Problem with ATAPI device on JMicron JMB363 on 7.0-BETA1

[韓家標 Bill Hacker]

Andrey V. Elsukov wrote:
> Nathan Butcher wrote:
>> I have a Gigabyte GA-G33-DS3R. FreeBSD 7.0-BETA1 detects the chipset,
>> but no devices attached to the SATA controllers find their relative /dev
>> entries. My SATA DVD-combo drive simply doesn't get /dev/acd0, and as a
>> result I cannot access the drive. I've tried hard drives too, with the
>> same results.
> 
> You can update your BETA1 to the fresh RELENG_7. Soren have fixed this 
> issue. Also, ATAPI support was added to the AHCI driver (only for 
> CURRENT yet).
> 
>> The IDE interface seems to work for me (I have an IDE -> CF converter in
>> there), but I don't have a PATA optical drive to test.
> 

There seem to also be certain 'proprietary' (I hope...) issues with that MB 
BIOS. We had tested combinations of BIOS settings extensively, using PATA CDR, 
PATA HDD, PATA <==> SATA converters, and SATA HDD - even an aged IDE Zip.

To cut to the chase -

When finished, we attached 2 SATA drives (only), wherein they appeared as 
/dev/ad10 and /dev/ad12, despite using the lower-numbered silkscreened-labeled 
SATA ports.

Further tuning had us move from all IDE/ legacy BIOS settings to attempt AHCI 
enabling after a recent csup & makebuild*

Nothing to do with FreeBSD - as we were unable to get out of the BIOS POST
OR into 'setup'.  Board hung-forever in the part where it tries to detect drives.

BUT - after a screwdriver-blade CMOS clear, then BIOS reset to the *same* 
conservative/legacy settings as used with ad10 / ad12

..what had been ad10 and ad12 appeared as ad0 and ad2

And gmirror complaineth not (thanks, PHK, PJD and others!)

Bottom Line:  BIOS is 'suboptimal', to put it kindly, and a CMOS reset is 
worthwhile before going further.

HTH,

Bill Hacker


> For Vladimir:
> here is a description for your issue (russian)
> http://unix1.jinr.ru/~lavr/cd_dvd_intel965.html
>