Under current CVS, release build, running regrtest.py crashes very soon after entering test___all__.py for me, on two different machines (but both Windows). The C stack has gotten lost by this point, and the program counter is pointing into static data(!), about a dozen bytes beyond the start of Python's static PyFloat_Type type. Alas, there is no problem in a debug build. There's also no problem under the release build if I run the debug build first and leave the .pyc files behind. Removing the .pyc files and then running the release build dies every time. So maybe it's something to do with compiling Python programs, or maybe with a vagary of when cyclic gc triggers. The latter is high on my suspect list, because the location of the death is affected by regrtest's -t option, and the release build runs the tests to completion with -t0. If it's in gc, I probably caused it. So I'm not asking you to fix it <wink>. It would help to know if anyone is having problems under Linux, and especially if you are and the debugger there is more helpful in a release build.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4