From c360666adc9cb1d441dc273e9561254d9170b811 Mon Sep 17 00:00:00 2001 From: Jakob Unterwurzacher Date: Sat, 16 Sep 2017 16:46:57 +0200 Subject: reverse mode: uppercase DerivePathIV --- docs/reverse_mode_crypto.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/reverse_mode_crypto.md b/docs/reverse_mode_crypto.md index d81da07..e7af8b5 100644 --- a/docs/reverse_mode_crypto.md +++ b/docs/reverse_mode_crypto.md @@ -27,7 +27,7 @@ SHA512. All values that are random in forward mode (File ID, Block IV) are instead deterministically derived from the encrypted path, -essentially using a salted hash (detailed in the section "derivePathIV"). +essentially using a salted hash (detailed in the section *DerivePathIV*). As all derived values are explicitely stored in the ciphertext file, decryption does not depend on knowledge of the derivation. @@ -51,7 +51,7 @@ File Names File name encryption is identical to forward mode, with the exception that the directory IV (stored in `gocryptfs.diriv`) is not random. -It is deterministically derived, using derivePathIV, from the encrypted +It is deterministically derived, using *DerivePathIV*, from the encrypted path to the directory. Because the encrypted path to the root directory is "" (the empty string), @@ -61,9 +61,9 @@ this means that the directory IV in the root directory is always DerivePathIV: Derive IVs from Encrypted Paths ---------------------------------------------- -DerivePathIV concatenates the encrypted path with a null byte and a +The *DerivePathIV* function concatenates the encrypted path with a null byte and a salt string (one of "DIRIV", "FILEID", "BLOCK0IV"). This is -is hashed with SHA256 and truncated to 128 bits (source code +hashed with SHA256 and truncated to 128 bits (source code [ref](https://github.com/rfjakob/gocryptfs/blob/f0e29d9b90b63d5fbe4164161ecb0e1035bb4af4/internal/pathiv/pathiv.go#L26)). ![](img/reverse-derivePathIV.svg) -- cgit v1.2.3