Re: [RFC] using "optimizaiton 1"

[Eli Zaretskii]

> Date: Fri, 14 Dec 2012 20:47:28 +0400
> From: Dmitry Antipov <address@hidden>
> CC: address@hidden
> 
> On 12/14/2012 06:15 PM, Eli Zaretskii wrote:
> 
> IOW, I suspect that this code:
> 
>   13248    /* If showing the region, and mark has changed, we must redisplay
>   13249       the whole window.  The assignment to this_line_start_pos 
> prevents
>   13250       the optimization directly below this if-statement.  */
>   13251    if (((!NILP (Vtransient_mark_mode)
>   13252          && !NILP (BVAR (XBUFFER (w->buffer), mark_active)))
>   13253         != !NILP (w->region_showing))
>   13254        || (!NILP (w->region_showing)
>   13255            && !EQ (w->region_showing,
>   13256                    Fmarker_position (BVAR (XBUFFER (w->buffer), 
> mark)))))
>   13257      CHARPOS (this_line_start_pos) = 0;
> 
> is obsolete (heh, it's there since 1993). If showing the region, and
> both PT and mark are on the same line

How do you conclude that both point and mark are on the same line?
The above just tests whether mark has changed, it doesn't say anything
about point, AFAICT.

In any case, is this situation frequent enough to be worth optimizing?