GNU bug report logs - #45303
28.0.50; [feature/native-comp] comp.c compilation error on Windows 10

Previous Next

Full log

View this message in rfc822 format

From: Andy Moreton <andrewjmoreton <at> gmail.com>
To: 45303 <at> debbugs.gnu.org
Subject: bug#45303: #45303 [feature/native-comp] building error on Windows
Date: Mon, 21 Dec 2020 00:53:40 +0000

On Sun 20 Dec 2020, Andrea Corallo via "Bug reports for GNU Emacs, the Swiss army knife of text editors" wrote:

> Andrea Corallo via "Bug reports for GNU Emacs, the Swiss army knife of
> text editors" <bug-gnu-emacs <at> gnu.org> writes:
>
> [...]
>
>> Thanks, hope 18ca9ea08a fix it.
>                ^^^^^
>                3bb2fd0c58

Thanks Andrea, that problem is fixed. There are still some warnings from
comp.c that indicate other troubles:

C:/emacs/git/emacs/native/src/comp.c: In function 'Fcomp__compile_ctxt_to_file':
C:/emacs/git/emacs/native/src/comp.c:4461:12: warning: unused variable 'oldset' [-Wunused-variable]
 4461 |   sigset_t oldset;
      |            ^~~~~~
C:/emacs/git/emacs/native/src/comp.c: In function 'eln_load_path_final_clean_up':
C:/emacs/git/emacs/native/src/comp.c:4626:12: warning: passing argument 7 of 'internal_condition_case_5' from incompatible pointer type [-Wincompatible-pointer-types]
 4626 |        Qt, return_nil, Qnil);
      |            ^~~~~~~~~~
      |            |
      |            struct Lisp_X * (*)(struct Lisp_X *)
In file included from C:/emacs/git/emacs/native/src/comp.c:23:
C:/emacs/git/emacs/native/src/lisp.h:4161:195: note: expected 'Lisp_Object' {aka 'struct Lisp_X *'} but argument is of type 'struct Lisp_X * (*)(struct Lisp_X *)'
 4161 | extern Lisp_Object internal_condition_case_5 (Lisp_Object (*) (Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object), Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object (*) (Lisp_Object));
      |                                                                                                                                                                                                   ^~~~~~~~~~~
In file included from C:/emacs/git/emacs/native/src/lisp.h:944,
                 from C:/emacs/git/emacs/native/src/comp.c:23:
./globals.h:6980:14: warning: passing argument 8 of 'internal_condition_case_5' from incompatible pointer type [-Wincompatible-pointer-types]
 6980 | #define Qnil builtin_lisp_symbol (0)
      |              ^~~~~~~~~~~~~~~~~~~~~~~
      |              |
      |              Lisp_Object {aka struct Lisp_X *}
C:/emacs/git/emacs/native/src/comp.c:4626:24: note: in expansion of macro 'Qnil'
 4626 |        Qt, return_nil, Qnil);
      |                        ^~~~
In file included from C:/emacs/git/emacs/native/src/comp.c:23:
C:/emacs/git/emacs/native/src/lisp.h:4161:208: note: expected 'struct Lisp_X * (*)(struct Lisp_X *)' but argument is of type 'Lisp_Object' {aka 'struct Lisp_X *'}
 4161 | extern Lisp_Object internal_condition_case_5 (Lisp_Object (*) (Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object), Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object, Lisp_Object (*) (Lisp_Object));
      |                                                                                                                                                                                                                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~


The internal_condition_case_5 usage looks like the last two arguments
might be swapped.

The return_nil function is also declared as a nested function: that is a
gcc extension, so it is more portable to define it as an ordinary static
function.

    AndyM

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.