From eliswilson at hushmail.com Wed May 1 01:22:05 2013 From: eliswilson at hushmail.com (eliswilson at hushmail.com) Date: Tue, 30 Apr 2013 19:22:05 -0400 Subject: [pypy-issue] Biggest Fake Conference in Computer Science Message-ID: <20130430232205.69352E6736@smtp.hushmail.com> Biggest Fake Conference in Computer Science We are researchers from different parts of the world and conducted a study on the world?s biggest bogus computer science conference WORLDCOMP http://sites.google.com/site/worlddump1 organized by Prof. Hamid Arabnia from University of Georgia, USA. We submitted a fake paper to WORLDCOMP 2011 and again (the same paper with a modified title) to WORLDCOMP 2012. This paper had numerous fundamental mistakes. Sample statements from that paper include: (1). Binary logic is fuzzy logic and vice versa (2). Pascal developed fuzzy logic (3). Object oriented languages do not exhibit any polymorphism or inheritance (4). TCP and IP are synonyms and are part of OSI model (5). Distributed systems deal with only one computer (6). Laptop is an example for a super computer (7). Operating system is an example for computer hardware Also, our paper did not express any conceptual meaning. However, it was accepted both the times without any modifications (and without any reviews) and we were invited to submit the final paper and a payment of $500+ fee to present the paper. We decided to use the fee for better purposes than making Prof. Hamid Arabnia richer. After that, we received few reminders from WORLDCOMP to pay the fee but we never responded. This fake paper is different from the two fake papers already published (see https://sites.google.com/site/worlddump4 for details) in WORLDCOMP. We MUST say that you should look at the above website if you have any thoughts of participating in WORLDCOMP. DBLP and other indexing agencies have stopped indexing WORLDCOMP?s proceedings since 2011 due to its fakeness. See http://www.informatik.uni-trier.de/~ley/db/conf/icai/index.html for of one of the conferences of WORLDCOMP and notice that there is no listing after 2010. See Section 2 of http://sites.google.com/site/dumpconf for comments from well-known researchers about WORLDCOMP. The status of your WORLDCOMP papers can be changed from scientific to other (i.e., junk or non-technical) at any time. Better not to have a paper than having it in WORLDCOMP and spoil the resume and peace of mind forever! Our study revealed that WORLDCOMP is money making business, using University of Georgia mask, for Prof. Hamid Arabnia. He is throwing out a small chunk of that money (around 20 dollars per paper published in WORLDCOMP?s proceedings) to his puppet (Mr. Ashu Solo or A.M.G. Solo) who publicizes WORLDCOMP and also defends it at various forums, using fake/anonymous names. The puppet uses fake names and defames other conferences to divert traffic to WORLDCOMP. He also makes anonymous phone calls and threatens the critiques of WORLDCOMP (See Item 7 of Section 5 of above website). That is, the puppet does all his best to get a maximum number of papers published at WORLDCOMP to get more money into his (and Prof. Hamid Arabnia?s) pockets. Prof. Hamid Arabnia makes a lot of tricks. For example, he appeared in a newspaper to fool the public, claiming him a victim of cyber-attack (see Item 8 in Section 5 of above website). Monte Carlo Resort (the venue of WORLDCOMP for more than 10 years, until 2012) has refused to provide the venue for WORLDCOMP?13 because of the fears of their image being tarnished due to WORLDCOMP?s fraudulent activities. That is why WORLDCOMP?13 is taking place at a different resort. WORLDCOMP will not be held after 2013. The draft paper submission deadline is over but still there are no committee members, no reviewers, and there is no conference Chairman. The only contact details available on WORLDCOMP?s website is just an email address! We ask Prof. Hamid Arabnia to publish all reviews for all the papers (after blocking identifiable details) since 2000 conference. Reveal the names and affiliations of all the reviewers (for each year) and how many papers each reviewer had reviewed on average. We also ask him to look at the Open Challenge (Section 6) at https://sites.google.com/site/moneycomp1 and respond if he has any professional values. Sorry for posting to multiple lists. Spreading the word is the only way to stop this bogus conference. Please forward this message to other mailing lists and people. We are shocked with Prof. Hamid Arabnia and his puppet?s activities at http://worldcomp-fake-bogus.blogspot.com Search Google using the keyword worldcomp fake for additional links. From tracker at bugs.pypy.org Wed May 1 10:50:36 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Wed, 01 May 2013 08:50:36 +0000 Subject: [pypy-issue] [issue1472] ast.Class are user-defined, modifiable on CPython In-Reply-To: <1367398236.47.0.812279223241.issue1472@bugs.pypy.org> Message-ID: <1367398236.47.0.812279223241.issue1472@bugs.pypy.org> New submission from Armin Rigo : In CPython, the classes in the 'ast' module are defined in C but as user classes. This makes them user-modifiable. See http://stackoverflow.com/questions/16310661 . Should we fix it? And, as importantly: is this trick used elsewhere? ---------- messages: 5655 nosy: arigo, pypy-issue priority: bug status: unread title: ast.Class are user-defined, modifiable on CPython ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 1 13:56:19 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Wed, 01 May 2013 11:56:19 +0000 Subject: [pypy-issue] [issue1472] ast.Class are user-defined, modifiable on CPython In-Reply-To: <1367398236.47.0.812279223241.issue1472@bugs.pypy.org> Message-ID: <1367409379.75.0.711279792061.issue1472@bugs.pypy.org> Amaury Forgeot d Arc added the comment: Logilab's astng (used by pylint) is very flexible, and it works on PyPy. But I just checked that astng has a completely custom set of classes, and makes a deep copy of the tree returned by compile(). ---------- nosy: +amaury status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 1 15:32:44 2013 From: tracker at bugs.pypy.org (Benjamin Peterson) Date: Wed, 01 May 2013 13:32:44 +0000 Subject: [pypy-issue] [issue1472] ast.Class are user-defined, modifiable on CPython In-Reply-To: <1367398236.47.0.812279223241.issue1472@bugs.pypy.org> Message-ID: <1367415164.61.0.320915730033.issue1472@bugs.pypy.org> Benjamin Peterson added the comment: Probably AST classes should just become applevel. ---------- nosy: +benjamin ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 1 16:11:19 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Wed, 01 May 2013 14:11:19 +0000 Subject: [pypy-issue] [issue1472] ast.Class are user-defined, modifiable on CPython In-Reply-To: <1367398236.47.0.812279223241.issue1472@bugs.pypy.org> Message-ID: <1367417479.29.0.586992217782.issue1472@bugs.pypy.org> Amaury Forgeot d Arc added the comment: And have like CPython two sets of classes, Nodes and AST? ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 1 16:14:52 2013 From: tracker at bugs.pypy.org (Benjamin Peterson) Date: Wed, 01 May 2013 14:14:52 +0000 Subject: [pypy-issue] [issue1472] ast.Class are user-defined, modifiable on CPython In-Reply-To: <1367398236.47.0.812279223241.issue1472@bugs.pypy.org> Message-ID: <1367417692.42.0.226698163716.issue1472@bugs.pypy.org> Benjamin Peterson added the comment: Yes, we could probably simplify the compiler's internal ones then. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 1 17:30:03 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Wed, 01 May 2013 15:30:03 +0000 Subject: [pypy-issue] [issue1472] ast.Class are user-defined, modifiable on CPython In-Reply-To: <1367398236.47.0.812279223241.issue1472@bugs.pypy.org> Message-ID: <1367422203.95.0.530247903072.issue1472@bugs.pypy.org> Amaury Forgeot d Arc added the comment: +1, especially if we have a custom allocator for those nodes (CPython uses a PyArena) Importing large .py files is very slow with pypy (5x for 30k lines) and I measured that it's because of GC pressure. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 2 01:29:30 2013 From: tracker at bugs.pypy.org (Philip Jenvey) Date: Wed, 01 May 2013 23:29:30 +0000 Subject: [pypy-issue] [issue1233] str() types can't be used with ctypes's py_object() In-Reply-To: <1344763493.64.0.519996205744.issue1233@bugs.pypy.org> Message-ID: <1367450970.06.0.815305890362.issue1233@bugs.pypy.org> Philip Jenvey added the comment: another use case, albeit not a great one: Py3's test_unicode test_from_format uses py_object to test capi calls to PyUnicode_FromFormat ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 4 04:19:48 2013 From: tracker at bugs.pypy.org (Michael Dale Long) Date: Sat, 04 May 2013 02:19:48 +0000 Subject: [pypy-issue] [issue1473] Unexpected behavior when inheriting from a ctypes.Structure In-Reply-To: <1367633988.57.0.76005816529.issue1473@bugs.pypy.org> Message-ID: <1367633988.57.0.76005816529.issue1473@bugs.pypy.org> New submission from Michael Dale Long : I have some code that relies on inheriting from SDL's SDL_Rect class. This works in CPython 2.x and 3.x, but fails in PyPy 2.0 beta and the most recent nightly build as well. There seems to be some problem passing arguments to the superclass constructor. See the attached sdl_inherit_test.py for a simplified example (SDL libraries not required, it just declares the Structure), and output.txt for my tests with various Python interpreters on Gentoo Linux. ---------- nosy: +link status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ -------------- next part -------------- link at brinstar /work $ python3 --version Python 3.2.4 link at brinstar /work $ python3 ./sdl_inherit_test.py SDL_Rect: 0 0 10 10 Our_Rect: 5 5 15 15 link at brinstar /work $ python2 --version Python 2.7.3 link at brinstar /work $ python2 ./sdl_inherit_test.py SDL_Rect: 0 0 10 10 Our_Rect: 5 5 15 15 link at brinstar /work $ pypy-c2.0 --version Python 2.7.3 (07e08e9c885ca67d89bcc304e45a32346daea2fa, Apr 03 2013, 12:48:25) [PyPy 2.0.0-beta1] link at brinstar /work $ pypy-c2.0 ./sdl_inherit_test.py Traceback (most recent call last): File "app_main.py", line 52, in run_toplevel File "./sdl_inherit_test.py", line 24, in subrect = Our_Rect(5, 5, 15, 15) File "./sdl_inherit_test.py", line 16, in __init__ SDL_Rect.__init__(self, args[0], args[1], args[2], args[3]) File "/usr/lib64/pypy2.0/lib_pypy/_ctypes/structure.py", line 228, in __init__ raise TypeError("too many initializers") TypeError: too many initializers link at brinstar /work $ ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy --version ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy: /usr/lib64/libssl.so.1.0.0: no version information available (required by ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy) ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy: /usr/lib64/libcrypto.so.1.0.0: no version information available (required by ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy) Python 2.7.3 (c16cd00e82ef, May 03 2013, 22:00:19) [PyPy 2.0.0-beta2 with GCC 4.6.3] link at brinstar /work $ ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy ./sdl_inherit_test.py ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy: /usr/lib64/libssl.so.1.0.0: no version information available (required by ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy) ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy: /usr/lib64/libcrypto.so.1.0.0: no version information available (required by ./pypy-c-jit-63819-c16cd00e82ef-linux64/bin/pypy) Traceback (most recent call last): File "app_main.py", line 72, in run_toplevel File "./sdl_inherit_test.py", line 24, in subrect = Our_Rect(5, 5, 15, 15) File "./sdl_inherit_test.py", line 16, in __init__ SDL_Rect.__init__(self, args[0], args[1], args[2], args[3]) File "/work/pypy-c-jit-63819-c16cd00e82ef-linux64/lib_pypy/_ctypes/structure.py", line 228, in __init__ raise TypeError("too many initializers") TypeError: too many initializers From tracker at bugs.pypy.org Sat May 4 11:26:49 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Sat, 04 May 2013 09:26:49 +0000 Subject: [pypy-issue] [issue1473] Unexpected behavior when inheriting from a ctypes.Structure In-Reply-To: <1367633988.57.0.76005816529.issue1473@bugs.pypy.org> Message-ID: <1367659609.91.0.0211323018451.issue1473@bugs.pypy.org> Amaury Forgeot d Arc added the comment: Fixed with 1236a55828ce, thanks a lot for the report! ---------- nosy: +amaury status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 10:03:15 2013 From: tracker at bugs.pypy.org (Raymond Hettinger) Date: Sun, 05 May 2013 08:03:15 +0000 Subject: [pypy-issue] [issue1474] Random seed() and random() not always matching CPython In-Reply-To: <1367740995.74.0.139717490677.issue1474@bugs.pypy.org> Message-ID: <1367740995.74.0.139717490677.issue1474@bugs.pypy.org> New submission from Raymond Hettinger : Running on Mac OS X 10.8, I observed CPython vs PyPy differences in seeded random numbers: $ pypy --version Python 2.7.2 (341e1e3821ff, Jun 07 2012, 15:42:54) [PyPy 1.9.0 with GCC 4.2.1] $ pypy -c "from random import *; seed('iss'); print random()" 0.128149120625 $ python -c "from random import *; seed('iss'); print random()" 0.418550760801 $ pypy -c "from random import *; seed('now'); print random()" 0.60810839303 $ python -c "from random import *; seed('now'); print random()" 0.60810839303 ---------- messages: 5664 nosy: pypy-issue, rhettinger priority: bug release: 1.9 status: unread title: Random seed() and random() not always matching CPython ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 13:10:09 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sun, 05 May 2013 11:10:09 +0000 Subject: [pypy-issue] [issue1474] Random seed() and random() not always matching CPython In-Reply-To: <1367740995.74.0.139717490677.issue1474@bugs.pypy.org> Message-ID: <1367752209.12.0.266611134725.issue1474@bugs.pypy.org> Armin Rigo added the comment: Reproduced on a recent pypy on Linux64. Same problem on Linux32 (with different values of the seed though). ---------- nosy: +arigo status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 13:21:25 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sun, 05 May 2013 11:21:25 +0000 Subject: [pypy-issue] [issue1474] Random seed() and random() not always matching CPython In-Reply-To: <1367740995.74.0.139717490677.issue1474@bugs.pypy.org> Message-ID: <1367752885.12.0.912679496232.issue1474@bugs.pypy.org> Armin Rigo added the comment: Fixed in 0a3ad482ff72. ---------- status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 14:41:59 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sun, 05 May 2013 12:41:59 +0000 Subject: [pypy-issue] [issue1475] Needed: __pypy__.add_memory_pressure()? In-Reply-To: <1367757719.47.0.0892814490429.issue1475@bugs.pypy.org> Message-ID: <1367757719.47.0.0892814490429.issue1475@bugs.pypy.org> New submission from Armin Rigo : With CFFI, we need occasionally to say that some cdata object keeps alive a large amount of memory (even though it doesn't contain it). This could be done either with an explicit call at app-level to a new function __pypy__.add_memory_pressure(), or with a more CFFI-specific solution, like an extra argument to ffi.gc(). ---------- messages: 5667 nosy: arigo, pypy-issue priority: feature status: unread title: Needed: __pypy__.add_memory_pressure()? ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 14:42:56 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sun, 05 May 2013 12:42:56 +0000 Subject: [pypy-issue] [issue1475] Needed: __pypy__.add_memory_pressure()? In-Reply-To: <1367757719.47.0.0892814490429.issue1475@bugs.pypy.org> Message-ID: <1367757776.32.0.999990670873.issue1475@bugs.pypy.org> Armin Rigo added the comment: Use case in mind: http://bazaar.launchpad.net/~tolot-solar-empire/+junk/gmpy_cffi/files ---------- status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 16:49:01 2013 From: tracker at bugs.pypy.org (Ian Delaney) Date: Sun, 05 May 2013 14:49:01 +0000 Subject: [pypy-issue] [issue1476] apparent regression in import in py2.0 b2 In-Reply-To: <1367765341.24.0.0682227490432.issue1476@bugs.pypy.org> Message-ID: <1367765341.24.0.0682227490432.issue1476@bugs.pypy.org> New submission from Ian Delaney : on installing the pypy2.0 beta2, the troublesome test.py pulls a peculiar fail. lxml-cffi-3.0.2 $ PYTHONPATH=src/ pypy-c2.0 test.py Traceback (most recent call last): File "app_main.py", line 52, in run_toplevel File "test.py", line 607, in exitcode = main(sys.argv) File "test.py", line 570, in main test_cases = get_test_cases(test_files, cfg, tracer=tracer) File "test.py", line 267, in get_test_cases module = import_module(file, cfg, tracer=tracer) File "test.py", line 210, in import_module mod = __import__(modname) File "/mnt/gen2/TmpDir/portage/dev-python/lxml-cffi-3.0.2/work/lxml-cffi-3.0.2/src/lxml/html/__init__.py", line 42, in from lxml import etree File "/mnt/gen2/TmpDir/portage/dev-python/lxml-cffi-3.0.2/work/lxml-cffi-3.0.2/src/lxml/etree.py", line 2798, in from .xpath import XPath, ETXPath, XPathElementEvaluator, XPathDocumentEvaluator File "/mnt/gen2/TmpDir/portage/dev-python/lxml-cffi-3.0.2/work/lxml-cffi-3.0.2/src/lxml/xpath.py", line 15, in from .xslt import LIBXSLT_VERSION File "/mnt/gen2/TmpDir/portage/dev-python/lxml-cffi-3.0.2/work/lxml-cffi-3.0.2/src/lxml/xslt.py", line 165, in XSLT_DOC_DEFAULT_LOADER = xslt.xsltDocDefaultLoader AttributeError: 'module' object has no attribute 'xsltDocDefaultLoader' work/lxml-cffi-3.0.2 $ find . -name xslt.py ./src/lxml/xslt.py ./src/lxml/includes/xslt.py work/lxml-cffi-3.0.2 $ PYTHONPATH=src/ python2.7 test.py Comparing with ElementTree 1.3.0 TESTED VERSION: 3.0.2 Python: sys.version_info(major=2, minor=7, micro=3, releaselevel='final', serial=0) lxml.etree: (3, 0, 2, 0) libxml used: (2, 9, 0) libxml compiled: (2, 9, 0) libxslt used: (1, 1, 28) libxslt compiled: (1, 1, 28) FAIL: txt (element_classes) Traceback (most recent call last): File "/usr/lib64/python2.7/unittest/case.py", line 327, in run testMethod() File "/usr/lib64/python2.7/doctest.py", line 2201, in runTest raise