00:02 <bbot2> Failure: http://buildbot.pypy.org/builders/own-win-x86-64/builds/373 [default]

00:27 <bbot2> Failure: http://buildbot.pypy.org/builders/pypy-c-jit-linux-aarch64/builds/1064 [default]

01:31 <bbot2> Success: http://buildbot.pypy.org/builders/own-linux-aarch64/builds/969 [default]

11:49 <arigato> smarr: in normal compilation, the bound check is performed by a macro which compiles to nothing at all---i.e. the bound check does not exist

16:37 <muke> is anyone able to share how to use gdb on running test cases with pytest?

16:42 <ronan> muke: gdb -args python -m pytest ...

16:45 <muke> oh it's the straight forward? sorry for some reason I thought there was something special involved

21:30 <smarr> arigato: hm this is the generated code I am seeing: https://gist.github.com/smarr/9a628e44e949b2f2149bd360425523ec

21:33 <smarr> I annotated it a little. The initial comment is the RPython code. If I am not missreading things, I would hope that line 23 to 50 are unnecessary (there's a little bit of address and possibly GC stuff in there, which I don't quite follow, but most of it seems like it would ideally not be needed and simply checks that my stack pointer is in the stack.

21:34 <smarr> arigato: I am just thinking wishfully here, though, I would be curious whether there's anything in RPython that I could use to perhaps get closer to the "unsafe" version