#pypy IRC log for Thursday, 2012-01-26

htoothrot (~mux@unaffiliated/htoothrot) joined #pypy.00:00
[Arfrever] (~Arfrever@apache/committer/Arfrever) joined #pypy.00:01
nickik (~nick@217.150.233.216) left irc: Quit: Verlassend00:01
nickik (~nick@217.150.233.216) joined #pypy.00:01
amaury (~amaury_@74.125.122.49) left irc: Ping timeout: 240 seconds00:01
amaury (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.00:05
Arfrever (~Arfrever@apache/committer/Arfrever) joined #pypy.00:06
amaury (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 255 seconds00:10
[Arfrever] (~Arfrever@apache/committer/Arfrever) left irc: Quit: leaving00:14
[Arfrever] (~Arfrever@apache/committer/Arfrever) joined #pypy.00:16
dmalcolm (~david@18.111.29.28) joined #pypy.00:18
nigelb_ (~nigel@nigelb.me) left irc: Changing host00:22
nigelb_ (~nigel@ubuntu/member/nigelb) joined #pypy.00:22
Nick change: nigelb_ -> nigelb00:22
dmalcolm (~david@18.111.29.28) left irc: Read error: Operation timed out00:26
nedbat (nedbat@python/psf/nedbat) joined #pypy.00:28
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: 00:30
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.00:37
rguillebert (~rguillebe@2a01:e34:eea7:c690:21f:c6ff:fe12:4dee) left irc: Quit: Quitte00:44
rguillebert (~rguillebe@2a01:e34:eea7:c690:21f:c6ff:fe12:4dee) joined #pypy.00:44
nickik (~nick@217.150.233.216) left irc: Quit: Verlassend00:47
bbot24Failure: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/103201:02
bbot24Failure: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/78701:03
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Read error: Connection reset by peer01:03
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.01:10
whyking (~quassel@ip68-14-13-208.ri.ri.cox.net) joined #pypy.01:11
Turicas (~alvaro@mail.justen.eng.br) left irc: Quit: Leaving01:14
Tobu (~Tobu@unaffiliated/tobu) left irc: Remote host closed the connection01:20
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.01:28
etrepum (~bob@75-101-96-144.dsl.static.sonic.net) joined #pypy.01:28
Tobu (~Tobu@unaffiliated/tobu) left irc: Remote host closed the connection01:29
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 240 seconds01:31
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.01:32
Tobu (~Tobu@unaffiliated/tobu) left irc: Remote host closed the connection01:34
nettok (~quassel@190.148.246.190) joined #pypy.01:34
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Quit: pnguyen01:37
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.01:43
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.01:46
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Read error: Connection reset by peer01:59
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/194002:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/35002:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/69002:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/34002:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/68702:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/121202:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/87202:00
bbot2Started: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/153102:00
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/80102:00
nedbat (nedbat@python/psf/nedbat) left irc: Ping timeout: 276 seconds02:04
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.02:10
__main__ (~main@adsl-99-173-15-158.dsl.pltn13.sbcglobal.net) left irc: Ping timeout: 252 seconds02:15
voidspace (~voidspace@87-194-212-65.bethere.co.uk) left irc: Quit: Goodbye...02:31
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/4402:33
etrepum (~bob@75-101-96-144.dsl.static.sonic.net) left irc: Quit: etrepum02:37
RhyoliteOkay, so Pyrolog interpreter translates and runs the examples in the doc02:44
bbot23Success: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/21602:49
nettok (~quassel@190.148.246.190) left irc: Remote host closed the connection02:50
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection02:50
Shanita (~John@osbk-4db15e46.pool.mediaWays.net) left irc: Ping timeout: 240 seconds03:03
Shanita (~John@osbk-4db141e3.pool.mediaWays.net) joined #pypy.03:03
ericflo (~ericflo@75.103.8.110) left irc: Quit: ericflo03:08
_pyre_ (~stian@137.241.189.109.customer.cdi.no) joined #pypy.03:17
_pyre_Pypy got a problem03:17
_pyre_bytes(rsa.core.decrypt_int(int(stream.encode("hex"), 16), D, N)) hangs03:18
_pyre_https://bitbucket.org/sybren/python-rsa/src/5d834ee3e7e5/rsa/core.py (rsa.core module)03:18
_pyre_basicly just a pow(cyphertext, dkey, n)03:19
timotimohuge pow?03:19
timotimoi believe pypys handling of big numbers isn't efficient at all03:19
_pyre_Well, it's a 1024 RSA03:19
_pyre_but CPython have some magic when you supply the last parameter03:20
_pyre_c ** d % n will hang even CPython03:21
_pyre_what is suspicious is that pypy use 0% cpu03:23
_pyre_seems like its stuck doing nothing03:24
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-win-x86-32/builds/34003:25
timotimooh, ok03:25
timotimothat's not so good, no03:26
Telugodu (~rambaho@202.3.77.231) left irc: Read error: Connection reset by peer03:28
_pyre_perhaps the existance of a huge number is somehow broken (tested using latest nightly doh)03:29
nettok (~quassel@190.148.246.190) joined #pypy.03:31
_pyre_maybe not, tried running: pow(0xFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF\03:31
_pyre_FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF, 0xFFFFFFFFFFFFFFFFFFF\03:31
_pyre_FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF\03:31
_pyre_FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF, 0xFFFFFFFFFFFFF) in pypy shell, it gave the result instantly03:31
DasIch (~DasIch@p4FFDFC57.dip.t-dialin.net) left irc: Quit: DasIch03:34
timotimohehe03:35
timotimothat's pretty strange03:35
timotimohave you tried stepping through it in a debugger? or is thin in rpython code?03:35
kvda (~kvda@124-169-134-221.dyn.iinet.net.au) joined #pypy.03:44
mat^2 (~mathias@212.130.113.35) left irc: 03:46
_pyre_wierd, inlining the pow call directly seemed to solve it, bad python-rsa module i guess03:50
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/194003:52
bbot2Started: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1941 [py3k]03:52
_pyre_Now it's the mystery why inlineCallbacks (yield a generator) doesn't work in Pypy, broken twisted? :p03:52
_pyre_naa, broken MySQLdb. Guess I can live with it :)03:55
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) left irc: Ping timeout: 245 seconds03:55
csenger` (~user@dslb-084-058-144-070.pools.arcor-ip.net) joined #pypy.03:57
intchanter (~intchante@c-76-27-100-38.hsd1.ut.comcast.net) joined #pypy.03:57
csenger (~user@dslb-084-058-185-001.pools.arcor-ip.net) left irc: Ping timeout: 255 seconds04:01
csenger` (~user@dslb-084-058-144-070.pools.arcor-ip.net) left irc: Quit: ERC Version 5.3 (IRC client for Emacs)04:11
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Quit: pnguyen04:51
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-32/builds/153104:54
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-Ojit-no-jit-linux-x86-32/builds/87204:56
_pyre_ (~stian@137.241.189.109.customer.cdi.no) left irc: Quit: Leaving.04:58
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.04:58
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-app-level-linux-x86-64/builds/68705:01
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Quit: pnguyen05:09
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-32/builds/121205:15
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.05:16
bbot23Success: 15http://buildbot.pypy.org/builders/pypy-c-jit-linux-x86-64/builds/69005:29
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) left irc: Ping timeout: 245 seconds05:35
anish_ (~anish@123.201.34.16) joined #pypy.05:37
AlekSi (~AlekSi@85.235.191.82) joined #pypy.05:38
synesp (~Frank@c-50-137-68-161.hsd1.ct.comcast.net) joined #pypy.05:40
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: Read error: Connection reset by peer05:42
synesp (Frank@c-50-137-68-161.hsd1.ct.comcast.net) left #pypy.05:42
nettok (~quassel@190.148.246.190) left irc: Remote host closed the connection05:45
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.05:45
nirbheek (~nirbheek@gentoo/developer/flyingspaghettimonster/nirbheek) joined #pypy.05:50
AlekSi (~AlekSi@85.235.191.82) left irc: Remote host closed the connection05:54
myf (myf@li245-175.members.linode.com) left #pypy.05:58
AlekSi (~AlekSi@85.235.191.82) joined #pypy.05:59
DasIch (~DasIch@p4FFDFED3.dip.t-dialin.net) joined #pypy.06:03
DasIch (~DasIch@p4FFDFED3.dip.t-dialin.net) left irc: Quit: DasIch06:17
amaury (~amaury_@46-127-23-192.dynamic.hispeed.ch) joined #pypy.06:25
AlekSi (~AlekSi@85.235.191.82) left irc: Remote host closed the connection06:26
AlekSi (~AlekSi@85.235.191.82) joined #pypy.06:27
