Re: [Mingw-cross-env-list] Congrats - short question.

[Nagaev Boris]

On Sun, Jan 31, 2016 at 11:20 AM, Jos De Laender <address@hidden> wrote:
> Thanks Tony and team !
>
> Does that analysis match with the x86_64 targets working as expected ? Can I
> rely reasonably on such build ?
>
> Can I address also a totally different minor point but which I believe is
> nevertheless very confusing for people like me who are just using your
> toolchain without being familiar with all the stuff behind it. It was only
> by trial and error I found out that x86_64 generates* Windows64 (PE32+)
> executables and that i686* generates Windows32 (PE32) executables. My
> initial understanding based on Google searches was that the first element of
> the triplet indicates the platform on which the compiler is running (and you
> could imagine generating PE32/PE32+ as well on i686 as on x86_64). So that's
> where I got pretty confused. I would like to suggest you getting a table in
> the documentation that clearly indicates what is the meaning of all the
> triplet stuff you are using and what's the practical use. (I guess that your
> users either are running on i686 either on x86_64 and want to generate
> either PE32 or PE32+. In both cases static or shared. So the list shouldn't
> be longer than 8 entries ?)

Actually, there are 3 triplets in MXE:

 * BUILD, e.g. x86_64-unknown-linux-gnu.
 * HOST, e.g. x86_64-unknown-linux-gnu.
 * TARGET, e.g., i686-w64-mingw32.static

All of them are in form of cpu-vendor-os.

BUILD describes the machine which builds the cross-compiler, HOST is
the machine on which the cross-compiler works (in case of MXE
BUILD=HOST). TARGET describes where the result of cross-compilation is
supposed to run.

See also 
https://www.gnu.org/savannah-checkouts/gnu/autoconf/manual/autoconf-2.69/html_node/Specifying-Target-Triplets.html

>
> Sorry for the nitpicking, I'd like to re-iterate I am in the first place
> very impressed by the results of your efforts !
>
> Jos
>
>
> Op 31/01/2016 om 8:49 schreef Tony Theodore:
>
>>> On 31 Jan 2016, at 18:21, Tony Theodore <address@hidden> wrote:
>>>
>>>
>>>> On 31 Jan 2016, at 04:57, Nagaev Boris <address@hidden> wrote:
>>>>
>>>> Something is wrong with your libstdc++-6.dll. Here there is the
>>>> following symbol in my libstdc++-6.dll:
>>>>
>>>> 6fec1330 T __ZNSt8ios_base4InitC1Ev
>>>
>>> Something is wrong on OSX 10.10.5 also:
>>>
>>> $ i686-w64-mingw32.shared-nm
>>> usr/i686-w64-mingw32.shared/bin/libstdc++-6.dll | grep InitC1
>>>          U __ZNSt8ios_base4InitC1Ev
>>
>> Appears to be a bug in binutils[1], reverting the recent update[2] seems
>> to fix it[3]:
>>
>> $ i686-w64-mingw32.shared-nm
>> usr/i686-w64-mingw32.shared/bin/libstdc++-6.dll | grep InitC1
>> 6fec1310 T __ZNSt8ios_base4InitC1Ev
>>
>> Cheers,
>>
>> Tony
>>
>> [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=813144
>> [2]
>> https://github.com/mxe/mxe/commit/bd5f44b810f8278f446d53186ebcdabad253f7c5
>> [3] https://github.com/mxe/mxe/pull/1206
>>
>>
>
> --
> Jos De Laender
>
> http://www.jodela.be
> https://www.de-laender.be:9961
>
>
> ---
> Dit e-mailbericht is gecontroleerd op virussen met Avast antivirussoftware.
> https://www.avast.com/antivirus
>
>



-- 


Best regards,
Boris Nagaev