From 3dc839cecdf5ea11336d51c9dbe4b7cd83b0f3f7 Mon Sep 17 00:00:00 2001 From: Mark Wooding Date: Sat, 29 Apr 2017 13:55:40 +0100 Subject: [PATCH] secnet.8, magic.h: Rephrase documentation of `capab-num' settings. In particular, I've abolished the idea of a specific class of `transform capabilities'. They're all just capabilities, and they need to mean the same thing at both ends. Signed-off-by: Mark Wooding --- magic.h | 16 ++++++++-------- secnet.8 | 12 ++++++------ 2 files changed, 14 insertions(+), 14 deletions(-) diff --git a/magic.h b/magic.h index 5b0407e..e0ea8aa 100644 --- a/magic.h +++ b/magic.h @@ -42,15 +42,15 @@ /* remaining bits are unused */ /* - * The transform capability mask is a set of bits, one for each - * transform supported. The transform capability numbers are set in - * the configuration (and should correspond between the two sites), - * although there are sensible defaults. + * The capability mask is a set of bits, one for each optional feature + * supported. The capability numbers for transforms are set in the + * configuration (and should correspond between the two sites), although + * there are sensible defaults. * - * Advertising a nonzero transform capability mask promises that - * the receiver understands LABEL_MSG3BIS messages, which - * contain an additional byte specifying the transform capability - * number actually chosen by the MSG3 sender. + * Advertising a nonzero capability mask promises that the receiver + * understands LABEL_MSG3BIS messages, which contain an additional byte + * specifying the transform capability number actually chosen by the MSG3 + * sender. * * Aside from that, an empty bitmask is treated the same as * 1u<