bbot23Success: 15http://buildbot.pypy.org/builders/own-linux-x86-64/builds/80106:30
CIA-603papercrane 07roundup * 10#1020/segfault when running test_greenlet.py through pypy pytest.py: (log message trimmed)06:42
CIA-6On both OSX and linux:06:42
CIA-6$ PYTHONPATH=. pypy/translator/goal/pypy-c.min pytest.py06:42
CIA-6pypy/module/test_lib_pypy/test_greenlet.py06:42
CIA-6..ends with.. * 14https://bugs.pypy.org/issue102006:42
amaury (~amaury_@46-127-23-192.dynamic.hispeed.ch) left irc: Ping timeout: 276 seconds06:54
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.07:01
asmeurer__ (~asmeurer@dhcp-altamirano-166.resnet.nmt.edu) left irc: Quit: asmeurer__07:10
stefanop (~stefanop@77.72.196.122) joined #pypy.07:12
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 240 seconds07:21
bbot24Failure: 15http://buildbot.pypy.org/builders/pypy-c-jit-macosx-x86-64/builds/35007:22
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.07:22
JaRoel|4d (~jaroel|4d@sink.jaroel.nl) left irc: Remote host closed the connection07:28
bbot24Failure: 15http://buildbot.pypy.org/builders/own-linux-x86-32/builds/1941 [py3k]07:33
Tobu (~Tobu@unaffiliated/tobu) left irc: Ping timeout: 272 seconds07:40
pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) left irc: Remote host closed the connection08:03
anish_ (~anish@123.201.34.16) left irc: Ping timeout: 252 seconds08:10
Telugodu (~rambaho@202.3.77.231) joined #pypy.08:14
Telugodu (~rambaho@202.3.77.231) left irc: Remote host closed the connection08:14
Telugodu (~rambaho@202.3.77.231) joined #pypy.08:19
pedronis (~pedronis@73-53.195-178.cust.bluewin.ch) joined #pypy.08:22
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.08:27
antocuni (~antocuni@host155-125-dynamic.25-79-r.retail.telecomitalia.it) joined #pypy.08:37
stefanop (~stefanop@77.72.196.122) left irc: Read error: Connection reset by peer08:42
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: Ping timeout: 252 seconds08:46
ericflo (~ericflo@c-50-131-58-97.hsd1.ca.comcast.net) joined #pypy.08:59
kenaan12arigo default 117b1eec04c0f9 15/pypy/: Add some reprs to Symbolic subclasses, and show them in the error messages.09:00
kenaan12arigo default 1166a54c1c81de 15/pypy/rlib/objectmodel.py: Fix.09:00
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.09:14
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Quit: pnguyen09:15
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) joined #pypy.09:19
fijalRhyolite: ping?09:22
kvdaHey fijal09:33
kvdaWas that you that added me to the benchmarks repo? 09:34
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection09:40
pjdelportis anyone else getting "pypy: error while loading shared libraries: libtinfo.so.5: wrong ELF class: ELFCLASS64" with the latest Ubuntu PPA build of PyPy?09:41
tumbleweedpjdelport: the PPA only builds 32bit09:44
tumbleweedit's wrong09:44
pjdelporthmm09:44
tumbleweedwe have better builds in the main archive for precise09:44
bivab (~david@ip-62-143-59-5.unitymediagroup.de) joined #pypy.09:44
pjdelporttumbleweed: i'm not sure exactly when this stopped working, but previous builds from the PPA definitely worked here09:44
tumbleweed(and barry's python PPA)09:44
kenaan12bivab arm-backend-2 111aea54f9875c 15/pypy/jit/backend/arm/: remove some XXX and update some comments09:45
pjdelportmaybe i had a 32-bit libtinfo5, but don't anymore09:45
kenaan12bivab arm-backend-2 11b2886fd16756 15/pypy/jit/backend/arm/codebuilder.py: implement the BX operation for jumps09:45
kenaan12bivab arm-backend-2 1125e1511591e1 15/pypy/jit/backend/x86/test/test_runner.py: backout b2fff2c474ff09:45
kenaan12bivab arm-backend-2 11e61421452c3f 15/pypy/jit/backend/test/runner_test.py: Backout 839659291f0309:45
kenaan12bivab arm-backend-2 11fb43a79417b7 15/pypy/jit/backend/: refactor test_compile_asmlen a bit to make it work the ARM backend09:45
kenaan12bivab arm-backend-2 11e13d8ab613df 15/pypy/jit/backend/arm/: Add methods and attributes required for test_compile_asmlen in the ARM test clases09:45
kenaan12bivab arm-backend-2 112d0405522a48 15/pypy/jit/backend/arm/assembler.py: move code a bit around to make things clearer09:45
kenaan12bivab arm-backend-2 11bdd2be87eb46 15/pypy/jit/backend/arm/opassembler.py: refactor call_assembler to use conditional_execution and the B instruction for the relative jumps09:45
kenaan12bivab arm-backend-2 11375bd2545643 15/pypy/jit/backend/arm/helper/assembler.py: kill unused code09:45
kenaan12bivab arm-backend-2 119c12b4103f60 15/pypy/jit/backend/arm/opassembler.py: add spaces09:45
kenaan12bivab arm-backend-2 115e16783d7519 15/pypy/jit/backend/arm/opassembler.py: replace a conditional jump with conditional execution09:45
kenaan12bivab arm-backend-2 11fc0c1822721d 15/pypy/jit/backend/arm/codebuilder.py: remove redundant saves09:45
kenaan12bivab arm-backend-2 11db27ab55d51b 15/pypy/jit/backend/arm/: add more checks for floating point support09:45
kenaan12bivab arm-backend-2 1123ebba29e2fc 15/pypy/jit/backend/arm/opassembler.py: remove unused import09:45
kenaan12bivab arm-backend-2 111f07ea8076c9 15/pypy/jit/backend/arm/helper/regalloc.py: add asserts09:45
tumbleweedpjdelport: probably09:47
zain (~textual@bzq-79-182-196-50.red.bezeqint.net) left irc: Quit: Textual IRC Client: http://www.textualapp.com/09:49
zain (~textual@bzq-79-182-196-50.red.bezeqint.net) joined #pypy.09:52
cwillu (~cwillu@cwillu.com) left irc: Ping timeout: 240 seconds09:56
G2P (~G2P@fw-asn1.ornis.com) joined #pypy.09:58
Tobu (~Tobu@unaffiliated/tobu) left irc: Ping timeout: 248 seconds09:59
apoirier (~apr@210.168.3.109.rev.sfr.net) joined #pypy.10:00
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.10:07
antocunigood, own tests in the py3k branch are run nightly now :-)10:11
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) joined #pypy.10:20
cwillu (~cwillu@cwillu.com) joined #pypy.10:20
cfbolzhello10:20
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) left irc: Ping timeout: 240 seconds10:32
antocuniuhm10:38
antocuniif I have a PyCode object at interp-level10:39
antocuniis there any easy way to disassemble it?10:39
AlekSi (~AlekSi@85.235.191.82) left irc: Read error: Connection reset by peer10:43
AlekSi (~AlekSi@85.235.191.82) joined #pypy.10:44
antocunipffffff10:44
squiddy (~squiddy@f053087106.adsl.alicedsl.de) joined #pypy.10:44
antocuniI have been spent half a day debugging this unicode-vs-str issue10:44
antocuniand now I *think* that the original problem is that on default it worked just by chance10:44
AlekSi (~AlekSi@85.235.191.82) left irc: Read error: Connection reset by peer10:45
mattip (2e7483e6@gateway/web/freenode/ip.46.116.131.230) joined #pypy.10:45
mattiphi10:45
AlekSi (~AlekSi@85.235.191.82) joined #pypy.10:45
antocuniand the actual problem is that our "raises" function in applevel tests does not handle unicode messages correctly10:45
dmitrey (~quassel@77.121.195.200) joined #pypy.10:46
mattipfijal: ping10:49
antocuni_ (~antocuni@host224-47-dynamic.116-80-r.retail.telecomitalia.it) joined #pypy.10:49
mattipnumppy-flatitter is ready for review10:49
antocuni (~antocuni@host155-125-dynamic.25-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds10:52
dmitreyfijal: does numpypy work with index ..., i.g. a[4,...]? Seems like no. It prevents of concatenate() porting. On the other side, maybe someone knows how to get slice of array with indexes  a:b along n-th axis? I.e. suppose I have to get A[:,:,3:6,...] Here n=2 thus I can use :,:,3:6,... but what if n is coming as argument(axis)? Currently I have         tmp = \10:54
dmitrey        r[ind:ind+ArrConcatAxisLengths[i], ...] if axis == 0 else \10:54
dmitrey        r[:, ind:ind+ArrConcatAxisLengths[i], ...] if axis == 1 else \10:54
dmitrey        r[:, :, ind:ind+ArrConcatAxisLengths[i], ...] if axis == 2 else \10:54
dmitrey        r[:, :, :, ind:ind+ArrConcatAxisLengths[i], ...] if axis == 3 else \10:54
dmitrey        r[:, :, :, :, ind:ind+ArrConcatAxisLengths[i], ...] if axis == 4 else \10:54
dmitrey        r[:, :, :, :, :, ind:ind+ArrConcatAxisLengths[i], ...] if axis == 5 else \10:54
dmitrey        r[:, :, :, :, :, :, ind:ind+ArrConcatAxisLengths[i], ...] if axis == 6 else \10:54
dmitrey        numpypy_unimplemented_error('concatenation is not implemented for axis > 6 yet')10:54
dmitreybut there should be a better solution (and as I've mentioned above indexing by ... doesn't work yet) 10:54
Tobu (~Tobu@unaffiliated/tobu) left irc: Read error: Operation timed out10:54
dmitreycurrently my concatenation works in numpy (tests are ok) but not numpypy10:56
ronnydmitrey: i wonder if something lile [slice(None)]*axis + [slice(ind,ind+ArrConcatAxisLengths[i])] would work as index10:57
dmitreyronny: thanks, I got it working11:05
mattip (2e7483e6@gateway/web/freenode/ip.46.116.131.230) left irc: Quit: Page closed11:06
squiddy (~squiddy@f053087106.adsl.alicedsl.de) left irc: Remote host closed the connection11:07
