[LEDE-DEV] [PATCH] ramips: fix AR670W partition alignment

[Mathias Kresin]

05.02.2017 10:15, John Crispin:
>
>
> On 05/02/2017 09:59, Claudio Leite wrote:
>> mtdsplit_lzma requires that the rootfs be aligned to a block boundary.
>> Pad the kernel partition to make this so.
>>
>> Signed-off-by: Claudio Leite <leitec at gmail.com>
>> ---
>>  target/linux/ramips/image/rt288x.mk | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/target/linux/ramips/image/rt288x.mk b/target/linux/ramips/image/rt288x.mk
>> index 4a0d6d4..86a5063 100644
>> --- a/target/linux/ramips/image/rt288x.mk
>> +++ b/target/linux/ramips/image/rt288x.mk
>> @@ -14,7 +14,7 @@ define Device/ar670w
>>    BLOCKSIZE := 64k
>>    DEVICE_TITLE := Airlink AR670W
>>    IMAGE_SIZE := $(ralink_default_fw_size_4M)
>> -  KERNEL := $(KERNEL_DTB)
>> +  KERNEL := $(KERNEL_DTB) | pad-to $$(BLOCKSIZE)
>>    IMAGES += factory.bin
>>    IMAGE/factory.bin := $$(sysupgrade_bin) | check-size $$$$(IMAGE_SIZE) | \
>>  	wrg-header wrgn16a_airlink_ar670w
>>
>
> Hi
>
> although technically correct, the lzma splitter needs to be fixed to not
> have this requirement, instead of adding these patches for all boards.
> there is no need for any padding beyond 4byte padding.

I'm in doubt that the lzma splitter can be fixed.

The dependency to align the rootfs isn't specific to the lzma splitter. 
mtd_find_rootfs_from() in mtdsplit.c uses the erase block size as step size.

For the other splitter an not erase block size aligned rootfs works only 
because the headers provide informations about the kernel size and we 
can pass the kernel size as offset to mtd_find_rootfs_from() to prevent 
stepping over the mtd.

But the lzma header doesn't provide any informations about 
kernel/compressed size. Only the uncompressed size is stored in the 
header. According to the LZMA SDK, the LZMA decoder simply stops 
decompressing as soon as the amount of decompressed data matches the 
value stored in the header.

The only fix I can think of, would be to change the step size in 
mtd_find_rootfs_from() to 1 byte. But that would introduce a huge 
performance penalty.

Hopefully I missed the obvious solution for the problem.

Mathias