Dice Random Numbers

Generating data with entropy, or random number generation (RNG), is a well-known difficult problem. Many crypto algorithms and protocols assumes random data is available. There are many implementations out there, including /dev/random in the BSD and Linux kernels and API calls in crypto libraries such as GnuTLS or OpenSSL. How they work can be understood by reading the source code. The quality of the data depends on actual hardware and what entropy sources were available — the RNG implementation itself is deterministic, it merely convert data with supposed entropy from a set of data sources and then generate an output stream.

In some situations, like on virtualized environments or on small embedded systems, it is hard to find sources of sufficient quantity. Rarely are there any lower-bound estimates on how much entropy there is in the data you get. You can improve the RNG issue by using a separate hardware RNG, but there is deployment complexity in that, and from a theoretical point of view, the problem of trusting that you get good random data merely moved from one system to another. (There is more to say about hardware RNGs, I’ll save that for another day.)

For some purposes, the available solutions does not inspire enough confidence in me because of the high complexity. Complexity is often the enemy of security. In crypto discussions I have said, only half-jokingly, that about the only RNG process that I would trust is one that I can explain in simple words and implement myself with the help of pen and paper. Normally I use the example of rolling a normal six-sided dice (a D6) several times. I have been thinking about this process in more detail lately, and felt it was time to write it down, regardless of how silly it may seem.

A die with six sides produces a random number between 1 and 6. It is relatively straight forward to intuitively convinced yourself that it is not clearly biased: inspect that it looks symmetric and do some trial rolls. By repeatedly rolling the die, you can generate how much data you need, time permitting.

I do not understand enough thermodynamics to know how to estimate the amount of entropy of a physical process, so I need to resort to intuitive arguments. It would be easy to just assume that a die produces 2.5 bits of entropy, because log_2(6)~=2.584. At least I find it easy to convince myself intuitively that 2.5 bits is an upper bound, there appears to me to be no way to get out more entropy than that out looking at a die roll outcome. I suspect that most dice have some form of defect, though, which leads to a very small bias that could be found with a large number of rolls. Thus I would propose that the amount of entropy of most D6’s are slightly below 2.5 bits on average. Further, to establish a lower bound, and intuitively, it seems easy to believe that if the entropy of particular D6 would be closer to 2 bits than to 2.5 bits, this would be noticeable fairly quickly by trial rolls. That assumes the die does not have complex logic and machinery in it that would hide the patterns. With the tinfoil hat on, consider a die with a power source and mechanics in it that allowed it to decide which number it would land on: it could generate seamingly-looking random pattern that still contained 0 bits of entropy. For example, suppose a D6 is built to produce the pattern 4, 1, 4, 2, 1, 3, 5, 6, 2, 3, 1, 3, 6, 3, 5, 6, 4, … this would mean it produces 0 bits of entropy (compare the numbers with the decimals of sqrt(2)). Other factors may also influence the amount of entropy in the output, consider if you roll the die by just dropping straight down from 1cm/1inch above the table. There could also be other reasons why the amount of entropy in a die roll is more limited, intuitive arguments are sometimes completely wrong! With this discussion as background, and for simplicity, going forward, I will assume that my D6 produces 2.5 bits of entropy on every roll.

We need to figure out how many times we need to roll it. I usually find myself needing a 128-bit random number (16 bytes). Crypto algorithms and protocols typically use power-of-2 data sizes. 64 bits of entropy results in brute-force attacks requiring about 2^64 tests, and for many operations, this is feasible with today’s computing power. Performing 2^128 operations does not seem possible with today’s technology. To produce 128 bits of entropy using a D6 that produces 2.5 bits of entropy per roll, you need to perform ceil(128/2.5)=52 rolls.

We also need to design an algorithm to convert the D6 output into the resulting 128-bit random number. While it would be nice from a theoretical point of view to let each and every bit of the D6 output influence each and every bit of the 128-bit random number, this becomes difficult to do with pen and paper. Update:This blog post used to include an algorithm here, however it was clearly wrong (written too late in the evening…) so I’ve removed it — I need to come back and think more about this.

So what’s the next step? Depends on what you want to use the random data for. For some purposes, such as generating a high-quality 128-bit AES key, I would be done. The key is right there. To generate a high-quality ECC private key, you need to generate somewhat more randomness (matching the ECC curve size) and do a couple of EC operations. To generate a high-quality RSA private key, unfortunately you will need much more randomness, at the point where it makes more sense to implement a PRNG seeded with a strong 128-bit seed generated using this process. The latter approach is the general solution: generate 128 bits of data using the dice approach, and then seed a CSPRNG of your choice to get large number of data quickly. These steps are somewhat technical, and you lose the pen-and-paper properties, but code to implement these parts are easier to verify compared to verifying that you get good quality entropy out of your RNG implementation.

