20,741
edits
Philosopher (talk | contribs) mNo edit summary |
m (→August 14) |
||
| Line 64: | Line 64: | ||
: Would my explore.py script help here? You can move it to $FG_HOME (or link -s), run it, and type "list by number of calls / 20" to get the top 20 most-called functions. It's quite helpful, actually ^.^ [[User:Philosopher|—Philosopher]] ([[User talk:Philosopher|talk]]) 21:14, 14 August 2013 (UTC) | : Would my explore.py script help here? You can move it to $FG_HOME (or link -s), run it, and type "list by number of calls / 20" to get the top 20 most-called functions. It's quite helpful, actually ^.^ [[User:Philosopher|—Philosopher]] ([[User talk:Philosopher|talk]]) 21:14, 14 August 2013 (UTC) | ||
:: Tried running it, but getting syntax errors - in a bunch of places, probably due to a different python version here. Also, looking at it, it would seem like another candidate to be eventually reimplemented as a Nasal background task once performance has improved, because we'd ideally show canvas-dialogs in FG with profiling results.--[[User:Hooray|Hooray]] ([[User talk:Hooray|talk]]) 21:23, 14 August 2013 (UTC) | |||
It's confirmed: As previously suggested, your setupDebugExtras() call results in the entire sim session being profiled, see my earlier comments - ideally, just passing extras to call() should MERELY profile the specified naFunc, nothing else - according to the profile, I am seeing the entire sim session profiled during the run, not just the test func - so that's where all the data and refs are coming from. | It's confirmed: As previously suggested, your setupDebugExtras() call results in the entire sim session being profiled, see my earlier comments - ideally, just passing extras to call() should MERELY profile the specified naFunc, nothing else - according to the profile, I am seeing the entire sim session profiled during the run, not just the test func - so that's where all the data and refs are coming from. | ||