Messages in this thread | | | Date | Fri, 21 May 1999 11:52:49 +0100 | From | Giuliano P Procida <> | Subject | bogus IDE disk geometry, what to do? |
| |
I know this came up recently, so apologies for the question. Just what should I do (LILO and fdisk) with this disk (IBM "8.4"Gb) to ensure maximum future and other OS compatibility?
I have set this oldish BIOS to "auto". These are CHS values reported by default, using kernel 2.2.9 and hdparm 3.5:
kernel 1027/255/63 (kernel 2.0.36 1024/255/63)
hdparm -I (raw) CurCHS 16383/16/63 LBA CHS 1023/256/63 CurSects 16514064 LBAsects 16514064
hdparm -i (cooked) CurCHS 1027/255/63 LBA CHS 513/510/63 CurSects 16514064 LBAsects 16514064
1024*255*63 = 16450560 513*510*63 = 16482690 1027*255*63 = 16498755 1023*256*63 = 16498944 16383*16*63 = 16514064
It would _seem_ that all the dividing down that hdparm does is a bit bogus when applied to odd numbers, as it results in the loss of the end of the disk.
The drives work fine when forced to 16383,16,63. This is also the CHS given by IBM in the datasheet (though they say, "as reported by the XYZ command") and the number of sectors also matches that. Going for 16384*16*63 gives seek errors, so that's too big.
Thanks for any reply, Giuliano Procida.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |