[ale] boot parms ide=???

Dow Hurst dhurst at kennesaw.edu
Wed Feb 19 15:02:00 EST 2003


This might be more what your after:
http://kurp-www.hut.fi/vlbi/instr/pcwr/pcwr.html


Dow Hurst wrote:

> Geof,
> In single user mode first test throughput with:
>
> hdparm -tT /dev/hdx
>
> do this a few times to get a decent average.
>
> then,
> hdparm /dev/hdx
>
> will give you the current settings
>
> after that,
> hdparm -d 1 /dev/hdx
>
> will turn on dma.
>
> That is the simple howto hdparm.  Now, to use the -u option you'd have 
> to experiment with a file system in read only mode unless you don't 
> care about what is on it.  I didn't touch that as the throughput I 
> have is great.  Do you want to see just how much performance you can get?
>
> I searched Google for
> high performance computing Linux IDE hdparm and got a bunch of hits.  
> Here is a link with some data on hdparm but not the -u option:
> http://www.networkcomputing.com/1122/1122ws2.html?ls=NCJS_1122bt
>
> Here is a link with some info down at the bottom about hparm but isn't 
> something you don't already know.
> http://slashdot.org/askslashdot/01/03/07/0512206.shtml
>
> Hope this helps a little,
> Dow
>
>
>
>
> Geoffrey wrote:
>
>> I'm still trying to figure this one out.  Can't find much 
>> documentation on all this.  I've since removed the boot parm 
>> reference and it appears that using hdparm is sufficient to get the 
>> most bang out of my drives.
>>
>> Cory T. Echols wrote:
>>
>>> On 02/19, Geoffrey wrote:
>>>
>>>> Interesting, I've not seen this.  I guess the kernel has not kept 
>>>> up with the latest technology, since the highest available value is 
>>>> 66, as noted above.
>>>>
>>>> Thanks.
>>>
>>>
>>>
>>>
>>> We may be talking about two different things here, but it is my
>>> understanding that the clock speed specified by the "idebus" parameter
>>> is not the same as the frequency that is being referred to on ide drive
>>> controllers.  In other words, setting "idebus=100" is not required when
>>> using a 100Mhz drive controller.  I've read the idebus parameter has
>>> something to do with PIO modes, and that PIO modes aren't used if 
>>> you're
>>> running a 100Mhz drive controller properly.
>>>
>>> I know for a fact that some Promise brand ide controllers, and the
>>> 100Mhz controllers on some ASUS boards are fully supported in 2.4
>>> kernels.  How you configure that support, I'm not sure.  I don't own 
>>> any
>>> 100Mhz controllers.
>>>
>>>
>>
>

-- 
__________________________________________________________
Dow Hurst                  Office: 770-499-3428
Systems Support Specialist    Fax: 770-423-6744
1000 Chastain Rd., Bldg. 12
Chemistry Department SC428  Email:dhurst at kennesaw.edu
Kennesaw State University         Dow.Hurst at mindspring.com
Kennesaw, GA 30144
*********************************
*Computational Chemistry is fun!*
*********************************


_______________________________________________
Ale mailing list
Ale at ale.org
http://www.ale.org/mailman/listinfo/ale






More information about the Ale mailing list