mentioned this in my other mail: OSKit Mach doesn't crash. Thanks, Marcus -- `Rhubarb is no Egyptian god.' Debian http://www.debian.org [EMAIL PROTECTED] Marcus Brinkmann GNU http://www.gnu.org [EMAIL PROTECTED] [EMAIL PROTECTED] http://www.marcus-brinkmann.de _______________________________________________ Bug-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-hurd
