config-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Σχετ: Re: x86_64-solaris


From: Rainer Orth
Subject: Re: Σχετ: Re: x86_64-solaris
Date: Thu, 20 Dec 2018 22:51:08 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/26.1 (usg-unix-v)

Hi Earnie,

> On 12/20/2018 4:20 PM, Rainer Orth wrote:
>> Hi Apostolos,
>>
>>> If I understand correctly you do notlike the fact config.sub
>>> correctlyguesses that a system is 32 or 64bit?
>>
>> quite the contrary: if you boot a 64-bit kernel (the default since
>> Solaris 10/SPARC and Solaris 11/x86), the *system* is *both* 32 and
>> 64-bit.  What's of interest to the build process and thus the configure
>> triplet is if the *compiler* used generates 32 or 64-bit code: if using
>> a 64-bit compiler (either because that's the default or because you use
>> -m64), you get the 64-bit configure triplet, while with a 32-bit
>> compiler (again by default or due to -m32), you get the 32-bit triplet
>> instead.  Your change lost that distinction.
>
> Using switches such as -m32/-m64 or -mwindows (which was removed) for
> Cygwin causes user issues. It makes tool maintenance nearly unattainable
> when all the possibilities are to be considered.  A decision was made to
> use the uname specifics to determine the guessed system.  If something
> other is required the user has the ability to specify the system.  You may
> want to review [1] to see if it might help you.
>
> [1]
> https://www.gnu.org/savannah-checkouts/gnu/autoconf/manual/autoconf-2.69/html_node/Site-Defaults.html#Site-Defaults

I'm not proposing to change anything for Cygwin or any other system,
only to revert an incompatible change on Solaris.  I know that part of
the autoconf docs perfectly well, but this is not about some site issue,
but a question about an individual configure run/compilation and
configure automatically determining the correct triplet, just as it has
done before.

        Rainer

-- 
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University



reply via email to

[Prev in Thread] Current Thread [Next in Thread]