Re: hdparm configuration help
On Thu, 30 Aug 2007 02:08:28 -0700, Joris Huizer wrote:
> Hello,
>
> After the recent udev + hdparm problems, I'm thinking of reconfiguring
> hdparm (hdparm currently is not configured, just reinstalled, so I'm
> assuming it's currently using default settings)
>
>
> This is the output of `hdparm -v -i /dev/hda`:
>
>
> /dev/hda:
> multcount = 0 (off)
> IO_support = 1 (32-bit)
> unmaskirq = 1 (on)
> using_dma = 0 (off)
> keepsettings = 0 (off)
> readonly = 0 (off)
> readahead = 256 (on)
> geometry = 65535/16/63, sectors = 78165360, start
> = 0
<snip>
>
> /dev/hdb:
> multcount = 0 (off)
> IO_support = 1 (32-bit)
> unmaskirq = 1 (on)
> using_dma = 0 (off)
> keepsettings = 0 (off)
> readonly = 0 (off)
> readahead = 256 (on)
> geometry = 65535/16/63, sectors = 80293248, start
> = 0
<snip>
> Could someone tell me what settings are appropiate for my hard discs?
I
> tried before with, what didn't really seem to push things, but
problems
> with modules not loading reappeared
>
You need to turn on at least dma. I suppose the output of hdparm
-t /dev/yourdisk is not very cheerful with dma turned off.
I recently built a new system and it defaults to dma turned on. I've not
edited /etc/defaults/hdparm or /etc/hdparm/.conf. So as to why yours
seems to be turned off I do not know.
I get this:
cevnet:~#hdparm /dev/hda
/dev/hda:
multcount = 0 (off)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 2048 (on)
geometry = 65535/16/63, sectors = 80293248, start = 0
cevnet:~#hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads: 92 MB in 3.14 seconds = 29.31 MB/sec
the other disc does better than that though:
cevnet:~#hdparm /dev/hdb
/dev/hdb:
multcount = 0 (off)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 2048 (on)
geometry = 24792/255/63, sectors = 398297088, start = 0
cevnet:~#hdparm -t /dev/hdb
/dev/hdb:
Timing buffered disk reads: 172 MB in 3.02 seconds = 56.95 MB/sec
--
Chris <nws@cevnet.mine.nu>
Reply to: