From tracker at bugs.pypy.org  Mon Mar  3 15:19:21 2014
From: tracker at bugs.pypy.org (coli)
Date: Mon, 03 Mar 2014 14:19:21 +0000
Subject: [pypy-issue] [issue1696] Can not install pip on Windows
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>


New submission from coli <coli at rocketmail.com>:

I ran "pypy get-pip.py" after installing the windows binary.

Downloading/unpacking pip
  Getting page https://pypi.python.org/simple/pip/
  URLs to search for versions for pip:
  * https://pypi.python.org/simple/pip/
  Analyzing links from page https://pypi.python.org/simple/pip/
    Found link https://pypi.python.org/packages/py2.py3/p/pip/pip-1.5.1-py2.py3-
none-any.whl#md5=237164a09943d823b954bb9e1b2a8f67 (from 
https://pypi.python.org/simple/pip/), version: 1.5.1
    Found link https://pypi.python.org/packages/py2.py3/p/pip/pip-1.5.2-py2.py3-
none-any.whl#md5=445a893564065937c0f31ac2cc8e2f35 (from 
https://pypi.python.org/simple/pip/), version: 1.5.2
    Found link https://pypi.python.org/packages/py2.py3/p/pip/pip-1.5.3-py2.py3-
none-any.whl#md5=78126f6352f3d5f29d077be0aa7f9efa (from 
https://pypi.python.org/simple/pip/), version: 1.5.3
    Found link https://pypi.python.org/packages/py2.py3/p/pip/pip-1.5.4-py2.py3-
none-any.whl#md5=f20bc8f31e322375ce06b26d73ce7b4f (from 
https://pypi.python.org/simple/pip/), version: 1.5.4
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.2.1.tar.gz#md5=d2af655c1a87e03799442d045c1d6743 (from 
https://pypi.python.org/simple/pip/), version: 0.2.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.2.tar.gz#md5=9eda07c8be7105aa774c7eb51c023294 (from 
https://pypi.python.org/simple/pip/), version: 0.2
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.3.1.tar.gz#md5=78102ddbb040a183dd361b5d432cdf88 (from 
https://pypi.python.org/simple/pip/), version: 0.3.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.3.tar.gz#md5=8fccb5b49c6377cbfb1949ccd7be43b3 (from 
https://pypi.python.org/simple/pip/), version: 0.3
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.4.tar.gz#md5=b45714d04f8fd38fe8e3d4c7600b91a2 (from 
https://pypi.python.org/simple/pip/), version: 0.4
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.5.1.tar.gz#md5=d4bdaa5f5f5bf8c6263ace75a0882232 (from 
https://pypi.python.org/simple/pip/), version: 0.5.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.5.tar.gz#md5=63eed8673e40628534cc0aa9c98e8f3d (from 
https://pypi.python.org/simple/pip/), version: 0.5
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.6.1.tar.gz#md5=7560e3055c66afb99ac4a7892389a237 (from 
https://pypi.python.org/simple/pip/), version: 0.6.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.6.2.tar.gz#md5=9a43e0a2ce8833069f41c347932bdb25 (from 
https://pypi.python.org/simple/pip/), version: 0.6.2
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.6.3.tar.gz#md5=0602fa9179cfaa98e41565d4a581d98c (from 
https://pypi.python.org/simple/pip/), version: 0.6.3
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.6.tar.gz#md5=083ff408ca5314cf0561ff79a048cd9a (from 
https://pypi.python.org/simple/pip/), version: 0.6
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.7.1.tar.gz#md5=420c83ad67bdcb542f772eb64392cce6 (from 
https://pypi.python.org/simple/pip/), version: 0.7.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.7.2.tar.gz#md5=cfe73090aaa0d3b0c104179a627859d1 (from 
https://pypi.python.org/simple/pip/), version: 0.7.2
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.7.tar.gz#md5=8d4b4f7266fe0808569182e6832d74ac (from 
https://pypi.python.org/simple/pip/), version: 0.7
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.8.1.tar.gz#md5=5d40614774781b118dd3f10c0d038cbc (from 
https://pypi.python.org/simple/pip/), version: 0.8.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.8.2.tar.gz#md5=df1eca0abe7643d92b5222240bed15f6 (from 
https://pypi.python.org/simple/pip/), version: 0.8.2
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.8.3.tar.gz#md5=0603337a81f83df2b1d2f1151565efac (from 
https://pypi.python.org/simple/pip/), version: 0.8.3
    Found link https://pypi.python.org/packages/source/p/pip/pip-
0.8.tar.gz#md5=468d9adc309f33ad51cee38f0d455429 (from 
https://pypi.python.org/simple/pip/), version: 0.8
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.0.1.tar.gz#md5=28dcc70225e5bf925532abc5b087a94b (from 
https://pypi.python.org/simple/pip/), version: 1.0.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.0.2.tar.gz#md5=47ec6ff3f6d962696fe08d4c8264ad49 (from 
https://pypi.python.org/simple/pip/), version: 1.0.2
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.0.tar.gz#md5=327fc4a03df189506966e15021730550 (from 
https://pypi.python.org/simple/pip/), version: 1.0
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.1.tar.gz#md5=62a9f08dd5dc69d76734568a6c040508 (from 
https://pypi.python.org/simple/pip/), version: 1.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.2.1.tar.gz#md5=db8a6d8a4564d3dc7f337ebed67b1a85 (from 
https://pypi.python.org/simple/pip/), version: 1.2.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.2.tar.gz#md5=99e44d1d7f88b182459e7c19c45aee9f (from 
https://pypi.python.org/simple/pip/), version: 1.2
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.3.1.tar.gz#md5=cbb27a191cebc58997c4da8513863153 (from 
https://pypi.python.org/simple/pip/), version: 1.3.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.3.tar.gz#md5=918559b784e2aca9559d498050bb86e7 (from 
https://pypi.python.org/simple/pip/), version: 1.3
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.4.1.tar.gz#md5=6afbb46aeb48abac658d4df742bff714 (from 
https://pypi.python.org/simple/pip/), version: 1.4.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.4.tar.gz#md5=ca790be30004937987767eac42cfa44a (from 
https://pypi.python.org/simple/pip/), version: 1.4
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.5.1.tar.gz#md5=4678c2ae5cce4e9234c3923d7dcb32f0 (from 
https://pypi.python.org/simple/pip/), version: 1.5.1
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.5.2.tar.gz#md5=5da30919f732d68b1c666e484e7676f5 (from 
https://pypi.python.org/simple/pip/), version: 1.5.2
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.5.3.tar.gz#md5=78871c5f84ea5fca61900347ce7864ad (from 
https://pypi.python.org/simple/pip/), version: 1.5.3
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.5.4.tar.gz#md5=834b2904f92d46aaa333267fb1c922bb (from 
https://pypi.python.org/simple/pip/), version: 1.5.4
    Found link https://pypi.python.org/packages/source/p/pip/pip-
1.5.tar.gz#md5=6969b8a8adc4c7f7c5eb1707118f0686 (from 
https://pypi.python.org/simple/pip/), version: 1.5
    Skipping https://github.com/pypa/pip/tarball/develop#egg=pip-dev (from 
https://pypi.python.org/simple/pip/) because it is externally hosted.
  Using version 1.5.4 (newest of versions: 1.5.4, 1.5.4, 1.5.3, 1.5.3, 1.5.2, 
1.5.2, 1.5.1, 1.5.1, 1.5, 1.4.1, 1.4, 1.3.1, 1.3, 1.2.1, 1.2, 1.1, 
1.0.2, 1.0.1, 1.0, 0.8.3, 0.8.2, 0.8.1, 0.8, 0.7.2, 0.7.1, 0.7, 0.6.3, 0.6.2, 
0.6.1, 0.6, 0.5.1, 0.5, 0.4, 0.3.1, 0.3, 0.2.1, 0.2)
Cleaning up...
  Removing temporary dir c:\users\owl\appdata\local\temp\pip_build_owl...
Exception:
Traceback (most recent call last):
  File "c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/basecommand.py", 
line 122, in main
    status = self.run(options, args)
  File 
"c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/commands/install.py", 
line 274, in run
    requirement_set.prepare_files(finder, force_root_egg_info=self.bundle, 
bundle=self.bundle)
  File "c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/req.py", line 
1173, in prepare_files
    self.unpack_url(url, location, self.is_download)
  File "c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/req.py", line 
1320, in unpack_url
    retval = unpack_http_url(link, location, self.download_cache, 
self.download_dir, self.session)
  File "c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/download.py", line 
555, in unpack_http_url
    resp = session.get(target_url, stream=True)
  File 
"c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/_vendor/requests/sessions
.py", line 395, in get
    return self.request('GET', url, **kwargs)
  File "c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/download.py", line 
237, in request
    return super(PipSession, self).request(method, url, *args, **kwargs)
  File 
"c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/_vendor/requests/sessions
.py", line 383, in request
    resp = self.send(prep, **send_kwargs)
  File 
"c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/_vendor/requests/sessions
.py", line 486, in send
    r = adapter.send(request, **kwargs)
  File 
"c:\users\owl\appdata\local\temp\tmpa8zmqn\pip.zip\pip/_vendor/requests/adapters
.py", line 378, in send
    raise ConnectionError(e)
ConnectionError: HTTPSConnectionPool(host='pypi.python.org', port=443): Max 
retries exceeded with url: /packages/py2.py3/p/pip/pip-1.5.4-py2.py3-none-
any.whl (Caused by <class 'httplib.BadStatusLine'>: '')

----------
messages: 6559
nosy: coli, pypy-issue
priority: critical
status: unread
title: Can not install pip on Windows

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Mon Mar  3 15:20:40 2014
From: tracker at bugs.pypy.org (coli)
Date: Mon, 03 Mar 2014 14:20:40 +0000
Subject: [pypy-issue] [issue1696] Can not install pip on Windows
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1393856440.85.0.882243738744.issue1696@bugs.pypy.org>


coli <coli at rocketmail.com> added the comment:

This is 2.2.1, I also tried latest trunk.

----------
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Mon Mar  3 16:29:22 2014
From: tracker at bugs.pypy.org (Kyle Lahnakoski)
Date: Mon, 03 Mar 2014 15:29:22 +0000
Subject: [pypy-issue] [issue1697] utcfromtimestamp raises ValueError: time
	argument too	large on Windows7
In-Reply-To: <1393860562.96.0.0357954768575.issue1697@bugs.pypy.org>
Message-ID: <1393860562.96.0.0357954768575.issue1697@bugs.pypy.org>


New submission from Kyle Lahnakoski <kyle at lahnakoski.com>:

utcfromtimestamp(9999999999) raises a ValueError in PyPy.  It does not happen in
cPython.


c:\Users\klahnakoski\git\Bugzilla-ETL>python
Python 2.7.4 (default, Apr  6 2013, 19:54:46) [MSC v.1500 32 bit (Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import datetime
>>> datetime.datetime.utcfromtimestamp(9999999999)
datetime.datetime(2286, 11, 20, 17, 46, 39)
>>> exit()



c:\Users\klahnakoski\git\Bugzilla-ETL>pypy
Python 2.7.3 (2cec7296d7fb, Nov 12 2013, 13:24:40)
[PyPy 2.2.0 with MSC v.1500 32 bit] on win32
Type "help", "copyright", "credits" or "license" for more information.
And now for something completely different: ``ctypes has a spin of 1/3''
>>>> import datetime
>>>> datetime.datetime.utcfromtimestamp(9999999999)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "c:\PyPy27\lib_pypy\datetime.py", line 1530, in utcfromtimestamp
    y, m, d, hh, mm, ss, weekday, jday, dst = _time.gmtime(t)
ValueError: time argument too large
>>>> exit()

----------
messages: 6561
nosy: klahnakoski at mozilla.com, pypy-issue
priority: bug
status: unread
title: utcfromtimestamp raises ValueError: time argument too large on Windows7

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1697>
________________________________________

From tracker at bugs.pypy.org  Tue Mar  4 23:35:39 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Tue, 04 Mar 2014 22:35:39 +0000
Subject: [pypy-issue] [issue1630] Matrix dot product ~5x slower in PyPy
In-Reply-To: <1383654269.15.0.0619090956137.issue1630@bugs.pypy.org>
Message-ID: <1393972539.2.0.769987261825.issue1630@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

I spent some time optimizing dot/iterators, we are now ~1.5x cpython+numpy C dot 
loop in this example. Added an improved benchmark script in 
pypy/module/micronumpy/bench/dot.py.

----------
nosy: +bdk

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1630>
________________________________________

From tracker at bugs.pypy.org  Wed Mar  5 20:13:35 2014
From: tracker at bugs.pypy.org (dmlockhart)
Date: Wed, 05 Mar 2014 19:13:35 +0000
Subject: [pypy-issue] [issue1698] cffi.FFI.cdef fails when source string
	contains $ symbols
In-Reply-To: <1394046815.3.0.470269397227.issue1698@bugs.pypy.org>
Message-ID: <1394046815.3.0.470269397227.issue1698@bugs.pypy.org>


New submission from dmlockhart <dmlockhart+pypy at gmail.com>:

I get the following error when trying to parse C source containing $ symbols:

E       CDefError: cannot parse "unsigned int   _top_in_$000_msg,"
E       :6:25: Illegal character '$'

This works using cffi-0.8.1 in CPython, so I suspect newer version of 
pycparser/cffi fix this issue.  Unfortunately, there doesn't seem to be a way to 
force pypy to use newer versions of cffi since it is integrated in lib_pypy.

----------
messages: 6563
nosy: dmlockhart, pypy-issue
priority: bug
status: unread
title: cffi.FFI.cdef fails when source string contains $ symbols

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1698>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  6 12:36:47 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Thu, 06 Mar 2014 11:36:47 +0000
Subject: [pypy-issue] [issue1698] cffi.FFI.cdef fails when source string
	contains $ symbols
In-Reply-To: <1394046815.3.0.470269397227.issue1698@bugs.pypy.org>
Message-ID: <1394105807.52.0.664294005402.issue1698@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

It seems to me that the $ character is illegal in C.  I don't understand the issue.

----------
nosy: +arigo
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1698>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  6 12:39:38 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Thu, 06 Mar 2014 11:39:38 +0000
Subject: [pypy-issue] [issue1698] cffi.FFI.cdef fails when source string
	contains $ symbols
In-Reply-To: <1394046815.3.0.470269397227.issue1698@bugs.pypy.org>
Message-ID: <1394105978.63.0.70168800724.issue1698@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Ok, sorry.  I didn't know that you can use the $ character freely in
identifiers. :-/  As a temporary workaround, just update
lib_pypy/cffi/_pycparser from a more recent version of pycparser.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1698>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  6 12:49:23 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Thu, 06 Mar 2014 11:49:23 +0000
Subject: [pypy-issue] [issue1698] cffi.FFI.cdef fails when source string
	contains $ symbols
In-Reply-To: <1394046815.3.0.470269397227.issue1698@bugs.pypy.org>
Message-ID: <1394106563.58.0.392032907684.issue1698@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Updated to pycparser 2.10.  You can do it manually too by copying the files
pycparser-2.10/pycparser/* to lib_pypy/cffi/_pycparser/.

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1698>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  6 14:27:02 2014
From: tracker at bugs.pypy.org (lesshaste)
Date: Thu, 06 Mar 2014 13:27:02 +0000
Subject: [pypy-issue] [issue1699] 'module' object has no attribute
	'correlate'
In-Reply-To: <1394112422.1.0.0844440275921.issue1699@bugs.pypy.org>
Message-ID: <1394112422.1.0.0844440275921.issue1699@bugs.pypy.org>


New submission from lesshaste <drraph at gmail.com>:

When trying to run this simple (if slightly dumb) script using Thu Mar  6 
nightly I get

AttributeError: 'module' object has no attribute 'correlate'

----------
files: detectkbit.py
messages: 6567
nosy: lesshaste, pypy-issue
priority: feature
release: ???
status: unread
title: 'module' object has no attribute 'correlate'

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1699>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  6 19:25:20 2014
From: tracker at bugs.pypy.org (mattip)
Date: Thu, 06 Mar 2014 18:25:20 +0000
Subject: [pypy-issue] [issue1653] performance regressions in benchmarks
In-Reply-To: <1386292390.88.0.0820526190869.issue1653@bugs.pypy.org>
Message-ID: <1394130320.2.0.64958032978.issue1653@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

here is the results of using  
PYPYLOG=jit-summary:logfile pypy go.py -n 5
and agregating the logs over 10 seperate runs of the command
Note that before the jit-counter branch, the accumulated jit statistics are very 
consistent, and after they not only vary widely but never reach the level of 
pre-merge.

----------
nosy: +mattip

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1653>
________________________________________

From tracker at bugs.pypy.org  Thu Mar  6 22:16:16 2014
From: tracker at bugs.pypy.org (dmlockhart)
Date: Thu, 06 Mar 2014 21:16:16 +0000
Subject: [pypy-issue] [issue1680] ast.NodeTransformer throws AttributeError
	in pypy only
In-Reply-To: <1390424807.78.0.389244299554.issue1680@bugs.pypy.org>
Message-ID: <1394140576.79.0.90922061629.issue1680@bugs.pypy.org>


dmlockhart <dmlockhart+pypy at gmail.com> added the comment:

A temporary work around is to modify the Python ast library to set attributes to 
None, instead of deleting them.

In lib-python/2.7/ast.py:

308                     #delattr(node, field)
309                     setattr(node, field, None)


I'm currently working on a fix so that delattr works properly on nodes in _ast.

----------
release:  -> 2.2
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1680>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 08:18:25 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 07 Mar 2014 07:18:25 +0000
Subject: [pypy-issue] [issue1653] performance regressions in benchmarks
In-Reply-To: <1386292390.88.0.0820526190869.issue1653@bugs.pypy.org>
Message-ID: <1394176705.27.0.64065445406.issue1653@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

mattip: can you also give a similar log for trunk?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1653>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 08:41:27 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 07 Mar 2014 07:41:27 +0000
Subject: [pypy-issue] [issue1696] Can not install pip on Windows
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1394178087.11.0.0416085458441.issue1696@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

The same bug occurs on Linux.

----------
nosy: +arigo

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 09:05:14 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 07 Mar 2014 08:05:14 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1394179514.88.0.0839584565062.issue1696@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Can you double-check the version numbers?  I'm getting exactly the same bug
whenever I try (on Linux, various machines) pypy 2.2; but this seems to have
been resolved in 2.2.1.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 12:39:53 2014
From: tracker at bugs.pypy.org (mattip)
Date: Fri, 07 Mar 2014 11:39:53 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1394192393.28.0.940471924358.issue1696@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

I am getting the BadStatusLine error for the second call to the ssl wrapped 
socket on many versions of win32, including latest nightly. It seems we don't 
actually have a test (yet) for two consecutive calls to a ssl server where the 
first one returns 1265 bytes of payload. So far I have put debug printing into 
httplib.py but cannot find why the second call is timimg out, The connection 
still has _connected==True, but that doesn't mean much apparantly

----------
nosy: +mattip

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 13:13:10 2014
From: tracker at bugs.pypy.org (mattip)
Date: Fri, 07 Mar 2014 12:13:10 +0000
Subject: [pypy-issue] [issue1653] performance regressions in benchmarks
In-Reply-To: <1386292390.88.0.0820526190869.issue1653@bugs.pypy.org>
Message-ID: <1394194390.25.0.836941248054.issue1653@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

done. The accumulated statistics seem more stable, but the best run time still 
varies ~%10
Perhaps this is OK?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1653>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 15:27:24 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Fri, 07 Mar 2014 14:27:24 +0000
Subject: [pypy-issue] [issue1653] performance regressions in benchmarks
In-Reply-To: <1386292390.88.0.0820526190869.issue1653@bugs.pypy.org>
Message-ID: <1394202444.25.0.380105748616.issue1653@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

There is still a regression demonstrated by the script here: 
https://bugs.pypy.org/issue1678. Look at timings pre-jit-counter, post, and trunk. 
Trunk hasn't improved the regression seen in the move from pre to post-jit-
counter.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1653>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 16:34:38 2014
From: tracker at bugs.pypy.org (Carl Friedrich Bolz)
Date: Fri, 07 Mar 2014 15:34:38 +0000
Subject: [pypy-issue] [issue1700] int_floordiv is badly named
In-Reply-To: <1394206478.62.0.51620380011.issue1700@bugs.pypy.org>
Message-ID: <1394206478.62.0.51620380011.issue1700@bugs.pypy.org>


New submission from Carl Friedrich Bolz <cfbolz at gmx.de>:

int_floordiv should really be int_truncdiv everywhere, as it does not round
down, but towards 0.

----------
messages: 6576
nosy: cfbolz, pypy-issue
priority: bug
status: unread
title: int_floordiv is badly named

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1700>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 16:40:08 2014
From: tracker at bugs.pypy.org (Carl Friedrich Bolz)
Date: Fri, 07 Mar 2014 15:40:08 +0000
Subject: [pypy-issue] [issue1701] a / b with integers could be faster in the
	JIT
In-Reply-To: <1394206808.89.0.474094609347.issue1701@bugs.pypy.org>
Message-ID: <1394206808.89.0.474094609347.issue1701@bugs.pypy.org>


New submission from Carl Friedrich Bolz <cfbolz at gmx.de>:

a / b with integers is currently doing something like:

r = a / b
p = r * y
if y < 0: u = p - x
else:     u = x - p
return r + (u >> INT_BITS_1)

the first two operations could be done with just one IDIV instruction on X86.
For the rest it's maybe better to write it branch-free.


discussion:

16:08:06 <cfbolz> indeed, before rtyping it's just "floordiv"
16:08:22 <arigato> and then "floordiv" is turned into... not "int_floordiv"
16:09:11 <cfbolz> int_truncdiv + a call
16:09:19 <arigato> yes
16:10:33 <cfbolz> arigato: might not get around to it now, but will keep it in mind
16:10:37 <arigato> ok
16:10:50 <arigato> I *think* the current slightly strange situation gives good
results for the jit
16:11:45 <cfbolz> no clue
16:12:13 <arigato> at least I remember when int_floordiv was meant to have
Python semantics too
16:12:54 <arigato> it was more messy in the jit, and then sometimes for nothing
if you really wanted a C-like behavior in the end
16:13:14 <cfbolz> right
16:14:11 <cfbolz> arigato: still, you get thiskind of code:
16:14:17 <cfbolz> https://www.irccloud.com/pastebin/x2jo6KoG
16:15:10 <arigato> right, but one way or another, the RPython "/" operator needs
to translate to this
16:15:35 <cfbolz> arigato: I am worried by the last guard a bit
16:15:45 <arigato> (probably someone with the right mindset can come up with a
different set of instructions that is somehow better)
16:15:57 <arigato> cfbolz: ah, I see
16:15:59 <cfbolz> arigato: because it can really make bridges
16:16:08 <arigato> ok
16:16:48 <arigato> it comes out of ll_correct_int_floordiv()
16:17:00 <cfbolz> can we write this branch free
16:17:19 <arigato> first step is to remember why this works at all :-)
16:17:44 <cfbolz> :-)
16:17:59 <arigato> right, I think the branch was put there specifically because
it's the less likely to have different values at run-time
16:18:18 <arigato> "x/y" where y is sometimes positive and sometimes negative...
is probably a bit rare
16:18:28 <cfbolz> ah, no
16:18:42 <cfbolz> arigato: it's maybe even because then if you know that y is
positive you can constant fold
16:19:00 <arigato> yes, that too
16:19:14 <cfbolz> arigato: because division by a constant factor
16:20:25 <cfbolz> still confused how it works
16:21:09 <arigato> (u >> INT_BITS_1) is -1 if u < 0, or 0 otherwise
16:21:37 <cfbolz> yes
16:21:41 <cfbolz> but what is u?
16:22:22 <arigato> u is computed to be 0 if the division was exact, or < 0 if not
16:22:37 <cfbolz> right
16:23:05 <arigato> a branch-free variant would be: return r + (((x - p) ^ y) >>
INT_BITS_1)
16:23:34 ? willingc joined (~willingc at cpe-75-80-19-24.san.res.rr.com)
16:23:43 <cfbolz> arigato: which means we would replace a guard by an xor
16:23:44 <arigato> the trade-off is if we already know that y > 0
16:23:49 <arigato> yes
16:24:20 <arigato> this kind of code would be all constant-folded away, right?
if jit.isconstant(y > 0):...
16:24:21 <cfbolz> no clue :-(
16:24:41 <cfbolz> why all?
16:24:49 <cfbolz> only the guard would go away, no?
16:24:54 ? Wessie joined (~Wessie at ip5651e009.adsl-surfen.hetnet.nl)
16:25:05 <arigato> yes yes, I mean the "jit.isconstant()" part
16:25:24 <cfbolz> yes
16:25:28 <arigato> but it's probably not going to ever say True
16:25:38 <arigato> y is known to be positive later
16:25:42 <arigato> not during tracing
16:25:48 <cfbolz> yes, e.g. if it's loop invariant
16:26:01 <arigato> hum
16:26:16 <arigato> the alternative is to add a correct (but very special-case)
arithmetic optmization:
16:26:33 <arigato> i3 = i1 ^ i2; i4 = i3 >> INT_BITS_1
16:26:55 <arigato> if i2 is known to be positive, then we can substitute i3 for
i1 in the computation of i4
16:27:07 <cfbolz> arigato: very special case :-)
16:27:15 <cfbolz> arigato: hm, in x86 assembler you could indeed do better,
because IDIV computes both x / y and the remainder in one instruction, no?
16:27:25 <arigato> ah, yes
16:27:27 <cfbolz> so you wouldn't need the multiply
16:27:38 ? zer0def quit (~null at 5.254.138.19) Ping timeout: 240 seconds
16:27:39 <arigato> indeed
16:28:13 <cfbolz> so we would need an int_divrem (which we can't have, because
it returns two things)
16:28:34 <arigato> it's probably not a problem:
16:29:10 <arigato> we can arrange things such that in the end the jit backend
sees a INT_TRUNCDIV immediately followed by a INT_REM
16:29:35 <cfbolz> I see
16:30:56 <arigato> (btw, same about 'mod' vs 'int_mod')
16:31:03 <cfbolz> yes

----------
messages: 6577
nosy: cfbolz, pypy-issue
priority: performance bug
status: unread
title: a / b with integers could be faster in the JIT

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1701>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 17:36:07 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 07 Mar 2014 16:36:07 +0000
Subject: [pypy-issue] [issue1701] a / b with integers could be faster in the
	JIT
In-Reply-To: <1394206808.89.0.474094609347.issue1701@bugs.pypy.org>
Message-ID: <1394210167.71.0.326610956419.issue1701@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Sorry, I said something wrong in that IRC extract: we can't just use one xor,
it's a bit more complicated.

The branch is the "cheapest" I could reasonably come up with.  A way to use the
combined div-and-mod x86 operation would be to recognize, in the x86 backend, a
combination of two resoperations, INT_TRUNCDIV followed by INT_TRUNCMOD.  It
could be written by using this kind of logic in rtyper/rint.py:

d = int_truncdiv(a, b)
m = int_truncmod(a, b)
if b > 0:
    d += (a&m)>>INT_BITS_1
else:
    d += ((-m)&(-a)) >>INT_BITS_1

Or without any branch:

d = int_truncdiv(a, b)
m = int_truncmod(a, b)
d += ((a^b)&(m|-m))>>INT_BITS_1

It goes far into the land of obscurity if you ask me...

----------
nosy: +arigo
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1701>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 18:02:17 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 07 Mar 2014 17:02:17 +0000
Subject: [pypy-issue] [issue1678] example 20x slower after the first time
In-Reply-To: <1390301150.19.0.0852731519756.issue1678@bugs.pypy.org>
Message-ID: <1394211737.98.0.0513598590815.issue1678@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Found out by bdk that this got much worse when the jit-counter branch was merged.

----------
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1678>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 18:38:12 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Fri, 07 Mar 2014 17:38:12 +0000
Subject: [pypy-issue] [issue1678] example 20x slower after the first time
In-Reply-To: <1390301150.19.0.0852731519756.issue1678@bugs.pypy.org>
Message-ID: <1394213892.21.0.430518035858.issue1678@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

ubuntu at leno:~$ ./pypy-c-jit-67776-748abf94a70d-linux64/bin/pypy pypy/x.py
0.161542892456
0.619294881821
0.613882780075
^C^[[ATraceback (most recent call last):
  File "app_main.py", line 72, in run_toplevel
  File "pypy/x.py", line 10, in <module>
    fibonacci(32)
KeyboardInterrupt
ubuntu at leno:~$ ./pypy-c-jit-67802-b96a176fed01-linux64/bin/pypy pypy/x.py
0.179089069366
2.2699611187
2.27039408684
^CTraceback (most recent call last):
  File "app_main.py", line 72, in run_toplevel
  File "pypy/x.py", line 10, in <module>
    fibonacci(32)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
  File "pypy/x.py", line 6, in fibonacci
    return fibonacci(n - 2) + fibonacci(n - 1)
...

----------
nosy: +bdk

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1678>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 21:37:32 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 07 Mar 2014 20:37:32 +0000
Subject: [pypy-issue] [issue1697] utcfromtimestamp raises ValueError: time
	argument too	large on Windows7
In-Reply-To: <1393860562.96.0.0357954768575.issue1697@bugs.pypy.org>
Message-ID: <1394224652.31.0.892011095885.issue1697@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

I committed 3f834e4349e4, which should fix this problem.  Please feel free to
reopen this bug report if it doesn't.

----------
nosy: +arigo
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1697>
________________________________________

From tracker at bugs.pypy.org  Fri Mar  7 21:41:38 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 07 Mar 2014 20:41:38 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1394224898.96.0.774361983703.issue1696@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Hi Matti!  If you give up on this one, please remember to post the small example
you came up with (two consecutive calls to ssl) so that we can take over.  But
thanks a lot for looking into it!

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Sat Mar  8 10:51:05 2014
From: tracker at bugs.pypy.org (iElectric)
Date: Sat, 08 Mar 2014 09:51:05 +0000
Subject: [pypy-issue] [issue1702] Transient failures of
	test	test.test_multiprocessing.WithProcessesTestCondition
In-Reply-To: <1394272265.97.0.487971766294.issue1702@bugs.pypy.org>
Message-ID: <1394272265.97.0.487971766294.issue1702@bugs.pypy.org>


New submission from iElectric <domen at dev.si>:

Our build farm get's a test failure on 64bit linux about every 10 builds with 
following traceback using pypy 2.2.1:


======================================================================
FAIL: test_notify_all (test.test_multiprocessing.WithProcessesTestCondition)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/tmp/nix-build-pypy-2.2.1.drv-0/pypy-2.2.1-src/lib-
python/2.7/test/test_multiprocessing.py", line 794, in test_notify_all
    self.assertReturnsIfImplemented(6, get_value, woken)
  File "/tmp/nix-build-pypy-2.2.1.drv-0/pypy-2.2.1-src/lib-
python/2.7/test/test_multiprocessing.py", line 149, in 
assertReturnsIfImplemented
    return self.assertEqual(value, res)
AssertionError: 6 != 3


Example build log: http://hydra.nixos.org/build/8577476/nixlog/1/raw
List of all builds: http://hydra.nixos.org/job/nixos/trunk-
combined/nixpkgs.pypy.x86_64-linux

----------
messages: 6583
nosy: iElectric, pypy-issue
priority: bug
status: unread
title: Transient failures of test test.test_multiprocessing.WithProcessesTestCondition

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1702>
________________________________________

From tracker at bugs.pypy.org  Sun Mar  9 13:28:48 2014
From: tracker at bugs.pypy.org (Johan Rade)
Date: Sun, 09 Mar 2014 12:28:48 +0000
Subject: [pypy-issue] [issue1703] Crash when overriding PyType_Type.tp_new
In-Reply-To: <1394368128.04.0.456413417264.issue1703@bugs.pypy.org>
Message-ID: <1394368128.04.0.456413417264.issue1703@bugs.pypy.org>


New submission from Johan Rade <johan.rade at gmail.com>:

If you use the C-API to derive a new metaclass from PyType_Type, override the
tp_new slot, and have the overriding tp_new call PyType_Type.tp_new, then PyPy
gets into an infinite loop and crashes.

I have attached files that show the problem.

Instructions:
    Compile Foo3.c as a python extension module named Foo3.
    Set up the paths so that Test3.py can find Foo3.
    Run Test3.py

Expected behavior and observed behavior with CPython 2.7.6:
    Test3.py runs

Observed behavior with PyPy 2.2.1:
    Test3.py crashes.

Fixing this issue is probably needed to get PySide to run on PyPy using the
Python C-API.

----------
files: Foo3.zip
messages: 6584
nosy: jrade, pypy-issue
priority: bug
status: unread
title: Crash when overriding PyType_Type.tp_new

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1703>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 10 07:53:50 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Mon, 10 Mar 2014 06:53:50 +0000
Subject: [pypy-issue] [issue1703] Crash when overriding PyType_Type.tp_new
In-Reply-To: <1394368128.04.0.456413417264.issue1703@bugs.pypy.org>
Message-ID: <1394434430.94.0.0950294355361.issue1703@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Checked in the tests as
cpyext/test/test_typeobject.py:test_tp_new_in_subclass_of_type.  The issue is
that PyType_Type.tp_new just calls slot_tp_new, which calls again
space.getattr(w_type, space.wrap("__new__")), leading to infinite recursion.

Unsure how much there is really to fix.  It seems to me that *any* slot has the
same problem.  But then probably not, otherwise this issue would have shown up
much earlier...

----------
nosy: +arigo
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1703>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 10 09:50:36 2014
From: tracker at bugs.pypy.org (Johan Rade)
Date: Mon, 10 Mar 2014 08:50:36 +0000
Subject: [pypy-issue] [issue1703] Crash when overriding PyType_Type.tp_new
In-Reply-To: <1394368128.04.0.456413417264.issue1703@bugs.pypy.org>
Message-ID: <1394441436.27.0.190752695224.issue1703@bugs.pypy.org>


Johan Rade <johan.rade at gmail.com> added the comment:

I tried to run some tests to see what happens when you override the other slots.
But then I discovered something surprising:

The derived metaclass crashes even if you do not override any slots.

I have uploaded an example Foo4 that demonstrates this behavior. You compile and
run Foo4 the same way as Foo3. With CPython 2.7.6 Foo4.py runs, which is the
expected behavior. With PyPy 2.2.1 and PyPy -r69793 it crashes.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1703>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 10 12:33:12 2014
From: tracker at bugs.pypy.org (Johan Rade)
Date: Mon, 10 Mar 2014 11:33:12 +0000
Subject: [pypy-issue] [issue1703] Crash when overriding PyType_Type.tp_new
In-Reply-To: <1394368128.04.0.456413417264.issue1703@bugs.pypy.org>
Message-ID: <1394451192.07.0.247736822744.issue1703@bugs.pypy.org>


Johan Rade <johan.rade at gmail.com> added the comment:

I tried to run some tests to see what happens when you override the other slots.
But then I discovered something surprising:

The derived metaclass crashes even if you do not override any slots.

I have uploaded an example Foo4 that demonstrates this behavior. You compile and
run Foo4 the same way as Foo3. With CPython 2.7.6 Test4.py runs, which is the
expected behavior. With PyPy 2.2.1 and PyPy -r69793 it crashes.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1703>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 10 13:56:14 2014
From: tracker at bugs.pypy.org (lesshaste)
Date: Mon, 10 Mar 2014 12:56:14 +0000
Subject: [pypy-issue] [issue1704] 'module' object has no attribute 'svd_n'
In-Reply-To: <1394456174.45.0.709129735348.issue1704@bugs.pypy.org>
Message-ID: <1394456174.45.0.709129735348.issue1704@bugs.pypy.org>


New submission from lesshaste <drraph at gmail.com>:

When I run the following simple code np.linalg.matrix_rank(M) on March 6 nightly
it fails with 

AttributeError: 'module' object has no attribute 'svd_n'

----------
files: circ-singular.py
messages: 6588
nosy: lesshaste, pypy-issue
priority: feature
status: unread
title: 'module' object has no attribute 'svd_n'

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1704>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 10 14:02:44 2014
From: tracker at bugs.pypy.org (Christoph Reiter)
Date: Mon, 10 Mar 2014 13:02:44 +0000
Subject: [pypy-issue] [issue1705] numpy: numpy.linalg.inv missing
In-Reply-To: <1394456564.81.0.503489208695.issue1705@bugs.pypy.org>
Message-ID: <1394456564.81.0.503489208695.issue1705@bugs.pypy.org>


New submission from Christoph Reiter <reiter.christoph at gmail.com>:

This is needed for matplotlib

import numpy

A = numpy.array([
    [1, 3, 3],
    [1, 4, 3],
    [1 ,3, 4],
])

A1 = numpy.array([
    [7, -3, -3],
    [-1, 1, 0],
    [-1, 0, 1],
])

assert numpy.array_equal(numpy.linalg.inv(A), A1)

----------
messages: 6589
nosy: lazka, pypy-issue
priority: feature
status: unread
title: numpy: numpy.linalg.inv missing

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1705>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 10 14:04:15 2014
From: tracker at bugs.pypy.org (Christoph Reiter)
Date: Mon, 10 Mar 2014 13:04:15 +0000
Subject: [pypy-issue] [issue1706] numpy: numpy.searchsorted missing
In-Reply-To: <1394456655.49.0.838090641581.issue1706@bugs.pypy.org>
Message-ID: <1394456655.49.0.838090641581.issue1706@bugs.pypy.org>


New submission from Christoph Reiter <reiter.christoph at gmail.com>:

This is needed for matplotlib

##############

import numpy

assert numpy.searchsorted([1,2,3,4,5], 3) == 2
assert numpy.searchsorted([1,2,3,4,5], 3, side='right') == 3

assert numpy.array_equal(
    numpy.searchsorted([1,2,3,4,5], [-10, 10, 2, 3]),
                        numpy.array([0, 5, 1, 2]))

##############

----------
messages: 6590
nosy: lazka, pypy-issue
priority: feature
status: unread
title: numpy: numpy.searchsorted missing

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1706>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 11 22:49:12 2014
From: tracker at bugs.pypy.org (mattip)
Date: Tue, 11 Mar 2014 21:49:12 +0000
Subject: [pypy-issue] [issue1707] speed regression on simple numpy loop
In-Reply-To: <1394574552.15.0.201193966192.issue1707@bugs.pypy.org>
Message-ID: <1394574552.15.0.201193966192.issue1707@bugs.pypy.org>


New submission from mattip <matti.picus at gmail.com>:

As reported on the pypy/numpy repo issue tracker, looping over a numpy ndarray 
has taken a big hit. The original report was for Linux, here I am testing Win32
D:\pypy_stuff>pypy-c-jit-68585-0dcf482a5cb8-win32\pypy.exe c:\temp\test_pypy.py
total 500
total 500
total 500
total 500
total 500
total 500
total 500
total 500
total 500
[2.5512983611464493, 2.32676067179098, 2.320774419773148]

D:\pypy_stuff>pypy-c-jit-69845-3eb83f4075a3-win32\pypy.exe c:\temp\test_pypy.py
total 500
total 500
total 500
total 500
total 500
total 500
total 500
total 500
total 500
[64.27379266845556, 64.29791385595621, 64.82441196908306]

----------
files: test_pypy.py
messages: 6591
nosy: bdk, mattip, pypy-issue
priority: performance bug
status: chatting
title: speed regression on simple numpy loop

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1707>
________________________________________
-------------- next part --------------

try:
    import numpy as np
except ImportError:
    import _numpypy.multiarray as np

tiles = np.zeros((256, 256), 'uint8')  # also tried uint32 - same result
tiles[5, 5] = 1

def benchmark_raw():
    total = 0
    for _ in xrange(500):
        for x in xrange(256):
            for y in xrange(256):
                total += tiles[x, y]
    print 'total',total

if __name__ == "__main__":
    import timeit
    t = timeit.Timer("benchmark_raw()", setup="from __main__ import benchmark_raw")
    times = t.repeat(repeat=3, number = 3)
    print times

From tracker at bugs.pypy.org  Tue Mar 11 23:57:20 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Tue, 11 Mar 2014 22:57:20 +0000
Subject: [pypy-issue] [issue1707] speed regression on simple numpy loop
In-Reply-To: <1394574552.15.0.201193966192.issue1707@bugs.pypy.org>
Message-ID: <1394578640.25.0.541975767443.issue1707@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

even more interesting:

$ ./pypy-c-jit-69353-a55cda9fb045-linux64/bin/pypy pypy/l.py
[2.5020031929016113, 2.4976398944854736, 2.4837629795074463]

$ ./pypy-c-jit-69421-f3e717c94913-linux64/bin/pypy pypy/l.py
[0.2546401023864746, 0.23397111892700195, 0.2335798740386963]

$ ./pypy-c-jit-69468-0e5972cfb8d4-linux64/bin/pypy pypy/l.py
[0.24732208251953125, 0.23346686363220215, 0.23294711112976074]

$ ./pypy-c-jit-69524-8df8a01902f8-linux64/bin/pypy pypy/l.py
[49.25207495689392, 49.31953716278076, 49.26526212692261]

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1707>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 12 09:14:12 2014
From: tracker at bugs.pypy.org (YAMAMOTO Takashi)
Date: Wed, 12 Mar 2014 08:14:12 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1394612052.2.0.380751419934.issue1514@bugs.pypy.org>


YAMAMOTO Takashi <yamt at mwd.biglobe.ne.jp> added the comment:

ping!

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 13 23:42:56 2014
From: tracker at bugs.pypy.org (Johannes Baiter)
Date: Thu, 13 Mar 2014 22:42:56 +0000
Subject: [pypy-issue] [issue1695] complexity of bytearray  operations
In-Reply-To: <1392460960.82.0.973503417088.issue1695@bugs.pypy.org>
Message-ID: <1394750576.83.0.53971181884.issue1695@bugs.pypy.org>


Johannes Baiter <johannes.baiter at gmail.com> added the comment:

I got bitten by this when running my jpegtran bindings under PyPy.
To extract EXIF data from JPEG images, I'm using indexed access on bytearrays a lot.
To illustrate the issue, the following (using IPython):

[1] import jpegtran
[2] %timeit -n10 -r5 jpegtran.JPEGImage('someimage.jpg').exif_thumbnail.as_blob()

gives 49.1ms as a 'best case' on CPython while it's ~2570ms on PyPy.
The source for the module can be found here:
https://github.com/jbaiter/jpegtran-cffi/blob/master/jpegtran/lib.py#L150

I also attached profiling data for PyPy.

----------
nosy: +jbaiter
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1695>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 14 00:12:23 2014
From: tracker at bugs.pypy.org (Johannes Baiter)
Date: Thu, 13 Mar 2014 23:12:23 +0000
Subject: [pypy-issue] [issue1695] complexity of bytearray  operations
In-Reply-To: <1392460960.82.0.973503417088.issue1695@bugs.pypy.org>
Message-ID: <1394752343.18.0.236585527896.issue1695@bugs.pypy.org>


Johannes Baiter <johannes.baiter at gmail.com> added the comment:

To clarify, the below numbers were obtained with a PyPy nightly build from
mid-March of the armhf version, running on a non-overclocked Raspberry Pi.

With PyPy 2.2.1, the 'benchmark' takes around 1090ms in the best case.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1695>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 15 03:51:02 2014
From: tracker at bugs.pypy.org (Sorawee Porncharoenwase)
Date: Sat, 15 Mar 2014 02:51:02 +0000
Subject: [pypy-issue] [issue1708] import * doesn't work
In-Reply-To: <1394851862.95.0.185828377009.issue1708@bugs.pypy.org>
Message-ID: <1394851862.95.0.185828377009.issue1708@bugs.pypy.org>


New submission from Sorawee Porncharoenwase <nullzero.free at gmail.com>:

In python, I do this: 
from flask.ext.wtf import *

It works perfectly.

In pypy, after I run that command, the error was raised: ImportError: No module 
named flask.ext.wtf.*

I can import flask.ext.wtf, though.

----------
messages: 6596
nosy: nullzero, pypy-issue
priority: bug
status: unread
title: import * doesn't work

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1708>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 15 08:04:39 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Sat, 15 Mar 2014 07:04:39 +0000
Subject: [pypy-issue] [issue1708] import * doesn't work
In-Reply-To: <1394851862.95.0.185828377009.issue1708@bugs.pypy.org>
Message-ID: <1394867079.32.0.949912901976.issue1708@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

I get the same error.  However, this fails as well for me:

    import flask.ext.wtf

Are you sure you can import flask.ext.wtf?  If so, which exact versions of pypy
and of flask?

----------
nosy: +arigo
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1708>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 15 08:22:06 2014
From: tracker at bugs.pypy.org (Sorawee Porncharoenwase)
Date: Sat, 15 Mar 2014 07:22:06 +0000
Subject: [pypy-issue] [issue1708] import * doesn't work
In-Reply-To: <1394851862.95.0.185828377009.issue1708@bugs.pypy.org>
Message-ID: <1394868126.49.0.939839376462.issue1708@bugs.pypy.org>


Sorawee Porncharoenwase <nullzero.free at gmail.com> added the comment:

This screenshot explains everything.

Note that this happens with only flask.ext.wtf; other packages such as standard 
libraries or flask itself have no problem for from ... import *

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1708>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 15 08:36:39 2014
From: tracker at bugs.pypy.org (Sorawee Porncharoenwase)
Date: Sat, 15 Mar 2014 07:36:39 +0000
Subject: [pypy-issue] [issue1708] import * doesn't work
In-Reply-To: <1394851862.95.0.185828377009.issue1708@bugs.pypy.org>
Message-ID: <1394868999.75.0.620510394494.issue1708@bugs.pypy.org>


Sorawee Porncharoenwase <nullzero.free at gmail.com> added the comment:

@arigo: you have to install flask and flask-wtf first in order to import that one.

Although it seems that the error is caused by special settings(?) in flask-wtf, my 
point is that pypy should be able to run whatever python can run. Therefore, since 
python can handle this case, pypy should be able to handle this case too.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1708>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 15 13:52:21 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Sat, 15 Mar 2014 12:52:21 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1394887941.62.0.0770996661812.issue1514@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Can this logic be put somewhere in pypy/module/imp/importing.py?  Or at least be
activated only in that case, which is "force_init=True"?  I'm hesitant to change
the default behavior of getbuiltinmodule().

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 16 14:49:04 2014
From: tracker at bugs.pypy.org (mattip)
Date: Sun, 16 Mar 2014 13:49:04 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1394977744.42.0.539426316353.issue1696@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

I get ssltimeout on the stdlib test_ftplib TestTLS_FTPClassMixin tests, as far 
back as 2.2.1 on win32.
Adding self.client.debugging = 1 jsut before self.client.prot_p() yields:
*cmd* 'PBSZ 0'
*resp* '200 PBSZ=0 successful.'
*cmd* 'PROT P'
SSLError: SSLError...med out')

cpython 2.7.5 passes the test with no errors.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 17 21:10:12 2014
From: tracker at bugs.pypy.org (Philip Jenvey)
Date: Mon, 17 Mar 2014 20:10:12 +0000
Subject: [pypy-issue] [issue1541] py3k: Re-enable rpython int sized
	app-level int
In-Reply-To: <1374013849.16.0.768623384444.issue1541@bugs.pypy.org>
Message-ID: <1395087013.0.0.868168208568.issue1541@bugs.pypy.org>


Philip Jenvey <pjenvey at underboss.org> added the comment:

