98e617d874
startup: use custom reset_handler
...
+ group confidential data in one place
+ zero all SRAM where needed
2017-08-16 13:52:01 +02:00
a01ba51a2a
storage: copy only required bytes from old storage, bump storage version
...
+ backup_device: ask for pin, always use correct number of words
2017-08-16 13:11:31 +02:00
f0d2e7a371
storage: bump version
2017-07-31 22:43:28 +02:00
c778d7b9c3
storage: wipe storage after 15 wrong pins
2017-07-31 02:26:28 +02:00
3d7d0f0734
use C99 for loop where possible
2017-07-23 22:20:51 +02:00
cf04279774
implement Storage.flags and related stuff
2017-07-17 18:36:09 +02:00
ccb0cd82b6
firmware: add changelog, bump version to 1.5.1
2017-07-12 14:19:05 +02:00
97fa33de46
reset: show needs backup signed when storage is not backed up
2017-06-29 14:14:20 +02:00
73708aa47e
refactor recovery.is_same_mnemonic function to storage_containsMnemonic
2017-06-26 15:44:01 +02:00
c6fd70b471
gettext: mark localizable strings as proof-of-concept
2017-06-18 22:47:32 +02:00
25b9bfd97b
timer: Use Cortex-M3 SysTick timers
...
Removed `usbDelay(uint32_t cycles)`, added `usbSleep(uint32_t millis)`
The same method signature could cause silent code breakage at runtime,
as opposed to noisy code breakage at compile time which is the better
kind.
2016-11-17 02:07:46 +01:00
9287dd7e04
Poll USB during BIP39 derivation.
...
This patch adds calls to usbPoll in the progress callback. This
should address #98 .
We call usbDelay instead of Poll, to call usbd_poll several times.
Otherwise it would only handle one event instead of handling all
events that were pending so far. The ugly magic number 5 is a guess.
Note that we also need to set usbTiny, so that we don't recursively
process messages. Since we don't know whether usbTiny is set, we
need to store the old value (especially true for u2f).
This fix also relies on another fix in libopencm3.
2016-11-09 16:17:50 +01:00
4ce4cc5605
halt the device when encountered an invalid mnemonic (skip if device loaded or recovered without enforced wordlist)
2016-11-09 16:06:50 +01:00
71890e4edf
implement u2f_counter in LoadDevice, Recoverydevice and ResetDevice messages
2016-10-31 16:20:15 +01:00
6d65551b82
More alignment fixes
2016-08-30 12:39:37 +02:00
ae4dff6e5f
Only compute pubkey on demand.
...
Changed all hdnode callers to call hdnode_fill_public_key if
they need the public key.
2016-06-27 10:13:18 +02:00
120cfc148f
new message SetU2FCounter
2016-06-12 23:44:36 +02:00
b1e3c52b08
remove DialogIcon enum, use bitmap structure directly
2016-06-08 19:20:07 +02:00
c4e8bd0d0e
More robust storage recycle sector
...
Clear storage marker first before clearing the second sector to prevent
leaving a state where only PIN failures have been cleared but storage is
still present.
2016-05-27 14:03:20 +02:00
e093371129
Added storage area for u2f counter.
...
To prevent flashing for every u2f operation just clear one bit in
the u2f area to indicate an increased counter.
2016-05-26 20:59:16 +02:00
8e7896456a
Merge branch 'origin/master' into u2f
2016-05-26 20:58:52 +02:00
87bfd5a829
Bugfix: restore storage.
...
Storage restore was broken due to my previous patch.
2016-05-25 01:14:32 +02:00
bc92fb95a5
Clear pinarea on storage_init if upgrade fails
...
This also cleans up the code a bit and resets storage_uuid if upgrade fails.
2016-05-20 20:13:43 +02:00
46119bd007
clear pin failures on wipe and when in debug mode
2016-05-20 17:00:10 +02:00
96f30a0ba7
Don't ask for passphrase with u2f.
2016-05-20 01:49:20 +02:00
c123db71b9
Merge branch 'master' into u2f
2016-05-16 18:19:31 +02:00
def4a9a02c
fix pbkdf2 usage in storage.c
2016-05-12 21:05:17 +02:00
cf38291ca4
Updated branch u2f
2016-04-29 16:19:54 +02:00
630e26dd20
use less stack memory in storage_commit
2016-04-27 19:23:02 +02:00
da98a3a6fd
Don't reflash storage after each PIN entry
...
Instead of reflashing the whole storage, we use a designated area
in the second storage block, where we mark each PIN failure by a
single zero bit. This is because one can set bits in flash to zero but
not to one. If the PIN was entered successfully the whole word is
set to zero and the next word stores the new PIN failure counter.
2016-04-27 18:39:04 +02:00
51f02ff763
ed25519 support
2016-04-22 18:22:45 +02:00
03c501d9e3
Do not use hardcoded string for secp256k1.
2016-04-20 15:39:15 +02:00
56238e63fc
Added curve type to HD node.
...
Create a different root node for every curve type to separate the key
space.
2016-04-19 18:23:12 +02:00
1d3c7ee3f2
cleanup Waking up screen usage
2016-01-19 15:36:43 +01:00
ed76d030ef
check for sessionPassphraseCached
2015-12-14 22:53:14 +01:00
9328cad7f1
Add U2F support
2015-11-03 16:47:12 +01:00
c08ff09f0f
storage: add compile-time assert for sizeof(Storage) validation
2015-09-06 21:14:23 +03:00
e99aafd949
fixed salt passing to pbkdf2_hmac_sha512
2015-05-05 12:01:34 -07:00
00ccf6a8ce
bump storage version
2015-04-01 19:43:36 +02:00
4cbf29505d
don't clear PIN on Initialize
2015-03-31 16:31:29 +02:00
40e174ac87
bump storage version
2015-03-18 13:34:09 +01:00
286ee0525c
PIN handling - constant time.
...
This diff contains three changes.
1. Make timing isPinCorrect independent of storage.pin, to avoid timing attacks
2. Only update failed PIN counter if the user entered a PIN.
Of course, the fail counter is still incremented, before the PIN is checked.
3. Don't cache the PIN, but just the fact that the PIN was entered. The
cache should be in sync with storage.pin in any case.
2015-02-22 15:42:31 +01:00
f2f50aa188
bump version to 1.3.1
2015-02-16 15:23:30 +01:00
b5221ce2e9
introduce homescreen
2015-02-04 21:27:07 +01:00
40efefc571
rework pin handling
2015-01-27 13:00:25 +01:00
7dacfd69ee
check for flash operation failure
2015-01-26 21:10:51 +01:00
2a2eba7de5
rework layoutProgress functions
2014-12-21 18:58:56 +01:00
0e92d4c588
error checking of hdnode functions return values
2014-12-16 18:28:46 +01:00
961566c9b7
implement ApplySetttings.use_passphrase
2014-12-13 19:29:27 +01:00
c051dfde9d
update trezor-crypto
2014-10-30 01:38:40 +01:00
Jochen Hoenicke
Pavol Rusnak
Saleem Rashid
Mark Bryars
Roman Zeyde
Darin Stanchfield