2013/8/3 Jonas Gorski <[email protected]>:
> On Sat, Aug 3, 2013 at 11:14 AM, Rafał Miłecki <[email protected]> wrote:
>> Hi,
>>
>> Personally I'm playing with Broadcom devices, but my friend asked me
>> for help with WNDR4300. It's a device very similar to the WNDR3700v4.
>>
>> http://wikidevi.com/wiki/Netgear_WNDR3700v4
>> http://wikidevi.com/wiki/Netgear_WNDR4300
>>
>> http://wiki.openwrt.org/toh/netgear/wndr3700
>> http://wiki.openwrt.org/toh/netgear/wndr4300
>>
>> It seems OpenWrt has a profile for this device:
>> https://dev.openwrt.org/changeset/34650/trunk/target/linux/ar71xx/generic/profiles/netgear.mk
>>
>> but it says about some problems with squashfs/jffs2.
>>
>> Could someone describe that issue, please? What's the problem with
>> that filesystems and NAND? I use Broadcom based WNDR4500 which keeps
>> squashfs and jffs2 on NAND, and it seems to work OK.
>> It this some Atheros specific problem?
>
> No, this is a generic NAND flash problem.
>
> The issue is that squashfs has no bad block management at all and
> requires all blocks on order; but for proper NAND bad block management
> you also need to be able to skip bad blocks and occasionally relocate
> blocks (see [1]). That's why raw squashfs is a bad idea on NAND (it
> works if you use a FTL like ubi).
>
> Regards,
> Jonas
>
> [1] http://www.infradead.org/pipermail/linux-mtd/2006-April/015386.html

What is your suggestion on this?
1) Should squashfs be rewritten to work on device with bad blocks?
2) Should squashfs be modified to use UBI?
3) Should we use different overlay for NAND? JFFS2 + JFFS2 maybe? NAND
devices are huge, so we probably could waste few MiB and change
squashfs to JFFS2 (and still use 2nd partition overlaying the first
one, to allow "rescue" mode working).

-- 
Rafał
_______________________________________________
openwrt-devel mailing list
[email protected]
https://lists.openwrt.org/cgi-bin/mailman/listinfo/openwrt-devel

Reply via email to