questions on S-ATA and ICH5 (now owns hardware :)

David Leimbach leimy2k at
Thu Jul 24 05:02:40 PDT 2003

>>   atapci1: <Intel ICH5 SATA150 controller> port 
>> 0xd000-0xd00f,0xcc00-0xcc03,0xc800-0xc807,0xc400-0xc403,0xc000-0xc007 
>> irq 9 at device 31.2 on pci0
>>   ...
>>   ata2: at 0xc000 on atapci1
>>   ad4: success setting UDMA133 on Intel ICH5 chip
>>   ad4: <ST3120023AS/3.01> ATA-6 disk at ata2-master
>>   ad4: 114473MB (234441648 sectors), 232581 C, 16 H, 63 S, 512 B
>>   ad4: 16 secs/int, 1 depth queue, UDMA133
>>   ad4: piomode=12 dmamode=34 udmamode=70 cblid=1
>> Shouldn't this drive be found as a SATA150 device?
> Well, technically yes, but in practice the modes the drives reports
> back as supported are the old UDMA ones, however the interface will
> run at SATA150 speed no matter what. I've not found a surefire way
> to tell this apart yet that also gives resonable results if you
> use a SATA->PATA dongle and other wierd comboes now possible...

Yeah.  My drive shows up as UDMA133 also.  What I did notice is that
my WD Raptor was slightly outperformed a few times on UFS2 by my actual
ATA-100 Western Digital drive.  This seems somewhat bad as the Raptor
costs a hell of a lot more and one would hope that it would pound the
ATA-100 drive pretty thoroughly.

Even the CPU overheads on both drives were about the same.  Maybe its 
8MB caching :).  I haven't found a good reason yet.

Soeren, do you actually have SATA drives to test with or do you just 
SATA adapters with SATA->PATA dongles?  Do you need more hardware to run
some benchmarks yourself?  [I don't know that I can afford to buy you
a SATA disk but I wonder anyway :)]


> -Søren
> _______________________________________________
> freebsd-current at mailing list
> To unsubscribe, send any mail to 
> "freebsd-current-unsubscribe at"

More information about the freebsd-current mailing list