<phlebas>
I'm getting compilation errors in the 3.6-v7.1.0 release sources in libbacktrace of rvmprof. I already disabled the vmprof module, but it's still trying to build these. What else do I need to disable?
Rhy0lite has joined #pypy
adamholmberg has joined #pypy
<phlebas>
ah, _continuations is the (to me) non-obvious one...
marky1991 has joined #pypy
Zaab1t has joined #pypy
marky1991 has quit [Ping timeout: 246 seconds]
<arigato>
phlebas: if you tried to disable rvmprof using a command-line option to "rpython" that is supposed to do that, and it didn't work, then please report a bug
<phlebas>
I disabled the _vmprof module and assumed that this wouldn't pull in rvmprof's backtrace sources. But the _continuation and faulthandler modules also pull these in - I just didn't know
<arigato>
ah ok
<arigato>
ideally there should be a general option to disable the rpython rvmprof
<mattip>
we should only trigger py3.6 builds if there are changes, like on default and rpython
<mattip>
looking at the aarch64 aws machines, it costs ~$3 to translate pypy
<arigato>
I vote "no"
demonimin has joined #pypy
demonimin has quit [Ping timeout: 252 seconds]
demonimin has joined #pypy
demonimin has quit [Ping timeout: 245 seconds]
darkman6_ has quit [Remote host closed the connection]
<mattip>
arigato: you want the redundant builds?
<arigato>
I think we should only look at expensive solutions if we can't use one of the machines that fijal has got for the arm64 contract
<arigato>
and if we have to pay expensive solutions I'd look if it makes sense to bill them to them
<mattip>
ahh, I thought you were responding to only building py3.6 when there are changes
<arigato>
ah sorry
<arigato>
no, if we can avoid redudant builds it's of course a good idea
<mattip>
the aws machine I have is one from someone at amazon, they asked what we need and gave me some credits to try it out
<mattip>
hopefully it will lead to them putting the cost on a budget line that gets forgotten about as a rounding error
Garen has quit [Read error: Connection reset by peer]
Garen has joined #pypy
darkman66 has joined #pypy
darkman6_ has joined #pypy
darkman66 has quit [Ping timeout: 255 seconds]
darkman6_ has quit [Remote host closed the connection]
petronny has quit [Read error: Connection reset by peer]
petronny has joined #pypy
JStoker has quit [Ping timeout: 250 seconds]
JStoker has joined #pypy
Zaab1t has quit [Quit: bye bye friends]
darkman66 has joined #pypy
marky1991 has quit [Ping timeout: 240 seconds]
darkman66 has quit [Remote host closed the connection]
darkman66 has joined #pypy
darkman66 has quit [Remote host closed the connection]
adamholm_ has joined #pypy
Rhy0lite has quit [Quit: Leaving]
adamholmberg has quit [Ping timeout: 252 seconds]
darkman66 has joined #pypy
darkman66 has quit [Remote host closed the connection]
darkman66 has joined #pypy
darkman66 has quit [Ping timeout: 250 seconds]
marky1991 has joined #pypy
darkman66 has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
xcm has quit [Remote host closed the connection]
xcm has joined #pypy
darkman66 has quit [Remote host closed the connection]
darkman66 has joined #pypy
tsaka__ has quit [Ping timeout: 246 seconds]
tsaka__ has joined #pypy
<RemoteFox>
is there a way how I can tell rpython to .. generate class with properties with different types without actually having to create duplicit code?
<RemoteFox>
I have a container class used to hold one data type and I would like to use it for others as well
<RemoteFox>
in java, I would use <template>
<RemoteFox>
I think maybe "specialization" is the right word?
<simpson>
RemoteFox: You can use standard Python code generation; RPython imports your application as standard Python 2.7 first, running every module, and then traces the generated bytecode.
<RemoteFox>
what do you mean by "standard python code generation"?
<RemoteFox>
I've tried subclassing, but that clashes with shared methods