On Wed, Aug 22, 2012 at 11:18 AM, Spencer Oliver <[email protected]> wrote:
> On 21 August 2012 23:16, Andreas Fritiofson
> <[email protected]> wrote:
>>
>> I found the problem and made a proper fix instead:
>> http://openocd.zylin.com/789
>>
>> Please try this on both arm7 and arm9. I've tested on an str750 but I
>> don't have any arm9 available.
>>
>
> this fixes issues on arm7, however does not work for arm9.
> Just tested current master on arm9 and it is working ok for me, so
> perhaps the issue only effected arm7.
>
> Spen

Ah, OK, then I'm pretty sure my comment in gerrit is correct. The real
and the discarded scan fields should be in the reverse order for arm9,
because apparently the DR is in the reverse bit order compared to
arm7. Could you try reversing them like the following:

fields[0].num_bits = size * 8;
fields[0].out_value = NULL;
fields[0].in_value = in;

fields[1].num_bits = 32 - size * 8;
fields[1].out_value = NULL;
fields[1].in_value = NULL;

On the other hand, it could be safe to do a partial scan with only the
needed bits. Depends on whether the data scanned into DR has any
effect or not. If it's safe, the last field of 3+32 bits could also be
removed as an optimization and only one field of size*8 would have to
be scanned.

I suggest you try with reversing the order and keeping the discarded
fields for now and we can look into the optimization later. I'll
update the patch this evening if it works.

/Andreas

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
OpenOCD-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openocd-devel

Reply via email to