17:28 <khem> infact we have too much compression if we let lose these algos each invocation wants to use every last resource on the machine

17:28 <khem> you have to bridle them

17:28 <khem> what would be interesting is to find which one suits us best

17:30 <khem> and the optimum parallelism

21:04 <mbulut> when force-refreshing a patch (which is in a 3rd party layer) because of fuzz using devtool, is there a way to place the updated patch in my own layer's append recipe such that the original patch is not applied?

21:04 <mbulut> or do I have to update the patch in the original layer?

22:58 <JaMa> mbulut: read about FILESPATH

23:19 <mbulut> JaMa, the way I understand FILESPATH and FILESEXTRAPATHS is that when prepending to the latter in custom layer (and supplying some file to be overridden), the upstream layer's recipe will pick up my version of the file because it matches first (due to prepend)

23:20 <mbulut> This always worked fine for files used in the upstream recipes, however when I tried to do the same to a patch that I wanted to have overridden, the build log still showed the patch fuzz

23:21 <mbulut> that's why I was unsure if the same rules apply for do_patch

23:21 <mbulut> It looked like it was double applying the patch (upstream first, then mine) which is not what I want, of course

23:23 <mbulut> I worked around the problem by now though...