Re: dc(4) related misalignment trap

From: Marcel Moolenaar <marcel_at_xcllnt.net>
Date: Mon, 21 Nov 2005 17:27:48 -0800
On Nov 21, 2005, at 1:51 PM, John Baldwin wrote:

> On Monday 21 November 2005 04:23 pm, John Baldwin wrote:
>> On Monday 21 November 2005 04:14 pm, John Baldwin wrote:
>>> On Monday 21 November 2005 09:49 am, John Baldwin wrote:
>>>>> Register s3 holds the sp variable ($2 in kgdb)
>>>>> Register t0 holds the result the left-hand side of the statement,
>>>>> which is not properly aligned for a 16-bit load ($7 = register  
>>>>> t0 +
>>>>> 8).
>>>>>
>>>>> Rough patch to fix the problem attached:
>>>>
>>>> I think de(4) would need the same fix as well at least.
>>>
>>> Once I fixed another bug with de(4) due to ru_at_'s recent changes I  
>>> wasn't
>>> able to reproduce an alignment trap with de(4) on my alpha, so  
>>> although I
>>> have a patch for de(4) I can't test it.  How are you triggering the
>>> alignment fault? Do you have something other than  
>>> 'ifconfig_dc0="DHCP"'
>>> in /etc/rc.conf?
>>
>> Check that, it would help to let the multiuser boot finish before  
>> firing
>> off the e-mail.  I did indeed get the alignment fault during dhcp,  
>> so I'll
>> be testing that alignment patch in a minute or so.
>
> Ok, so the same change Marcel posted for dc(4) fixed de(4) as well,  
> so Marcel,
> I think you should be good to go to commit your dc(4) patch.

Will do. Thanks for taking care of de(4).

-- 
  Marcel Moolenaar         USPA: A-39004          marcel_at_xcllnt.net
Received on Tue Nov 22 2005 - 00:28:13 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:48 UTC