cfbolz 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 | if a pep adds a mere 25-30 [C-API] functions or so, it's a drop in the ocean (cough) - Armin
lritter has joined #pypy
lritter has quit [Ping timeout: 245 seconds]
lritter has joined #pypy
jcea has quit [Ping timeout: 260 seconds]
xcm_ has joined #pypy
xcm has quit [Killed (orwell.freenode.net (Nickname regained by services))]
xcm_ is now known as xcm
trfl_ has joined #pypy
trfl has quit [Remote host closed the connection]
Taggnostr has quit [Ping timeout: 258 seconds]
graingert is now known as graingert[m]
graingert[m] is now known as graingert
forgottenone has joined #pypy
jacob22_ has quit [Ping timeout: 265 seconds]
<mattip>
tumbleweed: thanks for looking into this. Is there more info somewhere about "pip devendoring"?
<mattip>
it sounds like something that should be changed
oberstet has joined #pypy
<cfbolz>
mattip: sorry about the memfd breakage. Does it mean that our portable PyPy 3.8 can never expose memfd_create?
<mattip>
hmm
<mattip>
manylinux2010 is based on CenOS 6, which went EOL 3 months ago
<mattip>
so we could move to manylinux2014, based on CentOS 7, which will EOL in 2024, and is based on glibc 2.17
<mattip>
but we need 2.27, which is not even part of the next manylinux standard: manylinux_2_24
<cfbolz>
mattip: I wonder whether we should implement it using ctypes in app level
<cfbolz>
And check whether the symbol is there
<mattip>
is it really that important an interface?
<cfbolz>
Yeah, maybe not
<cfbolz>
Maybe enough to have it in 5 years
<mattip>
or just use conda's builds and not ours
<cfbolz>
Heh
<mattip>
.. which doesn't have it, even on CPython 3.9
<mattip>
:(
<mattip>
so if conda is happy to leave it out, I think we can too
<cfbolz>
Yeah
forgottenone has quit [Quit: Konversation terminated!]
Taggnostr has joined #pypy
<cfbolz>
mattip: OK, fixed on my laptop, will push later
<mattip>
cool
asmeurer has joined #pypy
<Dejan>
CentOS is dead, long live Rocky Linux!!
lritter has quit [Ping timeout: 245 seconds]
asmeurer has quit [Ping timeout: 246 seconds]
asmeurer has joined #pypy
forgottenone has joined #pypy
asmeurer has quit [Ping timeout: 265 seconds]
<cfbolz>
mattip: eh, thanks for paying attention
* cfbolz
is not getting enough sleep
<mattip>
it is easier to review smaller changes, I wish I could look over the bigger ones as carefully
jcea has joined #pypy
wleslie has quit [Quit: ~~~ Crash in JIT!]
commandoline_ has joined #pypy
vext01 has joined #pypy
WGH_ has joined #pypy
dustinm has joined #pypy
sknebel_ has joined #pypy
nulano_ has joined #pypy
dustinm- has quit [Ping timeout: 256 seconds]
commandoline has quit [Ping timeout: 256 seconds]
nulano has quit [Ping timeout: 256 seconds]
sknebel has quit [Ping timeout: 256 seconds]
ebarrett has quit [Ping timeout: 256 seconds]
gutworth has quit [Ping timeout: 256 seconds]
WGH has quit [Ping timeout: 256 seconds]
gutworth has joined #pypy
ammar2 has quit [Ping timeout: 256 seconds]
alcarithemad has quit [Ping timeout: 256 seconds]
nulano_ is now known as nulano
alcarithemad has joined #pypy
runciter_ has quit [Ping timeout: 264 seconds]
runciter has joined #pypy
ammar2 has joined #pypy
sknebel_ is now known as sknebel
sknebel has quit [Max SendQ exceeded]
sknebel has joined #pypy
gutworth has quit [Changing host]
gutworth has joined #pypy
<exarkun>
mattip: puny human brains capable of managing so little complexity
<cfbolz>
mattip: the tornado thing found a real JIT problem :-(
<mattip>
levels of complexity: (0) bug (1) path to solution is clear, but it needs a branch (2) arigo needed (3) masters student needed (4) worthy of a paper
<cfbolz>
1-2
<antocuni>
"level of complexity: arigo needed" :)
<Dejan>
:D
<cfbolz>
Armin is worth more than a master student though
todda7 has joined #pypy
<mattip>
:)
<mattip>
my ranking is due to my perception of difficulty of getting someone to do the work
<cfbolz>
Ah
<cfbolz>
Getting Master students to do work is not to hard. It's finding them in the first place that is the problem
asmeurer has joined #pypy
<antocuni>
it might be easier to write a paper than finding master students, then 😅
jacob22_ has joined #pypy
mattip has quit [Ping timeout: 240 seconds]
mattip has joined #pypy
<larstiq>
is the problem finding interested students, good ones, both?
<cfbolz>
interested + some prior knowledge of runtimey stuff
<larstiq>
are there events introducing the research groups to the students? I fondly remember when those started up at my faculty
<cfbolz>
larstiq: here it's more that your lectures are used to recruit students
<cfbolz>
but I don't give master-level ones atm, so that's the problem right there
<larstiq>
from what I've observed it helps to get started when they're bachelor students
<larstiq>
though the Finnish and Dutch systems I've seen were different in how much could be offered to them
<cfbolz>
sure, but I am not giving bachelor lectures either ;-)
<cfbolz>
only teaching non-majors atm
<larstiq>
ah
<larstiq>
the Dutch system seemed more rigid, having to do courses in a particular order and bachelor students not expected/encouraged to dip their feet in research
<cfbolz>
sort of similar here. the first couple of semesters are relatively rigid
danchr_ is now known as danchr
<larstiq>
cfbolz: that's a shame :/
<cfbolz>
I don't know, it makes some sense. some basic info is needed, and from the lecturer pov it's much easier if you know what you can assume the student's have already taken
<larstiq>
that part makes sense, but I prefer still allowing students to take courses they might lack the qualifications for. Ime the ones that stick work on catching up. It can be a struggle but that's instructive too
<cfbolz>
sure
<cfbolz>
that's always up to the instructor
<cfbolz>
(here, I mean)
<larstiq>
right
<larstiq>
I've also experienced the "the system will not allow you to enroll, if you're not enrolled you can't attend the course"
<larstiq>
cfbolz: are the non-majors more enthusiastic because they specifically chose this (minor?), or less because it's mandatory but not seen as core to them?
<larstiq>
and if I'm asking too many questions, I can go to bed :)
<cfbolz>
no worries ;-)
<cfbolz>
larstiq: they are very enthusiastic, because it's completely optional for them
<cfbolz>
but they don't write cs theses by definition
<larstiq>
do they have options to do smaller projects, independent study with a supervisor?
commandoline_ has quit [Changing host]
commandoline_ has joined #pypy
commandoline_ is now known as commandoline
<cfbolz>
larstiq: yeah. but they won't do compilers, usually something related to their minor
<cfbolz>
major
forgottenone has quit [Quit: Konversation terminated!]