Ricardo Wurmus <rek...@elephly.net> skribis: > These patches look great to me and they work. > My real-world test case is now down to about 12 seconds.
Good! Fixed the typo you mentioned on IRC and pushed as e778910bdfc68c60a5be59aac93049d32feae904. To summarize, the INFERIOR=y case still takes about twice as long as the INFERIOR=n case (before that it was actually 9 times slower). I suppose this is mostly due to the round trips between the inferior and the parent (one per package). We’d have to analyze more closely, for example with ‘perf timechart’, where the wait times are. It’ll always be slower than without an inferior though. Last, we should improve the baseline (4s here for the manifest you gave). That’s the tricky part. Thanks, Ludo’.