witulski (~stupsi@fwstups.cs.uni-duesseldorf.de) joined #pypy.11:08
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) joined #pypy.11:08
AlekSi (~AlekSi@85.235.191.82) left irc: Ping timeout: 252 seconds11:08
witulski (stupsi@fwstups.cs.uni-duesseldorf.de) left #pypy.11:08
AlekSi (~AlekSi@85.235.191.82) joined #pypy.11:09
AlekSi (~AlekSi@85.235.191.82) left irc: Read error: Connection reset by peer11:10
Nick change: ojii -> ojii|eatingbabie11:12
AlekSi (~AlekSi@85.235.191.82) joined #pypy.11:16
voidspace (~voidspace@python/psf/voidspace) joined #pypy.11:16
AlekSi_ (~AlekSi@85.235.191.82) joined #pypy.11:19
AlekSi (~AlekSi@85.235.191.82) left irc: Read error: Connection reset by peer11:21
dmitreyfijal: initial version of concatenate is available at http://pastebin.com/mfr7936Z, have you any suggestions?11:24
dmitreyI go AFK11:24
stefanop (~stefanop@77.72.196.122) joined #pypy.11:31
fijalcfbolz: hi11:41
cfbolzhello11:41
azanella (~azanella@189.6.80.131) joined #pypy.11:44
csenger (~user@dslb-084-058-144-070.pools.arcor-ip.net) joined #pypy.11:45
kenaan12fijal numppy-flatitter 11977157a7ea64 15/pypy/module/micronumpy/REVIEW: review11:46
stakkars_ (~tismer@i59F76506.versanet.de) joined #pypy.11:47
stakkars_ (~tismer@i59F76506.versanet.de) left irc: Client Quit11:48
stakkars_ (~tismer@i59F76506.versanet.de) joined #pypy.11:50
stakkars_ (~tismer@i59F76506.versanet.de) left irc: Client Quit11:50
Nick change: ojii|eatingbabie -> ojii11:51
stefanop (~stefanop@77.72.196.122) left irc: Remote host closed the connection11:53
dmitrey (~quassel@77.121.195.200) left irc: Remote host closed the connection11:56
kenaan12fijal numpy-back-to-applevel 11e96e525cc103 15/pypy/module/micronumpy/test/test_dtypes.py: skip that test11:59
kenaan12fijal numpy-back-to-applevel 114209800a8a51 15/pypy/module/micronumpy/: take for 1d arrays11:59
ericflo (~ericflo@c-50-131-58-97.hsd1.ca.comcast.net) left irc: Quit: ericflo11:59
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) left irc: Quit: Leaving12:01
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.12:06
fijaleh12:09
fijalso numpy docs are a lie :/12:09
fijaldoes anyone have 1.6.1 installed?12:09
fijalnumpy 1.6.112:09
fijaluh I see12:09
JStoker (jstoker@unaffiliated/jstoker) left irc: Ping timeout: 252 seconds12:15
AlekSi_ (~AlekSi@85.235.191.82) left irc: Remote host closed the connection12:16
nedbat (~nedbat@python/psf/nedbat) joined #pypy.12:17
cwillu (~cwillu@cwillu.com) got netsplit.12:18
kvda (~kvda@124-169-134-221.dyn.iinet.net.au) got netsplit.12:18
aboSamoor (~rmyeid@130.245.191.56) got netsplit.12:18
stakkars_ (~tismer@i59F76506.versanet.de) joined #pypy.12:20
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) joined #pypy.12:21
cwillu (~cwillu@cwillu.com) returned to #pypy.12:23
kvda (~kvda@124-169-134-221.dyn.iinet.net.au) returned to #pypy.12:23
aboSamoor (~rmyeid@130.245.191.56) returned to #pypy.12:23
AlekSi (~AlekSi@85.235.191.82) joined #pypy.12:24
JStoker (jstoker@unaffiliated/jstoker) joined #pypy.12:33
amaury (amaury_@nat/google/x-wpdzvboeajzrwpdi) joined #pypy.12:47
kenaan12fijal buildbot 1129b4d8f8ccf0 15/bot2/pypybuildbot/builds.py: I believe this is correct :)12:48
kenaan12fijal buildbot 112e2a10de4ae5 15/bot2/pypybuildbot/master.py: merge12:48
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/1033 [12fijal]12:49
fijalsomeone broke benchmarks!12:50
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.12:58
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 240 seconds12:59
DasIch (~DasIch@p4FFDFED3.dip.t-dialin.net) joined #pypy.13:03
csengerfijal: I think the changed build factory for CPython and the changes to the benchmark runner can be merged. Did you come around to look at it or is there something else I can do?13:19
fijalcsenger: I did not look at it yet13:21
fijalcsenger: I plan to today though13:21
csengerokay, no hurry. I'll post to python-speed to find out what we should record (beside python 3 :)13:23
Taggnostr2_ (~quassel@agw-sparknet.utu.fi) joined #pypy.13:27
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.13:33
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.13:33
bbot24Failure: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/1033 [12fijal]13:41
mat^2 (~mathias@212.130.113.35) joined #pypy.13:43
Taggnostr2_ (~quassel@agw-sparknet.utu.fi) left irc: Remote host closed the connection13:49
fijalAlex_Gaynor: ping13:50
fijalAlex_Gaynor: can you retweet nightly numpy status from @pypyproject?13:50
fijalantocuni_: or maybe you know the password?13:51
antocuni_fijal: it's the same as the bitbucket's one13:51
antocuni_I'm sure you *know* it, right? :-)13:51
fijalyes13:51
fijalI have it in a private chat with you13:51
Nick change: antocuni_ -> antocuni13:51
fijaland pypy-z list moderator password?13:51
antocunino clue, it's saved in my firefox13:52
antocuniI never have to type it :-)13:52
fijalok13:52
fijalcan you discard things in the queue?13:52
antocuniI already did13:52
fijalthat's my fault13:52
fijalool13:52
fijalcool13:52
fijalI posted a bug reported logged in as pypy :/13:52
antocuniyes, at first I was confused13:52
antocunibut then I understood what happened13:52
fijaloh my ffox knows password13:53
fijalgood :)13:53
espes (~espes@123-243-221-248.tpgi.com.au) left irc: Read error: Connection reset by peer13:55
espes (~espes@123-243-221-248.tpgi.com.au) joined #pypy.13:56
jnoller (~jnoller@vpn.nasuni.com) joined #pypy.14:04
fprimex (~fprimex@brent-macbook.sc.fsu.edu) joined #pypy.14:09
kenaan12fijal numpy-back-to-applevel 115ab418b60c8c 15/pypy/module/micronumpy/test/test_numarray.py: more tests14:15
kenaan12fijal numpy-back-to-applevel 110dc95b060dca 15/pypy/module/micronumpy/: least effort take for no axis14:15
kenaan12fijal numpy-back-to-applevel 11233034cea174 15/pypy/module/micronumpy/: more tests14:15
kenaan12fijal numpy-back-to-applevel 113c1c293c3fa4 15/pypy/module/micronumpy/interp_numarray.py: proper error14:15
kenaan12fijal numpy-back-to-applevel 11ecec9d981978 15/pypy/module/micronumpy/: basic compress + fix a weird bug14:15
stakkars_ (~tismer@i59F76506.versanet.de) left irc: Quit: schnarch14:19
kenaan12fijal numpy-back-to-applevel 11b70485f468be 15/pypy/module/micronumpy/: compress support14:25
Tobu (~Tobu@unaffiliated/tobu) left irc: Ping timeout: 272 seconds14:25
kirmaI have the impression that swipe-to-close started to accept swipes which were not coherent with the display orientation if the phone was somewhere between landscape and portrait orientation14:27
fijalkirma: ENOCONTEXT14:27
fijalor EWRONGCHANNEL14:27
fijalmore likely14:27
kirmaargh :)14:27
kirmatrue14:28
derdon (~derdon@pD9E1DF07.dip.t-dialin.net) joined #pypy.14:28
kirmaif I say something on this window like once couple weeks, likelihood of it going to wrong channel seems to be over 50% :)14:28
kenaan12fijal numpy-back-to-applevel 111a4c04c232fb 15/lib_pypy/numpypy/core/: cheat a tiny bit. One test passes (we have to export any and all though and implement axis there)14:31
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.14:32
Rhy0lite (dje@nat/ibm/x-dwzxuwcrwnfvjjgv) joined #pypy.14:32
fijalRhyolite: hi14:32
EnCuKou (~encukou@ip-94-113-220-25.net.upcbroadband.cz) left irc: Ping timeout: 240 seconds14:32
fprimex (~fprimex@brent-macbook.sc.fsu.edu) left irc: Quit: http://www.fprimex.com14:33
Rhy0litebivab: ping14:37
bivabRhy0lite: pong14:38
Rhy0litebivab: I don't know if you saw any of the scrollback from last night14:38
Rhy0liteI was able to translate Prolog, but not PyPy14:39
fijalRhy0lite: congrats btw!14:39
fijalwhat's blocking you from the full pypy?14:39
bivabRhy0lite: no, but I saw your tweet. Yes congrats14:39
Rhy0litefijal: the same problem we discussed last night14:39
Rhy0litethe bug on the ARM branch and PPC branch14:39
Rhy0litetranslation_time_error undefined14:39
gtaylor (~gtaylor@108-196-160-156.lightspeed.tukrga.sbcglobal.net) joined #pypy.14:40
fijalRhy0lite: no, it's not a bug14:41
fijalit's a missing feature14:41
Rhy0litefijal: and the other changes to runner that need to be fixed14:41
bivabRhy0lite: the translation_time_error?14:42
Rhy0litefijal: I consider a missing feature that causes an error a bug14:42
fijalRhy0lite: :)14:42
Rhy0litebivab: you made some changes relative to default14:42
Rhy0litebivab: in jit/backend/llsupport/llmodel.py14:43
Rhy0litedef get_on_leave_jitted_hook14:44
bivabwhere the backend needs to provide a leave_jitted_hook function?14:46
fijalbivab: what do you mean?14:47
kenaan12fijal numpy-back-to-applevel 11ecfe72e6275e 15/pypy/module/micronumpy/: expose name14:48
kenaan12fijal numpy-back-to-applevel 1172b8bf6e9019 15/lib_pypy/numpypy/core/numeric.py: pfff14:48
bivabi mean def get_on_leave_jitted_hook that needs to be overwritten in the subclasses of AbstractLLCPU14:48
EnCuKou (~encukou@ip-94-113-220-25.net.upcbroadband.cz) joined #pypy.14:49
fijalbivab: yes, what about that?14:49
