arigato changed the topic of #pypy to: PyPy, the flexible snake (IRC logs: https://quodlibet.duckdns.org/irc/pypy/latest.log.html#irc-end ) | use cffi for calling C | mac OS and Fedora are not Windows
jcea has quit [Ping timeout: 252 seconds]
forgottenone has joined #pypy
dddddd has quit [Read error: Connection reset by peer]
forgottenone has quit [Ping timeout: 258 seconds]
bogner has quit [Remote host closed the connection]
forgottenone has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
speeder39_ has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
glyph has quit [Remote host closed the connection]
glyph has joined #pypy
themsay has joined #pypy
speeder39_ has quit [Quit: Connection closed for inactivity]
brett-soric has joined #pypy
brett-soric has left #pypy [#pypy]
micisuta has quit [Ping timeout: 245 seconds]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
micisuta has joined #pypy
micisuta1 has joined #pypy
micisuta has quit [Ping timeout: 272 seconds]
micisuta1 is now known as micisuta
micisuta has quit [Ping timeout: 248 seconds]
<arigato> mattip: we're still getting EACCES: [Permission denied]: symlink('8149', '/tmp/ffi-203/.lock'). Do you have a clue? I failed to get one
<arigato> the_drow (for the logs): this issue was left dormant for 5 years, as often the case in the cpython world
<arigato> at some point I made sure our GIL would handle various cases well
<arigato> but also cpython changed its own GIL in the meantime
<arigato> so this issue may be fixed, or the bad behaviour changed in some way
<arigato> (at least I think? maybe it changed before 2014 actually and this issue is about the new behaviour)
<mattip> arigato: on bencher4?
<mattip> the failures don't appear on the last run
<mattip> after I cleared out the /tmp directory
<arigato> (the_drow: indeed, sorry, this is a problem in the "new GIL" of cpython, and I took care of that case in pypy as far as I'm aware)
<arigato> mattip: ah, indeed, sorry got confused by the missing "." in that line
<arigato> sorry in that column
<mattip> "success" is not reported in extra tests with a ".", since it is a different pytest
themsay has quit [Ping timeout: 252 seconds]
<arigato> looking at the source of py.path, I think it's a regression---the older version of py.path.local included with pypy doesn't have that problem, but the newer version installed from pip crashes that way if there are temporary directories owned by someone else
<arigato> (or more likely, the exact condition is more complicated)
<arigato> (and then it may not be a regression for all I know)
<arigato> (the_drow: sorry, 9 years, not 5)
Graypup__ has quit [Quit: ZNC 1.6.1 - http://znc.in]
<mattip> arigato: is there an issue open about permissions?
Graypup_ has joined #pypy
<arigato> I don't think so, didn't check
<arigato> the condition is indeed more obscure, btw
<arigato> I just don't understand why they switched from "we got any OSError trying to do this symlink, so let's skip this directory" to "crash if we get some kinds of OSErrors"
themsay has joined #pypy
themsay has quit [Ping timeout: 252 seconds]
themsay has joined #pypy
micisuta has joined #pypy
<kenaan> mattip py3.6 6c4d7cc5d649 /pypy/objspace/std/: test, fix for non-byte n-d setitem
gaze___ has quit [Quit: ZNC 1.7.1 - https://znc.in]
gaze__ has joined #pypy
<kenaan> fijal arm64 caa3d6c5a7fe /rpython/jit/backend/aarch64/: guard_not_invalidated
<kenaan> fijal arm64 03ea4ff1aa71 /rpython/jit/backend/aarch64/: math sqrt
<kenaan> fijal arm64 d11024632dda /rpython/jit/backend/aarch64/: a few more obscure operations
<kenaan> fijal arm64 8003825bd555 /rpython/jit/backend/: slow progress
<kenaan> fijal arm64 58ffd4498657 /rpython/jit/backend/aarch64/assembler.py: jump with float constants
