Messages in this thread |  | | From | jbradford@dial ... | Subject | Re: Heuristic readahead for filesystems | Date | Tue, 17 Sep 2002 11:37:56 +0100 (BST) |
| |
> > Well they could read contiguous sectors if the sector interleave > > was correctly determined and the correct interleave was set > > while low-level formatting. Now-days, interleave is either ignored > > or unavailable because there is a sector buffer that can contain > > an entire track of data. Some SCSI drives have sector buffers > > that can contain a whole cylinder of data. > > When I say contiguous I mean contiguous not interleaved, sonny. I had > CP/M (and UCSD Pascal) reading physically contiguous sectors on the > disk with no lost speed. That means I read, with my DSSD format of > 9 sectors each 512 bytes in size per side 18 full tracks 19 revolutions > of the disk. I did skew the sector numbers to allow for seeks. But I > did not interleave the tracks. It was not necessary with clean and > correct code. I rather resent the presumption that I am a dumb bitch > here.
Ah, but the *really* clever thing to do at the time, on systems where you couldn't optimally achieve 1:1 interleave on a floppy, was to allocate sectors on alternating sides of the disk. So, you could, for example, read two tracks in 3 revolutions, instead of 6, in the case of 3:1 interleave :-).
John. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |