* Glenn Enright <[EMAIL PROTECTED]> [2005-03-03 07:45]:
> > > gandalf root # hdparm -t /dev/hde /dev/sdc
> > >
> > > /dev/hde:
> > >  Timing buffered disk reads:  156 MB in  3.02 seconds =  51.73 MB/sec
> > >
> > > /dev/sdc:
> > >  Timing buffered disk reads:  170 MB in  3.01 seconds =  56.41 MB/sec
> > >
> > > hdparm is for IDE drives, not "SCSI", so you can't really expect it to
> > > work properly.
> > > I get a HDIO_GET_MULTCOUNT error when I view the current settings of a
> > > SATA disk.
> >
> > Hmmmm.., after all the /dev/sda should have 57 MB/sec, this is diamond 9
> > with 8 MB cache, hda is diamond 8 with 2 MB.
> >
> > So right now I do not whith on is faster :-(
> >
> > Is there something like hdparm for scsi on libata?
> 
> As far as I know there is not. The technology is till to new for any cool 
> stuff like that to be available on open-source. Hell the drivers are still a 
> moving target so can really expect any apps for them yet.

Date: Wed, 02 Mar 2005 21:23:34 -0500
From: Jeff Garzik <[EMAIL PROTECTED]>
To: Damian Kolkowski <[EMAIL PROTECTED]>
Cc: [email protected]
Subject: Re: [BUG] - SATA / ioctl(). (HDIO_GET_IDENTITY failed...)

> Is there any patch to correct libata working with ioctl()?

libata works fine with ioctl(2).

> .~. # hdparm -t /dev/hda /dev/sda
> /dev/hda:
>  Timing buffered disk reads:  174 MB in  3.03 seconds =  57.36 MB/sec
> /dev/sda:
>  Timing buffered disk reads:  152 MB in  3.03 seconds =  50.11 MB/sec
> HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl
> for device
>.~. #

Arbitrary command execution is only supported in libata-dev tree for now.

-- 
### Damian Kołkowski (dEiMoS) ## http://kolkowski.no-ip.org/ ###
# echo teb.cv-ba.vxfjbxybx.anvznq | rot13 | rev | sed s/\\./@/ #

--
[email protected] mailing list

Reply via email to