Re: DragonFly-2.2.1 installation problem

[Archimedes Gaviola <archimedes.gaviola@xxxxxxxxx>]

On Fri, May 8, 2009 at 10:03 AM, Archimedes Gaviola
<archimedes.gaviola@gmail.com> wrote:
> On Fri, May 8, 2009 at 12:35 AM, Matthew Dillon
> <dillon@apollo.backplane.com> wrote:
>> :> Basically this solves the problem dealing with USB keyboard. Somehow,
>> :> another problem get stuck on me in proceeding the installation. I
>> :> encountered this problem 'panic: command is on another queue'. Details
>> :> on the panic logs can be seen on the attached JPEG files. I also
>> :> provide dmesg log taken from FreeBSD-8.0 CURRENT (successfully
>> :> installed on this machine) for reference on the hardware specs.
>> :>
>> :> Thanks
>> :> Archimedes
>> :
>> :Here we go... instead of sending the large image files, I just uploaded it =
>> :here
>> :
>> :http://www.flickr.com/photos/38158047@N03/3510136106/sizes/l/
>> :http://www.flickr.com/photos/38158047@N03/3510136108/sizes/l/
>> :
>> :Any idea on the panic problem?
>> :
>> :Thanks,
>> :Archimedes
>>
>>    This is really odd.  The flags are 0x3 from the picture.  Those
>>    flags are fine, they shouldn't cause any panic.
>>
>>    The flags which cause the panic are:
>>
>>    #define AAC_ON_AACQ_MASK        ((1<<5)|(1<<6)|(1<<7)|(1<<8)|(1<<10)|(1<11))
>>
>>    And as you can see a flags value of 0x3 does not match any of those.
>>
>>    Did you compile the kernel with any sort of weird optimizations?
>>
>>                                        -Matt
>>
>>
>
> Hi Matt,
>
> No, I didn't change anything on the installers. In fact, I was using
> both 2.2.1-RELEASE and 2.0.1-RELEASE on installation to prove if it
> occurs on the previous releases and it did, it still encounter the
> same panic behavior.
>
> Thanks,
> Archimedes
>

In addition, while reading aac(4) manual, I don't know if the bugs
described are still true? Especially no. 2 because this machine is
having a 10GB RAM capacity.

BUGS
     (1) This driver is not compatible with controllers that have version 1. x
     firmware.  The firmware version is the same as the kernel version printed
     in the BIOS POST and driver attach messages.

     (2) This driver will not work on systems with more than 4GB of memory.

     (3) The controller is not actually paused on suspend/resume.

     (4) Unloading driver is not supported at this time.