[PATCH v2 00/29] implement KASLR for ARM

[Ard Biesheuvel]

On 6 September 2017 at 18:53, Tony Lindgren <tony at atomide.com> wrote:
> * Ard Biesheuvel <ard.biesheuvel at linaro.org> [170906 10:31]:
>> On 6 September 2017 at 18:12, Tony Lindgren <tony at atomide.com> wrote:
>> > Kernel image @ 0x81000000 [ 0x000000 - 0x426300 ]
>> > ## Flattened Device Tree blob at 84000000
>> >    Booting using the fdt blob at 0x84000000
>> >    Loading Device Tree to 86feb000, end 86fff2d5 ... OK
>> >
>> > Starting kernel ...
>> >
>> > regions.image_size:00e00000
>> > regions.pa_start:80000000
>> > regions.pa_end:87200000
>> > regions.zimage_start:81000000
>> > regions.zimage_size:00437320
>> > regions.dtb_start:86feb000
>> > regions.dtb_size:00012000
>> > regions.initrd_start:00000000
>> > regions.initrd_size:00000000
>> > num:00000028
>> > num:00000025
>> > *kaslr_offset:05e00000
>> > Uncompressing Linux... done, booting the kernel.
>> > Warning: Neither atags nor dtb found
>>
>> OK, so in this case, 80000000 + 00e00000 + 05e00000 == 86c00000, which
>> is still below the DTB, but apparently, it has corrupted it anyway.
>>
>> I will try to figure out what's going on here.
>
> Do you need to have kaslr_offset beyond the uncompressed
> kernel size maybe?
>

I think the problem is in the rounding of region.pa_start.

I have now changed this to

regions.image_size = image_base % SZ_128M + round_up(image_size, SZ_2M);
regions.pa_start = round_down(image_base, SZ_128M);


>> Thanks again for taking the time,
>
> No problem, I'm happy to test these changes.
>

I have updated my arm-kaslr-v3 with all the fixes from this discussion
(and more)

Thanks,