<kenaan> fijal arm64 c89b27574833 /rpython/jit/backend/aarch64/: zero_array, increment debug counter and exception handling
themsay has quit [Read error: Connection reset by peer]
themsay has joined #pypy
<kenaan> fijal arm64 a524391486ad /rpython/jit/backend/aarch64/: Small changes from test_random
<kenaan> fijal arm64 dc101f8cce32 /rpython/jit/backend/aarch64/callbuilder.py: start working on call_release_gil
dddddd has joined #pypy
tos9 has quit [Ping timeout: 252 seconds]
micisuta has quit [Quit: micisuta]
micisuta has joined #pypy
tos9 has joined #pypy
tos9 has quit [Ping timeout: 252 seconds]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
<kenaan> fijal arm64 62f3abe7cb9a /rpython/jit/backend/aarch64/: (arigo, fijal) call_release_gil first steps
tos9 has joined #pypy
Rhy0lite has joined #pypy
tsaka__ has quit [Ping timeout: 245 seconds]
<mattip> this numpy test segfaults on pypy2
<mattip> passes on pypy3
<mattip> attr = PyDict_GetItemString(iface, "data");
<mattip> where the difference between pypy2 and pypy3 is the contents of iface['data']
<mattip> on pypy2 it is buffer(arr), on pypy3 it is memoryview(arr)
<mattip> I tried to write a failing test with PyDict_GetItem and buffers, but they all passed.
<mattip> maybe the problem is specific to buffer(arr) and cpyext ?
micisuta has quit [Quit: micisuta]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
jcea has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
<ronan> mattip: I'm looking, but I'm still at the "how can it possibly fail?" stage
_whitelogger has joined #pypy
xcm has quit [Read error: Connection reset by peer]
xcm has joined #pypy
forgottenone has quit [Quit: Konversation terminated!]
<kenaan> fijal arm64 72e700c17875 /rpython/jit/backend/: a bit inefficient, but fix calling
dddddd has quit [Remote host closed the connection]
dddddd has joined #pypy
exarkun has joined #pypy
<mattip> trying a lldebug0 build
xcm has quit [Remote host closed the connection]
<ronan> mattip: actually, if you run the test directly, it shows the RPython error
xcm has joined #pypy
<exarkun> "SyntaxError: UTF-8 BOM with utf-8-with-signature-unix coding cookie" https://circleci.com/gh/tahoe-lafs/tahoe-lafs/9643 :/
themsay has quit [Read error: Connection reset by peer]
themsay has joined #pypy
<ronan> mattip: the crash is caused by buffer_attach() attempting to raise NotImplementedError
<ronan> raising app-level exceptions from attach functions is not supported
bogner has joined #pypy
<kenaan> fijal arm64 841338ea51d6 /rpython/jit/backend/aarch64/callbuilder.py: typo
<kenaan> fijal arm64 fe76a43ae509 /rpython/jit/backend/aarch64/: minor adjustments
<mattip> ok, what is raising the error?
<mattip> ahh, the 'buffer flavor not supported'?
bogner has quit [Client Quit]
bogner has joined #pypy
bogner has quit [Client Quit]
<kenaan> fijal arm64 f8ea8c51ed28 /rpython/jit/backend/aarch64/test/test_call.py: add passing test
fling has quit [Ping timeout: 248 seconds]
<mattip> once I had pyinteractive working with cpyext but it no longer does
marvin has quit [Remote host closed the connection]
marvin has joined #pypy
Rhy0lite has quit [Quit: Leaving]
<mattip> the buffer is a CBuffer, the result of slotdefs.wrap_getreadbuffer or slotdefs.wrap_getwritebuffer
bogner has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
micisuta has joined #pypy
micisuta1 has joined #pypy
micisuta has quit [Ping timeout: 272 seconds]
micisuta1 is now known as micisuta
micisuta1 has joined #pypy
xcm has quit [Remote host closed the connection]
micisuta has quit [Ping timeout: 268 seconds]
xcm has joined #pypy
micisuta1 has quit [Ping timeout: 252 seconds]
mattip has left #pypy ["Leaving"]
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
agronholm has quit [Ping timeout: 250 seconds]
agronholm has joined #pypy
_whitelogger has joined #pypy