On Friday, May 03 2019, Santiago Vila wrote: > Awesome! Thanks a lot, Sergio!
No problem. > Just as a clarification: The test machine I offered to reproduce this > problem is not particularly slow. It's a n1-standard-1 instance from > Google Compute Engine, i.e. this is not me trying to build Debian on > a very old computer, but instead something that could happen to anybody. Hm, that's interesting, then. I mean, I confess I hadn't checked the machine specs in detail. But knowing that it's not a slow machine really makes things a bit more complicated. I wonder why this is happening... So, I spent more time investigating this, and I found that, when executing the testsuite, if you run Emacs with "--no-site-lisp", the problem doesn't manifest. Upon more investigation, I noticed that the problem happens after you load a certain number of Lisp files into Emacs. This makes little sense to me. I tried replicating the same Emacs environment on my machine, but the bug still doesn't happen here. I'm out of ideas (and energy!) now; maybe I'll continue tomorrow. In any case, the patch I provided is enough to stabilize the testsuite and make it pass again. I'll let you know if I find anything else. Thanks, -- Sergio GPG key ID: 237A 54B1 0287 28BF 00EF 31F4 D0EB 7628 65FC 5E36 Please send encrypted e-mail if possible http://sergiodj.net/
signature.asc
Description: PGP signature