fijalit has to provide a function that will be called each time JIT leaves14:49
voidspace (~voidspace@python/psf/voidspace) left irc: Quit: Bye...14:49
bivabi was asking if that is the issue that Rhy0lite mentioned14:50
fijalyes14:50
fijalbivab: you know what the function does btw?14:50
bivabit returns a function that tells the gc to consider the values in the fail_boxes_ptr as young pointers14:51
whitelynx (~whitelynx@63.241.75.144) joined #pypy.14:51
voidspace (~voidspace@87-194-212-65.bethere.co.uk) joined #pypy.14:51
voidspace (~voidspace@87-194-212-65.bethere.co.uk) left irc: Changing host14:51
voidspace (~voidspace@python/psf/voidspace) joined #pypy.14:51
fijalyes14:56
fijalI think it can be almost directly copied from the x86 backend14:56
fijaldepends where you store those pointers anyway14:56
bivabyes, I think the field as even the same name in the PPC backend14:57
dmitrey (~quassel@77.121.195.200) joined #pypy.15:05
Tobu (~Tobu@unaffiliated/tobu) left irc: Ping timeout: 272 seconds15:06
dmitrey (~quassel@77.121.195.200) left irc: Remote host closed the connection15:14
squiddy (~squiddy@f053087106.adsl.alicedsl.de) joined #pypy.15:19
kenaan12fijal default 118d9c1bd7e3be 15/pypy/module/micronumpy/: merge numpy-concatenate and move it to interp-level15:22
fijalnumpy.array([[3, 4], [5, 6]][::2]).take([1])15:32
fijalthis aint fuckin funny15:32
Rhy0litebivab: I tried some of Carl Friedrich's prolog, and all but the one that performs a lot of allocations works15:33
fijaluh no15:33
fijalsorry my bad15:33
bivabRhy0lite: Do you know why the last one you mentioned fails?15:34
Rhy0liteno, I haven't debugged it15:35
Rhy0liteit's iterate_findall15:35
Rhy0litethat Carl says performs a lot of allocations15:35
bivabRhy0lite: but nice progress to have a translatable and running interpreter15:35
AlekSi (~AlekSi@85.235.191.82) left irc: Remote host closed the connection15:35
bivabRhy0lite: which one is it btw?15:35
cfbolzbivab: pyrolog15:36
bivabcfbolz: I meant which test/benchmark fails on pyrolog?15:37
Rhy0litebivab: iterate_findall15:38
Rhy0litebut I think something is strange with the translation15:38
Rhy0litebecause --jit off does not have any effect15:39
bivabRhy0lite: why?15:39
fijalRhy0lite: well15:39
fijalRhy0lite: --jit off might not work for pyrolog15:39
fijalyou have to have interpreter support15:39
cfbolzfijal: it does15:39
fijalcfbolz: did you update to new interface?15:39
fijalor in other words - do you have more than one jitdriver?15:40
cfbolzno15:40
fijalok15:40
fijalthen indeed might work15:40
fijalthese days if you pass None to those hooks it means "all jit drivers"15:40
Rhy0liteI don't think the JIT is getting invoked15:41
Rhy0litelet me try translating again15:41
fijalRhy0lite: PYPYLOG=jit-summary:- ...15:41
fijalwould tell you whether it is invoked or not15:41
cfbolzah yes, jit-summary is better15:41
Rhy0litenope :-(15:42
cfbolzRhy0lite: what was your precise translation cmdline?15:43
Rhy0litePYTHONPATH=.. /home/dje/src/pypy-ppc/pypy/translator/goal/translate.py --opt=jit targetprologstandalone.py15:43
cfbolzok, really weird15:44
Rhy0liteis that correct?15:44
fijalyes15:44
fijaldid it say "generating JIT generator"15:44
fijalor something along these lines15:44
zain (~textual@bzq-79-182-196-50.red.bezeqint.net) left irc: Quit: Computer has gone to sleep.15:44
cfbolzRhy0lite: or equivalent qusteion, how long did it take?15:44
fijalcfbolz: it's PPC our estimates might be off15:45
Rhy0litePYTHONPATH=.. /home/dje/install/Python2.7/bin/python /home/dje/src/pypy-ppc/pypy/translator/goal/translate.py --opt=jit targetprologstandalone.py15:46
Rhy0liteI forgot to say exactly that I'm invoking CPython15:46
Rhy0liteI'm re-running it now15:46
Rhy0liteI think it took around 1000-2000 seconds15:46
Rhy0lite[translation:info] -- someobjectness  0% (0 of 1618 functions polluted by SomeObjects)15:46
Rhy0lite[translation:info] RTyping...15:46
Rhy0litespecialized 13500 blocks15:47
Rhy0liteinlined 6605 callsites15:47
Rhy0litesome pypy.rpython.memory.gc warnings15:48
Rhy0litec:writing15:48
Rhy0litejust prior to this phase:15:48
Rhy0lite[c]    11863 nodes  [ array: 1143  framework rtti: 121  func: 1327  group: 1  struct: 9271 ]15:48
Rhy0liteCompiling c source...15:49
Rhy0liteI don't see "generating JIT generator"15:50
Rhy0lite[Timer] Timings:15:50
Rhy0lite[Timer] annotate                       ---  65.5 s15:50
Rhy0lite[Timer] rtype_lltype                   ---  42.5 s15:50
Rhy0lite[Timer] backendopt_lltype              ---  29.1 s15:50
Rhy0lite[Timer] stackcheckinsertion_lltype     ---   1.0 s15:50
Rhy0lite[Timer] database_c                     ---  38.1 s15:50
Rhy0lite[Timer] source_c                       ---  69.9 s15:50
Rhy0lite[Timer] compile_c                      ---  16.8 s15:50
Rhy0lite[Timer] ==========================================15:50
Rhy0lite[Timer] Total:                         --- 262.9 s15:50
Alex_Gaynorat under 5 minutes there's no way it generated a JIT15:50
Rhy0litethat's why I wanted to test translation to shake out these other problems15:52
dmalcolm (david@nat/redhat/x-fasynncxemlgblpn) joined #pypy.15:55
fijalRhy0lite: that's clearly not with JIT15:56
fijal-Ojit instead of --opt=jit?15:56
fijalbut it all sounds obscure15:56
Rhy0liteokay, I think I mistyped something with the re-translation this morning15:57
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.15:57
Rhy0liteif I try again, at the very beginning I see15:58
Rhy0lite[translation] translate.py configuration:15:58
Rhy0lite[translation] [translate]15:58
Rhy0lite[translation]     opt = jit15:58
Rhy0lite[translation]     targetspec = targetprologstandalone15:58
Rhy0lite[translation] translation configuration:15:58
Rhy0lite[translation] [translation]15:58
Rhy0lite[translation]     backend = c15:58
Rhy0lite[translation]     gc = minimark15:58
Rhy0lite[translation]     gcrootfinder = asmgcc15:58
Rhy0lite[translation]     gctransformer = framework15:58
Rhy0lite[translation]     jit = True15:58
Rhy0lite[translation]     list_comprehension_operations = True15:58
Rhy0lite[translation]     type_system = lltype15:58
Rhy0lite[translation]     withsmallfuncsets = 515:58
fijalRhy0lite: use paste.pocoo.org please15:58
Rhy0liteopt = jit is correct?15:58
Rhy0liteyes, sorry15:58
fijalsounds good15:59
fijaland it still does not have a JIT?15:59
fijalas in there should be another step in timer15:59
fijalafter rtyping that takes quite a bit15:59
Rhy0litelet me try again15:59
Rhy0liteI think I mistyped something this morning15:59
dmitrey (~quassel@77.121.195.200) joined #pypy.16:00
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.16:00
Rhy0litenow I hit the translation_time_error but, so I must have done something wrong with the previous translations16:02
Rhy0litewhich makes the segfault with iterate_findall even stranger16:04
fijalyes16:05
dmitrey_ (~quassel@77.121.195.200) joined #pypy.16:06
cfbolzRhy0lite: could be a boehm bug?16:06
dmitrey (~quassel@77.121.195.200) left irc: Remote host closed the connection16:06
dmitrey_ (~quassel@77.121.195.200) left irc: Remote host closed the connection16:06
dmitrey (~quassel@77.121.195.200) joined #pypy.16:06
Rhy0litecfbolz: possible16:07
bivabRhyolite, cfbolz: just noticed, it looks like you are not translating with boehm16:07
bivabRhyolite: you need to pass --gc=boehm to the translator16:07
fijal-O116:08
fijaldoes that by default16:08
dmitreyfijal: have you looked at concatenate()?16:08
fijaldmitrey: I commited concatenate from other branch16:08
fijalI think your empty_like is slightly buggy16:09
dmitreyyou said noone works on it16:09
fijalwell, I had to rewrite it16:09
fijalI didn't even remember the branch16:09
Rhy0litehttp://paste.pocoo.org/show/541116/16:09
bivabfijal: yes but with -Ojit you need to pass that explicitly 16:09
dmitreyrewrite empty_like or concatenate?16:09
Rhy0liteit fails in pypy_g_throw()16:09
fijaldmitrey: concatenate16:09
fijaldmitrey: you could not do it at applevel16:09
dmitreywell, whats is wrong with empty_like?16:10
fijaldmitrey: can you bring back the paste?16:11
fijalI don't have it with me (my computer crashed)16:11
dmitreyI proposed justinitial implementation of concatenate, I intended to write in RPython-compatible. Empty_like is here: http://pastebin.com/Q9qWPMBy16:12
derdon_ (~derdon@93.232.154.245) joined #pypy.16:12
stakkars_ (~tismer@82.113.121.244) joined #pypy.16:14
dmalcolm (david@nat/redhat/x-fasynncxemlgblpn) left irc: Read error: Connection reset by peer16:14
dmalcolm (david@nat/redhat/x-ckfhdjnokrbyzfce) joined #pypy.16:15
fijalit's so convoluted16:15
fijalfirst, if it's a list you should not create an array of the same size just to read the shape16:15
derdon (~derdon@pD9E1DF07.dip.t-dialin.net) left irc: Ping timeout: 260 seconds16:16
dmalcolm (david@nat/redhat/x-ckfhdjnokrbyzfce) left irc: Read error: Connection reset by peer16:16
