Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > muc.lists.netbsd.tech.security > #262

crypt(3) manual page

From Jan Schaumann <jschauma@netmeister.org>
Newsgroups muc.lists.netbsd.tech.security
Subject crypt(3) manual page
Date 2025-11-16 13:19 -0500
Organization Newsgate at muc.de e.V.
Message-ID <aRoVuhOpS7bkdUo_@netmeister.org> (permalink)

Show all headers | View raw


[Multipart message — attachments visible in raw view] - view raw

Hello,

I think the current crypt(3) manual page is not as
helpful as it could be.  Based on the current
description, a user might think that a valid 'setting'
to select MD5 without a salt might be

crypt(pass, "$1")

It would need to be "$1$", or, obviously better,
"$1$salt".

The manual page also says

"The entire password string is passed as setting for
interpretation."

which... is confusing.  You pass the parameters via
'setting' to _yield_ the "entire password string" - I'm
not sure what is meant here by this phrase.  (You need
the entire string to perform _validation_, so that you
can construct the 'setting', then call crypt(3) with
the plaintext, then compare the result, but I think
that's outside the scope of this manual page.)

For the Argon2 and Blowfish functions, the manual page
ought to describe the version string, and for both MD5
and Blowfish we should note that the salt is silently
truncated if longer than allowed.

Attached are my suggested improvements for this manual
page - does that look ok?

-Jan

Back to muc.lists.netbsd.tech.security | Previous | NextNext in thread | Find similar


Thread

crypt(3) manual page Jan Schaumann <jschauma@netmeister.org> - 2025-11-16 13:19 -0500
  Re: crypt(3) manual page Valery Ushakov <uwe@stderr.spb.ru> - 2025-11-17 16:58 +0300
    Re: crypt(3) manual page Jan Schaumann <jschauma@netmeister.org> - 2025-11-17 09:19 -0500
      Re: crypt(3) manual page Valery Ushakov <uwe@stderr.spb.ru> - 2025-11-17 18:06 +0300

csiph-web