Patchwork fix PR bootstrap/44825, Werror=sign-compare error

login
register
mail settings
Submitter Nathan Froyd
Date July 5, 2010, 7:24 p.m.
Message ID <20100705192442.GM17877@codesourcery.com>
Download mbox | patch
Permalink /patch/57924/
State New
Headers show

Comments

Nathan Froyd - July 5, 2010, 7:24 p.m.
The patch below makes the obvious fix for PR 44825.  I'm asking for
approval, though, because I don't have an convenient way to do bootstrap on a
32-bit HWI host.  (H.J.'s suggestion in the PR does not work for me,
presumably because of Ubuntu weirdness.)  If somebody wanted to quickly
toss this into their tree, that would be much appreciated.

OK to commit?

-Nathan

	PR bootstrap/44825
	* class.c (make_class_data): Cast result of VEC_length calls to int.

Index: ChangeLog
===================================================================
H.J. Lu - July 5, 2010, 10:17 p.m.
On Mon, Jul 5, 2010 at 12:24 PM, Nathan Froyd <froydnj@codesourcery.com> wrote:
> The patch below makes the obvious fix for PR 44825.  I'm asking for
> approval, though, because I don't have an convenient way to do bootstrap on a
> 32-bit HWI host.  (H.J.'s suggestion in the PR does not work for me,
> presumably because of Ubuntu weirdness.)  If somebody wanted to quickly
> toss this into their tree, that would be much appreciated.
> OK to commit?
>
> -Nathan
>
>        PR bootstrap/44825
>        * class.c (make_class_data): Cast result of VEC_length calls to int.
>

This passed the previous failure. Can you check it in?

Thanks.
Andrew Haley - July 6, 2010, 8:23 a.m.
On 07/05/2010 08:24 PM, Nathan Froyd wrote:
> The patch below makes the obvious fix for PR 44825.  I'm asking for
> approval, though, because I don't have an convenient way to do bootstrap on a
> 32-bit HWI host.  (H.J.'s suggestion in the PR does not work for me,
> presumably because of Ubuntu weirdness.)  If somebody wanted to quickly
> toss this into their tree, that would be much appreciated.
> 
> OK to commit?
> 
> -Nathan
> 
> 	PR bootstrap/44825
> 	* class.c (make_class_data): Cast result of VEC_length calls to int.

Sure.  It's obvious, right?  :-)

Andrew.

Patch

Index: class.c
===================================================================
--- class.c	(revision 161839)
+++ class.c	(working copy)
@@ -1921,9 +1921,9 @@  make_class_data (tree type)
     }
 
   gcc_assert (static_field_count
-              == VEC_length (constructor_elt, static_fields));
+              == (int) VEC_length (constructor_elt, static_fields));
   gcc_assert (instance_field_count
-              == VEC_length (constructor_elt, instance_fields));
+              == (int) VEC_length (constructor_elt, instance_fields));
 
   if (field_count > 0)
     {