Re: another issue

[Harry Thijssen]

The most astonishing is that this only happens with the 64 bits build. The 32 bits build runs well.

Maybe it is a good idea to remove everything and start a fresh build to be sure we are not looking at a hitch in the system or some leftovers.

BTW I am running a build with make test now.  I expect it to run for at least 7 hours. Cross-compiling on a virtual machine. Is this an expected behavior?

Have fun

2017-04-17 16:57 GMT+02:00 Ben Pfaff <address@hidden>:

On Mon, Apr 17, 2017 at 03:57:31PM +0200, John Darrington wrote:
> On Mon, Apr 17, 2017 at 02:04:51PM +0200, Harry Thijssen wrote:
>      It isn't.
>
>      From my config.h:
>
>
>      /* Define to `int' if <sys/types.h> doesn't define. */
>      #define uid_t int
>
> So uid_t is defined as int ...
>
>
>      > On Mon, Apr 17, 2017 at 01:03:25PM +0200, Harry Thijssen wrote:
>      >
>      >      mv -f $depbase.Tpo $depbase.Plo
>      >      libtool: compile:  x86_64-w64-mingw32-gcc -DHAVE_CONFIG_H
>      > -DEXEEXT=\".exe\"
>      >      -I. -I.. -I/usr/x86_64-w64-mingw32/sys-root/mingw/include
> ....
>      >      -I/usr/x86_64-w64-mingw32/sys-root/mingw/include
>      >      -I/usr/x86_64-w64-mingw32/sys-root/mingw/include/libpng16 -O2 -g
>      > -pipe
>      >      -Wall -fexceptions --param=ssp-buffer-size=4 -mms-bitfields -MT
>      >      fatal-signal.lo -MD -MP -MF .deps/fatal-signal.Tpo -c fatal-signal.c
>      >      -DDLL_EXPORT -DPIC -o .libs/fatal-signal.o
>      >      In file included from fatal-signal.c:26:0:
>      >      ./signal.h:703:3: error: unknown type name 'uid_t'
>      >         uid_t si_uid;
>      >         ^~~~~
>
> ... so I don't understand why it thinks that uid_t is an unknown type.
> Can you check that fatal-signal.c #includes "config.h"

fatal-signal.c does include config.h, so this is extra strange.