public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Florian Weimer <fweimer@redhat.com>
To: Zack Weinberg <zackw@panix.com>
Cc: GNU C Library <libc-alpha@sourceware.org>,
	Carlos O'Donell <carlos@redhat.com>,
	Nikos Mavrogiannopoulos <nmav@redhat.com>
Subject: Re: [PATCH] New configure option --disable-libcrypt.
Date: Tue, 17 Apr 2018 10:21:00 -0000	[thread overview]
Message-ID: <ac70eb58-98fb-07b3-bf78-0e54ebcf382b@redhat.com> (raw)
In-Reply-To: <CAKCAbMgbEBd1beqZ-UrVXW0T51LpL-n5i2YQwmHVS0x51X1vVw@mail.gmail.com>

On 04/16/2018 05:56 PM, Zack Weinberg wrote:

> Let me see if I understand what you have in mind:  In
> --disable-libcrypt mode, we would preserve _XOPEN_CRYPT, with value 1,
> and the prototype for crypt in unistd.h.

My main proposal is preserving the prototype for crypt, under 
_GNU_SOURCE if necessary (without defining _XOPEN_CRYPT).

I just don't see value in the friction caused by dropping the definition.

> We would also preserve the
> prototypes for encrypt and setkey, but maybe with
> __attribute__((deprecated)).

I'm less concerned about these two functions.  We can do there whatever 
we think is best from a standard compliance perspective, I think.  There 
simply aren't enough users to make a difference, I think.  This is very 
different for the crypt function.

> We just wouldn't provide crypt.h or libcrypt.

Exactly.

> It's not a conformance violation for us to continue
> defining _XOPEN_CRYPT as 1 because, as long as you get libcrypt from
> _somewhere_, it will work at runtime.  Is that right?

Right.

It could still be a conformance violation for setkey/encrypt because 
current libxcrypt (at least the version in Fedora) does not provide the 
functions anymore for linking:

$ eu-readelf --symbols=.dynsym /lib64/libcrypt.so.1 | grep encrypt
    38: 000000000000b330     20 FUNC    GLOBAL DEFAULT       12 
encrypt_r@GLIBC_2.2.5
    52: 000000000000b360     16 FUNC    GLOBAL DEFAULT       12 
encrypt@GLIBC_2.2.5

Thanks,
Florian

  parent reply	other threads:[~2018-04-17 10:21 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-04-16  1:36 Zack Weinberg
2018-04-16 15:11 ` Florian Weimer
2018-04-16 15:56   ` Zack Weinberg
2018-04-16 18:41     ` Zack Weinberg
2018-04-17 10:21     ` Florian Weimer [this message]
2018-04-18 13:43       ` Zack Weinberg

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ac70eb58-98fb-07b3-bf78-0e54ebcf382b@redhat.com \
    --to=fweimer@redhat.com \
    --cc=carlos@redhat.com \
    --cc=libc-alpha@sourceware.org \
    --cc=nmav@redhat.com \
    --cc=zackw@panix.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).