Re: [Qemu-devel] [PATCH] readline: avoid memcpy() of overlapping regions

[Stefan Hajnoczi <stefanha@gmail.com>]

On Sat, Jan 12, 2013 at 10:46:18AM +0000, Blue Swirl wrote:
> On Thu, Jan 10, 2013 at 12:43 PM, Stefan Hajnoczi <stefanha@gmail.com> wrote:
> > On Wed, Jan 09, 2013 at 08:43:51PM +0000, Blue Swirl wrote:
> >> On Tue, Jan 8, 2013 at 9:03 AM, Stefan Hajnoczi <stefanha@gmail.com> wrote:
> >> > On Mon, Jan 07, 2013 at 03:38:39PM -0500, Nickolai Zeldovich wrote:
> >> >> memcpy() for overlapping regions is undefined behavior; use memmove()
> >> >> instead in readline_hist_add().
> >> >>
> >> >> Signed-off-by: Nickolai Zeldovich <nickolai@csail.mit.edu>
> >> >> ---
> >> >>  readline.c |    4 ++--
> >> >>  1 file changed, 2 insertions(+), 2 deletions(-)
> >> >
> >> > I made a slight modification: keep the tab characters since the
> >> > surrounding code still uses them.
> >>
> >> I think tabs should be fixed whenever possible, otherwise we may never
> >> get them converted.
> >
> > Not in a one-line patch when the surrounding lines still use them.  It
> > creates a mess.
> 
> Only if the reader messes with the tab width settings (and in that
> case they deserve what they get and they are probably also used to
> this), otherwise a line with tabs converted to spaces looks exactly
> the same.

You are oversimplifying how tab widths work.  The author and reader's
tab width must match in order for displayed text to appear correctly.

Tell me what you consider the "correct" tab width for readers and I'll
find a piece of QEMU code that was authored for a different tab width
:).

In other words, it's a mess and best not to perturb it further.

Stefan