Last week:

• Pushed the change for Ed448 to libgcrypt (no optimization at all): D504: ECC change for Ed448
  • And support *ph variant and "Context" by "label" attribute.
    • Test is included for Ed448
    • No test included for Ed25519 yet
• Created (and updated) the change for Ed448 to GnuPG: D505: Ed448 support for GnuPG
• Minor clean up in GnuPG to prepare accepting D505
• Tested OpenPGP card v3.4 implementation by Achim
  • flashing was done successfully with TTXS reader
  • no problem with current GnuPG
  • tested against Gnuk's test suite
  • found a minor difference: In V3.4, SEX DO factory setting is '0' (Unknown), while it used to be '9' (N/A)
    • ISO/IEC 5218 thing
    • I don't know if we need to modify GnuPG or not

This week:

• Change UI of GnuPG master "cv448" to specify X448 to align to cv25519
• Others: FSIJ accounting of FY2019 and AGM 2020

D505:

• I realized that the patch number looks like "SOS" :-)
• For "Ed25519", libgcrypt supports ECDSA as well as EdDSA
  • And Gnunet looks like using it (I don't know if it's intended or not)
  • I don't know if it works well, I'm afraid cofactor matters
  • but for "Ed448", ECDSA is not supported
  • Anyway, because of this, a key or sig-data with Ed25519 curve for EdDSA has "(flags eddsa)"
• Learning math for elliptic integral and Jacobi elliptic functions

Last week:

• Office tasks
• Looked at James' TPM patches and started to port them to master.