emacs-pretest-bug
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: emacs aborts (almost always?) when pressing C-g in gnus group mode


From: Timmy Douglas
Subject: Re: emacs aborts (almost always?) when pressing C-g in gnus group mode
Date: Wed, 04 May 2005 13:10:32 -0400
User-agent: Gnus/5.110003 (No Gnus v0.3) Emacs/22.0.50 (gnu/linux)

Richard Stallman <address@hidden> writes:

>     > The function still exists in the source.  Can you investigate
>     > why it does not exist in your binary?  I can only guess
>     > it was inlined and optimized out of existence.  But that's
>     > supposed to be prevented by NO_INLINE which should expand to
>     > __attribute__((noinline)).
>     >
>     >     static void NO_INLINE
>     >     x_error_quitter (display, error)
>
>     Yeah, I'm it's supposed to not get inlined, but I think it does
>     anyways. Atleast that's what I can see from looking at both asm
>     files. They are here:
>
> Can you verify that __attribute__((noinline)) is truly
> present in the CPP output?  If so, this seems to be a GCC bug.
> Are you using the latest GCC?  If not, please try it.
> If yes, please report the GCC bug.

with -E:

static void __attribute__(())
x_error_quitter (display, error)
     Display *display;
     XErrorEvent *error;
{
  char buf[256], buf1[356];





strace -f -s 2000000 -- i686-pc-linux-gnu-gcc -E -D_BSD_SOURCE   -Demacs 
-DHAVE_CONFIG_H -DUSE_GTK  -I. 
-I/var/tmp/portage/emacs-cvs-22.0.50/work/emacs/src -D_BSD_SOURCE -DXTHREADS 
-D_REENTRANT -DXUSE_MTSAFE_API -I/usr/include/gtk-2.0 
-I/usr/lib/gtk-2.0/include -I/usr/include/atk-1.0 -I/usr/include/pango-1.0 
-I/usr/include/freetype2 -I/usr/include/freetype2/config 
-I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -g -Wall xterm.c 2>&1 | 
grep noinline|less


shows that /usr/include/linux/compiler.h is the problem:

   154  #ifndef noinline
   155  #define noinline
   156  #endif

I guess I could try contacting the lkml? But maybe for now it would be
better to put an #undef in xterm.c or something.







reply via email to

[Prev in Thread] Current Thread [Next in Thread]