cvs commit: src/sys/dev/ata ata-chipset.c ata-pci.c ata-pci.h

[Marius Strobl]

On Sat, Apr 12, 2008 at 05:21:23PM +0000, Sren Schmidt wrote:
> sos         2008-04-12 17:21:23 UTC
> 
>   FreeBSD src repository
> 
>   Modified files:
>     sys/dev/ata          ata-chipset.c ata-pci.c ata-pci.h 
>   Log:
>   Dont call hw.status recursively.
>   
>   Spotted by: Marcel Moolenaar
>   
>   Revision  Changes    Path
>   1.215     +5 -7      src/sys/dev/ata/ata-chipset.c
>   1.125     +1 -1      src/sys/dev/ata/ata-pci.c
>   1.85      +1 -0      src/sys/dev/ata/ata-pci.h

Even with these fixes in place your latest round of ata(4)
changes causes acd0 to disappear on a sparc64 machine for
me. Reverting to ata(4) sources of April 9th fixes these.
Verbose before and after dmesg output is at:
http://people.freebsd.org/~marius/v100_1.txt
http://people.freebsd.org/~marius/v100_2.txt

Btw., `atacontrol list` is broken on archs where ata2 is the
first ata(4) device (probably all except amd64 and i386) as
atacontrol(8) assumes channels from 0 to IOCATAGMAXCHANNEL
exist:
# atacontrol list
atacontrol: ioctl(IOCATADEVICES): Device not configured

Marius