The Case for Short OpenPGP Key Validity Periods

After I moved to a new OpenPGP key (see key transition statement) I have received comments about the short life length of my new key. When I created the key (see my GnuPG setup) I set it to expire after 100 days. Some people assumed that I would have to create a new key then, and therefore wondered what value there is to sign a key that will expire in two months. It doesn’t work like that, and below I will explain how OpenPGP key expiration works; how to extend the expiration time of your key; and argue why having a relatively short validity period can be a good thing.
Continue reading The Case for Short OpenPGP Key Validity Periods

Wifi on S3 with Replicant

I’m using Replicant on my main phone. As I’ve written before, I didn’t get Wifi to work. The other day leth in #replicant pointed me towards a CyanogenMod discussion about a similar issue. The fix does indeed work, and allowed me to connect to wifi networks and to setup my phone for Internet sharing. You need to run the following commands after every boot, disable/enable Wifi, and then it should work.

echo murata > /data/.cid.info
chown system /data/.cid.info
chgrp wifi /data/.cid.info
chmod 0660 /data/.cid.info

Digging deeper, I found a CM Jira issue about it, and ultimately a code commit. It seems the issue is that more recent S3’s comes with a Murata Wifi chipset that uses MAC addresses not known back in the Android 4.2 (CM-10.1.3 and Replicant-4.2) days. Pulling in the latest fixes for macloader.cpp solves this problem for me, and there is no need for the workaround above. I still need to load the non-free firmware images that I get from CM-10.1.3. I’ve created a pull request fixing macloader.cpp for Replicant 4.2 if someone else is curious about the details. You have to rebuild your OS with the patch for things to work (if you don’t want to, the workaround using /data/.cid.info works fine), and install some firmware blobs as below.

adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_apsta.bin_b1 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_apsta.bin_b2 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_mfg.bin_b0 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_mfg.bin_b1 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_mfg.bin_b2 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_p2p.bin_b0 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_p2p.bin_b1 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_p2p.bin_b2 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_sta.bin_b0 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_sta.bin_b1 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/bcmdhd_sta.bin_b2 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/nvram_mfg.txt /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/nvram_mfg.txt_murata /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/nvram_mfg.txt_murata_b2 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/nvram_mfg.txt_semcosh /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/nvram_net.txt /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/nvram_net.txt_murata /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/nvram_net.txt_murata_b2 /system/vendor/firmware/
adb push cm-10.1.3-i9300/system/etc/wifi/nvram_net.txt_semcosh /system/vendor/firmware/

Replicant 4.2 0002 and NFC on I9300

I’m using Replicant on my Samsung SIII (i9300) phone (see my earlier posts). During my vacation the Replicant project released version 4.2-0002 as a minor update to their initial 4.2 release. I didn’t anticipate any significant differences, so I followed the installation instructions but instead of “wipe data/factory reset” I chose “wipe cache partition” and rebooted. Everything appeared to work fine, but I soon discovered that NFC was not working. Using adb logcat I could get some error messages:

E/NFC-HCI ( 7022): HCI Timeout - Exception raised - Force restart of NFC service
F/libc    ( 7022): Fatal signal 11 (SIGSEGV) at 0xdeadbaad (code=1), thread 7046 (message)
I/DEBUG   ( 1900): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
I/DEBUG   ( 1900): Build fingerprint: 'samsung/m0xx/m0:4.1.1/JRO03C/I9300XXDLIB:user/release-keys'
I/DEBUG   ( 1900): Revision: '12'
I/DEBUG   ( 1900): pid: 7022, tid: 7046, name: message  >>> com.android.nfc <<<

The phone would loop trying to start NFC and having the NFC sub-system die over and over. Talking on #replicant channel, paulk quickly realized and fixed the bug. I had to rebuild the images to get things to work, so I took the time to create a new virtual machine based on Debian 7.5 for building Replicant on. As a side note, the only thing not covered by Replicant build dependency documentation was that I needed the Debian xmllint package to avoid a build failure and the Debian xsltproc package to avoid a error message being printed in the beginning of every build. Soon I had my own fresh images and installed them and NFC was working again, after installing the non-free libpn544_fw.so file.

