SII3512 rev0 ?

[Peter B]

>Peter B a écrit :
>> In /usr/src/sys/dev/ata/ata-chipset.c the table (ata_sii_ident) for
>> Silicon Image chips:
>>
>>                   pcirev
>>  { ATA_SII3512,   0x02, SIIMEMIO, 0,         ATA_SA150, "SiI 3512" },
>>  { ATA_SII3512,   0x00, SIIMEMIO, SIIBUG,    ATA_SA150, "SiI 3512" },
>>
>> Indicate there is a revision 0 of the chip SII3512. I wonder if there exist
>> such chip, or if this is just an assumption from the problems with
>> SII3112 rev0 ..?
>>
>> http://www.freebsd.org/cgi/cvsweb.cgi/src/sys/dev/ata/ata-chipset.c#rev1.62
>>
>> Any input on the reliability of SII3512 is also welcomed.

didier derny wrote:
>
>I recently added a sata board with a  SiI 3512 to my computer to bypass 
>a problem with the VT8732a of the

What was the problem with VIA VT8732a..?

>mother board, but if FreeBSD 6.2 booted from this controller it was 
>totally unstable, crashing as soon as the
>disk activity became high  (occuring in less than 10mn after boot)

Can you check which pci revision it has?   (scanpci -v or pciconf -l)
Look for 'REVISION' or 'rev='

I have the suspicion that despite possible fixes after the
SII3112A v1.1 (rev1) chip. Silicon Image still lacks in engineering reliable
equipment. As they infact have failed before with the famous CMD640 chip bug.
However it's hard to evade the sii chips as Sweex, Lycom, ST Labs, InnoDV,
Belkin, Sunix, Adaptec/1210SA all use their chips.

>I replaced this board by a Sata Controller with a SiI 3114 and now 
>FreeBSD is extremely stable
>no crashes at all in 2 days

What was the tradename for your SII 3114 equiped card .. ?

   Regards /Peter