self.failureException(self.format_failure(new.getvalue())) AssertionError: Failed doctest test for element_classes.txt File "/mnt/gen2/TmpDir/portage/dev-python/lxml-cffi-3.0.2/work/lxml-cffi-3.0.2/doc/element_classes.txt", line 0, in txt ---------------------------------------------------------------------- File "/mnt/gen2/TmpDir/portage/dev-python/lxml-cffi-3.0.2/work/lxml-cffi-3.0.2/doc/element_classes.txt", line 304, in element_classes.txt Failed example: a.honking # id does not match ! Expected: Traceback (most recent call last): AttributeError: '_Element' object has no attribute 'honking' Got: Traceback (most recent call last): File "/usr/lib64/python2.7/doctest.py", line 1289, in __run compileflags, 1) in test.globs File "", line 1, in a.honking # id does not match ! AttributeError: 'lxml.etree._Element' object has no attribute 'honking' py2.7 doesn't have problem with it, and I'm sure with pypy2.0 beta 1 installed it had no issue with import paths instead failing on the 3 vs 4 args issue in a data structure 'struct. ./src/lxml/xslt.py imports from with '.includes import xslt' but it doesn't. ---------- assignedto: amaury messages: 5669 nosy: amaury, idella5, pypy-issue priority: performance bug status: chatting title: apparent regression in import in py2.0 b2 ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 17:31:41 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Sun, 05 May 2013 15:31:41 +0000 Subject: [pypy-issue] [issue1475] Needed: __pypy__.add_memory_pressure()? In-Reply-To: <1367757719.47.0.0892814490429.issue1475@bugs.pypy.org> Message-ID: <1367767901.28.0.616401863175.issue1475@bugs.pypy.org> Amaury Forgeot d Arc added the comment: Another use case: a pygame.Surface contains a large cdata, and successive operations on Surfaces generate many intermediate objects which keep the memory alive until the next gc. ---------- nosy: +amaury ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 17:36:36 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Sun, 05 May 2013 15:36:36 +0000 Subject: [pypy-issue] [issue1476] apparent regression in import in py2.0 b2 In-Reply-To: <1367765341.24.0.0682227490432.issue1476@bugs.pypy.org> Message-ID: <1367768196.29.0.796478998983.issue1476@bugs.pypy.org> Amaury Forgeot d Arc added the comment: It's a regression in cffi I filed yesterday: https://bitbucket.org/cffi/cffi/issue/81/exported-variables-are-not-listed-in-dir ---------- priority: performance bug -> bug status: chatting -> duplicate ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 5 19:28:23 2013 From: tracker at bugs.pypy.org (Raymond Hettinger) Date: Sun, 05 May 2013 17:28:23 +0000 Subject: [pypy-issue] [issue1474] Random seed() and random() not always matching CPython In-Reply-To: <1367740995.74.0.139717490677.issue1474@bugs.pypy.org> Message-ID: <1367774903.88.0.383286219839.issue1474@bugs.pypy.org> Raymond Hettinger added the comment: Thanks Armin! ---------- status: resolved -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Mon May 6 07:50:38 2013 From: tracker at bugs.pypy.org (Ian Delaney) Date: Mon, 06 May 2013 05:50:38 +0000 Subject: [pypy-issue] [issue1476] apparent regression in import in py2.0 b2 In-Reply-To: <1367765341.24.0.0682227490432.issue1476@bugs.pypy.org> Message-ID: <1367819438.86.0.559528326934.issue1476@bugs.pypy.org> Ian Delaney added the comment: I see, thanks amaury ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 02:19:14 2013 From: tracker at bugs.pypy.org (Jim Kelly) Date: Tue, 07 May 2013 00:19:14 +0000 Subject: [pypy-issue] [issue1477] datetime issues in CPython vs PyPy In-Reply-To: <1367885954.82.0.578353818258.issue1477@bugs.pypy.org> Message-ID: <1367885954.82.0.578353818258.issue1477@bugs.pypy.org> New submission from Jim Kelly : So I had mentioned on twitter that I was seeing that CPython was significantly faster than PyPy in some performance testing I was doing. Essentially I was just pushing data as fast as possible through this app to get a sense of its throughput limits. Romain Guillebert requested I file a bug. I tracked this down a bit, and it seems to me the issue was actually the scaffolding around my tests. I basically had a set of sample data and was runnign a while look for a set period of time using datetime, then randomly selected sample data and ran it through my program, rinse, repeat. At any rate, looking at CProfile, it seems that datetime calls are taking a long time to execute. I tried this both on 1.9 and 2.0beta2, and this is in comparison to CPython 2.7.2. I've attached the simple test case I created. You'll note the "datetime.datetime.now() < (startTime + datetime.timedelta(minutes=2))" call is no longer part of the while loop, so as to make it more apparent what was running longer, while still executing it. Here are the various results. As you can see 2 beta2 is substantially faster than 1.9, but both are slower than CPython. Thanks! CPython 2.7.2: 1000003 function calls in 4.818 seconds Ordered by: cumulative time ncalls tottime percall cumtime percall filename:lineno(function) 1 2.878 2.878 4.818 4.818 mytest.py:6(TestFunc) 1000001 1.940 0.000 1.940 0.000 {built-in method now} 1 0.000 0.000 0.000 0.000 {method 'disable' of '_lsprof.Profiler' objects} PyPy 1.9: 87000029 function calls in 179.627 seconds Ordered by: cumulative time ncalls tottime percall cumtime percall filename:lineno(function) 1 0.489 0.489 179.627 179.627 mytest.py:6(TestFunc) 1000001 1.039 0.000 79.641 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:1543(now) 1000001 4.021 0.000 77.711 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:1492(fromtimestamp) 1000000 44.124 0.000 69.953 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:446(__new__) 2000001 5.676 0.000 48.972 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:1445(__new__) 1000001 5.792 0.000 44.650 0.000 {time.localtime} 1000001 30.022 0.000 38.858 0.000 /Users/pypy/buildslave/pypy-c-jit- macosx-x86-64/build/lib_pypy/_structseq.py:71(structseq_new) 2000001 27.699 0.000 38.829 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:759(__new__) 1000000 1.181 0.000 26.775 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:1838(__add__) 33000008 16.083 0.000 16.083 0.000 {isinstance} 3000001 10.666 0.000 10.666 0.000 {method '__new__' of 'type' objects} 9000000 5.756 0.000 5.756 0.000 {abs} 1000001 5.087 0.000 5.087 0.000 {method '__setattr__' of 'object' objects} 6000000 4.154 0.000 4.154 0.000 {divmod} 2000001 2.753 0.000 4.092 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:285(_check_time_fields) 2000001 3.785 0.000 3.785 0.000 {round} 1000001 3.431 0.000 3.431 0.000 {method '__new__' of '_structseq.structseqtype' objects} 1000000 0.516 0.000 2.769 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:1783(__lt__) 2000001 1.699 0.000 2.424 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:273(_check_date_fields) 1000000 0.647 0.000 2.065 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:1807(_cmp) 1000001 2.009 0.000 2.009 0.000 {min} 1000000 1.219 0.000 1.219 0.000 {cmp} 1000001 0.891 0.000 0.891 0.000 {time.time} 1000000 0.403 0.000 0.624 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:345(__init__) 2000002 0.318 0.000 0.318 0.000 {len} 3000001 0.048 0.000 0.048 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:56(_days_in_month) 3000002 0.043 0.000 0.043 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:298(_check_tzinfo_arg) 1000000 0.020 0.000 0.020 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:594(seconds) 1000000 0.020 0.000 0.020 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:599(microseconds) 1000000 0.019 0.000 0.019 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:589(days) 1000000 0.019 0.000 0.019 0.000 /usr/local/Cellar/pypy/1.9/lib_pypy/datetime.py:957(_checkOverflow) 1 0.000 0.000 0.000 0.000 {method 'disable' of '_lsprof.Profiler' objects} PyPy 2 beta2: 100000037 function calls in 97.047 seconds Ordered by: cumulative time ncalls tottime percall cumtime percall filename:lineno(function) 1 0.355 0.355 97.047 97.047 mytest.py:6(TestFunc) 1000000 42.643 0.000 67.528 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:436(__new__) 1000001 0.878 0.000 18.800 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:1534(now) 1000001 1.575 0.000 17.231 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:1489(fromtimestamp) 32000008 15.039 0.000 15.039 0.000 {isinstance} 2000001 4.595 0.000 13.873 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:1441(__new__) 1000000 1.085 0.000 9.230 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:1841(__add__) 1000001 2.137 0.000 7.435 0.000 {time.localtime} 14000007 3.853 0.000 6.044 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:272(_check_int_field) 1000001 3.542 0.000 5.297 0.000 /Users/pypy/buildslave/pypy-c-jit- macosx-x86-64/build/lib_pypy/_structseq.py:70(structseq_new) 8000000 4.659 0.000 4.659 0.000 {abs} 2000001 1.071 0.000 4.527 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:300(_check_time_fields) 7000001 4.163 0.000 4.163 0.000 {divmod} 3000001 3.951 0.000 3.951 0.000 {method '__new__' of 'type' objects} 2000001 0.389 0.000 3.012 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:287(_check_date_fields) 2000001 2.525 0.000 2.960 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:27(_round) 1000000 0.349 0.000 1.133 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:1787(__lt__) 1000001 0.777 0.000 0.777 0.000 {method '__new__' of '_structseq.structseqtype' objects} 1000001 0.699 0.000 0.699 0.000 {method '__setattr__' of 'object' objects} 1000001 0.692 0.000 0.692 0.000 {time.time} 1000000 0.381 0.000 0.614 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:360(__init__) 1000000 0.378 0.000 0.559 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:1811(_cmp) 2000001 0.435 0.000 0.435 0.000 {math.floor} 1000001 0.404 0.000 0.404 0.000 {min} 2000002 0.280 0.000 0.280 0.000 {len} 3000001 0.055 0.000 0.055 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:61(_days_in_month) 3000002 0.038 0.000 0.038 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:315(_check_tzinfo_arg) 1000000 0.024 0.000 0.024 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:24(_cmp) 1000000 0.020 0.000 0.020 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:575(seconds) 1000000 0.018 0.000 0.018 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:946(_checkOverflow) 1000000 0.018 0.000 0.018 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:580(microseconds) 1000000 0.017 0.000 0.017 0.000 /usr/local/pypy-2.0- beta2/lib_pypy/datetime.py:570(days) 1 0.000 0.000 0.000 0.000 {method 'disable' of '_lsprof.Profiler' objects} ---------- files: mytest.py messages: 5674 nosy: jkelly, pypy-issue priority: performance bug release: 2.0 status: unread title: datetime issues in CPython vs PyPy ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 08:36:02 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Tue, 07 May 2013 06:36:02 +0000 Subject: [pypy-issue] [issue1477] datetime issues in CPython vs PyPy In-Reply-To: <1367885954.82.0.578353818258.issue1477@bugs.pypy.org> Message-ID: <1367908562.55.0.842655653449.issue1477@bugs.pypy.org> Amaury Forgeot d Arc added the comment: datetime is a C module in CPython, but a pure-Python module in PyPy. This makes a huge difference for CProfile, which traces Python code and not C functions... timeit yields very different results: pypy is twice faster than CPython2.7! python -m timeit -s "import datetime; startTime=datetime.datetime.now()" "datetime.datetime.now() < (startTime + datetime.timedelta(minutes=2))" ---------- nosy: +amaury status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 12:03:23 2013 From: tracker at bugs.pypy.org (Fijal) Date: Tue, 07 May 2013 10:03:23 +0000 Subject: [pypy-issue] [issue1477] datetime issues in CPython vs PyPy In-Reply-To: <1367885954.82.0.578353818258.issue1477@bugs.pypy.org> Message-ID: <1367921003.72.0.47884120818.issue1477@bugs.pypy.org> Fijal added the comment: running pypy under cProfile and comparing it with C code in CPython is not completely fair. Please run your benchmarks without cProfile for a comparison. ---------- nosy: +fijal ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 15:00:34 2013 From: tracker at bugs.pypy.org (Jim Kelly) Date: Tue, 07 May 2013 13:00:34 +0000 Subject: [pypy-issue] [issue1477] datetime issues in CPython vs PyPy In-Reply-To: <1367885954.82.0.578353818258.issue1477@bugs.pypy.org> Message-ID: <1367931634.6.0.598901807536.issue1477@bugs.pypy.org> Jim Kelly added the comment: Using the Profile class instead of CProfile makes the discrepancy even worse, and seems slower to execute. Is there a method you suggest for profiling my application to determine where (or if - I guess it's possible this is erroneous if cProfile is misreporting total time) the slowdown is that would be more equitable to PyPy? Thanks! ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 15:32:54 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Tue, 07 May 2013 13:32:54 +0000 Subject: [pypy-issue] [issue1476] apparent regression in import in py2.0 b2 In-Reply-To: <1367765341.24.0.0682227490432.issue1476@bugs.pypy.org> Message-ID: <1367933574.75.0.899052176797.issue1476@bugs.pypy.org> Armin Rigo added the comment: amaury: can you explain how missing a dir() can cause AttributeErrors like these? ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 15:37:22 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Tue, 07 May 2013 13:37:22 +0000 Subject: [pypy-issue] [issue1477] datetime issues in CPython vs PyPy In-Reply-To: <1367885954.82.0.578353818258.issue1477@bugs.pypy.org> Message-ID: <1367933842.25.0.776795151695.issue1477@bugs.pypy.org> Amaury Forgeot d Arc added the comment: A profiler is designed to show where execution time is spent within one run. And with PyPy, the line-by-line tracing is likely to kill any optimization made by the JIT compiler. To compare execution of different systems, don't use cProfile, just measure the time it takes to execute your function (in a loop, probably). For my part, I use the Python timeit module, or just the unix "time" command. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 16:42:14 2013 From: tracker at bugs.pypy.org (Jim Kelly) Date: Tue, 07 May 2013 14:42:14 +0000 Subject: [pypy-issue] [issue1477] datetime issues in CPython vs PyPy In-Reply-To: <1367885954.82.0.578353818258.issue1477@bugs.pypy.org> Message-ID: <1367937734.88.0.71870827276.issue1477@bugs.pypy.org> Jim Kelly added the comment: Ah... there it is. Sorry, once you said, "the line-by-line tracing is likely to kill any optimization made by the JIT compiler" it's pretty obvious. Sorry to waste your time. :/ Thanks for your help and the suggestion of the timeit module, I hadn't seen that before. ---------- status: chatting -> invalid ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 17:17:39 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Tue, 07 May 2013 15:17:39 +0000 Subject: [pypy-issue] [issue1476] apparent regression in import in py2.0 b2 In-Reply-To: <1367765341.24.0.0682227490432.issue1476@bugs.pypy.org> Message-ID: <1367939859.33.0.710208559041.issue1476@bugs.pypy.org> Amaury Forgeot d Arc added the comment: I write my cffi helper modules like this: xmllib.py:: ...cffi declarations... ffilib = ffi.verify(...) for name in dir(ffilib): # Export all public declared names if name.startswith(('xml', 'xslt')): globals()[name] = getattr(ffilib, name) Then you can simply access functions and defines with code like:: import xmllib xmllib.XML_TEXT_NODE This used to work with variables. Of course, the workaround is to do:: xsltDocDefaultLoader = mylib.xsltDocDefaultLoader but for each variable. [Now, I realize that "mylib" is itself a Module... maybe there is nice a pattern which hacks sys.modules.] ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 7 22:11:18 2013 From: tracker at bugs.pypy.org (Markus) Date: Tue, 07 May 2013 20:11:18 +0000 Subject: [pypy-issue] [issue1478] Pickle from file slow In-Reply-To: <1367957478.65.0.675408779357.issue1478@bugs.pypy.org> Message-ID: <1367957478.65.0.675408779357.issue1478@bugs.pypy.org> New submission from Markus : Reading pickled data by Pickle.load is extremely slow compared to a load upfront and then using a loads() on it. I will come up with a testcase for it ---------- messages: 5682 nosy: jokey, pypy-issue priority: performance bug release: 2.0 status: unread title: Pickle from file slow ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 05:13:00 2013 From: tracker at bugs.pypy.org (David Cramer) Date: Wed, 08 May 2013 03:13:00 +0000 Subject: [pypy-issue] [issue1479] Uses all my memory and slow In-Reply-To: <1367982780.12.0.260641386288.issue1479@bugs.pypy.org> Message-ID: <1367982780.12.0.260641386288.issue1479@bugs.pypy.org> New submission from David Cramer : http://github.com/getsentry/sentry https://gist.github.com/dcramer/5537828 You'll need to install pysycopg2cffi otherwise Sentry wont load at all. I just stuff the monkey patch commands in the sentry.conf.py This benchmark probably sucks, but I'd love to see if I could get some reasonable gains out of PyPy. Some notes: - the login view does nothing, so probably awful test - it does nothing, so it should be faster - it does nothing, so it should not eventually climb to 500mb of mem ? (req/s is my primary goal here, though .5gb per process is pretty intense on the memory use) ---------- messages: 5683 nosy: dcramer, pypy-issue priority: performance bug status: unread title: Uses all my memory and slow ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 05:13:29 2013 From: tracker at bugs.pypy.org (David Cramer) Date: Wed, 08 May 2013 03:13:29 +0000 Subject: [pypy-issue] [issue1479] Uses all my memory and slow In-Reply-To: <1367982780.12.0.260641386288.issue1479@bugs.pypy.org> Message-ID: <1367982809.6.0.00383065300354.issue1479@bugs.pypy.org> David Cramer added the comment: Actually I lied about psycopg2cffi, it uses sqlite by default so should be ok ---------- status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 07:48:37 2013 From: tracker at bugs.pypy.org (Ian Delaney) Date: Wed, 08 May 2013 05:48:37 +0000 Subject: [pypy-issue] [issue1480] PyThreadState members lacking for gevent/callbacks.c In-Reply-To: <1367992117.51.0.188429863767.issue1480@bugs.pypy.org> Message-ID: <1367992117.51.0.188429863767.issue1480@bugs.pypy.org> New submission from Ian Delaney : current 2.0 beta2 Can't build gevent-1.0rc2 with pypy2 running build_ext building 'gevent.core' extension creating /mnt/gen2/TmpDir/portage/dev-python/gevent-1.0_rc2/work/gevent-1.0rc2-pypy2_0/temp.linux-x86_64-2.7 creating /mnt/gen2/TmpDir/portage/dev-python/gevent-1.0_rc2/work/gevent-1.0rc2-pypy2_0/temp.linux-x86_64-2.7/gevent cc -march=athlon64 -pipe -fomit-frame-pointer -O2 -fno-strict-aliasing -fPIC -I/usr/lib64/pypy2.0/include -c gevent/gevent.core.c -o /mnt/gen2/TmpDir/portage/dev-python/gevent-1.0_rc2/work/gevent-1.0rc2-pypy2_0/temp.linux-x86_64-2.7/gevent/gevent.core.o In file included from gevent/gevent.core.c:38714:0: gevent/callbacks.c: In function ?gevent_handle_error?: gevent/callbacks.c:8:18: error: ?PyThreadState? has no member named ?curexc_type? gevent/callbacks.c:11:19: error: ?PyThreadState? has no member named ?curexc_value? gevent/callbacks.c:12:23: error: ?The ? has no member named ?curexc_traceback? error: command 'cc' failed with exit status 1 The PyThreadState appears lacking. Can this be done via cffi? ---------- messages: 5685 nosy: idella5, pypy-issue priority: performance bug release: 2.0 status: unread title: PyThreadState members lacking for gevent/callbacks.c ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 10:19:29 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Wed, 08 May 2013 08:19:29 +0000 Subject: [pypy-issue] [issue1480] PyThreadState members lacking for gevent/callbacks.c In-Reply-To: <1367992117.51.0.188429863767.issue1480@bugs.pypy.org> Message-ID: <1368001169.03.0.587862720936.issue1480@bugs.pypy.org> Amaury Forgeot d Arc added the comment: The pypy emulation of PyThreadState cannot support these members. There are public functions to get the "hot exception" state, PyErr_Fetch and PyErr_Restore; gevent could be updated to use them. Anyway you should use the cffi version of gevent, as described in http://doc.pypy.org/en/latest/release-2.0.0-beta2.html ---------- nosy: +amaury status: unread -> wontfix ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 11:22:51 2013 From: tracker at bugs.pypy.org (Ian Delaney) Date: Wed, 08 May 2013 09:22:51 +0000 Subject: [pypy-issue] [issue1476] apparent regression in import in py2.0 b2 In-Reply-To: <1367765341.24.0.0682227490432.issue1476@bugs.pypy.org> Message-ID: <1368004971.92.0.276550701368.issue1476@bugs.pypy.org> Ian Delaney added the comment: With the change in builtin cffi in pypy2.0 b2 from https://bitbucket.org/cffi/cffi/issue/81/exported-variables-are-not-listed-in-dir I get testuser at archtester ~/Downloads/lxml-cffi-3.0.2 $ time PYTHONPATH=src nosetests-pypy-c2.0 ./src/lxml/tests ............E..............S.................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................F..................................................................................................................................................................................................................................F...............................................................................................................................EEEF..E......E...E........................................EEEF.FFFEFF.FF.........EEEEEE.EEEEEE..EEE.EEEEEEEEEE....F...........EE.....F.EE.F.EEEEEE...........EFEEE.EE...FE.EEEE.EEEEEEF..EEE.............................E................................Segmentation fault real 1m31.792s user 1m30.932s sys 0m0.230s similar from testuser at archtester ~/Downloads/lxml-cffi-3.0.2 $ time PYTHONPATH=src pypy-c2.0 test.py Skipping tests in lxml.cssselect - external cssselect package is not installed Comparing with ElementTree 1.3.0 Comparing with cElementTree 1.3.0 TESTED VERSION: 3.0.2 Python: (major=2, minor=7, micro=3, releaselevel='final', serial=42) lxml.etree: (3, 0, 2, 0) libxml used: (2, 9, 0) libxml compiled: (2, 7, 0) libxslt used: (1, 1, 28) libxslt compiled: (1, 1, 28) followed by the mass of errors Segmentation fault real 1m31.622s user 1m30.896s sys 0m0.185s Segmentation fault real 1m31.622s user 1m30.896s sys 0m0.185s ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 11:32:59 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Wed, 08 May 2013 09:32:59 +0000 Subject: [pypy-issue] [issue1476] apparent regression in import in py2.0 b2 In-Reply-To: <1367765341.24.0.0682227490432.issue1476@bugs.pypy.org> Message-ID: <1368005579.63.0.432904350676.issue1476@bugs.pypy.org> Amaury Forgeot d Arc added the comment: Can we move this discussion elsewhere? This is not a PyPy issue, but a lxml-cffi one. https://github.com/amauryfa/lxml/issues ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 12:21:40 2013 From: tracker at bugs.pypy.org (Ian Delaney) Date: Wed, 08 May 2013 10:21:40 +0000 Subject: [pypy-issue] [issue1480] PyThreadState members lacking for gevent/callbacks.c In-Reply-To: <1367992117.51.0.188429863767.issue1480@bugs.pypy.org> Message-ID: <1368008500.94.0.57535887011.issue1480@bugs.pypy.org> Ian Delaney added the comment: hmm right. on it thanks ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 13:52:54 2013 From: tracker at bugs.pypy.org (Fijal) Date: Wed, 08 May 2013 11:52:54 +0000 Subject: [pypy-issue] [issue1480] PyThreadState members lacking for gevent/callbacks.c In-Reply-To: <1367992117.51.0.188429863767.issue1480@bugs.pypy.org> Message-ID: <1368013974.21.0.00531712159405.issue1480@bugs.pypy.org> Fijal added the comment: There is really no way you can get gevent working this way. I suggest looking at pypycore + pypy-hacks branch of gevent https://github.com/gevent-on-pypy/pypycore https://github.com/schmir/gevent/tree/pypy-hacks it's not finished, but it's a start ---------- nosy: +fijal ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 17:06:41 2013 From: tracker at bugs.pypy.org (tanguofu) Date: Wed, 08 May 2013 15:06:41 +0000 Subject: [pypy-issue] =?utf-8?q?=5Bissue1481=5D_complie_error_=3A=22_src/h?= =?utf-8?b?YW5kbGUuYzoyMDo1OiBlcnJvcjog4oCYX1B5X0RFQ19SRUZUT1RBTOKAmSB1?= =?utf-8?q?ndeclared_=28first_use_in_this_function=29=22_when_use_pip_inst?= =?utf-8?q?all_pyuv?= In-Reply-To: <1368025601.01.0.647150945549.issue1481@bugs.pypy.org> Message-ID: <1368025601.01.0.647150945549.issue1481@bugs.pypy.org> New submission from tanguofu : Installing collected packages: pyuv Running setup.py install for pyuv libuv needs to be compiled. Building libuv... building 'pyuv' extension cc -fPIC -Wimplicit -DMODULE_VERSION=0.10.3 -DLIBUV_REVISION=9b801d5 -I/home/tanguofu/pypy/include -Ideps/libuv/include -c src/pyuv.c -o build/temp.linux-x86_64-2.7/src/pyuv.o In file included from src/pyuv.c:7:0: src/handle.c: In function ?resurrect_object?: src/handle.c:20:5: error: ?_Py_DEC_REFTOTAL? undeclared (first use in this function) src/handle.c:20:5: note: each undeclared identifier is reported only once for each function it appears in error: command 'cc' failed with exit status 1 Complete output from command /home/tanguofu/pypy/bin/pypy -c "import setuptools;__file__='/tmp/pip-build-tanguofu/pyuv/setup.py';exec(compile(open(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" install --record /tmp/pip-T_11x2-record/install-record.txt --single-version-externally-managed: running install running build running build_ext libuv needs to be compiled. Building libuv... building 'pyuv' extension creating build creating build/temp.linux-x86_64-2.7 creating build/temp.linux-x86_64-2.7/src cc -fPIC -Wimplicit -DMODULE_VERSION=0.10.3 -DLIBUV_REVISION=9b801d5 -I/home/tanguofu/pypy/include -Ideps/libuv/include -c src/pyuv.c -o build/temp.linux-x86_64-2.7/src/pyuv.o In file included from src/pyuv.c:7:0: src/handle.c: In function ?resurrect_object?: src/handle.c:20:5: error: ?_Py_DEC_REFTOTAL? undeclared (first use in this function) src/handle.c:20:5: note: each undeclared identifier is reported only once for each function it appears in error: command 'cc' failed with exit status 1 ---------------------------------------- Command /home/tanguofu/pypy/bin/pypy -c "import setuptools;__file__='/tmp/pip-build-tanguofu/pyuv/setup.py';exec(compile(open(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" install --record /tmp/pip-T_11x2-record/install-record.txt --single-version-externally-managed failed with error code 1 in /tmp/pip-build-tanguofu/pyuv Storing complete log in /home/tanguofu/.pip/pip.log ---------- messages: 5691 nosy: pypy-issue, tanguofu priority: feature release: 2.0 status: unread title: complie error :" src/handle.c:20:5: error: ?_Py_DEC_REFTOTAL? undeclared (first use in this function)" when use pip install pyuv ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 8 17:27:29 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Wed, 08 May 2013 15:27:29 +0000 Subject: [pypy-issue] =?utf-8?q?=5Bissue1481=5D_complie_error_=3A=22_src/h?= =?utf-8?b?YW5kbGUuYzoyMDo1OiBlcnJvcjog4oCYX1B5X0RFQ19SRUZUT1RBTOKAmSB1?= =?utf-8?q?ndeclared_=28first_use_in_this_function=29=22_when_use_pip_inst?= =?utf-8?q?all_pyuv?= In-Reply-To: <1368025601.01.0.647150945549.issue1481@bugs.pypy.org> Message-ID: <1368026849.42.0.574210531933.issue1481@bugs.pypy.org> Amaury Forgeot d Arc added the comment: You could try to #define _Py_DEC_REFTOTAL Or completely remove the line; it's only for statistics in debug builds (sys.gettotalrefcount). This said, this 'resurrect object' function which plays with reference counts is deeply incompatible with PyPy and its garbage collection scheme. Don't expect the resulting code to work correctly. By the way, see https://github.com/saghul/pyuv/issues/49 """ ...it seems there is a problem with callbacks... """ You should look for a port of pyuv with the cffi interface. ---------- nosy: +amaury status: unread -> wontfix ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 9 07:34:15 2013 From: tracker at bugs.pypy.org (sn6uv) Date: Thu, 09 May 2013 05:34:15 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> New submission from sn6uv <16sn6uv at gmail.com>: Raised when running the Sympy tests (specifically sympy/geometry/tests/test_geometry.py). Here's the traceback for PyPy 2.0.0-beta2 (line numbers differ slightly for the nightly): sympy/geometry/tests/test_geometry.py[20] ..f.............RPython traceback: File "rpython_jit_metainterp_optimizeopt___init__.c", line 328, in optimize_trace File "rpython_jit_metainterp_optimizeopt_unroll.c", line 2350, in UnrollOptimizer_propagate_all_forward File "rpython_jit_metainterp_optimizeopt_optimizer.c", line 4062, in Optimizer_propagate_all_forward File "rpython_jit_metainterp_optimizeopt_intbounds.c", line 14218, in OptIntBounds_optimize_ARRAYLEN_GC File "rpython_jit_metainterp_optimizeopt_virtualize.c", line 979, in OptVirtualize_optimize_ARRAYLEN_GC Fatal RPython error: NotImplementedError Affects both 2.0.0-beta2: Python 2.7.3 (c13457babbfb, Apr 22 2013, 17:59:01) [PyPy 2.0.0-beta2 with GCC 4.8.0], and the latest nightly build: Python 2.7.3 (0423d458c98d, May 08 2013, 22:00:18) [PyPy 2.0.0-beta2 with GCC 4.6.3]. Both with and without the JIT (i.e. with and without '--jit off'). ---------- messages: 5693 nosy: pypy-issue, sn6uv priority: bug release: 2.0 status: unread title: Fatal RPython error ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 9 07:39:02 2013 From: tracker at bugs.pypy.org (Alex Gaynor) Date: Thu, 09 May 2013 05:39:02 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368077942.23.0.379076690604.issue1482@bugs.pypy.org> Alex Gaynor added the comment: Are you sure --jit off is correcly being applied to the proecss you're running? It should be totally impossibly for this to occur with --jit off. All that said this appears to be because getlength() is called on something that's not a virtual array class, my first guess is that it's a VRawSliceValue, but I have no evidence for that. ---------- nosy: +agaynor status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 9 23:31:29 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Thu, 09 May 2013 21:31:29 +0000 Subject: [pypy-issue] [issue1188] Support -OO In-Reply-To: <1340208529.43.0.593686088035.issue1188@bugs.pypy.org> Message-ID: <1368135089.26.0.510360323838.issue1188@bugs.pypy.org> Armin Rigo added the comment: The recent version is elsewhere: https://bitbucket.org/pypy/pypy/pull-request/149/implement-o-and-oo-flags/diff ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 10 02:14:45 2013 From: tracker at bugs.pypy.org (Tomer Chachamu) Date: Fri, 10 May 2013 00:14:45 +0000 Subject: [pypy-issue] [issue1483] io.IOBase is missing some stub implementations In-Reply-To: <1368144885.41.0.996527671116.issue1483@bugs.pypy.org> Message-ID: <1368144885.41.0.996527671116.issue1483@bugs.pypy.org> New submission from Tomer Chachamu : This is a minor issue that affected some unit tests I had. The Python 3.2 implementations are here: https://github.com/jonashaag/cpython/blob/v3.2.4/Modules/_io/iobase.c#L88 A patch is attached. ---------- files: pypy_io.patch messages: 5696 nosy: pypy-issue, tomer priority: bug status: unread title: io.IOBase is missing some stub implementations ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 10 04:16:59 2013 From: tracker at bugs.pypy.org (Philip Jenvey) Date: Fri, 10 May 2013 02:16:59 +0000 Subject: [pypy-issue] [issue1483] io.IOBase is missing some stub implementations In-Reply-To: <1368144885.41.0.996527671116.issue1483@bugs.pypy.org> Message-ID: <1368152219.21.0.390885822177.issue1483@bugs.pypy.org> Philip Jenvey added the comment: I actually can't reproduce the AttributeError's you described on IRC: Python 3.2.3 (9d49e7e670ec+, May 09 2013, 20:56:06) [PyPy 2.0.0-beta2 with GCC 4.2.1 Compatible Clang Compiler] on darwin Type "help", "copyright", "credits" or "license" for more information. And now for something completely different: ``Seeing as the laws of physics are against you, you have to carefully consider your scope so that your goals are reasonable.'' >>>> import _io; file = _io._IOBase() >>>> file.seek() Traceback (most recent call last): File "", line 1, in io.UnsupportedOperation: seek >>>> file.truncate() Traceback (most recent call last): File "", line 1, in io.UnsupportedOperation: truncate >>>> file.fileno() Traceback (most recent call last): File "", line 1, in io.UnsupportedOperation: fileno IOBase already has implementations of these methods that call self._unsupportedoperation. The only thing I noticed is the default seek didn't accept any args: >>>> file.seek(0) Traceback (most recent call last): File "", line 1, in TypeError: seek() takes exactly 1 argument (2 given) >>>> file.seek(0, 1) Traceback (most recent call last): File "", line 1, in TypeError: seek() takes exactly 1 argument (3 given) So I committed a subset of your tests w/ a fix for that in 55096ba173ab Let me know if I've missed something ---------- status: unread -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 10 12:28:46 2013 From: tracker at bugs.pypy.org (Tomer Chachamu) Date: Fri, 10 May 2013 10:28:46 +0000 Subject: [pypy-issue] [issue1483] io.IOBase is missing some stub implementations In-Reply-To: <1368144885.41.0.996527671116.issue1483@bugs.pypy.org> Message-ID: <1368181726.57.0.318732881197.issue1483@bugs.pypy.org> Tomer Chachamu added the comment: I was on the default branch, not the py3k branch. ---------- status: resolved -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 06:29:20 2013 From: tracker at bugs.pypy.org (sn6uv) Date: Sat, 11 May 2013 04:29:20 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368246560.04.0.563694643664.issue1482@bugs.pypy.org> sn6uv <16sn6uv at gmail.com> added the comment: I'm fairly sure it is. Is there a way to test at runtime whether the JIT is enabled or not? Also, this bug is new in 2.0 (i.e. it's not it 1.9). ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 11:00:42 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sat, 11 May 2013 09:00:42 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368262842.27.0.66218714017.issue1482@bugs.pypy.org> Armin Rigo added the comment: agaynor: ARRAYLEN_GC cannot be applied on a raw array as far as I see. ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 11:13:15 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sat, 11 May 2013 09:13:15 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368263595.41.0.111728899196.issue1482@bugs.pypy.org> Armin Rigo added the comment: sn6uv: I cannot reproduce. Running "py.test test_geometry.py": test_geometry.py .....Fx............ and the failure is printed correctly at the end. Can you tell us more precisely which system you're running on, and the exact command-line? ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 11:15:02 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sat, 11 May 2013 09:15:02 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368263702.09.0.223610777521.issue1482@bugs.pypy.org> Armin Rigo added the comment: Also, the version of sympy. I tried with the one that comes with "pip install sympy" in a pypy virtualenv. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 11:30:38 2013 From: tracker at bugs.pypy.org (sn6uv) Date: Sat, 11 May 2013 09:30:38 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368264638.44.0.838177738973.issue1482@bugs.pypy.org> sn6uv <16sn6uv at gmail.com> added the comment: You're right. It seems to only be happening in the (unreleased) development version of sympy. My platform: 64bit Arch Linux $> uname -a Linux desktop 3.8.11-1-ARCH #1 SMP PREEMPT Wed May 1 20:18:57 CEST 2013 x86_64 GNU/Linux $> pypy --version Python 2.7.3 (c13457babbfb, Apr 22 2013, 17:59:01) [PyPy 2.0.0-beta2 with GCC 4.8.0] Steps to reproduce (should have included this earlier): >$ git clone https://github.com/sympy/sympy.git ... >$ cd sympy/ >$ pypy ... >>>> import sympy >>>> sympy.test("sympy/geometry/tests/test_geometry.py") ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 12:23:20 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sat, 11 May 2013 10:23:20 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368267800.84.0.879334720486.issue1482@bugs.pypy.org> Armin Rigo added the comment: I could reproduce a crash by running just "py.test", i.e. running all tests. Then it crashes somewhere around test_geometry, sometimes in the next few tests. Re "--jit off" not working: I think it's because py.test or someone starts a subprocess in which to run the tests. The subprocess is not passed the "--jit" option, of course... ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 12:47:17 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sat, 11 May 2013 10:47:17 +0000 Subject: [pypy-issue] [issue1483] io.IOBase is missing some stub implementations In-Reply-To: <1368144885.41.0.996527671116.issue1483@bugs.pypy.org> Message-ID: <1368269237.45.0.11771253747.issue1483@bugs.pypy.org> Armin Rigo added the comment: The default branch should be compared with CPython 2.7, not 3.2. Your initial comment led us think that these methods are a 3.2 feature. Is that wrong and is it also a feature present in CPython 2.7? ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 12:49:56 2013 From: tracker at bugs.pypy.org (Tomer Chachamu) Date: Sat, 11 May 2013 10:49:56 +0000 Subject: [pypy-issue] [issue1483] io.IOBase is missing some stub implementations In-Reply-To: <1368144885.41.0.996527671116.issue1483@bugs.pypy.org> Message-ID: <1368269396.73.0.548222058581.issue1483@bugs.pypy.org> Tomer Chachamu added the comment: Ah sorry for the confusion. Yes it's present in CPython 2.7 as well. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 11 13:42:15 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sat, 11 May 2013 11:42:15 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368272535.66.0.979609427149.issue1482@bugs.pypy.org> Armin Rigo added the comment: Fixed in 5c4cc250d96c. ---------- status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 12 03:18:38 2013 From: tracker at bugs.pypy.org (sn6uv) Date: Sun, 12 May 2013 01:18:38 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368321518.61.0.697640663722.issue1482@bugs.pypy.org> sn6uv <16sn6uv at gmail.com> added the comment: Awesome! I'll give this a go and report back if there are any issues. In regards to --jit off, you were right about it being caused by not passing the options to a subprocess. Thanks heaps! ---------- status: resolved -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 12 03:29:34 2013 From: tracker at bugs.pypy.org (sn6uv) Date: Sun, 12 May 2013 01:29:34 +0000 Subject: [pypy-issue] [issue1482] Fatal RPython error In-Reply-To: <1368077655.68.0.412279761672.issue1482@bugs.pypy.org> Message-ID: <1368322174.82.0.454973804388.issue1482@bugs.pypy.org> sn6uv <16sn6uv at gmail.com> added the comment: I seemed to have accidentally opened this issue again. [Closing] ---------- status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 12 09:44:44 2013 From: tracker at bugs.pypy.org (mikefc) Date: Sun, 12 May 2013 07:44:44 +0000 Subject: [pypy-issue] [issue1484] numpy pickling In-Reply-To: <1368344684.93.0.745354939033.issue1484@bugs.pypy.org> Message-ID: <1368344684.93.0.745354939033.issue1484@bugs.pypy.org> New submission from mikefc : Trying to use a multiprocessing.Pool() to map a function onto a list of numpy arrays fails in latest pypy nightly (OSX). Code attached. In python 2.7.x: Not MP: [0.0, 0.0, 0.0, 0.0] MP: [0.0, 0.0, 0.0, 0.0] In pypy (OSX nightly latest) Not MP: [0.0, 0.0, 0.0, 0.0] Process PoolWorker-1: Traceback (most recent call last): File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker ---------- files: bugreport.txt messages: 5710 nosy: mikefc, pypy-issue priority: bug status: unread title: numpy pickling ________________________________________ PyPy bug tracker ________________________________________ -------------- next part -------------- Not MP: [0.0, 0.0, 0.0, 0.0] Process PoolWorker-1: Traceback (most recent call last): File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker Process PoolWorker-2: Traceback (most recent call last): File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap put((job, i, result)) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/queues.py", line 390, in put return send(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 1417, in dumps self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker Pickler(file, protocol).dump(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 224, in dump self.save(obj) put((job, i, result)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/queues.py", line 390, in put f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple return send(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 1417, in dumps save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save rv = reduce(self.proto) TypeError: unsupported operand type for __getnewargs__: 'float64' Pickler(file, protocol).dump(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 224, in dump self.save(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save rv = reduce(self.proto) TypeError: unsupported operand type for __getnewargs__: 'float64' MP: Process PoolWorker-3: Traceback (most recent call last): File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker put((job, i, result)) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/queues.py", line 390, in put Process PoolWorker-4: Traceback (most recent call last): File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap return send(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 1417, in dumps self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker Pickler(file, protocol).dump(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 224, in dump self.save(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save put((job, i, result)) f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/queues.py", line 390, in put self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save rv = reduce(self.proto) TypeError: unsupported operand type for __getnewargs__: 'float64' return send(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 1417, in dumps Pickler(file, protocol).dump(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 224, in dump self.save(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save rv = reduce(self.proto) TypeError: unsupported operand type for __getnewargs__: 'float64' From tracker at bugs.pypy.org Sun May 12 09:45:21 2013 From: tracker at bugs.pypy.org (mikefc) Date: Sun, 12 May 2013 07:45:21 +0000 Subject: [pypy-issue] [issue1484] numpy pickling In-Reply-To: <1368344684.93.0.745354939033.issue1484@bugs.pypy.org> Message-ID: <1368344721.61.0.121371406061.issue1484@bugs.pypy.org> mikefc added the comment: code is here. ---------- status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 12 20:38:49 2013 From: tracker at bugs.pypy.org (kostia.lopuhin) Date: Sun, 12 May 2013 18:38:49 +0000 Subject: [pypy-issue] [issue1485] all is slow when passed a short generator, much faster with short list In-Reply-To: <1368383929.7.0.449993251863.issue1485@bugs.pypy.org> Message-ID: <1368383929.7.0.449993251863.issue1485@bugs.pypy.org> New submission from kostia.lopuhin : Attached code is 3-10 times slower if all is used like this: all(v == v0 for v in values[1:]) than if it is passed a list: all([v == v0 for v in values[1:]]) where values is a list of 4 elements. jitviewer shows a lot of extra stuff around call of "all" with generator argument (like call_may_force) ---------- files: all_slow.py messages: 5712 nosy: kostia.lopuhin, pypy-issue priority: performance bug status: unread title: all is slow when passed a short generator, much faster with short list ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 12 20:40:39 2013 From: tracker at bugs.pypy.org (kostia.lopuhin) Date: Sun, 12 May 2013 18:40:39 +0000 Subject: [pypy-issue] [issue1485] all is slow when passed a short generator, much faster with short list In-Reply-To: <1368383929.7.0.449993251863.issue1485@bugs.pypy.org> Message-ID: <1368384039.45.0.734637316722.issue1485@bugs.pypy.org> kostia.lopuhin added the comment: Run log (first fast, than slow) pypy all_slow.py 0.462735891342 34737 pypy all_slow.py 1.52023792267 34737 pypy Python 2.7.3 (b9c3566aa017, May 09 2013, 15:27:31) [PyPy 2.0.0 with GCC 4.2.1 Compatible Apple LLVM 4.2 (clang-425.0.27)] on darwin ---------- status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Mon May 13 01:12:12 2013 From: tracker at bugs.pypy.org (YAMAMOTO Takashi) Date: Sun, 12 May 2013 23:12:12 +0000 Subject: [pypy-issue] [issue1409] netbsd support In-Reply-To: <1361857361.33.0.161819094579.issue1409@bugs.pypy.org> Message-ID: <1368400332.22.0.111451630568.issue1409@bugs.pypy.org> YAMAMOTO Takashi added the comment: hi, what's a requirement for a buildbot machine? ---------- status: resolved -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 14 04:41:34 2013 From: tracker at bugs.pypy.org (mikefc) Date: Tue, 14 May 2013 02:41:34 +0000 Subject: [pypy-issue] [issue1484] numpy pickling In-Reply-To: <1368344684.93.0.745354939033.issue1484@bugs.pypy.org> Message-ID: <1368499294.27.0.634682879925.issue1484@bugs.pypy.org> mikefc added the comment: Theres a weird bug with pickling and numpy operations. New code and output attached. #======================================================== # Code #======================================================== import sys try: import numpypy as np except: import numpy as np import multiprocessing def f1(x): pass def f2(x): return 1 def f3(x): return x def f4(x): return x + 1 def f5(x): return np.sum(x) a = [np.zeros(3) for _ in xrange(4)] # All five functions work with the builtin map # but a multiprocessing map (which has to pickle objects) fails on f5 for map in (map, multiprocessing.Pool(2).map): print "f1:", map(f1, a); sys.stdout.flush() print "f2:", map(f2, a); sys.stdout.flush() print "f3:", map(f3, a); sys.stdout.flush() print "f4:", map(f4, a); sys.stdout.flush() print "f5:", map(f5, a); sys.stdout.flush() # FAILS with multiprocessing.map #======================================================== # Output (trimmed. full output attached) #======================================================== f1: [None, None, None, None] f2: [1, 1, 1, 1] f3: [array([ 0., 0., 0.]), array([ 0., 0., 0.]), array([ 0., 0., 0.]), array([ 0., 0., 0.])] f4: [array([ 1., 1., 1.]), array([ 1., 1., 1.]), array([ 1., 1., 1.]), array([ 1., 1., 1.])] f5: [0.0, 0.0, 0.0, 0.0] f1: [None, None, None, None] f2: [1, 1, 1, 1] f3: [array([ 0., 0., 0.]), array([ 0., 0., 0.]), array([ 0., 0., 0.]), array([ 0., 0., 0.])] f4: [array([ 1., 1., 1.]), array([ 1., 1., 1.]), array([ 1., 1., 1.]), array([ 1., 1., 1.])] ... File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save rv = reduce(self.proto) TypeError: unsupported operand type for __getnewargs__: 'float64' ________________________________________ PyPy bug tracker ________________________________________ -------------- next part -------------- f1: [None, None, None, None] f2: [1, 1, 1, 1] f3: [array([ 0., 0., 0.]), array([ 0., 0., 0.]), array([ 0., 0., 0.]), array([ 0., 0., 0.])] f4: [array([ 1., 1., 1.]), array([ 1., 1., 1.]), array([ 1., 1., 1.]), array([ 1., 1., 1.])] f5: [0.0, 0.0, 0.0, 0.0] f1: [None, None, None, None] f2: [1, 1, 1, 1] f3: [array([ 0., 0., 0.]), array([ 0., 0., 0.]), array([ 0., 0., 0.]), array([ 0., 0., 0.])] f4: [array([ 1., 1., 1.]), array([ 1., 1., 1.]), array([ 1., 1., 1.]), array([ 1., 1., 1.])] Process PoolWorker-1: Traceback (most recent call last): File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap Process PoolWorker-2: Traceback (most recent call last): File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker put((job, i, result)) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/queues.py", line 390, in put put((job, i, result)) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/queues.py", line 390, in put return send(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 1417, in dumps return send(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 1417, in dumps Pickler(file, protocol).dump(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 224, in dump Pickler(file, protocol).dump(obj) self.save(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 224, in dump File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self self.save(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list rv = reduce(self.proto) self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends TypeError: unsupported operand type for __getnewargs__: 'float64' save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save rv = reduce(self.proto) TypeError: unsupported operand type for __getnewargs__: 'float64' f5: Process PoolWorker-3: Traceback (most recent call last): File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap Process PoolWorker-4: self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run Traceback (most recent call last): self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 258, in _bootstrap put((job, i, result)) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/queues.py", line 390, in put self.run() File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/process.py", line 114, in run self._target(*self._args, **self._kwargs) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/pool.py", line 99, in worker return send(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 1417, in dumps put((job, i, result)) File "/Users/mike/python/pypy-latest/lib-python/2.7/multiprocessing/queues.py", line 390, in put return send(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 1417, in dumps Pickler(file, protocol).dump(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 224, in dump self.save(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends Pickler(file, protocol).dump(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 224, in dump save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save self.save(obj) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save rv = reduce(self.proto) f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple TypeError: unsupported operand type for __getnewargs__: 'float64' save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 548, in save_tuple save(element) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 286, in save f(self, obj) # Call unbound method with explicit self File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 600, in save_list self._batch_appends(iter(obj)) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 636, in _batch_appends save(tmp[0]) File "/Users/mike/python/pypy-latest/lib-python/2.7/pickle.py", line 306, in save rv = reduce(self.proto) TypeError: unsupported operand type for __getnewargs__: 'float64' From tracker at bugs.pypy.org Tue May 14 15:50:05 2013 From: tracker at bugs.pypy.org (jayqhacker) Date: Tue, 14 May 2013 13:50:05 +0000 Subject: [pypy-issue] [issue791] Simple wordcount is significantly slower and fatter than CPython In-Reply-To: <1310485078.56.0.855118946444.issue791@bugs.pypy.org> Message-ID: <1368539405.26.0.751659924408.issue791@bugs.pypy.org> jayqhacker added the comment: My hardware has apparently changed (Xeon E5620 @ 2.4 GHz, 12 MB cache; still Red Hat 5.8 64-bit), but 2.0 unfortunately brings no improvements for this case. For reasons mysterious to me, the relative gap between CPython and pypy is considerably widened, to almost 2X. (Note CPython and pypy 1.9 and 2.0 are compiled from source with gcc 4.1.2; 1.8 is the binary from pypy.org.) I'm also throwing in Jython just for giggles. Running on Oracle JDK 1.7.0_21, performance and memory usage vary wildly with choice of heap options, but -J-Xmx4G -J-XX:+UseConcMarkSweepGC seem to give the best results. CPython 2.7.4 : 20.4s 1400 MB CPython 2.7.3 : 20.5s 1400 MB PyPy 2.0 : 39.0s 1720 MB PyPy 1.9 : 39.0s 1700 MB PyPy 1.8 : 34.7s 1540 MB Jython 2.7b1 : 44.8s 3070 MB ---------- release: 1.9 -> 2.0 ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 14 21:35:33 2013 From: tracker at bugs.pypy.org (Fijal) Date: Tue, 14 May 2013 19:35:33 +0000 Subject: [pypy-issue] [issue1409] netbsd support In-Reply-To: <1361857361.33.0.161819094579.issue1409@bugs.pypy.org> Message-ID: <1368560133.92.0.94249083853.issue1409@bugs.pypy.org> Fijal added the comment: Anything would do, as long as it's maintained and the owner is responsive. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 14 22:47:18 2013 From: tracker at bugs.pypy.org (Bart Wiegmans) Date: Tue, 14 May 2013 20:47:18 +0000 Subject: [pypy-issue] [issue1475] Needed: __pypy__.add_memory_pressure()? In-Reply-To: <1367757719.47.0.0892814490429.issue1475@bugs.pypy.org> Message-ID: <1368564438.76.0.846688983616.issue1475@bugs.pypy.org> Bart Wiegmans added the comment: This diff add add_memory_pressure to the __pypy__ module, as well as a test. ---------- nosy: +brrt ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 15 00:19:35 2013 From: tracker at bugs.pypy.org (YAMAMOTO Takashi) Date: Tue, 14 May 2013 22:19:35 +0000 Subject: [pypy-issue] [issue1409] netbsd support In-Reply-To: <1361857361.33.0.161819094579.issue1409@bugs.pypy.org> Message-ID: <1368569975.85.0.607455977325.issue1409@bugs.pypy.org> YAMAMOTO Takashi added the comment: what would it run? does it need to accept connections from some central management? or does it run independently and upload results to somewhere or such? ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 15 02:00:58 2013 From: tracker at bugs.pypy.org (Fijal) Date: Wed, 15 May 2013 00:00:58 +0000 Subject: [pypy-issue] [issue1475] Needed: __pypy__.add_memory_pressure()? In-Reply-To: <1367757719.47.0.0892814490429.issue1475@bugs.pypy.org> Message-ID: <1368576058.41.0.559159643189.issue1475@bugs.pypy.org> Fijal added the comment: Applied, thanks! ---------- nosy: +fijal status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 15 14:13:59 2013 From: tracker at bugs.pypy.org (shenlong) Date: Wed, 15 May 2013 12:13:59 +0000 Subject: [pypy-issue] =?utf-8?q?=5Bissue1486=5D_Pypy_disk_IO_is_slower_tha?= =?utf-8?b?biBweXRob24yLjbvvJ8=?