svn commit: r252376 - head/lib/libutil

[Konstantin Belousov]

On Sat, Jun 29, 2013 at 09:42:49AM -0700, mdf at FreeBSD.org wrote:
> On Sat, Jun 29, 2013 at 9:36 AM, Tim Kientzle <kientzle at freebsd.org> wrote:
> 
> >
> > On Jun 29, 2013, at 9:19 AM, Konstantin Belousov wrote:
> >
> > > On Sat, Jun 29, 2013 at 03:52:49PM +0000, Tim Kientzle wrote:
> > >> Author: kientzle
> > >> Date: Sat Jun 29 15:52:48 2013
> > >> New Revision: 252376
> > >> URL: http://svnweb.freebsd.org/changeset/base/252376
> > >>
> > >> Log:
> > >>  Fix -Wunsequenced warning
> > > What is this ? From the name of the warning, it sounds as if the problem
> > > is in the lack of sequence point between two modifications of the same
> > > variable in the expression ?
> > >
> > > But, there function' argument evaluation and function call are separated
> > > by seq point, AFAIR.  Could you, please, clarify ?
> >
> > I think you're right about that, though I'd have to
> > look at the spec to be sure.
> >
> > Not sure why clang would report this as a -Wunsequenced
> > warning.  The implied store here is certainly redundant, though.
Definitily, I said that the changes are good (not bad).

> >
> 
> It may be like other warnings (-Wmissing-field-initializers, I'm looking at
> you) that warn about currently correct, but potentially problematic
> behavior.
> 
> In particular, if any of the functions is re-implemented as a macro, the
> sequence point goes away, and this code is broken without the code's author
> having made any changes.  So it seems like a reasonable warning.

It is only after the functions reimplemented the code would be broken.
Right now, it seems that the warning is broken, not code.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 834 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/svn-src-head/attachments/20130629/88c54fca/attachment.sig>