On 06/04/11 18:18, David wrote: > On 5 June 2011 10:59, JD<jd1008@xxxxxxxxx> wrote: >> On 06/04/11 17:36, David wrote: >>> On 5 June 2011 10:19, Aaron Konstam<akonstam@xxxxxxxxxxxxx> wrote: >>>> It would be better if partition ends on a cylinder boundary but the >>>> system will still work. >>> I am curious about *how* specifically that "it would be better". >>> In what situation is a end "cylinder boundary" important or relevant? >> The filesystem builds it's map of blocks in groups >> of cylinders. >> You can see this clearly when you create a partition >> which will use up the rest of the disk (ie, to end of disk). >> often about 8mb always remains unused at end of disk >> because it is not a complete cylinder. > Thanks for your interest and reply. > > "because it is not a complete CYLINDER" (my emphasis added) > I am curious if you can cite a reference for this? > What filesystem are you referring to? > So (assuming some linux filesystem) if we looked at its mkfs code, we > would see that it cares about obsolete (and meaningless under linux, > not to mention irrevocably broken for large disks) MSDOS cylinders? > > Just asking because I feel that this is unlikely but would be pleased > to be corrected by any authoritative facts you can cite :-) I think you can read the kernel source code as well as anyone. So you have to do some homework :) -- users mailing list users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
