cfbolz changed the topic of #pypy to: #pypy PyPy, the flexible snake | IRC logs: and | the pypy angle is to shrug and copy the implementation of CPython as closely as possible, and staying out of design decisions
derpydoo has joined #pypy
jcea has quit [Ping timeout: 256 seconds]
<fijal> I have renewed certificate for buildbot
<cfbolz> Hodgestar: yeah, but I think it's fixed on cpy main
derpydoo has quit [Ping timeout: 255 seconds]
komasa has quit [Read error: Connection reset by peer]
glyph has quit [Quit: End of line.]
glyph has joined #pypy
<mgorny> i have another bug where pypy3.10 is "eating" warnings from matplotlib but i don't have the energy to reduce it :-(
<mgorny> it seems to have something to do with depth of function wrapping
<mgorny> on top of everything, it's a heisenbug :-(
<mgorny> ok, i'm moving forward with dumb reduction
<mgorny> and it seems to happen less often as i remove more code
<cfbolz> mgorny: ouch
<cfbolz> mgorny: but it's independent of the JIT?
<mgorny> i don't know
<mgorny> my best guess right now is that with large enough amount of code in between warnings.simplefilter("error") and the warning, it gets dropped
<mgorny> i wonder if GC may be eating it
<mgorny> cfbolz: happens with PYPY_DISABLE_JIT=1 too
<mgorny> (though becomes even rarer)
<cfbolz> ok
<cfbolz> sounds annoying :-(
<mgorny> yeah, i feel like reduction is not going to help here
<mgorny> cfbolz: adding gc.disable() seems to make the problem disappear
<mgorny> i guess i'll file a bug with that i know so far
<cfbolz> mgorny: and the warnings appear correctly when you do that?
<mgorny> yes
<mgorny> hmm, lemme try something dumb
<mgorny> i suppose pytest is part of the equation
<mgorny> hmm, but no, warnings.simplefilter("error") should take precedence over pytest warning filters
jcea has joined #pypy
<Hodgestar> cfbolz: Woot, glad it's fixed. I wonder if it would be helpful to offer Armin's example as a unit test for CPython somewhere.
<cfbolz> Hodgestar: I mean if you wanted to report it as a bug for 3.11, that's certainly an option
<mgorny> that's the best i could come up with
jinsun_ has joined #pypy
jinsun is now known as Guest5647
jinsun_ is now known as jinsun
Guest5647 has quit [Ping timeout: 276 seconds]
ruth2345345 has quit [Ping timeout: 255 seconds]
derpydoo has joined #pypy
ruth2345345 has joined #pypy
ruth2345345 has quit [Ping timeout: 255 seconds]
Julian has joined #pypy
Julian has quit [Quit: leaving]
jcea has quit [Ping timeout: 276 seconds]
ruth2345345 has joined #pypy
Julian has joined #pypy
jcea has joined #pypy
Julian has quit [Quit: leaving]