Hi,

nobu.nokada / softhome.net wrote:
>>	If you read the patch, please note change in defines.h
>>	FLUSH_REGISTER_WINDOWS is now defined everytime, but
>>	compilation with -O2 handles empty proc body quite well
>>	(ie. no code - no useless jumps).
> 
> 
> Is it needed?

	Try -Wall compilation. It warns badly (gcc-3.2.2) about the code has no 
effect and so on, which looks ugly to me. This patch has the same result 
(code has no effect), but without warnings from compiler.

> 
> 
>>Index: dln.c
>>===================================================================
>>RCS file: /src/ruby/dln.c,v
>>retrieving revision 1.48
>>diff -u -p -r1.48 dln.c
>>--- dln.c	31 Dec 2002 10:53:14 -0000	1.48
>>+++ dln.c	15 Jan 2003 20:52:42 -0000
>>@@ -1304,7 +1304,7 @@ dln_load(file)
>> 
>>     char *buf;
>>     /* Load the file as an object one */
>>-    init_funcname(&buf, file);
>>+    init_funcname(&buf, (char *)file);
> 
> 
> Second argument of init_funcname() is const char*, so this cast
> doesn't seem necessery.

	My bad. Sorry. It seems that prototype was changed. So this is really 
useless.

>>Index: eval.c
>>===================================================================
>>RCS file: /src/ruby/eval.c,v
>>retrieving revision 1.384
>>diff -u -p -r1.384 eval.c
>>--- eval.c	15 Jan 2003 08:07:41 -0000	1.384
>>+++ eval.c	15 Jan 2003 20:52:50 -0000
>>@@ -938,7 +938,7 @@ static void call_trace_func _((char*,NOD
>> #define SET_CURRENT_SOURCE() (ruby_sourcefile = ruby_current_node->nd_file, \
>> 			      ruby_sourceline = nd_line(ruby_current_node))
>> #else
>>-#define SET_CURRENT_SOURCE() 0
>>+#define SET_CURRENT_SOURCE() do { } while (0)
> 
> 
> What about ((void)0) ?

	Again - gcc warns about 'code has no effect'...

>>@@ -2181,6 +2181,8 @@ call_trace_func(event, node, self, id, k
>>     if (state) JUMP_TAG(state);
>> }
>> 
>>+VALUE rb_check_array_type _((VALUE));
>>+
> 
> Isn't it declared in intern.h?

	Ahh, again... When I did this cleanup, it wasn't there. So sorry once 
again :)

> 
>>Index: ext/readline/readline.c
>>===================================================================
>>RCS file: /src/ruby/ext/readline/readline.c,v
>>retrieving revision 1.11
>>diff -u -p -r1.11 readline.c
>>--- ext/readline/readline.c	30 Aug 2002 10:41:57 -0000	1.11
>>+++ ext/readline/readline.c	15 Jan 2003 20:53:08 -0000
>>@@ -3,6 +3,7 @@
>> 
>> #include <errno.h>
>> #include <stdio.h>
>>+#include <unistd.h>
>> #include <readline/readline.h>
>> #include <readline/history.h>
> 
> 
> It may be better to surround with HAVE_UNISTD_H.
> 

	True. Unisth.h is needed for isatty().

> 
>>Index: ext/stringio/stringio.c
>>===================================================================
>>RCS file: /src/ruby/ext/stringio/stringio.c,v
>>retrieving revision 1.12
>>diff -u -p -r1.12 stringio.c
>>--- ext/stringio/stringio.c	20 Dec 2002 11:23:40 -0000	1.12
>>+++ ext/stringio/stringio.c	15 Jan 2003 20:53:12 -0000
> 
> 
> This patch is OK.
> 

	:)

		Michal