Fritzbox 7630 v2 crypto drivers (ltq-deu)

For Developers
1

While cleaning up house I found an old Fritzbox 7360 probably version 2. According to this:
https://boxmatrix.info/wiki/FRITZ!Box_Fon_WLAN_7360_v2#Part-Numbers

The part number on the label show 20002536. If that's the case, it should be able to run OpenWrt and very likely could use the now disabled crypto engine: https://github.com/openwrt/openwrt/commit/e85180d90ed01ef4fb89675702622a9cabf3b092

Can someone please confirm that this device would be capable of running OpenWrt and use the ltd-deu driver. If so, I could have a look at it and run test on this device.

2

21.02 is based on the 5.4 kernel, where, if I understand the github link correctly, the module is still working.

3

There is a PR associated with ltq-deu.

1 Like
4

I started with flashing the official 21.02.1 version and I couldn't use LAN3 and LAN4.

I upgraded to snapshot with kernel 5.10 support. This uses the DSA driver which allowed me to use LAN3 and LAN4.

Going over the kernel log I noticed this:

[    0.606596] lantiq nor flash device: 02000000 at 10000000
[    0.610901] ltq_nor: Found 1 x16 devices at 0x0 in 16-bit bank. Manufacturer ID 0x0000c2 Chip ID 0x00227e
[    0.620231] Amd/Fujitsu Extended Query Table at 0x0040
[    0.625364]   Amd/Fujitsu Extended Query version 1.3.
[    0.630320] number of CFI chips: 1
[    0.633900] 4 fixed-partitions partitions found on MTD device ltq_nor
[    0.640150] Creating 4 MTD partitions on "ltq_nor":
[    0.645116] 0x000000000000-0x000000020000 : "urlader"
[    0.651793] 0x000000020000-0x000001f80000 : "firmware"
[    0.655532] mtd: partition "firmware" extends beyond the end of device "ltq_nor" -- size truncated to 0xfe0000
[    0.672083] 2 eva-fw partitions found on MTD device firmware
[    0.676304] 0x000000020000-0x0000002ba8ae : "kernel"
[    0.681310] mtd: partition "kernel" doesn't end on an erase/write block -- force read-only
[    0.690885] 0x0000002c0100-0x000001000000 : "rootfs"
[    0.694528] mtd: partition "rootfs" doesn't start on an erase/write block boundary -- force read-only
[    0.705174] mtd: device 3 (rootfs) set to be root filesystem
[    0.709487] 1 squashfs-split partitions found on MTD device rootfs
[    0.715679] 0x000000920000-0x000001000000 : "rootfs_data"
[    0.722899] 0x000001f80000-0x000001fc0000 : "tffs (1)"
[    0.726634] mtd: partition "tffs (1)" is out of reach -- disabled
[    0.734545] 0x000001fc0000-0x000002000000 : "tffs (2)"
[    0.738290] mtd: partition "tffs (2)" is out of reach -- disabled

So it seems that even the part number did suggest that I have version 2, the flash size suggests otherwise.
Flashing the the 7360SL version the log is the same except now obviously the NOR size is correct in the DTS.:

[    0.609847] lantiq nor flash device: 01000000 at 10000000
[    0.614076] ltq_nor: Found 1 x16 devices at 0x0 in 16-bit bank. Manufacturer ID 0x0000c2 Chip ID 0x00227e
[    0.623454] Amd/Fujitsu Extended Query Table at 0x0040
[    0.628647]   Amd/Fujitsu Extended Query version 1.3.
[    0.633730] number of CFI chips: 1
[    0.637102] 4 fixed-partitions partitions found on MTD device ltq_nor
[    0.643454] Creating 4 MTD partitions on "ltq_nor":
[    0.648402] 0x000000000000-0x000000020000 : "urlader"
[    0.655794] 0x000000020000-0x000000f80000 : "firmware"
[    0.666022] 2 eva-fw partitions found on MTD device firmware
[    0.670249] 0x000000020000-0x0000002ba8ad : "kernel"
[    0.675298] mtd: partition "kernel" doesn't end on an erase/write block -- force read-only
[    0.685054] 0x0000002c0100-0x000000f80000 : "rootfs"
[    0.688615] mtd: partition "rootfs" doesn't start on an erase/write block boundary -- force read-only
[    0.699320] mtd: device 3 (rootfs) set to be root filesystem
[    0.703775] 1 squashfs-split partitions found on MTD device rootfs
[    0.709793] 0x000000920000-0x000000f80000 : "rootfs_data"
[    0.717826] 0x000000f80000-0x000000fc0000 : "tffs (1)"
[    0.723110] 0x000000fc0000-0x000001000000 : "tffs (2)"

Edit:
PR is half functional (ciphers work, hashes don't). Will continue the discussion about this on GitHub.

5

You can add the Tested-by tag. This will increase the chance of merging this PR. Maintainers need feedback that everything works fine.

1 Like