On Mon, 2007-03-26 at 18:55 +0200, Tzahi Fadida wrote: > On Monday 26 March 2007 18:04, Erik Mouw wrote: > > On Mon, Mar 26, 2007 at 05:23:50PM +0200, Tzahi Fadida wrote: > > > On Monday 26 March 2007 16:27, burns.ethan@xxxxxxxxx wrote: > > > > On Mon, Mar 26, 2007 at 03:43:56PM +0200, Tzahi Fadida wrote: > > > > > See here for a description of a sector: > > > > > http://en.wikipedia.org/wiki/Cylinder-head-sector > > > > > > > > This article is sort of misleading. > > > > > > > > I didn't do more than skim the beginning of the first paragraph, but > > > > you may find that this is a better explanation: > > > > > > > > http://www.dewassoc.com/kbase/hard_drives/hard_disk_sector_structures.h > > > >tm > > > > > > I c. 10x! > > > However, in LDD3 they did not explain how to define a geometry with zone > > > bit recording. I.e. a geometry with different zone track groups. > > > > That's why CHS addressing has been obsoleted by LBA addressing: you > > just tell the drive "I want sector 1234567" and it will figure out by > > itself which sector on which track in which zone it is. > > That is not what i meant. Without specifying the geometry of the drive, how > the schedulers would know accurately which sector is near which and will know > how to order them correctly. Especially same sector position on different > platters and sides. AFAIK, this information isn't exported from the disk - you can only assume that sector numbers that are close to each other are close to each other on the disk. Avishay -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
