bug#10374: 3 test failures on fedora 16

[Jim Meyering]

Stefano Lattarini wrote:
...
>> Same with 60MB, but it passes with 70MB.
>> So I suggest changing s/50/70/ in both the patch and the log message.
>> Maybe even 90 or 100, just to be a little more future/bloat-proof.
>>
> Since we are being liberal already, let's go for 150 MB then ;-)
>
> What about the updated, attached patch?

Looks fine, modulo a typo in a new comment.

...
>>From 9b06cda000e86888abfc9999b4c66ab8df10873a Mon Sep 17 00:00:00 2001
> Message-Id: <address@hidden>
> From: Stefano Lattarini <address@hidden>
> Date: Tue, 27 Dec 2011 10:40:57 +0100
> Subject: [PATCH] tests: fix spurious failure of cond29.test
...
>  # or cpu time it can found.
> +# Limit memory available to spawned processes to ~ 150 MB.  This might
> +# seem too liberal, but it is needed to account for present and features

s/features/future/

I would say that 70MB is enough, but we're allowing for more.
The 150MB isn't really needed.

   # Limit memory available to spawned processes to ~ 150 MB.  Considering
   # that 70 MB is enough for x86_64 Fedora 16, that gives tools and libraries
   # plenty of room to grow.

> +# "bloating" of automake, perl, or even the C library (especially on 64
> +# bit machines).  Suggested by Jim Meyering in automake bug#10374.
>  (ulimit -v 1; sh -c ":") && skip_ "no adequate 'ulimit' builtin found"
> -(ulimit -v 20000; sh -c ":") || skip_ "no adequate 'ulimit' builtin found"
> -ulimit -v 20000
> -
> -for i in 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22; 
> do
> +(ulimit -v 150000; sh -c ":") || skip_ "no adequate 'ulimit' builtin found"
> +ulimit -v 150000
> +
> +for i in \
> +  01 02 03 04 05 06 07 08 09 10 \
> +  11 12 13 14 15 16 17 18 19 20 \
> +  21 22 23 24
> +do
>    unindent >>Makefile.am <<EOF
>      if C$i
>      bin_PROGRAMS += a$i