This is pretty much done

----------
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1541>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 17 21:56:01 2014
From: tracker at bugs.pypy.org (mattip)
Date: Mon, 17 Mar 2014 20:56:01 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1395089761.92.0.39413450755.issue1696@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

I have boiled this down to a 10 line test on using login(), passes on cpython, 
fails on pypy win32. It would be nice to be able to run this test untranslated. 
Naively trying pyinteractive with

c:\Python27\python.exe pypy\bin\pyinteractive.py --withmod-_ssl --withmod-
_socket --withmod-struct --withmod-binascii --withmod-select --withmod-time --
withmod-signal --withmod-thread c:\temp\logintest.py

unfortunately crashes before it runs the test

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________
-------------- next part --------------
import sys
sys.modules['_testcapi'] = sys

from test import test_ftplib
import ftplib

server = test_ftplib.DummyTLS_FTPServer(('127.0.0.1',0))
server.start()
client = ftplib.FTP_TLS(timeout=10)
client.connect(server.host, server.port)
client.debugging = 1
client.login()
client.close()
server.stop()

From tracker at bugs.pypy.org  Mon Mar 17 22:16:24 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Mon, 17 Mar 2014 21:16:24 +0000
Subject: [pypy-issue] [issue1709] check _cffi_backend on Windows
In-Reply-To: <1395090984.0.0.602689979075.issue1709@bugs.pypy.org>
Message-ID: <1395090984.0.0.602689979075.issue1709@bugs.pypy.org>


New submission from Armin Rigo <armin.rigo at gmail.com>:

Check _cffi_backend on Windows, and update translator/c/src/libffi_msvc from
hg/cffi.  Notably, I think the ABI auto-detection is not working with the JIT: a
call to either a stdcall ABI function (or maybe it is a cdecl ABI function)
would work fine only in interpreter mode and crash when JITted.  There is an XXX
about that in module/_cffi_backend/ctypefunc.py but the problem is not there.

----------
messages: 6604
nosy: arigo, pypy-issue
priority: bug
status: unread
title: check _cffi_backend on Windows

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1709>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 18 06:08:25 2014
From: tracker at bugs.pypy.org (YAMAMOTO Takashi)
Date: Tue, 18 Mar 2014 05:08:25 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395119305.39.0.136566797974.issue1514@bugs.pypy.org>


YAMAMOTO Takashi <yamt at mwd.biglobe.ne.jp> added the comment:

assuming that force_init=True is used only by load_module,
i think activating it only for force_init=True makes sense.
or probably add another argument to specify the behavior, rather than
abusing force_init.
do you want me to prepare a patch?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 18 07:02:38 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Tue, 18 Mar 2014 06:02:38 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395122558.37.0.317781676773.issue1514@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

conditional on force_init sounds reasonable. please make sure the patch enables 
and passes the two test_builtin_reimport tests in 
pypy/module/imp/test/test_app.py.

----------
nosy: +bdk

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 18 08:41:52 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Tue, 18 Mar 2014 07:41:52 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395128512.41.0.592031528391.issue1514@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

fixed in 0ecfb7242213

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 18 08:49:16 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Tue, 18 Mar 2014 07:49:16 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395128956.33.0.390489265792.issue1514@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

(turns out that broke things, but it is a place to start for a patch)

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 18 10:21:31 2014
From: tracker at bugs.pypy.org (Carl Friedrich Bolz)
Date: Tue, 18 Mar 2014 09:21:31 +0000
Subject: [pypy-issue] [issue1710] GC misbehaviour on large heaps
In-Reply-To: <1395134491.88.0.55171554682.issue1710@bugs.pypy.org>
Message-ID: <1395134491.88.0.55171554682.issue1710@bugs.pypy.org>


New submission from Carl Friedrich Bolz <cfbolz at gmx.de>:

All info provided by Martin Koch <mak at issuu.com> on pypy-dev

-------
https://mail.python.org/pipermail/pypy-dev/2014-March/012305.html

I'm running a server (written in python, executed with pypy) that holds a large
graph (55GB, millions of nodes and edges) in memory and responds to queries by
traversing the graph.The graph is mutated a few times a second, and there are
hundreds of read-only requests a second.

My problem is that I no control over garbage collection. Thus, a major GC might
kick in while serving a query, and with this amount of data, the GC takes around
2 minutes. I have tried mitigating this by guessing when a GC might be due, and
proactively starting the garbage collector while not serving a request (this is
ok, as duplicate servers will respond to requests while this one is collecting). 

What I would really like is to be able to disable garbage collection for the old
generation. This is because the graph is fairly static, and I can live with
leaking memory from the relatively few and small mutations that occur. Any
queries are only likely to generate objects in the new generation, and it is
fine to collect these. Also, by design, the process is periodically restarted in
order to re-synchronize it with an authoritative source (thus rebuilding the
graph from scratch), so slight leakage is not an issue here. 

I have tried experimenting with setting environment variables as well as the
'gc' module, but nothing seems to give me what I want.

If disabling gc for certain generations is not possible, it would be nice to be
able to get a hint when a major collection is about to occur, so I can stop
serving requests.



-------
https://mail.python.org/pipermail/pypy-dev/2014-March/012310.html:

We have hacked up a small sample that seems to exhibit the same issue. 

We basically generate a linked list of objects. To increase connectedness,
elements in the list hold references (dummy_links) to 10 randomly chosen
previous elements in the list.

We then time a function that traverses 50000 elements from the list from a
random start point. If the traversal reaches the end of the list, we instead
traverse one of the dummy links. Thus, exactly 50K elements are traversed every
time. To generate some garbage, we build a list holding the traversed elements
and a dummy list of characters. 

Timings for the last 100 runs are stored in a circular buffer. If the elapsed
time for the last run is more than twice the average time, we print out a line
with the elapsed time, the threshold, and the 90% runtime (we would like to see
that the mean runtime does not increase with the number of elements in the list,
but that the max time does increase (linearly with the number of object, i
guess); traversing 50K elements should be independent of the memory size).

We have tried monitoring memory consumption by external inspection, but cannot
consistently verify that memory is deallocated at the same time that we see slow
requests. Perhaps the pypy runtime doesn't always return freed pages back to the OS?

Using top, we observe that 10M elements allocates around 17GB after building,
20M elements 26GB, 30M elements 28GB (and grows to 35GB shortly after building).

Here is output from a few runs with different number of elements:


pypy mem.py 10000000
start build
end build 84.142424
that took a long time elapsed: 13.230586  slow_threshold: 1.495401 
90th_quantile_runtime: 0.421558
that took a long time elapsed: 13.016531  slow_threshold: 1.488160 
90th_quantile_runtime: 0.423441
that took a long time elapsed: 13.032537  slow_threshold: 1.474563 
90th_quantile_runtime: 0.419817

pypy mem.py 20000000
start build
end build 180.823105
that took a long time elapsed: 27.346064  slow_threshold: 2.295146 
90th_quantile_runtime: 0.434726
that took a long time elapsed: 26.028852  slow_threshold: 2.283927 
90th_quantile_runtime: 0.374190
that took a long time elapsed: 25.432279  slow_threshold: 2.279631 
90th_quantile_runtime: 0.371502

pypy mem.py 30000000
start build
end build 276.217811
that took a long time elapsed: 40.993855  slow_threshold: 3.188464 
90th_quantile_runtime: 0.459891
that took a long time elapsed: 41.693553  slow_threshold: 3.183003 
90th_quantile_runtime: 0.393654
that took a long time elapsed: 39.679769  slow_threshold: 3.190782 
90th_quantile_runtime: 0.393677
that took a long time elapsed: 43.573411  slow_threshold: 3.239637 
90th_quantile_runtime: 0.393654

Code below
--------------------------------------------------------------
import time
from random import randint, choice
import sys


allElems = {}

class Node:
    def __init__(self, v_):
        self.v = v_
        self.next = None
        self.dummy_data = [randint(0,100)
                           for _ in xrange(randint(50,100))]
        allElems[self.v] = self
        if self.v > 0:
            self.dummy_links = [allElems[randint(0, self.v-1)] for _ in xrange(10)]
        else:
            self.dummy_links = [self]

    def set_next(self, l):
        self.next = l


def follow(node):
    acc = []
    count = 0
    cur = node
    assert node.v is not None
    assert cur is not None
    while count < 50000:
        # return a value; generate some garbage
        acc.append((cur.v, [choice("abcdefghijklmnopqrstuvwxyz") for x in
xrange(100)]))