= In-Reply-To: <1368620039.71.0.0263891291277.issue1486@bugs.pypy.org> Message-ID: <1368620039.71.0.0263891291277.issue1486@bugs.pypy.org> New submission from shenlong : import os def main(): path = "XXX" dir_list=os.listdir(path) for i in dir_list: f=open(path+i) j=0 for line in f: j+=1 f.close if __name__=='__main__': main() with python2.6 time python26 -m profile tool3.py 2644 function calls in 1.898 CPU seconds Ordered by: standard name ncalls tottime percall cumtime percall filename:lineno(function) 1 0.000 0.000 1.897 1.897 :0(execfile) 1 0.001 0.001 0.001 0.001 :0(listdir) 2637 0.040 0.000 0.040 0.000 :0(open) 1 0.001 0.001 0.001 0.001 :0(setprofile) 1 0.000 0.000 1.897 1.897 :1() 1 0.000 0.000 1.898 1.898 profile:0(execfile('tool3.py')) 0 0.000 0.000 profile:0(profiler) 1 0.000 0.000 1.897 1.897 tool3.py:3() 1 1.856 1.856 1.897 1.897 tool3.py:5(main) real 0m1.968s user 0m1.518s sys 0m0.449s with pypy time pypy -m profile tool3.py 5 function calls in 3.809 seconds Ordered by: standard name ncalls tottime percall cumtime percall filename:lineno(function) 1 0.002 0.002 0.002 0.002 :0(listdir) 1 0.001 0.001 0.001 0.001 :0(setprofile) 1 0.001 0.001 3.809 3.809 profile:0(, file 'tool3.py', line 3>) 0 0.000 0.000 profile:0(profiler) 1 0.001 0.001 3.807 3.807 tool3.py:3() 1 3.804 3.804 3.806 3.806 tool3.py:5(main) real 0m4.111s user 0m3.477s sys 0m0.480s ---------- messages: 5721 nosy: pypy-issue, sd87271833 priority: performance bug status: chatting title: Pypy disk IO is slower than python2.6? ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 15 16:28:28 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Wed, 15 May 2013 14:28:28 +0000 Subject: [pypy-issue] [issue1465] cffi callback crashes in multithread environment In-Reply-To: <1366796639.81.0.357906871623.issue1465@bugs.pypy.org> Message-ID: <1368628108.21.0.129534120242.issue1465@bugs.pypy.org> Armin Rigo added the comment: A known issue, probably won't fix. PyPy only works with threads started from Python code. Other threads cause various internal states not to be initialized. ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 15 16:32:26 2013 From: tracker at bugs.pypy.org (Fijal) Date: Wed, 15 May 2013 14:32:26 +0000 Subject: [pypy-issue] [issue1465] cffi callback crashes in multithread environment In-Reply-To: <1366796639.81.0.357906871623.issue1465@bugs.pypy.org> Message-ID: <1368628346.24.0.11606461114.issue1465@bugs.pypy.org> Fijal added the comment: maybe we can export enough primitives to make it work? ---------- nosy: +fijal ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 15 16:33:40 2013 From: tracker at bugs.pypy.org (Fijal) Date: Wed, 15 May 2013 14:33:40 +0000 Subject: [pypy-issue] [issue1409] netbsd support In-Reply-To: <1361857361.33.0.161819094579.issue1409@bugs.pypy.org> Message-ID: <1368628420.69.0.120062744213.issue1409@bugs.pypy.org> Fijal added the comment: It connects to a central place and pops commands that can be essentially everything. Additionally it's of limited use if we can't get remote access there one way or another. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 15 16:34:57 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Wed, 15 May 2013 14:34:57 +0000 Subject: [pypy-issue] [issue1465] cffi callback crashes in multithread environment In-Reply-To: <1366796639.81.0.357906871623.issue1465@bugs.pypy.org> Message-ID: <1368628497.83.0.532454548979.issue1465@bugs.pypy.org> Amaury Forgeot d Arc added the comment: we should reconsider this (the "won't fix" part). There are many systems which create threads from C/C++ and still want to call into Python. fijal has started a simple C API for embedding, it could probably be used to implement some kind of PyGILState_Ensure. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 16 07:29:48 2013 From: tracker at bugs.pypy.org (YAMAMOTO Takashi) Date: Thu, 16 May 2013 05:29:48 +0000 Subject: [pypy-issue] [issue1487] workaround for eventlet In-Reply-To: <1368682188.33.0.840206844814.issue1487@bugs.pypy.org> Message-ID: <1368682188.33.0.840206844814.issue1487@bugs.pypy.org> New submission from YAMAMOTO Takashi : this patch allows nose+evenlet run with pypy. ---------- files: a.diff messages: 5726 nosy: pypy-issue, yamt priority: bug status: unread title: workaround for eventlet ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 16 09:08:03 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Thu, 16 May 2013 07:08:03 +0000 Subject: [pypy-issue] [issue1487] workaround for eventlet In-Reply-To: <1368682188.33.0.840206844814.issue1487@bugs.pypy.org> Message-ID: <1368688083.83.0.690321392068.issue1487@bugs.pypy.org> Armin Rigo added the comment: Marked as duplicate of issue1468. We're aware of the problem already, and it's a bit unclear if just not calling _decref_socketios() is a good idea, but thanks for the test case! ---------- nosy: +arigo status: unread -> duplicate ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 16 09:09:02 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Thu, 16 May 2013 07:09:02 +0000 Subject: [pypy-issue] [issue1468] PyPy eventlet trouble with _decref_socketios In-Reply-To: <1366885536.44.0.498230363361.issue1468@bugs.pypy.org> Message-ID: <1368688142.47.0.0837498605839.issue1468@bugs.pypy.org> Armin Rigo added the comment: yamt sumitted the following simple test case in issue1487: import eventlet.greenio as i import os r, w = os.pipe() wf = i.GreenPipe(w, 'wb') wf.close() ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 16 09:14:21 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Thu, 16 May 2013 07:14:21 +0000 Subject: [pypy-issue] [issue1468] PyPy eventlet trouble with _decref_socketios In-Reply-To: <1366885536.44.0.498230363361.issue1468@bugs.pypy.org> Message-ID: <1368688461.32.0.603083414514.issue1468@bugs.pypy.org> Armin Rigo added the comment: In PyPy the internal socket objects have methods "_reuse()" and "_drop()" emulating the refcounted behavior explicitly. Where does this "_decref_socketios()" come from, by the way? I don't find it anywhere in Python 2.7's standard library or C files. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 16 09:28:20 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Thu, 16 May 2013 07:28:20 +0000 Subject: [pypy-issue] [issue1468] PyPy eventlet trouble with _decref_socketios In-Reply-To: <1366885536.44.0.498230363361.issue1468@bugs.pypy.org> Message-ID: <1368689300.07.0.556033549258.issue1468@bugs.pypy.org> Armin Rigo added the comment: My current understanding: - the call to _decref_socketios() that crashes occurs in socket.py, not in eventlet at all. I believe this is because eventlet builds a instance that contains directly a <_socket.socket> instance, rather than (as done normally) a . - _reuse and _drop on <_socket.socket> are actually never used any more, and not tested anywhere I can find. In summary I've a bit no clue but until a better understanding is found I'll just adopt the solution of issue1487. ---------- status: chatting -> testing ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 16 10:00:08 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Thu, 16 May 2013 08:00:08 +0000 Subject: [pypy-issue] [issue1468] PyPy eventlet trouble with _decref_socketios In-Reply-To: <1366885536.44.0.498230363361.issue1468@bugs.pypy.org> Message-ID: <1368691208.95.0.503262209498.issue1468@bugs.pypy.org> Amaury Forgeot d Arc added the comment: _decref_socketios() comes from Python 3, where it was decided that sockets should not rely on reference counting (cf. ResourceWarning). IIRC this is also important for ssl sockets, which may create cycles. I also saw the small inconsistency of eventlets which use raw _socket.socket instead of socket._socketobject. IMO this is the one to fix, but I could not find the place. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 17 00:28:16 2013 From: tracker at bugs.pypy.org (mattip) Date: Thu, 16 May 2013 22:28:16 +0000 Subject: [pypy-issue] [issue1484] numpy pickling In-Reply-To: <1368344684.93.0.745354939033.issue1484@bugs.pypy.org> Message-ID: <1368743296.64.0.862012022288.issue1484@bugs.pypy.org> mattip added the comment: this can be simplified, it appears pickling is not yet completely supported a = np.zeros(3) pickle.dumps(np.sum(a)) => TypeError: can't pickle float64 objects ---------- nosy: +mattip, rguillebert status: chatting -> in-progress ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 17 02:04:44 2013 From: tracker at bugs.pypy.org (mikefc) Date: Fri, 17 May 2013 00:04:44 +0000 Subject: [pypy-issue] [issue1484] numpy pickling In-Reply-To: <1368344684.93.0.745354939033.issue1484@bugs.pypy.org> Message-ID: <1368749084.71.0.376039600897.issue1484@bugs.pypy.org> mikefc added the comment: ugh. yep. you're correct. thanks for simplifying the issue. I was trying all combinations of numpy *arrays* to try and reproduce the error, but it appears that pickling doesn't work for *scalars* (float64, float32, int32 etc) ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 17 09:29:11 2013 From: tracker at bugs.pypy.org (Nick Zorin) Date: Fri, 17 May 2013 07:29:11 +0000 Subject: [pypy-issue] [issue1488] Times slower in PyPy In-Reply-To: <1368775751.52.0.789251725482.issue1488@bugs.pypy.org> Message-ID: <1368775751.52.0.789251725482.issue1488@bugs.pypy.org> New submission from Nick Zorin <3.14l5926 at gmail.com>: This code works several times slower in PyPy comparing to Python. import time def timeit(f): def timed(*args, **kw): ts = time.time() result = f(*args, **kw) te = time.time() print 'func:%r args:[%r, %r] took: %2.4f sec' % \ (f.__name__, args, kw, te-ts) return result return timed def fib(): i = 1 a = b = c = 0 while True: if i == 1: a = 0 result = a elif i == 2: b = 1 result = 2 else: c = a + b result = c a, b = b, c i += 1 if '31415926' in str(result): yield i @timeit def xxx(f): f.next() f.next() f.next() f.next() f.next() f = fib() xxx(f) ---------- messages: 5734 nosy: 314l5926, pypy-issue priority: performance bug release: 2.0 status: unread title: Times slower in PyPy ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 17 11:05:04 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 17 May 2013 09:05:04 +0000 Subject: [pypy-issue] [issue1488] Times slower in PyPy In-Reply-To: <1368775751.52.0.789251725482.issue1488@bugs.pypy.org> Message-ID: <1368781504.92.0.629789348396.issue1488@bugs.pypy.org> Armin Rigo added the comment: Reduced the test case to: str() is seriously slower than on CPython. l = 3 ** 2234 for i in range(10000): l += 1 if 'foobar' in str(l): break ---------- nosy: +arigo status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 17 11:39:49 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 17 May 2013 09:39:49 +0000 Subject: [pypy-issue] [issue1488] Times slower in PyPy In-Reply-To: <1368775751.52.0.789251725482.issue1488@bugs.pypy.org> Message-ID: <1368783589.34.0.623456132048.issue1488@bugs.pypy.org> Armin Rigo added the comment: (The usual warning about longs: that's never going to be faster on PyPy than on CPython. Manipulating large values of type "long" is taking time and it's all done as C loops anyway. It wouldn't be significantly faster if you hand-coded it in C. But thanks for the report anyway, which shows that our str() can be improved.) ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 18 13:33:27 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Sat, 18 May 2013 11:33:27 +0000 Subject: [pypy-issue] [issue1488] 'str(long)' slower in PyPy In-Reply-To: <1368775751.52.0.789251725482.issue1488@bugs.pypy.org> Message-ID: <1368876807.61.0.219539607582.issue1488@bugs.pypy.org> Armin Rigo added the comment: Improved by 25% in af160213d701 5de7c3413954 f12e27274ff3. Still worse than CPython, but I think I won't try to improve even more myself. The numbers I get are 3.2 seconds on PyPy, 2.1 seconds on CPython. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 19 00:25:10 2013 From: tracker at bugs.pypy.org (Stefano Rivera) Date: Sat, 18 May 2013 22:25:10 +0000 Subject: [pypy-issue] [issue673] Linux L2 cache size In-Reply-To: <1300130407.54.0.711591322006.issue673@> Message-ID: <1368915910.47.0.258593705374.issue673@bugs.pypy.org> Stefano Rivera added the comment: Added detection for all the archs I can in https://bitbucket.org/pypy/pypy/commits/ed146ac82e63b2ff2903dba54b2ebf2c46ac1d61 ---------- status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Mon May 20 11:25:09 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Mon, 20 May 2013 09:25:09 +0000 Subject: [pypy-issue] [issue1487] workaround for eventlet In-Reply-To: <1368682188.33.0.840206844814.issue1487@bugs.pypy.org> Message-ID: <1369041909.55.0.12764761576.issue1487@bugs.pypy.org> Armin Rigo added the comment: The fix for now is like your diff. Thanks! ---------- status: duplicate -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 21 05:24:55 2013 From: tracker at bugs.pypy.org (dktc) Date: Tue, 21 May 2013 03:24:55 +0000 Subject: [pypy-issue] [issue1489] pystone regression In-Reply-To: <1369106695.44.0.810401206544.issue1489@bugs.pypy.org> Message-ID: <1369106695.44.0.810401206544.issue1489@bugs.pypy.org> New submission from dktc : I just built fresh PyPy 2.0.1 and decided to run pystones test (from getting started guide) and compare the performance difference. To my surprise there is a significant regression: v2.0.1 => This machine benchmarks at ~410000 pystones/second v2.0.0b2 => This machine benchmarks at ~420000 pystones/second v1.9.0 => This machine benchmarks at ~520000 pystones/second Everything is averaged over many runs on the same CentOS 5 box. What could be causing this? Is this a real regression or maybe there are some build options that influence performance? ---------- messages: 5740 nosy: pypy-issue priority: performance bug status: unread title: pystone regression ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 21 06:30:53 2013 From: tracker at bugs.pypy.org (dktc) Date: Tue, 21 May 2013 04:30:53 +0000 Subject: [pypy-issue] [issue1490] pyopenssl fails to install on pypy 2.0.1 In-Reply-To: <1369110653.32.0.866977030556.issue1490@bugs.pypy.org> Message-ID: <1369110653.32.0.866977030556.issue1490@bugs.pypy.org> New submission from dktc : Any idea what could be wrong / how to fix this? ==================================================== (pypy-2.0.1)[root at localhost bin]# pip install pyopenssl ................. cc -O2 -fPIC -Wimplicit -I/opt/pypy-2.0.1/include -c OpenSSL/ssl/connection.c -o build/temp.linux-x86_64- 2.7/OpenSSL/ssl/connection.o OpenSSL/ssl/connection.c: In function ?ssl_Connection_set_context?: OpenSSL/ssl/connection.c:289: warning: implicit declaration of function ?SSL_set_SSL_CTX? OpenSSL/ssl/connection.c: In function ?ssl_Connection_get_servername?: OpenSSL/ssl/connection.c:313: error: ?TLSEXT_NAMETYPE_host_name? undeclared (first use in this function) OpenSSL/ssl/connection.c:313: error: (Each undeclared identifier is reported only once OpenSSL/ssl/connection.c:313: error: for each function it appears in.) OpenSSL/ssl/connection.c:320: warning: implicit declaration of function ?SSL_get_servername? OpenSSL/ssl/connection.c:320: warning: assignment makes pointer from integer without a cast OpenSSL/ssl/connection.c: In function ?ssl_Connection_set_tlsext_host_name?: OpenSSL/ssl/connection.c:346: warning: implicit declaration of function ?SSL_set_tlsext_host_name? error: command 'cc' failed with exit status 1 ---------- messages: 5741 nosy: pypy-issue priority: bug status: unread title: pyopenssl fails to install on pypy 2.0.1 ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 21 08:44:55 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Tue, 21 May 2013 06:44:55 +0000 Subject: [pypy-issue] [issue1490] pyopenssl fails to install on pypy 2.0.1 In-Reply-To: <1369110653.32.0.866977030556.issue1490@bugs.pypy.org> Message-ID: <1369118695.11.0.7674941659.issue1490@bugs.pypy.org> Amaury Forgeot d Arc added the comment: The function SSL_set_SSL_CTX appeared somewhere "between 0.9.8n and 1.0.0", Which version of ssl is your system using? CPython has conditional code to cope with older version of openssl (search for HAVE_SNI in _ssl.c), pypy should probably do the same. tag:easy ---------- nosy: +amaury status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 21 09:53:52 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Tue, 21 May 2013 07:53:52 +0000 Subject: [pypy-issue] [issue1490] pyopenssl fails to install on pypy 2.0.1 In-Reply-To: <1369110653.32.0.866977030556.issue1490@bugs.pypy.org> Message-ID: <1369122832.35.0.330989144654.issue1490@bugs.pypy.org> Armin Rigo added the comment: This is not in pypy's _ssl module, as far as I can tell. This is in PyOpenSSL's source code. I believe that another answer is: the latest released version of PyOpenSSL is not really usable with PyPy, but the trunk of PyOpenSSL is, and comes with cffi bindings. ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 21 14:20:25 2013 From: tracker at bugs.pypy.org (dktc) Date: Tue, 21 May 2013 12:20:25 +0000 Subject: [pypy-issue] [issue1490] pyopenssl fails to install on pypy 2.0.1 In-Reply-To: <1369110653.32.0.866977030556.issue1490@bugs.pypy.org> Message-ID: <1369138825.82.0.78876145869.issue1490@bugs.pypy.org> dktc added the comment: Here is what I have - [root at localhost ~]# openssl version OpenSSL 0.9.8e-fips-rhel5 01 Jul 2008 I agree it would be good if PyPy could conditionally/automagically support older SSL versions. In the meantime, which older PyOpenSSL version should be compatible on CentOS 5? ________________________________________ PyPy bug tracker ________________________________________ From fijall at gmail.com Tue May 21 14:47:20 2013 From: fijall at gmail.com (Maciej Fijalkowski) Date: Tue, 21 May 2013 14:47:20 +0200 Subject: [pypy-issue] [issue1490] pyopenssl fails to install on pypy 2.0.1 In-Reply-To: <1369138825.82.0.78876145869.issue1490@bugs.pypy.org> References: <1369110653.32.0.866977030556.issue1490@bugs.pypy.org> <1369138825.82.0.78876145869.issue1490@bugs.pypy.org> Message-ID: On Tue, May 21, 2013 at 2:20 PM, dktc wrote: > > dktc added the comment: > > Here is what I have - > > [root at localhost ~]# openssl version > OpenSSL 0.9.8e-fips-rhel5 01 Jul 2008 > > I agree it would be good if PyPy could conditionally/automagically support older SSL versions. > > In the meantime, which older PyOpenSSL version should be compatible on CentOS 5? > a new pyopenssl release should natively support pypy (without cpyext, using cffi instead). You can try pyopenssl trunk From tracker at bugs.pypy.org Tue May 21 23:54:46 2013 From: tracker at bugs.pypy.org (dktc) Date: Tue, 21 May 2013 21:54:46 +0000 Subject: [pypy-issue] [issue1490] pyopenssl fails to install on pypy 2.0.1 In-Reply-To: <1369110653.32.0.866977030556.issue1490@bugs.pypy.org> Message-ID: <1369173286.49.0.613200326467.issue1490@bugs.pypy.org> dktc added the comment: === removed last message because it exposes email address === > a new pyopenssl release should natively support pypy (without cpyext, > using cffi instead). You can try pyopenssl trunk Is there a way to install pyopenssl trunk through pip command line? Looks like it was last updated on PyPI two years ago and Launchpad project is quite confusing. Are they planning to make a real release soon? ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 22 10:03:40 2013 From: tracker at bugs.pypy.org (sn6uv) Date: Wed, 22 May 2013 08:03:40 +0000 Subject: [pypy-issue] [issue1491] raw_input incorrectly places cursor when given escape sequences In-Reply-To: <1369209820.42.0.70338361307.issue1491@bugs.pypy.org> Message-ID: <1369209820.42.0.70338361307.issue1491@bugs.pypy.org> New submission from sn6uv <16sn6uv at gmail.com>: Description: raw_input seems to include the length of ANSI escape sequences in the prompt when positioning the cursor. This does not seem to happen when running the command from a script, only when entered into the PyPy shell. The entered text in put in the right place only the cursor's position is wrong, which is a bit confusing Example: >>> raw_input('\033[31m>>> \033[39m') Screenshot (attached): http://imgur.com/nqOhvfM System Info: $pypy --version Python 2.7.3 (fcb6b056f00e33a72cba5e6769b253b839fe90bb, Nov 24 2012, 02:30:32) [PyPy 2.0.0-beta1 with GCC 4.7.2] $ uname -a Linux angus-laptop 3.6.7-1-ARCH #1 SMP PREEMPT Sun Nov 18 10:11:22 CET 2012 x86_64 GNU/Linux $ xterm -v XTerm(287) $ aterm --version 1.00.01 ---------- files: pypy_escapeseq_cursor_bug.png messages: 5747 nosy: pypy-issue, sn6uv priority: bug status: chatting title: raw_input incorrectly places cursor when given escape sequences ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 22 16:41:58 2013 From: tracker at bugs.pypy.org (Drew Gulino) Date: Wed, 22 May 2013 14:41:58 +0000 Subject: [pypy-issue] [issue1492] eventlet example fails only on pypy In-Reply-To: <1369233718.16.0.38984718393.issue1492@bugs.pypy.org> Message-ID: <1369233718.16.0.38984718393.issue1492@bugs.pypy.org> New submission from Drew Gulino : The example on http://eventlet.net fails on pypy 2.0, 2.0.2 running on OSX 10.8.3, works on CPython pypy: Python 2.7.3 (5acfe049a5b0, May 21 2013, 13:47:22) [PyPy 2.0.2 with GCC 4.2.1 Compatible Apple LLVM 4.2 (clang-425.0.28)] on darwin CPython: Python 2.7.5 (default, May 19 2013, 13:26:46) [GCC 4.2.1 Compatible Apple Clang 4.1 ((tags/Apple/clang-421.11.66))] on darwin pip freeze (same on both): cffi==0.6 eventlet==0.12.1 greenlet==0.4.0 wsgiref==0.1.2 code: #!/usr/bin/env pypy urls = ["http://www.google.com/intl/en_ALL/images/logo.gif", "http://us.i1.yimg.com/us.yimg.com/i/ww/beta/y3.gif"] import eventlet from eventlet.green import urllib2 def fetch(url): return urllib2.urlopen(url).read() pool = eventlet.GreenPool() for body in pool.imap(fetch, urls): print "got body", len(body) error: Traceback (most recent call last): File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/hubs/poll.py", line 97, in wait readers.get(fileno, noop).cb(fileno) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib_pypy/greenlet.py", line 52, in switch return self.__switch('switch', args) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib_pypy/greenlet.py", line 81, in __switch args = unbound_method(current, *args, to=target) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib_pypy/greenlet.py", line 134, in _greenlet_start res = greenlet.run(*args) File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/greenthread.py", line 194, in main result = function(*args, **kwargs) File "./eventlet_crawl.py", line 11, in fetch return urllib2.urlopen(url).read() File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib-python/2.7/socket.py", line 401, in read data = self._sock.recv(rbufsize) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib-python/2.7/httplib.py", line 561, in read s = self.fp.read(amt) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib-python/2.7/socket.py", line 430, in read data = self._sock.recv(left) File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/greenio.py", line 238, in recv return fd.recv(buflen, flags) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib-python/2.7/socket.py", line 188, in recv return self._sock.recv(buffersize, flags=flags) error: [Errno 9] Bad file descriptor Removing descriptor: 6 Traceback (most recent call last): File "app_main.py", line 72, in run_toplevel File "./eventlet_crawl.py", line 15, in for body in pool.imap(fetch, urls): File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/greenpool.py", line 232, in next val = self.waiters.get().wait() File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/greenthread.py", line 168, in wait return self._exit_event.wait() File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/event.py", line 116, in wait return hubs.get_hub().switch() File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/hubs/hub.py", line 187, in switch return self.greenlet.switch() File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib_pypy/greenlet.py", line 52, in switch return self.__switch('switch', args) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib_pypy/greenlet.py", line 81, in __switch args = unbound_method(current, *args, to=target) File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/greenthread.py", line 194, in main result = function(*args, **kwargs) File "./eventlet_crawl.py", line 11, in fetch return urllib2.urlopen(url).read() File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib-python/2.7/socket.py", line 401, in read data = self._sock.recv(rbufsize) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib-python/2.7/httplib.py", line 561, in read s = self.fp.read(amt) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib-python/2.7/socket.py", line 430, in read data = self._sock.recv(left) File "/Users/drewgulino/Downloads/python/pypy/pypy_env/site-packages/eventlet/greenio.py", line 238, in recv return fd.recv(buflen, flags) File "/Users/drewgulino/Downloads/python/pypy/pypy-2.0.2/lib-python/2.7/socket.py", line 188, in recv return self._sock.recv(buffersize, flags=flags) error: [Errno 9] Bad file descriptor ---------- messages: 5748 nosy: dgulino, pypy-issue priority: bug release: 2.0 status: unread title: eventlet example fails only on pypy ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 22 21:22:34 2013 From: tracker at bugs.pypy.org (Jihyeok Seo) Date: Wed, 22 May 2013 19:22:34 +0000 Subject: [pypy-issue] [issue1392] _checkClosed method is misisng in io.RawIOBase In-Reply-To: <1360404665.6.0.144734969058.issue1392@bugs.pypy.org> Message-ID: <1369250554.62.0.26439919914.issue1392@bugs.pypy.org> Jihyeok Seo added the comment: Implement io.IOBase._checkClosed() ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 23 20:11:22 2013 From: tracker at bugs.pypy.org (Alex Gaynor) Date: Thu, 23 May 2013 18:11:22 +0000 Subject: [pypy-issue] [issue1493] Missing os.statvfs In-Reply-To: <1369332682.61.0.383363876987.issue1493@bugs.pypy.org> Message-ID: <1369332682.61.0.383363876987.issue1493@bugs.pypy.org> New submission from Alex Gaynor : This is needed for OpenStack Nova. ---------- nosy: +agaynor status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 23 23:39:55 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Thu, 23 May 2013 21:39:55 +0000 Subject: [pypy-issue] [issue1493] Missing os.statvfs In-Reply-To: <1369332682.61.0.383363876987.issue1493@bugs.pypy.org> Message-ID: <1369345195.26.0.240868420367.issue1493@bugs.pypy.org> Amaury Forgeot d Arc added the comment: tag:easy ---------- nosy: +amaury ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 24 05:33:41 2013 From: tracker at bugs.pypy.org (bigjools) Date: Fri, 24 May 2013 03:33:41 +0000 Subject: [pypy-issue] [issue1494] __future__ imports spanning multiple lines are a compile error In-Reply-To: <1369366421.8.0.71310493397.issue1494@bugs.pypy.org> Message-ID: <1369366421.8.0.71310493397.issue1494@bugs.pypy.org> New submission from bigjools : from __future__ import ( absolute_import, print_function, unicode_literals, ) results in SyntaxError: __future__ statements must appear at beginning of file ---------- nosy: +bigjools status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 24 09:02:17 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 24 May 2013 07:02:17 +0000 Subject: [pypy-issue] [issue1494] __future__ imports spanning multiple lines are a compile error In-Reply-To: <1369366421.8.0.71310493397.issue1494@bugs.pypy.org> Message-ID: <1369378937.27.0.259117660088.issue1494@bugs.pypy.org> Armin Rigo added the comment: The error only occurs if the first line ends with the opening parenthesis. ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 24 10:31:23 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 24 May 2013 08:31:23 +0000 Subject: [pypy-issue] [issue1494] __future__ imports spanning multiple lines are a compile error In-Reply-To: <1369366421.8.0.71310493397.issue1494@bugs.pypy.org> Message-ID: <1369384283.23.0.974181080115.issue1494@bugs.pypy.org> Armin Rigo added the comment: Fixed in 51490363a945. ---------- status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 24 10:32:36 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 24 May 2013 08:32:36 +0000 Subject: [pypy-issue] [issue1493] Missing os.statvfs In-Reply-To: <1369332682.61.0.383363876987.issue1493@bugs.pypy.org> Message-ID: <1369384356.16.0.698945680061.issue1493@bugs.pypy.org> Armin Rigo added the comment: Duplicate of issue 833. ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 24 16:23:58 2013 From: tracker at bugs.pypy.org (Romain Guillebert) Date: Fri, 24 May 2013 14:23:58 +0000 Subject: [pypy-issue] [issue1484] numpy pickling In-Reply-To: <1368344684.93.0.745354939033.issue1484@bugs.pypy.org> Message-ID: <1369405438.67.0.857391984674.issue1484@bugs.pypy.org> Romain Guillebert added the comment: FYI, I'm working on it right now. ---------- assignedto: -> rguillebert ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 25 01:47:28 2013 From: tracker at bugs.pypy.org (Ben Jolitz) Date: Fri, 24 May 2013 23:47:28 +0000 Subject: [pypy-issue] [issue1495] Crash in JIT from Bitbucket latest (checkout of 5/24/2013 at 1pm PST) In-Reply-To: <1369439248.28.0.958242326061.issue1495@bugs.pypy.org> Message-ID: <1369439248.28.0.958242326061.issue1495@bugs.pypy.org> New submission from Ben Jolitz : RPython traceback: File "rpython_jit_metainterp_pyjitpl.c", line 20079, in MIFrame_run_one_step File "rpython_jit_metainterp_pyjitpl_1.c", line 3376, in MIFrame_opimpl_jit_merge_point File "rpython_jit_metainterp_pyjitpl_1.c", line 34578, in MetaInterp_reached_loop_header File "rpython_jit_metainterp_pyjitpl_1.c", line 64302, in MetaInterp_compile_loop File "rpython_jit_metainterp_compile.c", line 14996, in compile_retrace Fatal RPython error: AssertionError Abort trap: 6 This is in a program that starts up seven threads of HTTPServer with the ThreadingMixin (mitmproxy) and shuts them down when a client dies. built as: python ../../rpython/bin/rpython -Ojit --gcrootfinder=shadowstack --make-jobs=7 targetpypystandalone.py --withmod-cpyext modules used: slimit, ply, pyOpenSSL (c-extension, built inside virtualenv with pypy), BeautifulSoup4. ---------- messages: 5757 nosy: benjolitz, pypy-issue priority: bug status: unread title: Crash in JIT from Bitbucket latest (checkout of 5/24/2013 at 1pm PST) ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 25 01:49:17 2013 From: tracker at bugs.pypy.org (Ben Jolitz) Date: Fri, 24 May 2013 23:49:17 +0000 Subject: [pypy-issue] [issue1495] Crash in JIT from Bitbucket latest (checkout of 5/24/2013 at 1pm PST) In-Reply-To: <1369439248.28.0.958242326061.issue1495@bugs.pypy.org> Message-ID: <1369439357.27.0.793804232803.issue1495@bugs.pypy.org> Ben Jolitz added the comment: RPython traceback: File "rpython_jit_metainterp_pyjitpl.c", line 20079, in MIFrame_run_one_step File "rpython_jit_metainterp_pyjitpl_1.c", line 3376, in MIFrame_opimpl_jit_merge_point File "rpython_jit_metainterp_pyjitpl_1.c", line 34578, in MetaInterp_reached_loop_header File "rpython_jit_metainterp_pyjitpl_1.c", line 64302, in MetaInterp_compile_loop File "rpython_jit_metainterp_compile.c", line 14996, in compile_retrace Fatal RPython error: AssertionError Abort trap: 6 This is in a program that starts up seven threads of HTTPServer with the ThreadingMixin (mitmproxy) and shuts them down when a client dies. built as: python ../../rpython/bin/rpython -Ojit --gcrootfinder=shadowstack --make-jobs=7 targetpypystandalone.py --withmod-cpyext modules used: slimit, ply, pyOpenSSL (c-extension, built inside virtualenv with pypy), BeautifulSoup4. Platform: OSX Mountain Lion 10.8.3 ---------- status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 25 10:44:20 2013 From: tracker at bugs.pypy.org (tonypypy) Date: Sat, 25 May 2013 08:44:20 +0000 Subject: [pypy-issue] [issue1496] pypy 2.0.x crash (stack trace included) In-Reply-To: <1369471460.16.0.652344394226.issue1496@bugs.pypy.org> Message-ID: <1369471460.16.0.652344394226.issue1496@bugs.pypy.org> New submission from tonypypy : While trying out latest pypy, I got consistent core dumps on one program; here's the stack trace from 2.0.2; I had the same problem yesterday with 2.0.1, but didn't get a change to generate the trace. Pypy 2.0.2; built from source; installed with virtualenv: $ ./rpython/bin/rpython -Ojit pypy/goal/targetpypystandalone.py $ virtualenv -p $PWD/pypy-c /opt/pypy-2.0.2 $ cd ~/myproject $ . /opt/pypy-2.0.2/bin/activate (pypy-2.0.2)$ pypy bin/program.py < data.json ................................................................................ ... ................................................................................ ... ....................................................zsh: segmentation fault (core dumped) Trying again with /usr/bin/time: 2.22user 0.14system 0:03.03elapsed 77%CPU (0avgtext+0avgdata 159568maxresident)k 0inputs+0outputs (0major+53786minor)pagefaults 0swaps Each period is printed at the end of a function call that creates a few thousand sets, frozensets, tuples, lists. The program is purely sequential, and pure python (no C/C++ parts). I use cPickle instead of pickle. --------------------------------------------------------------------- No problem pypy 1.9. Haven't tried stock python (too slow for this). * Environment: Fedora core 17; 8GB RAM; intel i7 $ uname -a Linux lenovo-t420s 3.8.12-100.fc17.x86_64 #1 SMP Wed May 8 15:36:14 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux # This pypy does not dump core. $ pypy -V Python 2.7.2 (341e1e3821fff77db3bb5cdb7a4851626298c44e, Jun 18 2012, 13:48:33) [PyPy 1.9.0 with GCC 4.7.0] ---------- messages: 5759 nosy: pypy-issue, tonypypy priority: critical status: unread title: pypy 2.0.x crash (stack trace included) ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 25 12:18:42 2013 From: tracker at bugs.pypy.org (alito) Date: Sat, 25 May 2013 10:18:42 +0000 Subject: [pypy-issue] [issue1497] Pypy doesn't compile when ncurses is compiled with --with-termlib In-Reply-To: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> Message-ID: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> New submission from alito : Issue 667 https://bugs.pypy.org/issue667 again I suppose. I noticed that that patch is no longer in the tree. On the other hand, that ncurses configuration is common, so I suspect that pypy is trying to spot it in some other way that doesn't work with my system (Gentoo but not using ebuilds to compile pypy). Grepping for tinfo doesn't give me anything relevant. How does pypy try to detect the existence of libtinfo? ---------- messages: 5760 nosy: alito, pypy-issue priority: critical release: 2.0 status: unread title: Pypy doesn't compile when ncurses is compiled with --with-termlib ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 25 15:48:09 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Sat, 25 May 2013 13:48:09 +0000 Subject: [pypy-issue] [issue1496] pypy 2.0.x crash (stack trace included) In-Reply-To: <1369471460.16.0.652344394226.issue1496@bugs.pypy.org> Message-ID: <1369489689.25.0.307785970377.issue1496@bugs.pypy.org> Amaury Forgeot d Arc added the comment: Where is the stack trace? ---------- nosy: +amaury status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 25 17:58:23 2013 From: tracker at bugs.pypy.org (Nicolas Lelong) Date: Sat, 25 May 2013 15:58:23 +0000 Subject: [pypy-issue] [issue1498] Problems subclassing from a defined ctypes Structure In-Reply-To: <1369497503.17.0.189737095715.issue1498@bugs.pypy.org> Message-ID: <1369497503.17.0.189737095715.issue1498@bugs.pypy.org> New submission from Nicolas Lelong : We currently try to run some of our code which relies heavily on ctypes. We define our C structures the usual way, and then 'augment' them by subclassing the C definition. Something along the lines of : import ctypes class cStruct( ctypes.Structure ): _fields_ = [ ( "x", ctypes.c_float ) , ( "y", ctypes.c_float ) ] class pyStruct( cStruct ): def __init__(self, *args): super(pyStruct, self).__init__(*args) Pypy behaviour is different from CPython. The following code runs fine with CPython : assert cStruct._fields_ == pyStruct._fields_ a = cStruct() b = pyStruct() print "a._fields_ =", a._fields_ print "b._fields_ =", b._fields_ b = pyStruct(1.0, 2.0) And gives the following output : a._fields_ = [('x', ), ('y', )] b._fields_ = [('x', ), ('y', )] With pypy, it seems that _fields_ is empty for 'b' ; see below. Is this to be expected ? Thanks == This gives the following output / exception : a._fields_ = [('x', ), ('y', )] b._fields_ = [] Traceback (most recent call last): File "app_main.py", line 72, in run_toplevel File "D:\rotoglup-scratchpad\experiments\pypy_ctypes\test_struct_inherit.py", line 18, in b = pyStruct(1.0, 2.0) File "D:\rotoglup-scratchpad\experiments\pypy_ctypes\test_struct_inherit.py", line 11, in __init__ super(pyStruct, self).