dmitreyand how do you do that? IIRC in numpy code they usually create array16:16
dmalcolm (david@nat/redhat/x-cronkmxakjnvmvmb) joined #pypy.16:16
cfbolz (~cfbolz@fwstups.cs.uni-duesseldorf.de) left irc: Quit: Leaving16:17
fijaldmitrey: you can do it in RPython at least16:17
fijaldmitrey: I'll implement empty_like correctly, just not rigth now16:17
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection16:17
fijaldmitrey: I also fixed the type bug16:20
cocoatomo (~cocoatomo@p2118-ipbf901souka.saitama.ocn.ne.jp) joined #pypy.16:22
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) joined #pypy.16:22
amaury (amaury_@nat/google/x-wpdzvboeajzrwpdi) left irc: Ping timeout: 245 seconds16:25
dmitreyfijal: what about the funcs hstack, vstack, dstack? have you committed them?16:33
fijalno16:33
dmitreythey are 2-3 lines of code16:33
fijalyeah16:33
dmitreycasting to concatenate16:33
fijalI'll commit them as soon as I get back16:34
dmitreyok16:34
antocuni (~antocuni@host224-47-dynamic.116-80-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds16:34
Action: fijal off for few h16:34
fijaldmitrey: thanks!16:34
antocuni (~antocuni@host186-122-dynamic.10-79-r.retail.telecomitalia.it) joined #pypy.16:35
dmitreyI would go for something else, but I'm not sure my code will not be rejected again. Which most essential funcs can be accepted in app-level, that are still missing? BTW, I see sort() already present in numpypy, but is absent is ndarray method. IIRC the single difference - ndarray method do it in-place.16:37
dmitrey"but is absent AS ndarray method"16:37
dmitreyI go AFK for several minutes16:39
Rhy0liteDoes the translator normally compile the source files with debugging enabled?16:42
arigato (~arigo@adsl-89-217-207-222.adslplus.ch) joined #pypy.16:45
jacob22 (~jacob@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) left irc: Ping timeout: 248 seconds16:45
jacob22 (~jacob@c-c4c4e055.1321-1-64736c11.cust.bredbandsbolaget.se) joined #pypy.16:48
bivabRhyolite: no, you can recompile them with debugging enabled if you go to /tmp/usession-branchname-xxx/testing_1 and perform a make clean lldebug16:51
Action: bivab off16:57
bivab (~david@ip-62-143-59-5.unitymediagroup.de) left irc: Quit: bivab16:57
Rhy0litehttp://paste.pocoo.org/show/541148/16:59
antocuni_ (~antocuni@host118-20-dynamic.13-87-r.retail.telecomitalia.it) joined #pypy.17:00
antocuni (~antocuni@host186-122-dynamic.10-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds17:01
zain (~textual@87.69.91.91) joined #pypy.17:06
amaury_ (amaury_@nat/google/x-impbtuejraqrkflv) joined #pypy.17:07
antocuni_ (~antocuni@host118-20-dynamic.13-87-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds17:09
G2P (~G2P@fw-asn1.ornis.com) left irc: Quit: Leaving.17:10
kenaan12fijal numpy-back-to-applevel 112b3f50dad1dd 15/pypy/: merge default17:13
cocoatomo (~cocoatomo@p2118-ipbf901souka.saitama.ocn.ne.jp) left irc: Remote host closed the connection17:13
kenaan12fijal numpy-back-to-applevel 11349ebdfc7cde 15/pypy/module/micronumpy/test/test_numarray.py: make sure it works on multidim arrays as well17:13
kenaan12fijal numpy-back-to-applevel 119698baa9d6b5 15/: start implementing iter17:13
fijaldmitrey: do you have vstack implementation somewhere?17:14
antocuni_ (~antocuni@host62-122-dynamic.6-79-r.retail.telecomitalia.it) joined #pypy.17:14
kenaan12fijal default 11b4faea8e796e 15/pypy/module/micronumpy/: obscure dtype equality17:22
antocuni_ (~antocuni@host62-122-dynamic.6-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds17:22
CIA-603fijal 07roundup * 10#1021/numpypy: incorrect dtype comparison (vs original numpy): [resolved] Fixed in b4faea8e796e thanks! * 14https://bugs.pypy.org/issue102117:22
fijaldmitrey: I can't find vstack etc. anywhere in bugs?17:23
dmitreyfijal: I haven't port vstack etc. cpython numpy vstack (numpy/core/shape_base.py): return _nx.concatenate(map(atleast_2d,tup),0)17:26
dmitreyhstack:     arrs = map(atleast_1d,tup)17:26
dmitrey    # As a special case, dimension 0 of 1-dimensional arrays is "horizontal"17:26
dmitrey    if arrs[0].ndim == 1:17:26
dmitrey        return _nx.concatenate(arrs, 0)17:26
dmitrey    else:17:26
dmitrey        return _nx.concatenate(arrs, 1)17:26
dmitreydstack: return _nx.concatenate(map(atleast_3d,tup),2)17:27
whyking (~quassel@ip68-14-13-208.ri.ri.cox.net) left irc: Ping timeout: 245 seconds17:30
apoirier (apr@210.168.3.109.rev.sfr.net) left #pypy ("Leaving...").17:30
Telugodu (~rambaho@202.3.77.231) left irc: Remote host closed the connection17:32
Telugodu (~rambaho@202.3.77.231) joined #pypy.17:33
lucian (~lucian@93-97-174-114.zone5.bethere.co.uk) left irc: Ping timeout: 248 seconds17:38
zain (~textual@87.69.91.91) left irc: Quit: Computer has gone to sleep.17:41
fijaldmitrey: cool, can you write or copy paste tests?17:47
fijalor I can do it17:47
fijaldmitrey: note that from what you write to actual patch is quite a bit of work17:47
dmitreytests for vstack etc?17:48
fijalyeah17:50
dmitreyWhich file have they be committed? Or plain text funcs will be enough?17:51
fijaltest_numeric.py probably17:52
dmitreywell, I'll copy-paste numpy doc examples as the tests ok? Or additional tests are required?17:53
whyking (~quassel@138.16.112.46) joined #pypy.17:57
zain (~textual@bzq-79-179-136-94.red.bezeqint.net) joined #pypy.17:57
dmitreyfijal: I see test_numeric hasn't been commited into repository yet?17:59
fijaldoc examples are usually not enough18:01
fijaljust create a new one18:01
fijalit's on a branch18:01
fijalI guess it does not really matter as long as those are applevel, doc examples are okish18:01
fijalunless you can think about obscure edge cases18:01
fijalI usually can :/18:01
Telugodu (~rambaho@202.3.77.231) left irc: Remote host closed the connection18:01
Telugodu (~rambaho@202.3.77.231) joined #pypy.18:02
dmalcolm (david@nat/redhat/x-cronkmxakjnvmvmb) left irc: Ping timeout: 245 seconds18:03
Telugodu (~rambaho@202.3.77.231) left irc: Client Quit18:06
Telugodu (~rambaho@202.3.77.231) joined #pypy.18:07
kenaan12fijal extradoc 114acf71058742 15/planning/micronumpy.txt: add an optimization18:19
kenaan12fijal extradoc 11da88aafb0de1 15/: merge18:19
amaury_ (amaury_@nat/google/x-impbtuejraqrkflv) left irc: Ping timeout: 260 seconds18:20
fprimex (~fprimex@brent-macbook.sc.fsu.edu) joined #pypy.18:21
zain (~textual@bzq-79-179-136-94.red.bezeqint.net) left irc: Quit: Computer has gone to sleep.18:23
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 252 seconds18:32
dmitreyfijal: I can't access that branch where test_numeric.py was committed so I guess my "hg diff"will be different from your. I put my tests for hstack etc to http://pastebin.com/cfc9KpLd  , I guess you can easily commit them by yourself18:32
stakkars_ (~tismer@82.113.121.244) left irc: Quit: schnarch18:32
fijaldmitrey: hg up <branch name>18:34
fijalshould work18:34
dmitreyso which branch test_numeric.py was committed?18:36
fijalnumpy-back-to-applevel18:37
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Quit: pnguyen18:37
kenaan12fijal default 119183b721128a 15/pypy/module/micronumpy/: import to_coords from numppy-flatitter branch and modify it. Thanks mattip!18:37
dmitreyfijal: which class name should I commit to test_numeric?18:42
dmitreyfijal: i.e. for further usage with hg diff18:43
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.18:45
fijal (~fijal@196-210-150-33.dynamic.isadsl.co.za) left irc: Ping timeout: 240 seconds18:45
aboudreault (~alanb@osgeo/member/aboudreault) joined #pypy.18:47
etrepum (~bob@75-101-96-144.dsl.static.sonic.net) joined #pypy.18:49
Tobu (~Tobu@unaffiliated/tobu) left irc: Ping timeout: 248 seconds18:53
etrepum (~bob@75-101-96-144.dsl.static.sonic.net) left irc: Quit: etrepum18:56
rguilleberthi19:00
arigatohi19:00
htoothrothi19:01
whyking (~quassel@138.16.112.46) left irc: Ping timeout: 240 seconds19:03
pnguyen_ (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) joined #pypy.19:14
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Ping timeout: 260 seconds19:15
Nick change: pnguyen_ -> pnguyen19:15
fijal (~fijal@196-210-150-33.dynamic.isadsl.co.za) joined #pypy.19:20
dmitreyfijal: which class name should I commit to test_numeric ( for further usage with hg diff)?19:22
fijalwhatever you feel like really19:23
dmitreyfor the funcs vstack etc19:23
dmitreyfijal: btw I see those tests for ones_like etc are absent in the file yet19:24
fijalyes, ones_like is not commited yet19:24
fijalbecause there is no empty_like19:24
kenaan12mattip numppy-flatitter 116d23b2d09136 15/pypy/module/micronumpy/: rework next_skip_x, there still is an AnnotationError19:25
mattip (~mattip@bzq-79-177-109-132.red.bezeqint.net) joined #pypy.19:25
mattipfijal: ping19:26
dmalcolm (david@nat/redhat/x-rdzxeemmexaqzysu) joined #pypy.19:27
mattipmy commit ^^^ gives me this AnnotationError http://paste.pocoo.org/show/541202 (scroll to the bottom)19:27
