First up i was told after many days of solution search that *hdparm* was
broken with 2.4 and some chipsets. I see a new release of it at metalab
- but have not tried it, maybe its fixed now.
In any case, man hdparm for the statutory warnings and:
Rajesh Fowkar wrote:
>
> ------------------------------------------
> /dev/hda:
> multcount = 0 (off)
Try -m16 = minor speed increase.
> debian:~# dmesg | grep hda
> ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:DMA, hdb:pio
> hda: ST317221A, ATA DISK drive
> hda: 33683328 sectors (17246 MB) w/512KiB Cache, CHS=2096/255/63, UDMA(33)
> hda: hda1 hda2 < hda5 hda6 hda7 hda8 hda9 hda10 hda11 >
> ------------------------------------------
UDMA(33) means you are not getting 66. Another way of checking this is
passing hdparm -X67 or 68 or 69 - if its not functional, u are stuck
with probly a 40 pinner. Get a 80 pin cable.
> Results of hdparm on Kernel 2.4.2 -> No improvements ;-)
>
> ------------------------------------------
> debian:~# hdparm -tT /dev/hda
> /dev/hda:
> Timing buffer-cache reads: 128 MB in 1.82 seconds = 70.33 MB/sec
> Timing buffered disk reads: 64 MB in 14.72 seconds = 4.35 MB/sec
Doubt if it will change. Try bonnie to test your hard drive.
HTH
Ashwin
----------------------------------------------
Find out more about this and other Linux India
mailing lists at http://lists.linux-india.org/
