Re: Make CVS HEAD version build out of the box

[Yongwei Wu]

On 10/10/2007, Paul Smith <address@hidden> wrote:
> On Tue, 2007-10-09 at 21:57 +0800, Yongwei Wu wrote:
> > All are silent?  Could we first do the undisputed patches, like this
> > one and the w32_kill change?
> >
> > And I think the CVS version should at least build.  So I think Eli's
> > patch to change `p[-1] =' is better than the current state, using or
> > not using alloca. --- What do you think?
>
> If you guys can come up with a patch everyone seems to agree on, I'll
> apply it.  I don't even own a Windows system, much less one that can
> compile GNU make, so I'm completely at the mercy of the porting folks
> for Windows-only issues.  I generally don't apply patches unless there
> is a clear and firm consensus and agreement from one or more of the
> "longstanding" Windows hackers.
>
> I'm not sure exactly what you're suggesting we apply: Eli's last
> patches, here:
>         http://www.mail-archive.com/make-w32%40gnu.org/msg01586.html
>
> and here:
>         http://www.mail-archive.com/make-w32%40gnu.org/msg01585.html
>
> ?

These two and:

Index: make.h
===================================================================
RCS file: /sources/make/make/make.h,v
retrieving revision 1.129
diff -u -p -r1.129 make.h
--- make.h      4 Jul 2007 19:35:19 -0000       1.129
+++ make.h      30 Sep 2007 03:15:35 -0000
@@ -329,7 +329,7 @@ char *strsignal (int signum);
 # define kill(_pid,_sig) w32_kill((_pid),(_sig))

 void sync_Path_environment (void);
-int kill (int pid, int sig);
+int w32_kill (int pid, int sig);
 char *end_of_token_w32 (const char *s, char stopchar);
 int find_and_set_default_shell (const char *token);

Best regards,

Yongwei

-- 
Wu Yongwei
URL: http://wyw.dcweb.cn/