The good news is that I believe that I’ve isolated the commit that introduced this issue. The bad news is that it seems to have been a pretty big one.
Specifically, in the “main” branch it appears to have been introduced in the second commit of November 27 of last year (2017). I think that the relevant link is this.
I found this by using the "git reset --hard " command to hop from one commit to another, searching for the last one that worked and the first that didn’t. I chose the “main” branch, as I recall, because I remembered having previously used one of the builds on the SDK download page (which I presume draw from the “main” branch) without encountering this issue. (Indeed, that memory was an inspiration for this experiment, I believe.)
Hmm… that is helpful, thanks. That commit changed the container used for storing tags, so I can imagine that something there might be causing a crash in some conditions.
However, I do notice that test.egg has a transform element. I remember some people reporting odd bugs on Linux with Eigen that only popped up when the .egg file has a matrix at the top. I build without Eigen myself (to get a faster build), so I wonder if that might be it. Are you building with Eigen enabled?
As to Eigen, my “makepanda” command includes “–no-eigen” (after “–everything”), so as long as it’s respecting that, it should be building without Eigen.
Speaking of which, a minor feature request (presuming that it’s not already there): a parameter (or pair of parameters) to “makepanda” that allows me to build without any of the “Maya” and “Max” plugins, without specifying each individually, and thus allowing a slightly less cluttered command line. I gather that it will automatically ignore them if they’re not present on the system, but I do like to have them explicitly excluded. ^^;
Nevertheless, I decided to nevertheless test for the crash without the transform. I deleted the transform section from the egg file, leaving it as simply this:
Doing so seems to have no effect: the program crashes in a build in which it crashed before, and runs in a build in which it ran before, it would seem. (The latter being the commit just before the one that I identified above, I believe.)
OK, this is a stab in the dark, but I identified a potential problem with the container type now used to store the tags; could you pull the latest master version (or cherry-pick 7ea16fd) and let me know whether that seems to fix it?
Ah, that seems to have done it! Wonderful, and thank you very much!
(I presume then that you’re a rogue: skilled at stabbing in the dark. ;P)
If I have further issues in 1.10, I’ll likely start new threads, so–unless you have more that you want to do in this matter–I think that we can consider this thread closed.
