[prev in list] [next in list] [prev in thread] [next in thread] 

List:       gcc-patches
Subject:    Re: [PATCH][RFC] Add a subset of -Warray-bounds warnings to C/C++ front ends
From:       "H.J. Lu" <hjl.tools () gmail ! com>
Date:       2008-05-02 20:01:18
Message-ID: 6dc9ffc80805021301u68091843k20334c3ce935d0f1 () mail ! gmail ! com
[Download RAW message or body]

On Fri, May 2, 2008 at 12:53 PM, Andrew Pinski <pinskia@gmail.com> wrote:
> On Fri, May 2, 2008 at 9:34 AM, Mark Mitchell <mark@codesourcery.com> wrote:
>  > Simon Baldwin wrote:
>  >
>  >
>  > > I think the wisest course for now is to roll back this change until I have
>  > a better understanding of the issues.
>  > >
>  >
>  >  If you want to roll back, that's fine.  You can do that without any further
>  > approval.
>
>  x86-darwin is also broken with the following warning:
>  /home/regress/tbox/svn-gcc/gcc/config/i386/i386.c:20877: error: array
>
> subscript is above array bounds
>
>  which is a false warning.
>  We have:
>  mode0 = insn_data[icode].operand[0].mode;
>
>  where insn_data is defined as:
>  extern const struct insn_data insn_data[];
>
>  operand is a pointer so that is not an issue.
>

May I suggest to revert it now and use a branch to implement it
properly before merging it with trunk? I can prepare a patch in half
an hour.

Thanks.

H.J.
[prev in list] [next in list] [prev in thread] [next in thread]