cfbolz changed the topic of #pypy to: PyPy, the flexible snake (IRC logs: https://botbot.me/freenode/pypy/ ) | use cffi for calling C | the secret reason for us trying to get PyPy users: to test the JIT well enough that we're somewhat confident about it
marky1991 has joined #pypy
nimaje has joined #pypy
marky1991 has quit [Read error: Connection reset by peer]
xorAxAx has quit [Remote host closed the connection]
xorAxAx has joined #pypy
xorAxAx has quit [Changing host]
xorAxAx has joined #pypy
dddddd has quit [Remote host closed the connection]
lritter has quit [Ping timeout: 252 seconds]
jcea has quit [Quit: jcea]
Blendify_i5 has joined #pypy
Blendify_i5 has quit [Remote host closed the connection]
AlwaysHigh25 has joined #pypy
AlwaysHigh25 has quit [Read error: Connection reset by peer]
hexa- has quit [Ping timeout: 252 seconds]
nimaje has quit [Remote host closed the connection]
nimaje has joined #pypy
hexa- has joined #pypy
nimaje has quit [Ping timeout: 252 seconds]
nimaje has joined #pypy
GTAXL24 has joined #pypy
GTAXL24 has quit [Ping timeout: 252 seconds]
redj_ has joined #pypy
redj has quit [Ping timeout: 268 seconds]
redj_ has left #pypy [#pypy]
redj has joined #pypy
jamesaxl has joined #pypy
forgottenone has joined #pypy
oberstet has joined #pypy
mattip has quit [Ping timeout: 246 seconds]
mattip has joined #pypy
oberstet has quit [Remote host closed the connection]
oberstet has joined #pypy
kg11 has joined #pypy
mattip has quit [Ping timeout: 245 seconds]
kg11 has quit [Remote host closed the connection]
mattip has joined #pypy
mattip has quit [Ping timeout: 252 seconds]
arigato has joined #pypy
Hink has joined #pypy
Hink has quit [Ping timeout: 245 seconds]
forgottenone has quit [Ping timeout: 244 seconds]
ronan has joined #pypy
cjwelborn has quit [Remote host closed the connection]
mattip has joined #pypy
dddddd has joined #pypy
cjwelborn has joined #pypy
nimaje has quit [Ping timeout: 252 seconds]
forgottenone has joined #pypy
nimaje has joined #pypy
WhitePhosphorus0 has joined #pypy
WhitePhosphorus0 has quit [Ping timeout: 246 seconds]
antocuni has joined #pypy
nopf has joined #pypy
arigato has quit [Quit: Leaving]
bendlas[m] has quit [Remote host closed the connection]
antocuni has quit [Ping timeout: 245 seconds]
bendlas[m] has joined #pypy
because has joined #pypy
because has quit [Remote host closed the connection]
jamesaxl has quit [Ping timeout: 245 seconds]
jamesaxl has joined #pypy
Hasimir has joined #pypy
TheAdversary has joined #pypy
Hasimir has quit [Read error: Connection reset by peer]
TheAdversary has quit [Remote host closed the connection]
sn0wmonster19 has quit [Read error: Connection reset by peer]
forgottenone has joined #pypy
<cfbolz>
_aegis_: what is -funroll-loops?
<_aegis_>
pypy -funroll-loops
bendlas[m] has quit [Write error: Connection reset by peer]
mattip has joined #pypy
bendlas[m] has joined #pypy
<cfbolz>
_aegis_: the only thing that does is: print("Vroom vroom, I'm a racecar!")
<_aegis_>
yes, but it's broken in pypy3 :P
<_aegis_>
ok, so why is there so much of what looks like dead code in unicodedb?
<_aegis_>
like megabytes
<_aegis_>
# the following dictionary is used by modules that take this as a base
<_aegis_>
_orig_names = {
<_aegis_>
but no references to _orig_names anywhere in pypy
<_aegis_>
it looks like the lookups are done with a trie instead of the orig dict
<cfbolz>
_aegis_: that is indeed a bit puzzling. note however that the final executable will not contain that dict, if it is not referenced from any code
<_aegis_>
(I'm looking into why libpypy-c is so huge, over half is the data section!)
<_aegis_>
it does
<cfbolz>
then it's used
<_aegis_>
it's not used by any code I can find
<cfbolz>
ok, but the translator never puts anything into the binary that's not referenced
<_aegis_>
is the `usession-py3.5-14/testing_1` tmpdir all the C code?
<cfbolz>
should
<_aegis_>
so 'LEFT CURLY BRACKET' is not in any files in that folder
<_aegis_>
but still ends up in the data section along with the rest of the unicode bloat dict
<cfbolz>
_aegis_: strings are encoded strangely in the C files sometimes
<cfbolz>
like as the charcodes maybe?
<_aegis_>
oh fair
<_aegis_>
I need to compile with debug symbols so I can trace references back to the source easily
<_aegis_>
not finding the name of the module or anything
<cfbolz>
_aegis_: one hack would be to rename _orig_names and see what breaks :-)
<_aegis_>
gonna generate with lldebug anyway, I'll just delete it
<_aegis_>
data_rpython_rlib_unicodedata.c doesn't seem to have the huge dict
<cfbolz>
_aegis_: anyway, if you can find something huge and superfluous, that would be amazing ;-)
<cfbolz>
so cool that you're looking into t
<cfbolz>
it
<_aegis_>
I've been dying over here making my app dmg 22mb
<_aegis_>
like I have a ton of optimizations in place already, I do a lot to the stdlib
<_aegis_>
I'm in here 'cause I'm trying to get stdlib zipimport working, but I happened to look at .data and I can't unsee
<_aegis_>
(here = the bloat weeds)
<cfbolz>
_aegis_: note that it's not quite like CPython. we have a lot in the main executable that CPython has in shared libs
<_aegis_>
yes I know
<_aegis_>
oh, I might be thrown off because the data is in the python file twice
<_aegis_>
there's both _stringtable and _orig_names
<cfbolz>
_aegis_: but indeed, when I rename _orig_names at least the unicodedata tests still pass
<_aegis_>
oh, I think the big blob I'm seeing is from unicode 8.0.0
<_aegis_>
not from 5.2.0
<_aegis_>
8.0.0 doesn't use the trie, so it seems to have a much less efficient encoding
<cfbolz>
_aegis_: 8.0.0 stores only those chars that are different from 5.2.0
<cfbolz>
not all of them
<_aegis_>
yes, but it does that using a dictionary instead of a strtab+trie
<cfbolz>
yes
<_aegis_>
which takes up 800kb.
siniStar has joined #pypy
siniStar has quit [Remote host closed the connection]
jcea has quit [Ping timeout: 260 seconds]
jcea has joined #pypy
<_aegis_>
how mutable are the emitted rpython C objects that end up in .data?