On Monday 2012-11-26 18:12, Romanator wrote:
On Sun 25 Nov 2012 08:15:15 AM EST, Jan Engelhardt wrote:
And you got the alignment. 2048, 146802688, 155187200, 155189248 and 870324224 are all multiples of 2048 (KByte), so where is the problem?
No problem. I realized that running the multiplication. The WD "Black" (it's using 4096 bytes) 2012 is different from the WD "Blue" that I am using from 2007.
Note that LBA addresses are in units of 512 bytes, and no physical sector size will change that.
It's very misleading to users. The Yast gui representation shows each partition's _start_ and _end_ cylinders increasing by one cylinder. Whereas upon completion "fdisk -lu=cylinders shows them overlapping.
Welcome to the Wonderful World of (IBM-)Microsoft. While I do not necessarily claim it was the first, MS-DOS was the most prominent one to date which used the stupid "1-based, last-inclusive" style, i.e. p1: start=cyl 1 end=cyl 1023 (for a total of 1024 cyls) p2: start=cyl 1024, end=cyl 2047 (for another 1024) Solaris, Apple and BSD, and even EFI GUID partition tables all use the much more sane "0-based, last-exclusive": p1: start=sector 0, length=1024 sects p2: start=sector 1024, length=1024 sects Sometimes people like to see "end" rather than length, and so, it becomes p1: start=sector 0, end=sector 1024 p2: start=sector 1024, end=sector 2048 Why does that make more sense than MS-DOS-style counted cyls, at least people with a bit of a math or CS clue? Because "end" minus "start" conveniently gives you the partition size. With the MS-DOS thing, you need to add a magic 1. And so, tools like parted and (modern versions of) `fdisk -lu=cylinders` use the 0-based last-exclusive scheme, even with MS-DOS partition tables. The cylinder thing in yast should die immediately. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org