[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: gcc or binutils bug?
From: |
Paul Novak |
Subject: |
Re: gcc or binutils bug? |
Date: |
Tue, 9 Jan 2007 11:59:24 -0600 |
On Tue, Jan 09, 2007 at 07:56:30AM +0000, Nick Clifton wrote:
> Hi Paul,
>
> >I'm experiencing some problem try to build binutils-2.17
> >using GCC 4.1.1 on an AMD Athon. By default,
> >binutils-2.17 is configured and compiled with -Werror.
> >When using -O2, binutils-2.17 compiles cleanly.
> >However, with -O3, I get some warnings which get
> >elevated to fatal errors.
>
> We enabled -Werror in order to help uncover any lurking bugs which do
> not normally trigger compiler error messages. (And in the process also
> to help tidy up the sources). Hence we are very much interested in
> warnings/errors produced when compiling at -O3, even though we do not
> normally do that ourselves.
>
> So, please feel free to report these warnings/errors on this list. Even
> better, please feel free to submit patches to fix them as well. In some
> cases however it is necessary to be a little bit cautious, as not all
> warnings reported at -O3 are actual problems with the code. Sometimes
> the compiler itself can be a little bit too over-enthusiastic. In such
> cases we will consider altering the code to avoid the warning as long as
> it does not introduce any performance penalties or unnecessary
> complexities into the code. Otherwise we would probably just ignore the
> warning and instead put a comment into the code explaining why we are
> not taking any other action.
I checked one more time, this time just with googel, and I did find
the bug already reported.
[Bug binutils/3039] New: Compile fails (file wrstabs.c)
Thank you for your explanations.
Paul.