__init__(*args) File "D:\pypy\pypy-2.0.2\lib_pypy\_ctypes\structure.py", line 232, in __init__ self.__setattr__(name, arg) File "D:\pypy\pypy-2.0.2\lib_pypy\_ctypes\structure.py", line 116, in __set__ obj._buffer.__setattr__(self.name, arg) AttributeError: C Structure has no attribute x pypy-2.0-beta1.1 failed with another exception : a._fields_ = [('x', ), ('y', )] b._fields_ = [] Traceback (most recent call last): File "app_main.py", line 52, in run_toplevel File "D:\rotoglup-scratchpad\experiments\pypy_ctypes\test_struct_inherit.py", line 18, in b = pyStruct(1.0, 2.0) File "D:\rotoglup-scratchpad\experiments\pypy_ctypes\test_struct_inherit.py", line 11, in __init__ super(pyStruct, self).__init__(*args) File "D:\pypy\pypy-2.0-beta1.1-win32\pypy-2.0- beta1\lib_pypy\_ctypes\structure.py", line 228, in __init__ raise TypeError("too many initializers") TypeError: too many initializers ---------- messages: 5762 nosy: pypy-issue, rotoglup priority: bug status: unread title: Problems subclassing from a defined ctypes Structure ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sat May 25 21:47:24 2013 From: tracker at bugs.pypy.org (Amaury Forgeot d Arc) Date: Sat, 25 May 2013 19:47:24 +0000 Subject: [pypy-issue] [issue1498] Problems subclassing from a defined ctypes Structure In-Reply-To: <1369497503.17.0.189737095715.issue1498@bugs.pypy.org> Message-ID: <1369511243.99.0.0886463805494.issue1498@bugs.pypy.org> Amaury Forgeot d Arc added the comment: The fix for issue1473 was incomplete. Fixed with changeset ff06e504ab9c, thanks for the report! Note that _fields_ are handled a bit differently than CPython: - with CPython, _fields_ is inherited from the base class. - PyPy considers that a subclass without _fields_ is equivalent to a class with _fields_=[], so it sets this _fields_=[] as a default value. The resulting classes should behave the same, of course. ---------- nosy: +amaury status: unread -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 26 01:20:24 2013 From: tracker at bugs.pypy.org (Philip Jenvey) Date: Sat, 25 May 2013 23:20:24 +0000 Subject: [pypy-issue] [issue1483] io.IOBase is missing some stub implementations In-Reply-To: <1368144885.41.0.996527671116.issue1483@bugs.pypy.org> Message-ID: <1369524024.67.0.777709479074.issue1483@bugs.pypy.org> Philip Jenvey added the comment: Applied to default, thanks ---------- status: in-progress -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 26 22:31:33 2013 From: tracker at bugs.pypy.org (Tom Prince) Date: Sun, 26 May 2013 20:31:33 +0000 Subject: [pypy-issue] [issue1499] Can't install setuptools on a fresh install of pypy_2.0.2+dfsg-1~ppa1_amd64.deb In-Reply-To: <1369600293.97.0.574868911376.issue1499@bugs.pypy.org> Message-ID: <1369600293.97.0.574868911376.issue1499@bugs.pypy.org> New submission from Tom Prince : $ pypy /opt/ez_setup.py Downloading http://pypi.python.org/packages/2.7/s/setuptools/setuptools-0.6c11-py2.7.egg error: can't create or remove files in install directory The following error occurred while trying to add or remove files in the installation directory: [Errno 2] No such file or directory: '/usr/lib/pypy/../../local/lib/pypy2.7/dist-packages/test-easy-install-19231.pth' The installation directory you specified (via --install-dir, --prefix, or the distutils default setting) was: /usr/lib/pypy/../../local/lib/pypy2.7/dist-packages/ This directory does not currently exist. Please create it and try again, or choose a different installation directory (using the -d or --install-dir option). ---------- messages: 5765 nosy: pypy-issue, tom.prince priority: bug status: unread title: Can't install setuptools on a fresh install of pypy_2.0.2+dfsg-1~ppa1_amd64.deb ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Sun May 26 23:06:58 2013 From: tracker at bugs.pypy.org (Tom Prince) Date: Sun, 26 May 2013 21:06:58 +0000 Subject: [pypy-issue] [issue1500] Ubuntu packages don't play with debian/ubuntu python infrastructure. In-Reply-To: <1369602418.5.0.351825317872.issue1500@bugs.pypy.org> Message-ID: <1369602418.5.0.351825317872.issue1500@bugs.pypy.org> New submission from Tom Prince : The ubuntu packages at https://launchpad.net/~pypy/+archive/ppa don't seem to interact with debian/ubuntu's support for installing python package debs against multiple versions of python. ---------- messages: 5766 nosy: pypy-issue, tom.prince priority: bug status: unread title: Ubuntu packages don't play with debian/ubuntu python infrastructure. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Mon May 27 10:04:39 2013 From: tracker at bugs.pypy.org (Fijal) Date: Mon, 27 May 2013 08:04:39 +0000 Subject: [pypy-issue] [issue1500] Ubuntu packages don't play with debian/ubuntu python infrastructure. In-Reply-To: <1369602418.5.0.351825317872.issue1500@bugs.pypy.org> Message-ID: <1369641879.78.0.726255221723.issue1500@bugs.pypy.org> Fijal added the comment: This is a well known issue and I don't think they have reasonable answers to that (they = debian and ubuntu package maintainers) ---------- nosy: +fijal status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Mon May 27 10:50:08 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Mon, 27 May 2013 08:50:08 +0000 Subject: [pypy-issue] [issue1495] Crash in JIT from Bitbucket latest (checkout of 5/24/2013 at 1pm PST) In-Reply-To: <1369439357.27.0.793804232803.issue1495@bugs.pypy.org> Message-ID: <1369644608.63.0.635903887991.issue1495@bugs.pypy.org> Armin Rigo added the comment: Likely, we need a way to reproduce it ourselves. Please post a sample script, as well as the version number of the PyPy you're using (the output saying just "pypy" is enough). ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Mon May 27 10:56:26 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Mon, 27 May 2013 08:56:26 +0000 Subject: [pypy-issue] [issue1496] pypy 2.0.x crash (stack trace included) In-Reply-To: <1369471460.16.0.652344394226.issue1496@bugs.pypy.org> Message-ID: <1369644986.23.0.455135840383.issue1496@bugs.pypy.org> Armin Rigo added the comment: Your report is mostly useless if we don't get to reproduce it ourselves. Please attach a sample program. (We don't care if it's big! You don't have to reduce the problem to a 10-lines example.) ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 12:29:12 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Tue, 28 May 2013 10:29:12 +0000 Subject: [pypy-issue] [issue1495] Crash in JIT from Bitbucket latest (checkout of 5/24/2013 at 1pm PST) In-Reply-To: <1369439357.27.0.793804232803.issue1495@bugs.pypy.org> Message-ID: <1369736952.91.0.901515265171.issue1495@bugs.pypy.org> Armin Rigo added the comment: To be clearer, we don't mind if the program is big. We can perform reduction ourselves if needed. The point is that without this information we cannot do anything from your bug report. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 12:32:18 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Tue, 28 May 2013 10:32:18 +0000 Subject: [pypy-issue] [issue667] Final link failure due to libtinfo dependency In-Reply-To: <1299767405.56.0.72258074824.issue667@> Message-ID: <1369737138.18.0.0187411939624.issue667@bugs.pypy.org> Armin Rigo added the comment: fijal: I'm not sure, but it seems that 2 years ago you closed this bug report *without applying the patch* that it contains. It was lost :-( ---------- nosy: +arigo status: resolved -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 12:36:12 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Tue, 28 May 2013 10:36:12 +0000 Subject: [pypy-issue] [issue667] Final link failure due to libtinfo dependency In-Reply-To: <1299767405.56.0.72258074824.issue667@> Message-ID: <1369737372.6.0.877862107682.issue667@bugs.pypy.org> Armin Rigo added the comment: Moved to issue 1497 ---------- status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 12:37:17 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Tue, 28 May 2013 10:37:17 +0000 Subject: [pypy-issue] [issue1497] Pypy doesn't compile when ncurses is compiled with --with-termlib In-Reply-To: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> Message-ID: <1369737437.67.0.468447685344.issue1497@bugs.pypy.org> Armin Rigo added the comment: The patch seems to never have been applied :-( Can you test if, instead of guessing, we could run the following commands instead? ncurses5-config --cflags ncurses5-config --libs ---------- nosy: +arigo status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 13:03:43 2013 From: tracker at bugs.pypy.org (Fijal) Date: Tue, 28 May 2013 11:03:43 +0000 Subject: [pypy-issue] [issue667] Final link failure due to libtinfo dependency In-Reply-To: <1299767405.56.0.72258074824.issue667@> Message-ID: <1369739023.77.0.110632419803.issue667@bugs.pypy.org> Fijal added the comment: bah ---------- nosy: +fijal status: resolved -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 13:52:37 2013 From: tracker at bugs.pypy.org (alito) Date: Tue, 28 May 2013 11:52:37 +0000 Subject: [pypy-issue] [issue1497] Pypy doesn't compile when ncurses is compiled with --with-termlib In-Reply-To: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> Message-ID: <1369741957.58.0.660569035808.issue1497@bugs.pypy.org> alito added the comment: It seems that ncurses5-config would work here (5.9) $ ncurses5-config --libs -L/usr/lib -lncurses -ltinfo $ ncurses5-config --cflags # gives me nothing $ ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 15:57:48 2013 From: tracker at bugs.pypy.org (Romain Guillebert) Date: Tue, 28 May 2013 13:57:48 +0000 Subject: [pypy-issue] [issue1484] numpy pickling In-Reply-To: <1368344684.93.0.745354939033.issue1484@bugs.pypy.org> Message-ID: <1369749468.74.0.834603302923.issue1484@bugs.pypy.org> Romain Guillebert added the comment: I merged the branch on default today, it should work now ---------- status: in-progress -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 18:43:33 2013 From: tracker at bugs.pypy.org (Ben Jolitz) Date: Tue, 28 May 2013 16:43:33 +0000 Subject: [pypy-issue] [issue1495] Crash in JIT from Bitbucket latest (checkout of 5/24/2013 at 1pm PST) In-Reply-To: <1369439357.27.0.793804232803.issue1495@bugs.pypy.org> Message-ID: <1369759413.64.0.461369584417.issue1495@bugs.pypy.org> Ben Jolitz added the comment: Hi Armin, Upon disabling logging in my script, all crashes disappeared. I consider this a non-issue as logging is optional in my program. It's a confidential script, so I cannot give the original. However, a primitive version of it would entail: a) main script with logger.critical(), logger.exception() b) using threaded class to start n threads with each thread making m threads for opening each of m sockets and closing it once data is read for m urls on a HTML page. c) logger.critical, logger.info, logger.exception within context of each socket thread I'll see if I can get around to making a test program soon. Definitely looks like acts of `logging' though, which I remember seeing in the wiki was a "no-no". You're free to close this if you'd like. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 21:37:14 2013 From: tracker at bugs.pypy.org (Edd Barrett) Date: Tue, 28 May 2013 19:37:14 +0000 Subject: [pypy-issue] [issue1501] Minor English bugs in rpython documentation In-Reply-To: <1369769834.64.0.443474862555.issue1501@bugs.pypy.org> Message-ID: <1369769834.64.0.443474862555.issue1501@bugs.pypy.org> New submission from Edd Barrett : Hi, I am just reading about rpython and I have found a couple of English nits. On this page: http://doc.pypy.org/en/latest/getting-started-dev.html#getting- started-with-rpython 1) "RPython is considered from live objects after the imports are done". Was the word "considered" supposed to be another word? I don't understand the meaning of "considered" here. 2) "A good introductory level articles are available:" should perhaps read "The following introductory level articles are available". As it stands the grammar is incorrect. That is all. Like I said, very minor :) Cheers. ---------- messages: 5778 nosy: pypy-issue, vext01 priority: bug status: unread title: Minor English bugs in rpython documentation ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 22:21:41 2013 From: tracker at bugs.pypy.org (Stefano Rivera) Date: Tue, 28 May 2013 20:21:41 +0000 Subject: [pypy-issue] [issue1499] Can't install setuptools on a fresh install of pypy_2.0.2+dfsg-1~ppa1_amd64.deb In-Reply-To: <1369600293.97.0.574868911376.issue1499@bugs.pypy.org> Message-ID: <1369772501.35.0.244657070866.issue1499@bugs.pypy.org> Stefano Rivera added the comment: This is a packaging bug rather than an upstream PyPy bug. Fixed in http://anonscm.debian.org/gitweb/?p=collab-maint/pypy.git;a=commitdiff;h=7bb4f4c289faa12bb05c54eb8064ba097b6eacee It'll be in the next PPA upload. ---------- assignedto: -> stefanor nosy: +stefanor status: unread -> in-progress ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 22:30:45 2013 From: tracker at bugs.pypy.org (Glyph) Date: Tue, 28 May 2013 20:30:45 +0000 Subject: [pypy-issue] [issue1499] Can't install setuptools on a fresh install of pypy_2.0.2+dfsg-1~ppa1_amd64.deb In-Reply-To: <1369600293.97.0.574868911376.issue1499@bugs.pypy.org> Message-ID: <1369773045.14.0.66085182962.issue1499@bugs.pypy.org> Glyph added the comment: This came to our attention because of https://github.com/twisted-infra/braid/issues/39 ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 23:40:40 2013 From: tracker at bugs.pypy.org (dktc) Date: Tue, 28 May 2013 21:40:40 +0000 Subject: [pypy-issue] [issue1492] More PyPy eventlet problems In-Reply-To: <1369233718.16.0.38984718393.issue1492@bugs.pypy.org> Message-ID: <1369777240.89.0.78629917815.issue1492@bugs.pypy.org> dktc added the comment: I have another possibly related issue with eventlet. When EVENTLET_THREADPOOL_SIZE env variable is set (see http://eventlet.net/doc/threading.html) eventlet crashes on PyPy 2.0.1. The crash does not happen on Python 2.7.3 or when the variable is not present. I do not have a compact code sample to reproduce this, since eventlet is being invoked from Spawning (https://pypi.python.org/pypi/Spawning). I just see that when Spawning settings cause creation of a subprocess (through os.execve call) with EVENTLET_THREADPOOL_SIZE env var, this subprocess fails: "(21432) Child died from signal 11 with code 0" Any thoughts why this could be happening on PyPy? ---------- status: unread -> chatting title: eventlet example fails only on pypy -> More PyPy eventlet problems ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Tue May 28 23:41:40 2013 From: tracker at bugs.pypy.org (dktc) Date: Tue, 28 May 2013 21:41:40 +0000 Subject: [pypy-issue] [issue1492] More PyPy eventlet problems In-Reply-To: <1369233718.16.0.38984718393.issue1492@bugs.pypy.org> Message-ID: <1369777300.51.0.180230141918.issue1492@bugs.pypy.org> dktc added the comment: I have another possibly related issue with eventlet. When EVENTLET_THREADPOOL_SIZE env variable is set ( see http://eventlet.net/doc/threading.html ) eventlet crashes on PyPy 2.0.1. The crash does not happen on Python 2.7.3 or when the variable is not present. I do not have a compact code sample to reproduce this, since eventlet is being invoked from Spawning ( https://pypi.python.org/pypi/Spawning ). I just see that when Spawning settings cause creation of a subprocess (through os.execve call) with EVENTLET_THREADPOOL_SIZE env var, this subprocess fails: "(21432) Child died from signal 11 with code 0" Any thoughts why this could be happening on PyPy? ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 29 10:25:03 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Wed, 29 May 2013 08:25:03 +0000 Subject: [pypy-issue] [issue1492] More PyPy eventlet problems In-Reply-To: <1369233718.16.0.38984718393.issue1492@bugs.pypy.org> Message-ID: <1369815903.58.0.734596414252.issue1492@bugs.pypy.org> Armin Rigo added the comment: Can you try 2.0.2? ---------- nosy: +arigo ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 29 11:07:54 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Wed, 29 May 2013 09:07:54 +0000 Subject: [pypy-issue] [issue1497] Pypy doesn't compile when ncurses is compiled with --with-termlib In-Reply-To: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> Message-ID: <1369818474.4.0.519907967996.issue1497@bugs.pypy.org> Armin Rigo added the comment: What, "ncurses5-config --cflags" works and gives you nothing? Bah, I get the same thing on Ubuntu 11.04. I'M ANNOYED. CRAPPY LIBRARIES. May I say. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 29 11:18:26 2013 From: tracker at bugs.pypy.org (alito) Date: Wed, 29 May 2013 09:18:26 +0000 Subject: [pypy-issue] [issue1497] Pypy doesn't compile when ncurses is compiled with --with-termlib In-Reply-To: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> Message-ID: <1369819106.39.0.650192226415.issue1497@bugs.pypy.org> alito added the comment: By would work, I meant that the results seem like they would compile, not that I've tried it. Nothing for --cflags sounds right to me (or -I/usr/include maybe?). Btw, pkg-config seems to return the same things as ncurses5-config. If this isn't reported often, then it's probably not as common a configuration as I thought it was. I've just recompiled my ncurses without terminfo, so -1 to this issue. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 29 11:36:15 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Wed, 29 May 2013 09:36:15 +0000 Subject: [pypy-issue] [issue1497] Pypy doesn't compile when ncurses is compiled with --with-termlib In-Reply-To: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> Message-ID: <1369820175.7.0.837021300918.issue1497@bugs.pypy.org> Armin Rigo added the comment: Ah, pkg-config. Yet another thing to try if all else fails. What is the exact command you have to run? ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 29 11:49:08 2013 From: tracker at bugs.pypy.org (alito) Date: Wed, 29 May 2013 09:49:08 +0000 Subject: [pypy-issue] [issue1497] Pypy doesn't compile when ncurses is compiled with --with-termlib In-Reply-To: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> Message-ID: <1369820948.13.0.550061127991.issue1497@bugs.pypy.org> alito added the comment: pkg-config --cflags ncurses and pkg-config --libs ncurses There's also more refined options. eg: pkg-config --libs-only-l ncurses (would skip the -L bit) Another useful one is: pkg-config --exists ncurses (returns 0 on success) ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Wed May 29 11:55:34 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Wed, 29 May 2013 09:55:34 +0000 Subject: [pypy-issue] [issue1497] Pypy doesn't compile when ncurses is compiled with --with-termlib In-Reply-To: <1369477122.56.0.189735082737.issue1497@bugs.pypy.org> Message-ID: <1369821334.44.0.477362686249.issue1497@bugs.pypy.org> Armin Rigo added the comment: Thanks! Done in c80297370acf, which should attempt a lot of stuff, trying them all: pkg-config ncurses, ncurses5-config, and various combination of built-in paths. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 30 09:56:38 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Thu, 30 May 2013 07:56:38 +0000 Subject: [pypy-issue] [issue1488] 'str(long)' slower in PyPy In-Reply-To: <1368775751.52.0.789251725482.issue1488@bugs.pypy.org> Message-ID: <1369900598.15.0.484404060828.issue1488@bugs.pypy.org> Armin Rigo added the comment: Getting this down to 50% slower is reasonable. That's on Linux. Please re-open this bug report if you see that for some reason Windows or OS/X is still seriously slower. ---------- status: chatting -> wontfix ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 30 12:23:17 2013 From: tracker at bugs.pypy.org (dktc) Date: Thu, 30 May 2013 10:23:17 +0000 Subject: [pypy-issue] [issue1502] trouble translating with older pypy In-Reply-To: <1369909397.15.0.78141707219.issue1502@bugs.pypy.org> Message-ID: <1369909397.15.0.78141707219.issue1502@bugs.pypy.org> New submission from dktc : First of all, here is why this is important. On CentOS 5.x the latest pre-built PyPy binary that's really stable is v1.8, available through Yum repo. Unfortunately trying to translate PyPy 2.0.2 on this PyPy 1.8 binary causes an error below. Could this be fixed? Without support for PyPy 1.8 it is problematic to translate PyPy on CentOS 5.x, since using CPython is way too slow and consumes way too much RAM... =============================== [translation:ERROR] File "/root/pypy/pypy-2.0.2-src/pypy/interpreter/baseobjspace.py", line 282, in _build [translation:ERROR] return self.build(key) [translation:ERROR] File "/root/pypy/pypy-2.0.2-src/pypy/interpreter/gateway.py", line 936, in build [translation:ERROR] defs = gateway._getdefaults(space) # needs to be implemented by subclass [translation:ERROR] File "/root/pypy/pypy-2.0.2-src/pypy/interpreter/gateway.py", line 884, in _getdefaults [translation:ERROR] self._code.identifier, name, defaultval)) [translation:ERROR] AssertionError: ("pypy.module.__pypy__.interp_signal-signals_exit-*: default value for 'w_ignored1' can only be None, got ''; use unwrap_spec(...=WrappedDefault(default))", < [translation:ERROR] v5 = getattr(w_dict_0, ('dstorage')) [translation:ERROR] In : [translation:ERROR] Happened at file /root/pypy/pypy-2.0.2-src/pypy/objspace/std/celldict.py line 53 [translation:ERROR] [translation:ERROR] ==> return self.unerase(w_dict.dstorage).get(key, None) [translation:ERROR] [translation:ERROR] Known variable annotations: [translation:ERROR] w_dict_0 = SomeInstance(can_be_None=False, classdef=pypy.objspace.std.dictmultiobject.W_DictMultiObject)>) [translation:ERROR] Processing block: [translation:ERROR] block at 3 is a [translation:ERROR] in (pypy.objspace.std.celldict:51)_getdictvalue_no_unwrapping_pure [translation:ERROR] containing the following operations: [translation:ERROR] v6 = getattr(self_0, ('unerase')) [translation:ERROR] v5 = getattr(w_dict_0, ('dstorage')) [translation:ERROR] v7 = simple_call(v6, v5) [translation:ERROR] v8 = getattr(v7, ('get')) [translation:ERROR] v9 = simple_call(v8, key_0, (None)) [translation:ERROR] --end-- ---------- messages: 5790 nosy: pypy-issue priority: bug status: unread title: trouble translating with older pypy ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 30 13:40:48 2013 From: tracker at bugs.pypy.org (Fijal) Date: Thu, 30 May 2013 11:40:48 +0000 Subject: [pypy-issue] [issue1502] trouble translating with older pypy In-Reply-To: <1369909397.15.0.78141707219.issue1502@bugs.pypy.org> Message-ID: <1369914048.89.0.0500198608006.issue1502@bugs.pypy.org> Fijal added the comment: it's a known bug in pypy 1.9. Use a newer pypy or cpython. ---------- nosy: +fijal status: unread -> wontfix ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Thu May 30 22:40:24 2013 From: tracker at bugs.pypy.org (dktc) Date: Thu, 30 May 2013 20:40:24 +0000 Subject: [pypy-issue] [issue1503] portability of PyPy binaries In-Reply-To: <1369946424.37.0.308088885419.issue1503@bugs.pypy.org> Message-ID: <1369946424.37.0.308088885419.issue1503@bugs.pypy.org> New submission from dktc : Since PyPy takes so much time and resources to build, I want to build/distribute it centrally within our environment (all same CentOS 5.8 systems). I built PyPy 2.0.2 on one of our boxes and then copied the build and virtualenv directories to the box where I want to use it. While I was able to run PyPy and install basic packages with PIP, I am running into trouble with the ones that compile native code (see log below). Any idea what might be missing? The same lxml package installs find on the initial build box. ==================================== (pypy-2.0.2)[root at localhost ~]# pip install lxml Downloading/unpacking lxml Running setup.py egg_info for package lxml Building lxml version 2.3.4. Building without Cython. Using build configuration of libxslt 1.1.17 Building against libxml2/libxslt in the following directory: /usr/lib64 Installing collected packages: lxml Running setup.py install for lxml Building lxml version 2.3.4. Building without Cython. Using build configuration of libxslt 1.1.17 Building against libxml2/libxslt in the following directory: /usr/lib64 building 'lxml.etree' extension cc -O2 -fPIC -Wimplicit -I/usr/include/libxml2 -I/opt/pypy-2.0.2/include -c src/lxml/lxml.etree.c -o build/temp.linux-x86_64-2.7/src/lxml/lxml.etree.o -w src/lxml/lxml.etree.c: In function ?__Pyx_PyObject_Pop?: src/lxml/lxml.etree.c:3684: error: ?PyListObject? has no member named ?allocated? src/lxml/lxml.etree.c: In function ?initetree?: src/lxml/lxml.etree.c:158525: error: ?PyWrapperDescrObject? undeclared (first use in this function) src/lxml/lxml.etree.c:158525: error: (Each undeclared identifier is reported only once src/lxml/lxml.etree.c:158525: error: for each function it appears in.) src/lxml/lxml.etree.c:158525: error: expected expression before ?)? token src/lxml/lxml.etree.c:158525: error: ?__pyx_wrapperbase_4lxml_5etree_8_Element_2__setitem__? has an incomplete type src/lxml/lxml.etree.c:158526: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158527: error: expected expression before ?)? token src/lxml/lxml.etree.c:158533: error: expected expression before ?)? token src/lxml/lxml.etree.c:158533: error: ?__pyx_wrapperbase_4lxml_5etree_8_Element_3__delitem__? has an incomplete type src/lxml/lxml.etree.c:158534: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158535: error: expected expression before ?)? token src/lxml/lxml.etree.c:158541: error: expected expression before ?)? token src/lxml/lxml.etree.c:158541: error: ?__pyx_wrapperbase_4lxml_5etree_8_Element_15__repr__? has an incomplete type src/lxml/lxml.etree.c:158542: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158543: error: expected expression before ?)? token src/lxml/lxml.etree.c:158549: error: expected expression before ?)? token src/lxml/lxml.etree.c:158549: error: ?__pyx_wrapperbase_4lxml_5etree_8_Element_16__getitem__? has an incomplete type src/lxml/lxml.etree.c:158550: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158551: error: expected expression before ?)? token src/lxml/lxml.etree.c:158557: error: expected expression before ?)? token src/lxml/lxml.etree.c:158557: error: ?__pyx_wrapperbase_4lxml_5etree_8_Element_17__len__? has an incomplete type src/lxml/lxml.etree.c:158558: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158559: error: expected expression before ?)? token src/lxml/lxml.etree.c:158565: error: expected expression before ?)? token src/lxml/lxml.etree.c:158565: error: ?__pyx_wrapperbase_4lxml_5etree_8_Element_19__contains__? has an incomplete type src/lxml/lxml.etree.c:158566: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158567: error: expected expression before ?)? token src/lxml/lxml.etree.c:158573: error: expected expression before ?)? token src/lxml/lxml.etree.c:158573: error: ?__pyx_wrapperbase_4lxml_5etree_8_Element_20__iter__? has an incomplete type src/lxml/lxml.etree.c:158574: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158575: error: expected expression before ?)? token src/lxml/lxml.etree.c:158670: error: expected expression before ?)? token src/lxml/lxml.etree.c:158670: error: ?__pyx_wrapperbase_4lxml_5etree_20__ContentOnlyElement_3__setitem__? has an incomplete type src/lxml/lxml.etree.c:158671: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158672: error: expected expression before ?)? token src/lxml/lxml.etree.c:158678: error: expected expression before ?)? token src/lxml/lxml.etree.c:158678: error: ?__pyx_wrapperbase_4lxml_5etree_20__ContentOnlyElement_4__getitem__? has an incomplete type src/lxml/lxml.etree.c:158679: error: invalid use of undefined type ?struct wrapperbase? src/lxml/lxml.etree.c:158680: error: expected expression before ?)? token src/lxml/lxml.etree.c:158686: error: expected expression before ?)? token src/lxml/lxml.etree.c:158686: error: ?__pyx_wrapperbase_4lxml_5etree_20__ContentOnlyElement_5__len__? has an incomplete type ..................... src/lxml/lxml.etree.c:164363: error: ?PyThreadState? has no member named ?exc_traceback? src/lxml/lxml.etree.c:164364: error: ?PyThreadState? has no member named ?exc_type? src/lxml/lxml.etree.c:164365: error: ?PyThreadState? has no member named ?exc_value? src/lxml/lxml.etree.c:164366: error: ?PyThreadState? has no member named ?exc_traceback? src/lxml/lxml.etree.c: In function ?__Pyx_ExceptionSave?: src/lxml/lxml.etree.c:164530: error: ?PyThreadState? has no member named ?exc_type? src/lxml/lxml.etree.c:164531: error: ?PyThreadState? has no member named ?exc_value? src/lxml/lxml.etree.c:164532: error: ?PyThreadState? has no member named ?exc_traceback? src/lxml/lxml.etree.c: In function ?__Pyx_ExceptionReset?: src/lxml/lxml.etree.c:164541: error: ?PyThreadState? has no member named ?exc_type? src/lxml/lxml.etree.c:164542: error: ?PyThreadState? has no member named ?exc_value? src/lxml/lxml.etree.c:164543: error: ?PyThreadState? has no member named ?exc_traceback? src/lxml/lxml.etree.c:164544: error: ?PyThreadState? has no member named ?exc_type? src/lxml/lxml.etree.c:164545: error: ?PyThreadState? has no member named ?exc_value? src/lxml/lxml.etree.c:164546: error: ?PyThreadState? has no member named ?exc_traceback? src/lxml/lxml.etree.c: In function ?__Pyx_Method_ClassMethod?: src/lxml/lxml.etree.c:164833: error: ?PyMethodDescrObject? undeclared (first use in this function) src/lxml/lxml.etree.c:164833: error: ?descr? undeclared (first use in this function) src/lxml/lxml.etree.c:164833: error: expected expression before ?)? token error: command 'cc' failed with exit status 1 ---------------------------------------- Command /opt/pypy-2.0.2/bin/pypy-c -c "import setuptools;__file__='/opt/pypy- 2.0.2/build/lxml/setup.py';exec(compile(open(__file__).read().replace('\r\n', '\n'), __file__, 'exec'))" install --record /tmp/pip-wyYlE9-record/install-record.txt --single- version-externally- managed --install-headers /opt/pypy-2.0.2/include/site/python2.7 failed with error code 1 in /opt/pypy-2.0.2/build/lxml Storing complete log in /root/.pip/pip.log ---------- messages: 5792 nosy: pypy-issue priority: bug status: unread title: portability of PyPy binaries ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 07:59:41 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 05:59:41 +0000 Subject: [pypy-issue] [issue1503] portability of PyPy binaries In-Reply-To: <1369946424.37.0.308088885419.issue1503@bugs.pypy.org> Message-ID: <1369979980.98.0.318842112454.issue1503@bugs.pypy.org> Armin Rigo added the comment: Compare the output of "pip install lxml" on the two machines? ---------- nosy: +arigo status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 08:01:46 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 06:01:46 +0000 Subject: [pypy-issue] [issue1492] More PyPy eventlet problems In-Reply-To: <1369233718.16.0.38984718393.issue1492@bugs.pypy.org> Message-ID: <1369980106.88.0.425352348171.issue1492@bugs.pypy.org> Armin Rigo added the comment: Ah, sorry dgulino. It seems your bug report was "hijacked" by dktc for a possibly different problem. For future references: this bug report now described *two* problems. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 08:51:16 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 06:51:16 +0000 Subject: [pypy-issue] [issue1504] More PyPy eventlet issues In-Reply-To: <1369983076.21.0.0696376437796.issue1504@bugs.pypy.org> Message-ID: <1369983076.21.0.0696376437796.issue1504@bugs.pypy.org> New submission from Armin Rigo : >From dktc: When EVENTLET_THREADPOOL_SIZE env variable is set ( see http://eventlet.net/doc/threading.html ) eventlet crashes on PyPy 2.0.1. The crash does not happen on Python 2.7.3 or when the variable is not present. I do not have a compact code sample to reproduce this, since eventlet is being invoked from Spawning ( https://pypi.python.org/pypi/Spawning ). I just see that when Spawning settings cause creation of a subprocess (through os.execve call) with EVENTLET_THREADPOOL_SIZE env var, this subprocess fails: "(21432) Child died from signal 11 with code 0" Any thoughts why this could be happening on PyPy? ---------- messages: 5795 nosy: arigo, pypy-issue priority: bug status: unread title: More PyPy eventlet issues ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 08:51:43 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 06:51:43 +0000 Subject: [pypy-issue] [issue1492] More PyPy eventlet problems In-Reply-To: <1369233718.16.0.38984718393.issue1492@bugs.pypy.org> Message-ID: <1369983103.67.0.425471814631.issue1492@bugs.pypy.org> Armin Rigo added the comment: Moved the dktc issue into https://bugs.pypy.org/issue1504 . ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 08:52:37 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 06:52:37 +0000 Subject: [pypy-issue] [issue1504] More PyPy eventlet issues In-Reply-To: <1369983076.21.0.0696376437796.issue1504@bugs.pypy.org> Message-ID: <1369983157.69.0.0487174990768.issue1504@bugs.pypy.org> Armin Rigo added the comment: Can you try 2.0.2? ---------- status: unread -> chatting ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 10:22:46 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 08:22:46 +0000 Subject: [pypy-issue] [issue1492] More PyPy eventlet problems In-Reply-To: <1369233718.16.0.38984718393.issue1492@bugs.pypy.org> Message-ID: <1369988566.22.0.857125487985.issue1492@bugs.pypy.org> Armin Rigo added the comment: Fixed in d439d104a605. Thanks for the report, it seems that eventlet never worked (at least with http urls), but you're the first to actually report it. ---------- status: chatting -> resolved ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 10:24:20 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 08:24:20 +0000 Subject: [pypy-issue] [issue1504] More PyPy eventlet issues In-Reply-To: <1369983076.21.0.0696376437796.issue1504@bugs.pypy.org> Message-ID: <1369988660.36.0.722586452037.issue1504@bugs.pypy.org> Armin Rigo added the comment: Works for me. Please give steps to reproduce. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 10:26:45 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 08:26:45 +0000 Subject: [pypy-issue] [issue1495] Crash in JIT from Bitbucket latest (checkout of 5/24/2013 at 1pm PST) In-Reply-To: <1369439357.27.0.793804232803.issue1495@bugs.pypy.org> Message-ID: <1369988805.55.0.548060741019.issue1495@bugs.pypy.org> Armin Rigo added the comment: Well, it's a real bug in the JIT. It's one of the many asserts in the rather obscure function compile_retrace() that fails. I'm overall unhappy about the state of the part of the JIT related to retracing/unrolling/etc. ________________________________________ PyPy bug tracker ________________________________________ From tracker at bugs.pypy.org Fri May 31 17:35:41 2013 From: tracker at bugs.pypy.org (Armin Rigo) Date: Fri, 31 May 2013 15:35:41 +0000 Subject: [pypy-issue] [issue1505] Removing multimethods slows the nojit case In-Reply-To: <1370014541.14.0.57716620094.issue1505@bugs.pypy.org> Message-ID: <1370014541.14.0.57716620094.issue1505@bugs.pypy.org> New submission from Armin Rigo : Removing multimethods had a big performance hit on the no-jit case, sometimes. I suspect it can be fixed relatively easily. We *should* probably fix it. Revision 93322cb79586 (hg merge remove-list-smm-2) makes nbody_modified much slower; we can see that it makes this kind of code almost twice as slow: lst = [1.2, 3.4] for i in range(BIG): lst[1] = lst[1] + n Revision d1459869ce85 (Merge remove-array-smm branch that kills multimethods from the array module) makes crypto_pyaes.py much slower; not surprizingly, it uses array. "Much slower" is almost a factor two in both cases. ---------- messages: 5801 nosy: arigo, pypy-issue priority: performance bug status: unread title: Removing multimethods slows the nojit case ________________________________________ PyPy bug tracker ________________________________________