During this, I noticed that there are multiple libpn544_fw.so files floating around. I have the following files:

version string source
libpn544_fw_C3_1_26_SP.so internet
libpn544_fw_C3_1_34_SP.so stock ROM on S3 bought in Sweden during 2013 and 2014 (two phones)
libpn544_fw_C3_1_39_SP.so internet

(For reference the md5sum's of these files are 682e50666effa919d557688c276edc48, b9364ba59de1947d4588f588229bae20 and 18b4e634d357849edbe139b04c939593 respectively.)

If you do not have any of these files available as /vendor/firmware/libpn544_fw.so you will get the following error message:

I/NfcService( 2488): Enabling NFC
D/NFCJNI  ( 2488): Start Initialization
E/NFC-HCI ( 2488): Could not open /system/vendor/firmware/libpn544_fw.so or /system/lib/libpn544_fw.so
E/NFCJNI  ( 2488): phLibNfc_Mgt_Initialize() returned 0x00ff[NFCSTATUS_FAILED]
E/NFC-HCI ( 2488): Could not open /system/vendor/firmware/libpn544_fw.so or /system/lib/libpn544_fw.so
W/NFCJNI  ( 2488): Firmware update FAILED
E/NFC-HCI ( 2488): Could not open /system/vendor/firmware/libpn544_fw.so or /system/lib/libpn544_fw.so
W/NFCJNI  ( 2488): Firmware update FAILED
E/NFC-HCI ( 2488): Could not open /system/vendor/firmware/libpn544_fw.so or /system/lib/libpn544_fw.so
W/NFCJNI  ( 2488): Firmware update FAILED
E/NFCJNI  ( 2488): Unable to update firmware, giving up
D/NFCJNI  ( 2488): phLibNfc_Mgt_UnConfigureDriver() returned 0x0000[NFCSTATUS_SUCCESS]
D/NFCJNI  ( 2488): Terminating client thread...
W/NfcService( 2488): Error enabling NFC

Using the first (26) file or the last (39) file does not appear to be working on my phone, I get the following error messages. Note that the line starting with 'NFC capabilities' has 'Rev = 34' in it, possibly indicating that I need the version 34 file.

I/NfcService( 5735): Enabling NFC
D/NFCJNI  ( 5735): Start Initialization
D/NFCJNI  ( 5735): NFC capabilities: HAL = 8150100, FW = b10122, HW = 620003, Model = 12, HCI = 1, Full_FW = 1, Rev = 34, FW Update Info = 8
D/NFCJNI  ( 5735): Download new Firmware
W/NFCJNI  ( 5735): Firmware update FAILED
D/NFCJNI  ( 5735): Download new Firmware
W/NFCJNI  ( 5735): Firmware update FAILED
D/NFCJNI  ( 5735): Download new Firmware
W/NFCJNI  ( 5735): Firmware update FAILED
E/NFCJNI  ( 5735): Unable to update firmware, giving up
D/NFCJNI  ( 5735): phLibNfc_Mgt_UnConfigureDriver() returned 0x0000[NFCSTATUS_SUCCESS]
D/NFCJNI  ( 5735): Terminating client thread...
W/NfcService( 5735): Error enabling NFC

Loading the 34 works fine.

I/NfcService( 2501): Enabling NFC
D/NFCJNI  ( 2501): Start Initialization
D/NFCJNI  ( 2501): NFC capabilities: HAL = 8150100, FW = b10122, HW = 620003, Model = 12, HCI = 1, Full_FW = 1, Rev = 34, FW Update Info = 0
D/NFCJNI  ( 2501): phLibNfc_SE_GetSecureElementList()
D/NFCJNI  ( 2501): 
D/NFCJNI  ( 2501): > Number of Secure Element(s) : 1
D/NFCJNI  ( 2501): phLibNfc_SE_GetSecureElementList(): SMX detected, handle=0xabcdef
D/NFCJNI  ( 2501): phLibNfc_SE_SetMode() returned 0x000d[NFCSTATUS_PENDING]
I/NFCJNI  ( 2501): NFC Initialized
D/NdefPushServer( 2501): start, thread = null
D/NdefPushServer( 2501): starting new server thread
D/NdefPushServer( 2501): about create LLCP service socket
D/NdefPushServer( 2501): created LLCP service socket
D/NdefPushServer( 2501): about to accept
D/NfcService( 2501): NFC-EE OFF
D/NfcService( 2501): NFC-C ON

What is interesting is, that my other S3 running CyanogenMod does not have the libpn544_fw.so file but still NFC works. The messages are:

I/NfcService( 2619): Enabling NFC
D/NFCJNI  ( 2619): Start Initialization
E/NFC-HCI ( 2619): Could not open /system/vendor/firmware/libpn544_fw.so or /system/lib/libpn544_fw.so
W/NFC     ( 2619): Firmware image not available: this device might be running old NFC firmware!
D/NFCJNI  ( 2619): NFC capabilities: HAL = 8150100, FW = b10122, HW = 620003, Model = 12, HCI = 1, Full_FW = 1, Rev = 34, FW Update Info = 0
D/NFCJNI  ( 2619): phLibNfc_SE_GetSecureElementList()
D/NFCJNI  ( 2619): 
D/NFCJNI  ( 2619): > Number of Secure Element(s) : 1
D/NFCJNI  ( 2619): phLibNfc_SE_GetSecureElementList(): SMX detected, handle=0xabcdef
D/NFCJNI  ( 2619): phLibNfc_SE_SetMode() returned 0x000d[NFCSTATUS_PENDING]
I/NFCJNI  ( 2619): NFC Initialized
D/NdefPushServer( 2619): start, thread = null
D/NdefPushServer( 2619): starting new server thread
D/NdefPushServer( 2619): about create LLCP service socket
D/NdefPushServer( 2619): created LLCP service socket
D/NdefPushServer( 2619): about to accept
D/NfcService( 2619): NFC-EE OFF
D/NfcService( 2619): NFC-C ON

Diffing the two NFC-relevant repositories between Replicant (external_libnfc-nxp and packages_apps_nfc) and CyanogenMod (android_external_libnfc-nxp and android_packages_apps_Nfc) I found a commit in Replicant that changes a soft-fail on missing firmware to a hard-fail. I manually reverted that patch in my build tree, and rebuilt and booted a new image. Enabling NFC now prints this on my Replicant phone:

I/NfcService( 2508): Enabling NFC
D/NFCJNI  ( 2508): Start Initialization
E/NFC-HCI ( 2508): Could not open /system/vendor/firmware/libpn544_fw.so or /system/lib/libpn544_fw.so
W/NFC     ( 2508): Firmware image not available: this device might be running old NFC firmware!
D/NFCJNI  ( 2508): NFC capabilities: HAL = 8150100, FW = b10122, HW = 620003, Model = 12, HCI = 1, Full_FW = 1, Rev = 34, FW Update Info = 0
D/NFCJNI  ( 2508): phLibNfc_SE_GetSecureElementList()
D/NFCJNI  ( 2508): 
D/NFCJNI  ( 2508): > Number of Secure Element(s) : 1
D/NFCJNI  ( 2508): phLibNfc_SE_GetSecureElementList(): SMX detected, handle=0xabcdef
D/NFCJNI  ( 2508): phLibNfc_SE_SetMode() returned 0x000d[NFCSTATUS_PENDING]
I/NFCJNI  ( 2508): NFC Initialized
D/NdefPushServer( 2508): start, thread = null
D/NdefPushServer( 2508): starting new server thread
D/NdefPushServer( 2508): about create LLCP service socket
D/NdefPushServer( 2508): created LLCP service socket
D/NdefPushServer( 2508): about to accept
D/NfcService( 2508): NFC-EE OFF
D/NfcService( 2508): NFC-C ON

And NFC works! At least YubiKey NEO with the Yubico Authenticator app. One less non-free blob on my phone.

I have double-checked that power-cycling the phone (even removing battery for a while) does not affect anything, so it seems the NFC chip has firmware loaded from the factory.

Question remains why that commit was added. Is it necessary on some other phone? I have no idea, other than if the patch is reverted, S3 owners will have NFC working with Replicant without non-free software added. Alternatively, make the patch apply only on the platform where it was needed, or even to all non-S3 builds.

Offline GnuPG Master Key and Subkeys on YubiKey NEO Smartcard

I have moved to a new OpenPGP key. There are many tutorials and blog posts on GnuPG key generation around, but none of them matched exactly the setup I wanted to have. So I wrote down the steps I took, to remember them if I need to in the future. Briefly my requirements were as follows:

  • The new master GnuPG key is on an USB stick.
  • The USB stick is only ever used on an offline computer.
  • There are subkeys stored on a YubiKey NEO smartcard for daily use.
  • I want to generate the subkeys using GnuPG so I have a backup.
  • Some non-default hash/cipher preferences encoded into the public key.

Continue reading Offline GnuPG Master Key and Subkeys on YubiKey NEO Smartcard