kicad-developers team mailing list archive

[Henner Zeller <h.zeller@xxxxxxx>]

On Thu, 2 May 2019 at 09:25, Seth Hillbrand <seth@xxxxxxxxxxxxx> wrote:
>
> Am 2019-05-02 12:18, schrieb Henner Zeller:
> > On Tue, 30 Apr 2019 at 20:24, Henner Zeller <h.zeller@xxxxxxx> wrote:
> >>
> >> Hi,
> >> so here one digit patch.
> >>
> >> Problem Symptom: in the schematic symbol chooser, if you search for an
> >> exact match of a component with multiple units, it is not selected.
> >> For instance, search for
> >>
> >>    74LS00
> >>
> >> The scored element is in the tree, but you need to manually unfold it
> >> (see before.png image). This usually works otherwise (I suspect it has
> >> to do with the fact that there are sub-units).
> >>
> >> The attached change will reliably select the first unit of that
> >> particular symbol and fix the problem (after.png image).
> >>
> >> Now it might be up for debate if the search should actually unfold to
> >> the first unit or if the tree unfolding should stop at the 74LS00 part
> >> - I guess if the latter is wished, something dependent on
> >> tree-children needs to be introduced. I leave that up to you.
> >>
> >> Attached: patch (against 5.1 branch),
> >
> > It also applies cleanly against current head, which has the same
> > problem.
>
>
> Hmmm... cache will be local to each core, so the we'll probably get
> speed up by dividing the image into quadrants based on the number of
> threads and having a morton sweep per quadrant.

so on my 16 core, 32 'thread' Ryzen Threadripper, I only see about
1-2% slower execution, so it is probably not a big deal even with a
lot of competing cores.

Having said that, I have some ideas for some further optimizations,
but that would go more into the guts of it (e.g. dividing memory areas
by cores etc.) that I didn't want to start it now for a (hopefully)
5.1 acceptable patch.

-h

>
> But even without, the center spiral patch is a nice touch.
>
> -Seth