>>> And in any case, the patch for gui_init_fringe is sub-optimal: it 
>>> unnecessarily loops over the standard bitmaps that were superseded. It 
>>> is better to leave the first loop go over the standard bitmaps, 
>>> whether superseded or not, and the second loop go over non-standard 
>>> bitmaps only.
>>
>> It is not sub-optimal, rif->define_fringe_bitmap will only be called 
>> for those standard bitmaps that were superseded.  For the 
>> non-superseded ones, fringe_bitmaps[bt] is NULL.
>
> I meant the first loop, not the second one.
>

Okay, the patch was meant only to fix the bug, but indeed the small 
optimization you mention makes sense.  Updated patch attached.