Re: gc.m4 and hard failure

[Ralf Wildenhues]

Hi Simon,

* Simon Josefsson wrote on Tue, Mar 07, 2006 at 02:17:13PM CET:
> Simon Josefsson <address@hidden> writes:
> > Ralf Wildenhues <address@hidden> writes:
> >
> >> It would be great if the gc.m4 module could be taught not to fail when
> >> no random devices exist (as seen on hppa2.0-hp-hpux10.20).  Passing
> >>
> >>   --disable-random-device --disable-pseudo-random-device \
> >>   --disable-nonce-device
> >>
> >> did not help.

> > I agree.  The functions using those random devices can return an
> > error, so I suggest turning the errors into warnings, and make the
> > code properly return an error if the device is unavailable.  I think
> > the code already does this though.
> 
> I have installed the patch below.  It separate the random stuff into a
> new module, gc-random.  It also turn the errors into warnings.

> What do you think?

Better, thanks.  But
  --disable-random-device  (--enable-random-devide=no)
could IMVHO still be given a useful meaning, which it currently does not
have.  Since I don't actually use this code in any project, I can't tell
you how realistic this usage case would be.

Meanwhile, I noted another small issue, see below.

Cheers,
Ralf

> Index: m4/gc-random.m4
> ===================================================================
> RCS file: m4/gc-random.m4
> diff -N m4/gc-random.m4
> --- /dev/null 1 Jan 1970 00:00:00 -0000
> +++ m4/gc-random.m4   7 Mar 2006 13:15:33 -0000
> @@ -0,0 +1,81 @@
> +# gc-random.m4 serial 1
> +dnl Copyright (C) 2005, 2006 Free Software Foundation, Inc.
> +dnl This file is free software; the Free Software Foundation
> +dnl gives unlimited permission to copy and/or distribute it,
> +dnl with or without modifications, as long as this notice is preserved.
> +
> +AC_DEFUN([gl_GC_RANDOM],
> +[
> +  # Devices with randomness.
> +  # FIXME: Are these the best defaults?
> +
> +  case "${target}" in

Why do you use $target?  Are you putting this in a compiler?
It'd be good to precede with
  AC_REQUIRE([AC_CANONICAL_HOST])dnl

and then you really want
   case $host in

> +    *-openbsd*)

       *-*-openbsd*)

You could also just test $host_os, and then match
       openbsd*)
only.

> +      NAME_OF_RANDOM_DEVICE="/dev/srandom"
> +      NAME_OF_PSEUDO_RANDOM_DEVICE="/dev/prandom"
> +      NAME_OF_NONCE_DEVICE="/dev/urandom"
> +      ;;
*snip*