fijalmattip: pong19:27
mattipbut only when I run the complete test suite 19:27
fijalthat's because you mix two objspaces19:28
fijalsomehow19:28
fijaloh19:29
fijalno this is easy19:29
fijalmattip: compile.py does not know how to create W_FlatIterator19:29
fijalbut has exposed elements19:29
dmitreyfijal: here's the tests for vstack etc http://pastebin.com/G6VeykJ619:29
fijalgo and write a new test for test_zjit that involves creating flat iterator and the bug will go away19:29
fijaldmitrey: ok, can you put it on some bug report?19:30
fijalso I know where to look for it later19:30
dmitreyok19:30
mattipthnks.19:30
fijalmattip: I merged to_coords to default btw19:30
mattipcool. 19:31
mattipat least one success.19:31
CIA-603Dmitrey 07roundup * 10#1025/numpypy: commit tests for vstack, hstack, dstack: [new] http://pastebin.com/G6VeykJ6 * 14https://bugs.pypy.org/issue102519:31
kenaan12arigo extradoc 1112399d963afe 15/planning/stm.txt: Start a draft about STM+GC.19:31
Rhy0litewho is the write person to ask about the code generated in interpreter_continuation.c ?19:35
mattipfijal: my version of to_coords returned "start" as well. :(19:35
fijalwhy?19:36
fijalRhy0lite: armin probably19:36
mattipotherwise I have to parse it out of the slice again for getitem19:36
mattipahh, it's ok. Ignore.19:37
fijalnothing stops you from adding this btw ;)19:37
mattipI didn't realize decode_index4 can accept a w_int19:37
fijalthat is the whole point ;-)19:37
mattipso to_coords can just do to_coords, not parsing slice as well.19:38
mattipgood.19:38
jnoller (~jnoller@vpn.nasuni.com) left irc: Quit: Leaving...19:38
Transformer (~Transform@ool-4a59e397.dyn.optonline.net) joined #pypy.19:44
Transformer (Transform@ool-4a59e397.dyn.optonline.net) left #pypy.19:44
dmitrey (~quassel@77.121.195.200) left irc: Read error: Operation timed out19:47
JaRoel|4d (~jaroel|4d@office.fourdigits.nl) left irc: Remote host closed the connection19:54
kenaan12fijal numpy-back-to-applevel 11e74b4e330798 15/pypy/module/micronumpy/: merge default. why oh why so many conflicts???19:55
kenaan12fijal numpy-back-to-applevel 11b1cb2bf3f898 15/pypy/module/micronumpy/: item, basic support19:55
kenaan12fijal numpy-back-to-applevel 11ae72e452c1fc 15/pypy/module/micronumpy/: make tests pass with -A and return scalars on zeros(()). A bit hard to test without is_scalar, but z...19:55
fijalmattip: I must say the print test suite is comprehensive19:56
Alex_Gaynorfijal: ours or theirs?19:57
fijalmattips19:58
kenaan12fijal numpy-back-to-applevel 1148f4c0d75310 15/pypy/module/micronumpy/: a test and a fix19:59
fijalAlex_Gaynor: I'm passing tests with numpy's array_print with minimal cheating19:59
fijalI have yet to understand why numpy is doing that much work in order to print something19:59
Alex_Gaynorfijal: that's someone else's problem20:00
fijala bit yes20:00
fijaladmitedly if array printing is the bottleneck of your computations you have other problems that we can't solve20:00
kenaan12fijal numpy-back-to-applevel 11ddf5b962aa39 15/pypy/module/micronumpy/interp_numarray.py: make the unsupported case applevel20:02
Alex_Gaynorfijal: your test for dtype.__eq__ is wrong, == is supposed to raise an error for something it can't equal on (stupid)20:03
Alex_Gaynorit should really do `if not isinstance(other, dtype): other = dtype(other) return self == other`20:03
pnguyen (~pnguyen@cpe-66-68-12-3.austin.res.rr.com) left irc: Quit: pnguyen20:03
fijaluh20:03
fijalno, it's supposed to return NotImplemented no?20:04
Alex_Gaynorfijal: It probably should, but it doesn't.20:04
fijaloh ok20:04
fijaltrue20:04
whyking (~quassel@138.16.110.209) joined #pypy.20:04
fijalfeel like fixing? ;-)20:04
Alex_Gaynorfijal: btw, you don't need to add wrap() bethod to GenericBox, tolist() already does that20:04
nickik (~nick@217.150.233.216) joined #pypy.20:05
fijalwhy it's called tolist then?20:05
Alex_Gaynorbecause tolist() is also a method on GenericBox and array20:05
fijalnice20:05
fijalcan I still name it wrap and expose it as tolist?20:06
nickik (~nick@217.150.233.216) left irc: Client Quit20:06
Alex_Gaynorname it item() or something, wrap is a tiny bit confusing IMO20:06
fijalok20:06
Alex_GaynorI'll fix the eq thing20:07
Alex_Gaynorwhich branch is it on?20:07
fijalnumpy-back-to-applevel20:08
kenaan12fijal numpy-back-to-applevel 11c3680933a008 15/pypy/module/micronumpy/: rename wrap to item and merge with tolist. Improve item() and test20:09
whyking (~quassel@138.16.110.209) left irc: Ping timeout: 255 seconds20:09
fijalAlex_Gaynor: while you're at it, can you please make int_w on W_IntegerBox RPython?20:10
fijalit's not now, because it merges .value attribute20:11
Alex_Gaynorfijal: yeah, I can take a look20:11
kenaan12alex_gaynor default 110e63f32cf93c 15/pypy/module/micronumpy/: Fix dtype.__eq__20:13
fijalok, sorry silly me20:13
fijalabout the dtype.__eq__20:13
Alex_Gaynorfijal: in fairness, dtype.__eq__ makes about 0 sense20:14
kenaan12arigo extradoc 1179baef447541 15/planning/stm.txt: More.20:14
Alex_Gaynorwhy the hell should dtype(bool) be equal to bool20:14
aboudreault (~alanb@osgeo/member/aboudreault) left irc: Ping timeout: 260 seconds20:15
kenaan12fijal numpy-back-to-applevel 110db7c34519db 15/pypy/module/micronumpy/interp_numarray.py: oops20:21
fijalAlex_Gaynor: are you fixing the int_w?20:22
Alex_Gaynorfijal: oh, yeah, whoops20:22
mwhudson (~mwh@120.136.5.22) left irc: Quit: Leaving20:23
florian2 (~work@dslb-088-078-099-200.pools.arcor-ip.net) joined #pypy.20:24
whyking (~quassel@138.16.110.209) joined #pypy.20:25
fijalpfff20:25
fijalsomeone feel like implementing bool on SingleFloat?20:25
mwhudson (~mwh@120.136.5.22) joined #pypy.20:25
mwhudson (~mwh@120.136.5.22) left irc: Changing host20:25
mwhudson (~mwh@linaro/mwhudson) joined #pypy.20:25
fijalmwhudson: hi20:26
florian2Hi. I am looking to get started with rpython (and pypy in general). Is there some "hello world" in rpython? 20:26
fijalflorian2: hi, why do you want to hack in RPython?20:26
kenaan12arigo extradoc 1165d56ccf97b0 15/planning/stm.txt: Add para20:27
florian2because I am interested in type inference and want to play with the framework that you pypy guys already did20:27
florian2and I think about using rpython within another project (so not for pypy) but pypy is a huge project and actually not just the translation framework is of interest to me :) But it seems like a good candidate to get started to me20:28
KristianDK (~KristianD@50-57-189-190.static.cloud-ips.com) joined #pypy.20:28
florian2fijal: does that answer your question?20:29
dmitrey (~quassel@77.121.195.200) joined #pypy.20:29
arigatofijal: can you write "x != 0" and hope that it will work for any type of x?20:29
fijalarigato: I think so?20:30
fijalflorian2: ok20:30
fijalflorian2: so rpython is python, hello world looks almost the same except some boilerplate20:30
fijallook in translator/goal/targetnopstandalone.py20:30
fijalit uses os.write, but print works as well :)20:30
arigatoyes, since at least a few years, but it didn't originally :-)20:31
fijalarigato: can I update it to confuse people less?20:32
arigatoyes :-)20:32
kenaan12alex_gaynor numpy-back-to-applevel 116d4db9f0618a 15/pypy/module/micronumpy/interp_boxes.py: make this method RPython20:32
fijalAlex_Gaynor: I think bool() is also not RPython?20:32
Alex_Gaynorfijal: uhh, I think it is20:32
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.20:32
vad (~vad@host221-110-dynamic.25-79-r.retail.telecomitalia.it) joined #pypy.20:33
fijalAlex_Gaynor: ok....20:33
florian2fijal: thanks, I think I got a starting point :) (and it compiles and works)20:33
Action: arigato discovers "targetwarningalone.py", finds that it was created by fijal in 2007 for experiments, and forget about it again :-)20:33
Alex_Gaynorfijal: I mean, apparently bool(X) isn't valid if X is a small integer20:33
Alex_Gaynorbut in general bool() is valid (apparently)20:33
fijalcan I remove it?20:34
fijalarigato: targetwarning20:34
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: Ping timeout: 252 seconds20:34
fijalAlex_Gaynor: you can start reviewing numpy-back-to-applevel20:35
fijalthere are few things I know, like a missing jitdriver20:35
kenaan12fijal default 11ddbfdbe2d419 15/pypy/translator/goal/targetnopstandalone.py: simplify targetnopstandalone20:35
kenaan12fijal default 110586c5404983 15/pypy/module/micronumpy/: merge20:35
whyking (~quassel@138.16.110.209) left irc: Ping timeout: 276 seconds20:35
Alex_Gaynorfijal: ok, probably won't start until tomorrow morning20:37
fijalok20:38
mwhudsonfijal: hi20:38
mwhudsonfijal: did you get your laptop working with oneiric in the end?20:38
whyking (~quassel@138.16.110.209) joined #pypy.20:38
fijalmwhudson: for some definition of working20:39
fijalmwhudson: unity crashes horribly20:39
fijalbut it's not the worst thing ever20:39
mwhudsonfijal: :/20:39
