[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Not-really-cross-compilation issues
From: |
Akim Demaille |
Subject: |
Re: Not-really-cross-compilation issues |
Date: |
22 Jan 2001 17:08:56 +0100 |
User-agent: |
Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.1 (Crater Lake) |
>>>>> "Peter" == Peter Eisentraut <address@hidden> writes:
Peter> Akim Demaille writes: Autoconf now assumes that if you can't
Peter> run a program then you must be cross-compiling, so you should
Peter> use --host.
>> No, it says `It doesn't work', and `if you were doing cross
>> compilation, then say it loud and clear'.
>>
>> What kind of message would you suggest?
Peter> No message, just continue.
I'm surprised!
You seem to prefer the old behavior? I certainly don't. The sooner
losing environment are diagnosed, the better.
Peter> Or maybe some message, but continue anyway. Notice how my
Peter> example didn't even use AC_CANONICAL_xxx, so the specification
Peter> of --host is completely pointless.
You seem to have missed one of the most important changes in 2.50: the
consistent and uniform handling of cross-compilation issues which *are
not related* to AC_CANON.
Peter> This will confuse users.
If you think the error message should be improved, please make
suggestions. But as far as the error is concerned, I don't think we
will change it.
- Not-really-cross-compilation issues, Peter Eisentraut, 2001/01/19
- Re: Not-really-cross-compilation issues, akim, 2001/01/21
- Re: Not-really-cross-compilation issues, Peter Eisentraut, 2001/01/21
- Re: Not-really-cross-compilation issues, Akim Demaille, 2001/01/22
- Re: Not-really-cross-compilation issues, Peter Eisentraut, 2001/01/22
- Re: Not-really-cross-compilation issues, Alexandre Oliva, 2001/01/22
- Re: Not-really-cross-compilation issues, Peter Eisentraut, 2001/01/22
- Re: Not-really-cross-compilation issues, Alexandre Oliva, 2001/01/22
- Re: Not-really-cross-compilation issues,
Akim Demaille <=