.align may cause data to be interpreted as instructions

[taras.kondratiuk@linaro.org (Taras Kondratiuk)]

On 10/16/2013 08:03 PM, Jon Medhurst (Tixy) wrote:
> On Wed, 2013-10-16 at 17:06 +0100, Jon Medhurst (Tixy) wrote:
>> On Wed, 2013-10-16 at 12:13 +0100, Ben Dooks wrote:
>>> On 15/10/13 23:38, Taras Kondratiuk wrote:
>>>> Hi
>>>>
>>>> I was debugging kprobes-test for BE8 and noticed that some data fields
>>>> are stored in LE instead of BE. It happens because these data fields
>>>> get interpreted as instructions.
>>>>
>>>> Is it a known issue?
>>>
>>> I reported the crashes to Tixy along with a different
>>> method of sovling the problem (changed to using pointers to
>>> the strings) a while ago.
>>
>> I found that fix in the list archives:
>> http://lists.infradead.org/pipermail/linux-arm-kernel/2013-July/186528.html
>>
>>>  However it seems that nothing has
>>> happened to fix this.
>>>
>>> Since kprobes seems to work with the fixed tests I forgot
>>> to follow up and prod Jon about looking into this problem.
>>
>> Sorry, I sorta forgot/ignored the issue, strange compiler problem with a
>> vague explanation on a big-endian kernel (who uses that ;-)
>>
>>> Jon, if you are not interested in fixing this, then please
>>> let me know and we can get a patch sorted to fix it.
>>
>> Looking at your old patch again, it looks good to me. 
> 
> Actually it's broken for thumb, the pointer to the title needs storing
> at a 4 byte alignment, which presumably will make the problem we're
> trying to workaround re-occur. I'll try and find some time to take a
> look for a different way of doing things...

Word alignment here doesn't cause the problem, so this patch should
workaround the issue.