mwhudsonfijal: what graphics card do you have?20:39
fijalI have what I want and actually it's better off with the amount of things that crashed20:39
fijalintel20:40
fijalit somehow froze my configuration tool and then won't boot again when I kileld it20:40
mwhudsonsounds unpleasant20:40
fijalwell no, it's not half bad20:40
fijalI did not like those features anyway20:40
fijaland the half-working state is kinda cool20:40
mwhudsonheh20:40
fijalhowever I wonder20:41
fijaldo ubuntu desktop people want to hear they're going nowhere?20:41
arigato(I have similar complains from a friend of mine, btw)20:41
fijalthe thing is I'm sure they heard it already20:42
arigato(me, I don't use any standard interface of course)20:42
mwhudsonfijal: well, it's the usual 'you can't please all of the people all of the time' sort of thing20:42
fijalarigato: I think mine does not qualify as standard any more20:42
fijalmwhudson: I think "not having bugs" would please most20:42
mwhudsonah yeah20:42
fijal(admitedly I'm happier with bugs)20:42
mwhudsonwell there is definitely a lot of work going into bug fixing20:43
fijalmwhudson: but yes, it seems they try to make it like a buggy mac20:43
mwhudsoncompiz is a large c++ program20:43
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.20:43
Alex_Gaynormwhudson: FWIW tell your people: I like unity20:43
mwhudsonwritten by people who didn't really know c++20:43
fijaland people who use ubuntu are pissed off because it changed and you can't change it back20:43
fijaland mac people might maybe like it20:43
mwhudson(actually, i think it started in c, and then got converted to "mostly c++, ish")20:43
mwhudsonwhat do you think it's chances of not crashing are?20:43
fijalmwhudson: note I don't mind editing config in obscure files20:43
fijalbut it's REALLY HARD20:43
fijalmwhudson: heh :)20:44
mattip (~mattip@bzq-79-177-109-132.red.bezeqint.net) left irc: Remote host closed the connection20:44
fijalok20:44
fijalI guess I got it20:44
fijalI think unity attacks a very serious problem20:44
fijalbut for example things like Alt-F2 are unusable20:44
fijalit used to be I can just type a command20:44
fijalnow I have to type it, wait until it wakes up20:44
fijalpush a few arrows20:44
fijalit does not respond to fast typing or fast anything really20:44
fijalyou have to do 1. wait 2. wait etc.20:45
mattip (~mattip@bzq-79-177-109-132.red.bezeqint.net) joined #pypy.20:45
mwhudsonhuh, that stuff is all basically instant for me20:46
fijalAlex_Gaynor: uh, Scalar thinks it's value field is a W_Root20:46
mwhudsonfijal: have you tried unity-2d?20:46
fijalmwhudson: yeah20:46
Alex_Gaynorfijal: go reivew my branch and add hacks to it so we can kill Scalar20:46
Alex_Gaynor;)20:46
fijalmwhudson: but the problem is as soon as you move off the defaults, bugs are 10x20:46
mwhudsoni have sandybridge gfx, they suck much less than previous intel efforts :)20:47
fijalmwhudson: I had nvidia quattro and it was equally slow20:47
fijalyou can't just press Alt-f2 type and press enter20:47
fijalyou have to wait at least twice20:47
fijalwell, I have fast fingers20:47
fijalmwhudson: I think unity can be a great thing, but it's just unusable as it is now20:48
Alex_Gaynorit's not that unusable, I'm using it!20:48
fijalI'm exagerrating for an effect20:48
mwhudsonfijal: i simply do not have the problem you are describing :/20:48
fijalI've used it20:48
fijalmwhudson: are you using oneiric or the never one?20:49
mwhudsonfijal: oneiric20:49
fijal:/20:49
fijalI can't really show you because this particular tab is crashed20:50
fijaland won't start again20:50
Rhy0lite (dje@nat/ibm/x-dwzxuwcrwnfvjjgv) left irc: Quit: Leaving20:50
mattipfijal: "someone feel like implementing bool on SingleFloat": I looked at the source for this, AFAICT numpy does ==020:53
mattipassuming I understood the question.20:53
ericflo (~ericflo@75.103.8.110) joined #pypy.20:54
dmalcolm (david@nat/redhat/x-rdzxeemmexaqzysu) left irc: Ping timeout: 252 seconds20:57
fijalarigato: you can't compare SHORT to 020:58
etrepum (~bob@75-101-96-144.dsl.static.sonic.net) joined #pypy.20:58
zain (~textual@bzq-79-179-136-94.red.bezeqint.net) joined #pypy.20:59
kenaan12fijal numpy-back-to-applevel 11f5834bdc4464 15/pypy/module/micronumpy/: a pretty important fix for item and more tests, bad fijal20:59
kenaan12fijal numpy-back-to-applevel 11101fcf0a6557 15/pypy/module/micronumpy/: soem rpython fixes20:59
whyking (~quassel@138.16.110.209) left irc: Ping timeout: 245 seconds21:05
kvda (~kvda@124-169-134-221.dyn.iinet.net.au) left irc: Quit: x___x21:05
fijal_ (~fijal@196-210-150-33.dynamic.isadsl.co.za) joined #pypy.21:05
arigatofijal: :-(21:05
fijal_http://paste.pocoo.org/show/541283/21:06
fijal_arigato: ever seen that?21:06
fijal (~fijal@196-210-150-33.dynamic.isadsl.co.za) left irc: Disconnected by services21:06
Nick change: fijal_ -> fijal21:06
zain (~textual@bzq-79-179-136-94.red.bezeqint.net) left irc: Quit: Computer has gone to sleep.21:08
mattip_ (~mattip@bzq-79-177-109-132.red.bezeqint.net) joined #pypy.21:09
fijalarigato: I'm slightly clueless21:10
fijaluh21:12
fijalAlex_Gaynor: bad alex no cookie21:12
Alex_Gaynorfijal: what'd I do?21:12
fijalyou messed up somehow dict.pop21:12
fijaltry running test_ztranslation now21:12
Alex_Gaynoruh oh21:12
mattip (~mattip@bzq-79-177-109-132.red.bezeqint.net) left irc: Ping timeout: 260 seconds21:12
Nick change: mattip_ -> mattip21:13
fijalAlex_Gaynor: didn't you merge all the dicts or something?21:14
Alex_Gaynorfijal: wtf, I have no idea21:14
Alex_Gaynorfijal: not intenteionally...21:14
fijalok21:14
fijalwell fix it21:14
arigato:-)21:14
Alex_GaynorI... I have no idea how :)21:15
Alex_Gaynorif you can write a failing tests I'll fix it21:15
vad (~vad@host221-110-dynamic.25-79-r.retail.telecomitalia.it) left irc: Remote host closed the connection21:16
fijaltest_ztranslation21:17
Alex_GaynorI meant a test in rpython/test/21:17
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.21:21
derdon_ (~derdon@93.232.154.245) left irc: Remote host closed the connection21:22
derdon (~derdon@93.232.154.245) joined #pypy.21:23
dmitreyfijal: I have wrote func for determing shape w/o casting to array (for example, for empty_like), you can see 2 implementations here: http://pastebin.com/N7sMWX08  ; 2nd implementation is more suitable for rpython (I haven't test it there although, I tested it only with CPython and PyPy)21:23
fijalAlex_Gaynor: where is rtype_pop?21:24
Alex_Gaynorfijal: {oo,ll}typesystem/rdict.py21:24
Alex_Gaynorfijal: ah, it needs self.recast_value(hop,llops, v_res)21:26
Alex_Gaynortest!21:26
derdon (~derdon@93.232.154.245) left irc: Ping timeout: 252 seconds21:27
fijalAlex_Gaynor: I think you're doing it wrong21:28
fijalmaybe not21:28
Alex_Gaynorfijal: look at get or sedefault21:28
Alex_Gaynorthey do the same gendirectcall21:29
fijalyeah21:29
fijalhow do we write test for that?21:29
kenaan12fijal numpy-back-to-applevel 11ecbc34a30eca 15/pypy/module/micronumpy/interp_ufuncs.py: fix translation?21:29
kenaan12fijal numpy-back-to-applevel 1177d1d7f10210 15/pypy/module/micronumpy/interp_ufuncs.py: another fix?21:29
kenaan12fijal numpy-back-to-applevel 11aa22b6b02b7e 15/pypy/rpython/lltypesystem/rdict.py: cheat21:29
Alex_Gaynorfijal: let me try21:30
fijalI commited the fxi ;-021:30
Alex_Gaynorfijal: with no test?21:30
fijalyou write a test21:30
fijalI'm trying to make other stuff work21:31
CIA-603Dmitrey 07roundup * 10#1026/numpypy: incorrect shape of array with zero-shape coords: 21:31
CIA-6[new] >>> np.array(np.ones((0,3,4))).shape21:31
CIA-6(0, 3, 4)21:31
CIA-6>>>> np.array(np.ones((0,3,4))).shape21:31
CIA-6(0,) * 14https://bugs.pypy.org/issue102621:31
fijal(I'm testing on tannit because my computer overheats)21:31
fijalAlex_Gaynor: that fixes it at least21:31
iko (iko@eskarina.iko.pp.se) left irc: Ping timeout: 252 seconds21:31
dmitrey (~quassel@77.121.195.200) left irc: Remote host closed the connection21:31
Alex_Gaynorfijal: your change to popitem is defintiely wrong too21:34
Alex_Gaynorfijal: I'm reverting it21:34
kenaan12alex_gaynor numpy-back-to-applevel 1145f2ee87bb74 15/pypy/rpython/: fix dict.pop() in RPython with instances.21:35
kenaan12alex_gaynor numpy-back-to-applevel 118fe3c92a0d5e 15/pypy/: merged upstream21:35
kenaan12alex_gaynor numpy-back-to-applevel 11ff1328acfae8 15/pypy/rpython/lltypesystem/rdict.py: revert this part of the change, ti's wrong21:35
fijalok21:37
fijalit's fine it was not tested ;-)21:37
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) joined #pypy.21:37
kenaan12fijal numpy-back-to-applevel 11857a38cda57f 15/pypy/module/micronumpy/test/test_base.py: fix test_base21:39
fijalAlex_Gaynor: I'm moving zeros and ones to applevel btw at some point21:40
fijalor "back to applevel"21:40
Alex_Gaynorfine with me21:40
fijalwell, it's applevel in numpy21:41
fijalI don't think we have *any* reason to keep in interp-level what's applevel in numpy21:41
