Exynos4Bootrom » History » Version 8
Kurtis Hanna, 10/02/2019 08:04 AM
added link to Vault 7 Cadmium PDF
1 | 1 | Denis 'GNUtoo' Carikli | h1. Exynos4 Bootrom |
---|---|---|---|
2 | |||
3 | h2. Background information |
||
4 | |||
5 | The Replicant project wants to support devices with free software bootloaders, but most/all the smartphones and tablets supported by Replicant do check the signature of the first stage bootloader. |
||
6 | |||
7 | A presentation on the situation of some of the devices supported by Replicant was made at the Replicant contributors meeting in July 2019. The "presentation slides":https://ftp.osuosl.org/pub/replicant/conferences/replicant-contributors-meeting-july-2019-france/replicant-and-bootloaders.pdf and "video":https://ftp.osuosl.org/pub/replicant/conferences/replicant-contributors-meeting-july-2019-france/replicant-and-bootloaders.webm are available. |
||
8 | |||
9 | h2. Exynos 4 signature check |
||
10 | |||
11 | The Exynos4 bootrom has a strange way to check the signatures: |
||
12 | * The first stage bootloader is encrypted |
||
13 | * The signature check is not very clear[1] |
||
14 | * The header that holds the key has a "func_ptr_BaseAddr" field[1]. |
||
15 | |||
16 | h2. Tests to attempt |
||
17 | |||
18 | 3 | Denis 'GNUtoo' Carikli | * Test with qemu[2] if func_ptr_BaseAddr is somehow used by the bootrom, when verifying the BL1. |
19 | 1 | Denis 'GNUtoo' Carikli | * Try to understand better the scheme used to check the signature. |
20 | 4 | Kurtis Hanna | * Try to see if the fuses can still be written (zeroed) and see whether it's computationally feasible to compute the private key for a zeroed fuses hash. |
21 | 2 | Denis 'GNUtoo' Carikli | * Try to understand why encryption is used. |
22 | 8 | Kurtis Hanna | * Try to see if "Cadmium":https://wikileaks.org/ciav7p1/cms/files/cadmium.pdf, released in Wikileak's Vault 7 documents, can be ported to devices we support |
23 | 1 | Denis 'GNUtoo' Carikli | |
24 | h2. Test setup |
||
25 | |||
26 | 3 | Denis 'GNUtoo' Carikli | Either qemu[2] or a development board with JTAG can be used to do the test. |
27 | 1 | Denis 'GNUtoo' Carikli | |
28 | 3 | Denis 'GNUtoo' Carikli | Testing with qemu[2] is probably way more easy. |
29 | 1 | Denis 'GNUtoo' Carikli | |
30 | fn1. https://fredericb.info/2018/03/emulating-exynos-4210-bootrom-in-qemu.html |
||
31 | 5 | Kurtis Hanna | |
32 | 3 | Denis 'GNUtoo' Carikli | fn2. https://github.com/frederic/qemu-exynos-bootrom |
33 | 6 | Denis 'GNUtoo' Carikli | |
34 | 7 | Kurtis Hanna | h2. Loading a bootloader from SD |
35 | 6 | Denis 'GNUtoo' Carikli | |
36 | When booting Parabola with a Replicant 9 kernel on a Galaxy SIII (i9300), it is possible to erase the bootloader to make the device boot from the microSD instead. |
||
37 | |||
38 | This could be used to do some testing, for instance to see if the BL1 signature can somehow be bypassed, however as no free software bootloaders do exist yet (u-boot relies on nonfree and non-redistributable software), this is not very useful yet. |
||
39 | |||
40 | If you really want to erase the bootloader (your device will be broken and will never boot anymore), you could run the following: |
||
41 | |||
42 | <pre> |
||
43 | # echo 0 > /sys/class/block/mmcblk2boot0/force_ro |
||
44 | # ddrescue -f /dev/zero /dev/mmcblk2boot0 |
||
45 | GNU ddrescue 1.24 |
||
46 | Press Ctrl-C to interrupt |
||
47 | ipos: 4194 kB, non-trimmed: 0 B, current rate: 4194 kB/s |
||
48 | opos: 4194 kB, non-scraped: 0 B, average rate: 4194 kB/s |
||
49 | non-tried: 9223 PB, bad-sector: 0 B, error rate: 0 B/s |
||
50 | rescued: 4194 kB, bad areas: 0, run time: 0s |
||
51 | pct rescued: 0.00%, read errors: 0, remaining time: n/a |
||
52 | time since last successful read: n/a |
||
53 | Copying non-tried blocks... Pass 1 (forwards) |
||
54 | ddrescue: Write error: No space left on device |
||
55 | </pre> |
||
56 | |||
57 | And then verify that it's erased: |
||
58 | <pre> |
||
59 | # hexdump -C /dev/mmcblk2boot0 |
||
60 | 00000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| |
||
61 | * |
||
62 | 00400000 |
||
63 | |||
64 | </pre> |
||
65 | |||
66 | Also verify that the following partitions are also erased: |
||
67 | * mmcblk2boot1 |
||
68 | * BOTA0 |
||
69 | * BOTA1 |
||
70 | |||
71 | I'm not sure what BOTA0 and BOTA1 are but they were already blank in my case. |