Re: [Nano-devel] compilation problem

[Benno Schulenberg]

[When you respond, please reply only to the mailing list.]


Op 02-09-19 om 14:00 schreef Ansuel Smith:
> I'm using linaro gcc 4.8

What *is* linaro gcc?

> and i think nano have some problem with it.

Have you compiled nano before on that system?  If yes, which was
the last nano you could compile successfully there?

> ansuel@Ansuel-Gaming:~/openwrt$ make package/nano/compile V=s
> fatal: Invalid revision range ee53a240ac902dc83209008a2671e7fdcf55957a..HEAD
> fatal: Invalid revision range

??  What are these errors?

Are you cross compiling?

> In file included from ./scratch_buffer.h:9:0,
>                  from malloc/scratch_buffer_grow.c:23:
> ./malloc/scratch_buffer.h:69:3: error: unknown type name 'max_align_t'

The error occurs while trying to compile a gnulib module.  My system
does not have 'max_align_t' defined in any /usr/include header file either,
but compilation succeeds.  Using gcc-7.4.0 or clang-6.0.0 here.

I don't know what's the problem.  Some googling brings up this:

  https://github.com/emscripten-core/emscripten/issues/6660


Grepping the gnulib source with:

   grep -A2 "lack max" lib/stddef.in.h

/* Some platforms lack max_align_t.  The check for _GCC_MAX_ALIGN_T is
   a hack in case the configure-time test was done with g++ even though
   we are currently compiling with gcc.  */

I don't understand any of that code, but maybe you can work around
the problem by removing the two #ifs from line 87 and 88 in that file
(and the corresponding #endifs on line 109 and 100, of course).

Benno

signature.asc
Description: OpenPGP digital signature