Alex_Gaynorare you sure?21:41
iko (iko@eskarina.iko.pp.se) joined #pypy.21:41
Alex_GaynorI'm pretty sure zeros is python21:41
fijalhm, no21:41
Alex_Gaynorerr C21:41
fijalones is applevel21:41
fijal!@$#$!@#21:41
Alex_Gaynorfijal: zeros as interp level makes a bit of sense, that way you call calloc() basically21:42
fijalgood21:42
fijalall tests pass21:42
mattipMy two bits: do both zeros() at applevel and _zeros at interp. That way the initial parsing is == to ones()21:44
fijalAlex_Gaynor: I need to add a jitdriver in take and the branch is ready to be merged21:46
fijalyou can still review it it's missing just this one thing IMO21:46
Alex_GaynorI'll reivewi it tommrow morning, I'm super tired ATM21:46
fprimex (~fprimex@brent-macbook.sc.fsu.edu) left irc: Quit: http://www.fprimex.com21:46
fijalAlex_Gaynor: yeah that's fine I'm about to go away anyway21:46
fijalmattip: what's the flatiter status btw?21:47
kenaan12mattip numppy-flatitter 114bd2c28ea1e4 15/pypy/: merge with default21:47
kenaan12mattip numppy-flatitter 1188475cf6b4ab 15/pypy/module/micronumpy/: cleanup merge, add failing test21:47
mattipfijal: I tried what you suggested and now broke zjit :(21:47
mattipfeel like taking a look?21:47
fijalyeah I can21:48
tilgovi (~randall@couchdb/developer/tilgovi) left irc: Remote host closed the connection21:48
fijaljust try test_zjit?21:48
mattipyes.21:48
fijalarigato: btw, did you break nightly benchmarks?21:49
fijalhm21:49
fijalno idea how did they break they look obscure :/21:50
fijalmattip: this message is super clear isn't it?21:50
fijaland you didn't do what I suggested :)21:51
verte-wleslie (~verte@python/site-packages/verte) joined #pypy.21:51
fijalthere is no next method on W_FlatIterator is there?21:51
fijalI suggested adding actual jit tests for flat iterator21:51
fijalso say support in compile.py21:51
mattipahh.21:52
fijaland not a random call to a method that's not there21:52
mattiptrue.21:52
mattipbut that's harder :)21:52
mattipOK, let me try again.21:52
fijalit also might yield better results IMO ;-)21:52
mattipfijal: I would appreciate it if you could take a look at my changes to the next_skip_x function21:58
mattipso I could fix that too.21:58
fijalok21:58
fijalmattip: now next_skip_x is unroll safe21:58
mattipI don't like the // operator21:58
fijalbecause all loops are there with a constant length21:59
fijalwhy?21:59
mattipisn't it expensive?21:59
mattipin terms of cpu cycles?21:59
fijalwell22:00
mattipbackstrides is much better zen.22:00
fijalI think a flatiterator accessed by view that has a step is *really* rare22:00
fijalas in I would not worry too much for now22:00
fijalyou have backstrides don't you there btw?22:01
fijalmattip: ah, this is complex enough logic I would write a few direct tests for that22:01
fijalI'm not smart enough to tell you without writing tests if this is correct or not22:01
Tobu (~Tobu@unaffiliated/tobu) left irc: Read error: Operation timed out22:01
mattipok, more tests is good.22:01
fijallike direct tests22:02
mattipyes22:02
fijalx = ViewIterator(some parameters)22:02
fijalnext = x.next()22:02
fijalassert about properties22:02
fijalright?22:02
fijalyou can create test_iter.py for it22:02
fijalmattip: btw, I removed all the string formatting22:02
mattipgood. I wasn't too proud of that.22:03
mattipbut it taught me alot22:03
fijalgood :)22:03
fijalit's applevel in numpy and it's plain insane22:03
fijalbut well, it's no longer our problem22:03
kenaan12fijal numpy-back-to-applevel 111d6e533a0aaa 15/pypy/module/micronumpy/interp_numarray.py: no longer needed22:03
fijalmattip: it means we don't have to maintain it as well (which is always good)22:04
mattipto support fpypy we will need order='F' pretty soon, so direct tests are good.22:04
fijalyes22:04
fijalwould be cool to have22:04
mattipmy plan: matrixmath-dot, then linalg via fpypy22:05
fijalcool22:05
fijalthat would already shut up some people22:05
mattipmaybe with a bypass to get win32 build to pass more tests22:05
fijalmattip: ok, I gonna sleep I think22:06
mattipgnight.22:06
csenger (~user@dslb-084-058-144-070.pools.arcor-ip.net) left irc: Quit: ERC Version 5.3 (IRC client for Emacs)22:14
arigato (~arigo@adsl-89-217-207-222.adslplus.ch) left irc: Quit: See you22:16
squiddy (~squiddy@f053087106.adsl.alicedsl.de) left irc: Quit: Leaving22:17
Nick change: verte-wleslie -> wleslie22:19
exnihilo (~nathan@descartes.its.unc.edu) left irc: Quit: Lost terminal22:24
Tobu (~Tobu@unaffiliated/tobu) joined #pypy.22:26
antocuni_ (~antocuni@host62-122-dynamic.6-79-r.retail.telecomitalia.it) joined #pypy.22:29
nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 248 seconds22:36
JaRoel|4d (~jaroel|4d@sink.jaroel.nl) joined #pypy.22:38
nedbat (~nedbat@python/psf/nedbat) joined #pypy.22:43
nedbat (~nedbat@python/psf/nedbat) left irc: Read error: Connection reset by peer22:44
tilgovi (~randall@couchdb/developer/tilgovi) joined #pypy.22:44
nedbat (~nedbat@python/psf/nedbat) joined #pypy.22:46
mattip (~mattip@bzq-79-177-109-132.red.bezeqint.net) left irc: Quit: Leaving22:48
florian2 (~work@dslb-088-078-099-200.pools.arcor-ip.net) left irc: Ping timeout: 276 seconds22:54
zain (~textual@bzq-79-179-136-94.red.bezeqint.net) joined #pypy.22:56
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-32/builds/103423:00
bbot2Started: 15http://buildbot.pypy.org/builders/own-macosx-x86-32/builds/78823:00
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64/builds/21723:00
bbot2Started: 15http://buildbot.pypy.org/builders/jit-benchmark-linux-x86-64-2/builds/4523:00
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: Ping timeout: 244 seconds23:03
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.23:06
azanella (~azanella@189.6.80.131) left irc: Ping timeout: 260 seconds23:11
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: Ping timeout: 252 seconds23:14
whitelynx (~whitelynx@63.241.75.144) left irc: Quit: Ex-Chat23:19
zain (~textual@bzq-79-179-136-94.red.bezeqint.net) left irc: Quit: Computer has gone to sleep.23:21
nedbat (~nedbat@python/psf/nedbat) left irc: Ping timeout: 255 seconds23:21
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.23:22
kenaan12antocuni py3k 11f93127358305 15/pypy/: after one day of investigation, add support for unicode characters inside the 'raises' function in apptests. The te...23:31
kenaan12antocuni py3k 11c08972109bbb 15/pypy/interpreter/astcompiler/test/test_compiler.py: fix test_constants on 32bit, by ignoring the L suffix that we get in python2 in the repr of longs23:31
kenaan12antocuni py3k 117c5c365ec355 15/pypy/interpreter/test/test_syntax.py: kill all the tests which use import * from within a function, which is no longer syntactically valid. Leave just on...23:31
kenaan12antocuni py3k 11d882e23cc8ae 15/pypy/interpreter/astcompiler/symtable.py: kill code to deal with 'import *' inside functions, as it's not longer valid syntax23:31
kenaan12antocuni py3k 11a8eb581e9e9a 15/pypy/interpreter/test/test_syntax.py: extend these test to make sure that the valid snippets are actually valid with cpython3. So far it expects to find ...23:31
kenaan12antocuni py3k 1164a00c269dce 15/pypy/interpreter/test/test_syntax.py: check for invalid syntax also on top of cpython323:31
kenaan12antocuni py3k 11003f9177cdb4 15/pypy/interpreter/test/test_syntax.py: don't try to run python3 if we cannot find it23:31
kenaan12antocuni py3k 114a4a774f432f 15/pypy/interpreter/test/test_syntax.py: exec is a function in python3, not a statement23:31
kenaan12antocuni py3k 11cbba65e7057f 15/pypy/interpreter/test/test_syntax.py: exec is a function in python3, not a statement (also in the INVALID cases, although some still fail)23:31
wleslieoh, we're not supporting both in the same codebase?23:32
antocuni_nope23:32
kenaan12mattip numppy-flatitter 115a2c53898154 15/pypy/module/micronumpy/: fix for z_jit test, now passes23:32
kenaan12mattip numppy-flatitter 11fc5fe1efe542 15/pypy/module/micronumpy/test/test_iter.py: add iter test, more to come23:32
nedbat (~nedbat@python/psf/nedbat) joined #pypy.23:36
kAlug (~kalug@189.65.128.182) joined #pypy.23:38
kAlug (kalug@189.65.128.182) left #pypy.23:39
antocuni_ (~antocuni@host62-122-dynamic.6-79-r.retail.telecomitalia.it) left irc: Ping timeout: 252 seconds23:40
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: Remote host closed the connection23:42
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.23:42
lucian (~lucian@cpc1-newc15-2-0-cust84.gate.cable.virginmedia.com) left irc: Ping timeout: 240 seconds23:43
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) left irc: Ping timeout: 255 seconds23:46
voidspac_ (~voidspace@87-194-212-65.bethere.co.uk) joined #pypy.23:47
inhahe (~inhahe@adsl-074-186-083-219.sip.mia.bellsouth.net) joined #pypy.23:48
--- Fri Jan 27 201200:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!