2002-07-27 17:03:55

by Lionel Bouton

[permalink] [raw]
Subject: Re: Driver sis5513.c / sis630(ET)

Ralf wrote:

>Hi Lionel,
>
>the driver works well with the motherboard "ASUS TUSI-M", bios
>revision 1015.
>
>http://www.asus.com.tw/mb/socket370/tusi-m/specification.htm
>
>This board has the SIS630ET chipset (reported as 630 by 'lspci'):
>
>00:00.0 Host bridge: Silicon Integrated Systems [SiS] 630 Host (rev
>30)
>00:00.1 IDE interface: Silicon Integrated Systems [SiS] 5513 [IDE]
>(rev d0)
>00:01.0 ISA bridge: Silicon Integrated Systems [SiS] 85C503/5513
>00:01.1 Ethernet controller: Silicon Integrated Systems [SiS] SiS900
>10/100 Ethernet (rev 84)
>00:01.2 USB Controller: Silicon Integrated Systems [SiS] 7001 (rev
>07)
>00:01.3 USB Controller: Silicon Integrated Systems [SiS] 7001 (rev
>07)
>
>The chipset SIS630ET supports UDMA100:
>
>http://www.sis.com/products/chipsets/integrated/socket370/630chart.htm
>
>With the driver version 0.13 it is set to UDMA66 and 'hdparm -t'
>results in 35 MB/s.
>
>With this little change in the driver source the chip is set to
>UDMA100 and 'hdparm -t' results in 45 MB/s.
>
>- { "SiS630", PCI_DEVICE_ID_SI_630, ATA_100,
>SIS5513_LATENCY },
>+ { "SiS630", PCI_DEVICE_ID_SI_630, ATA_66,
>SIS5513_LATENCY
>
>
>

Hum, my guess would have been ATA_100a.
6xy/7xy chipsets tends to be the same IDE wise and 730 is of ATA_100a type.
ATA_100a is an evolution of the chip design based on ATA_66, ATA_100 is
a new design (timing registers moved).
Be careful, I can't guarantee that you are using correct timings, it
might work, but it may also throw your data by the window if you don't
pick the right timings configuration.
I'll ask precisions to SiS engineers.

hdparm -t only test read performance, and read timings are controlled by
the drive, the IDE controller only sets *write* timings. So hdparm
working does not guarantee anything...
If I'm wrong, the linux-kernel mailing list is CCed, I'd be corrected in
an eye's blink...

LB.