>Then, it is in a loop to write first 32 bit data (32bit int) to the
>beginning of the page and
>write another 32 bit data starting at the two bytes before the page ends.
>The second
>32 bit data is written with intention to cause the page overflow for the
>page boundary
>handling testing. Repeats this process on every page until the allocated
>memory
>reaches to the end.
What kernel are you using? It's true that a user space program shouldn't be
able to crash the machine or damage the kernel, though it seems that we are
still some way from that ideal at the moment. If you have a kernel that does
unaligned fixups for user code then my guess would be that the problem lies
there.
p
unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]
