Web lists-archives.com

Re: [PATCH v2 2/4] config.mak.uname: support for modern HPE NonStop config.

On Thu, Dec 27, 2018 at 1:35 PM <randall.s.becker@xxxxxxxxxx> wrote:
> A number of configuration options are not automatically detected by
> configure mechanisms, including the location of Perl and Python.
> There was a problem at a specific set of operating system versions
> that caused getopt to have compile errors. Accounted for this by


> providing emulation defines for those versions.
> Signed-off-by: Randall S. Becker <rsbecker@xxxxxxxxxxxxx>
> ---
> diff --git a/config.mak.uname b/config.mak.uname
> @@ -441,26 +441,45 @@ ifeq ($(uname_S),NONSTOP_KERNEL)
>         # Our's are in ${prefix}/bin (perl might also be in /usr/bin/perl).
> -       PERL_PATH = ${prefix}/bin/perl
> -       PYTHON_PATH = ${prefix}/bin/python
> +       PERL_PATH = /usr/bin/perl
> +       PYTHON_PATH = /usr/bin/python

Referring to your answer[1] to my question about ${prefix}, should the
comment above these lines, which talks about ${prefix}, be removed by
this patch?

[1]: https://public-inbox.org/git/000601d49e0b$e11d7520$a3585f60$@rogers.com/

> +       # The current /usr/coreutils/rm at lowest support level does not work
> +       # with the git test structure. Long paths cause nftw as in
> +       # 'trash directory...' break at OS releases J06.21 and L17.02.

"break" in what fashion? What is the actual incorrect behavior?

I'm not asking for my own edification particularly, but rather to help
the person who comes after you supporting this platform. That person
will need to understand the exact nature of the problem in order to
determine if this work-around is still needed or if it can be retired
or somehow conditionalized. Stated another way, the reason behind this
change still seems like a black-box, mysterious and under-explained.
There is nothing concrete here upon which to grasp to gain an
understanding of the problem.

> +       # Default to the older rm until those two releases are deprecated.
> +       RM = /bin/rm -f