        # if we have reached the end, chose a random link
        cur = choice(cur.dummy_links) if cur.next is None else cur.next
        count += 1

    return acc


def build(num_elems):
    start = time.time()
    print "start build"
    root = Node(0)
    cur = root
    for x in xrange(1, num_elems):
        e = Node(x)
        cur.next = e
        cur = e
    print "end build %f" % (time.time() - start)
    return root


num_timings = 100
if __name__ == "__main__":
    num_elems = int(sys.argv[1])
    build(num_elems)
    total = 0
    timings = [0.0] * num_timings # run times for the last num_timings runs
    i = 0
    beginning = time.time()
    while time.time() - beginning < 600:
        start = time.time()
        elem = allElems[randint(0, num_elems - 1)]
        assert(elem is not None)

        lst = follow(elem)

        total += choice(lst)[0] # use the return value for something

        end = time.time()

        elapsed = end-start
        timings[i % num_timings] = elapsed
        if (i > num_timings):
            slow_time = 2 * sum(timings)/num_timings # slow defined as > 2*avg
run time
            if (elapsed > slow_time):
                print "that took a long time elapsed: %f  slow_threshold: %f 
90th_quantile_runtime: %f" % \
                    (elapsed, slow_time, sorted(timings)[int(num_timings*.9)])
        i += 1
    print total

----------
messages: 6609
nosy: cfbolz, pypy-issue
priority: performance bug
status: unread
title: GC misbehaviour on large heaps

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1710>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 18 11:28:55 2014
From: tracker at bugs.pypy.org (Dmitrey)
Date: Tue, 18 Mar 2014 10:28:55 +0000
Subject: [pypy-issue] [issue1711] numpy.copy: ValueError: Unknown order: K
In-Reply-To: <1395138535.24.0.0759444635711.issue1711@bugs.pypy.org>
Message-ID: <1395138535.24.0.0759444635711.issue1711@bugs.pypy.org>


New submission from Dmitrey <dmitrey15 at ukr.net>:

32bit ubuntu:

>>>> copy(1.0)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File
"/home/dmitrey/Install/pypy-c-jit-70035-049d8d5677f9-linux/site-packages/numpy/lib/function_base.py",
line 830, in copy
    return array(a, order=order, copy=True)
ValueError: Unknown order: K
>>>> array(1.0)
array(1.0)
>>>> copy(array(1.0))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File
"/home/dmitrey/Install/pypy-c-jit-70035-049d8d5677f9-linux/site-packages/numpy/lib/function_base.py",
line 830, in copy
    return array(a, order=order, copy=True)
ValueError: Unknown order: K

----------
messages: 6610
nosy: Dmitrey, pypy-issue
priority: bug
status: unread
title: numpy.copy: ValueError: Unknown order: K

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1711>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 19 11:07:11 2014
From: tracker at bugs.pypy.org (YAMAMOTO Takashi)
Date: Wed, 19 Mar 2014 10:07:11 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395223631.5.0.232863335602.issue1514@bugs.pypy.org>


YAMAMOTO Takashi <yamt at mwd.biglobe.ne.jp> added the comment:

here's a new version of patch. (pypy-import.diff2)
"./pypy-c ./pytest.py -A pypy/module/imp" passed.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 19 12:07:10 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Wed, 19 Mar 2014 11:07:10 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395227230.1.0.536711304022.issue1514@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

simplified/applied in 6002c93c0cc0, thanks.

----------
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 19 12:07:26 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Wed, 19 Mar 2014 11:07:26 +0000
Subject: [pypy-issue] [issue1712] Meta: crash the issue tracker in one view
In-Reply-To: <1395227246.54.0.0217504830943.issue1712@bugs.pypy.org>
Message-ID: <1395227246.54.0.0217504830943.issue1712@bugs.pypy.org>


New submission from Armin Rigo <armin.rigo at gmail.com>:

The pypy issue tracker crashes when we ask for "sort on: activity; descending; 
group on: activity; not descending".

----------
messages: 6613
nosy: arigo, pypy-issue
priority: bug
status: unread
title: Meta: crash the issue tracker in one view

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1712>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 20 02:42:51 2014
From: tracker at bugs.pypy.org (wenzhu man)
Date: Thu, 20 Mar 2014 01:42:51 +0000
Subject: [pypy-issue] [issue1713] dict of ast node misbehaving
In-Reply-To: <1395279771.98.0.61496930017.issue1713@bugs.pypy.org>
Message-ID: <1395279771.98.0.61496930017.issue1713@bugs.pypy.org>


New submission from wenzhu man <manwenzhu at gmail.com>:

I added tests to test dict of ast node misbehaving:
 def test_dict_astNode:
I also add three tests for issue 1673:
deep copy works fine for num that's fully initialized but not working if ast 
node is not fully initialized.
I tested the Num and Str class

----------
files: testAst.patch
messages: 6614
nosy: pypy-issue, wman
priority: bug
status: unread
title: dict of ast node misbehaving

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1713>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 20 03:59:13 2014
From: tracker at bugs.pypy.org (wenzhu man)
Date: Thu, 20 Mar 2014 02:59:13 +0000
Subject: [pypy-issue] [issue1713] dict of ast node misbehaving
In-Reply-To: <1395279771.98.0.61496930017.issue1713@bugs.pypy.org>
Message-ID: <1395284353.72.0.687667624406.issue1713@bugs.pypy.org>


wenzhu man <manwenzhu at gmail.com> added the comment:

https://bitbucket.org/pypy/pypy/pull-request/207/add-tests-to-prove-issue-1713-
and-issue/diff
 this is better for review tests

----------
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1713>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 20 08:00:53 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Thu, 20 Mar 2014 07:00:53 +0000
Subject: [pypy-issue] [issue1713] dict of ast node misbehaving
In-Reply-To: <1395279771.98.0.61496930017.issue1713@bugs.pypy.org>
Message-ID: <1395298853.58.0.0500795172478.issue1713@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Thanks!  The tests are fine.  As long as they fail, though, we should not accept
them into the "default" mercurial branch.  The two options are: either you make
a new branch; or we accept the tests with an additional line:
py.test.skip("in-progress").

----------
nosy: +arigo

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1713>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 20 20:27:12 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Thu, 20 Mar 2014 19:27:12 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395343632.38.0.472985097464.issue1514@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

had to back out this patch -- failures in pypy/module/zipimport tests

also would like the fix to include changes to imp.init_builtin to match cpython

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 21 00:57:37 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Thu, 20 Mar 2014 23:57:37 +0000
Subject: [pypy-issue] [issue1714] forcing a raw malloc virtual doesn't
	reinsert	guard_no_exception
In-Reply-To: <1395359857.74.0.552988295192.issue1714@bugs.pypy.org>
Message-ID: <1395359857.74.0.552988295192.issue1714@bugs.pypy.org>


New submission from Brian Kearns <bdkearns at gmail.com>:

see XXX in test_optimizeopt.py test_virtual_raw_malloc_force and related tests 
with XXX

----------
messages: 6618
nosy: bdk, pypy-issue
priority: bug
status: unread
title: forcing a raw malloc virtual doesn't reinsert guard_no_exception

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1714>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 21 20:40:43 2014
From: tracker at bugs.pypy.org (Dmitrey)
Date: Fri, 21 Mar 2014 19:40:43 +0000
Subject: [pypy-issue] [issue1715] numpy: ones_like yields incorrect type
In-Reply-To: <1395430843.6.0.596739030349.issue1715@bugs.pypy.org>
Message-ID: <1395430843.6.0.596739030349.issue1715@bugs.pypy.org>


New submission from Dmitrey <dmitrey15 at ukr.net>:

>>>> ones_like(array(True))
array(1.0)

>>> ones_like(array(True))
array(True, dtype=bool)

----------
messages: 6619
nosy: Dmitrey, pypy-issue
priority: bug
status: unread
title: numpy: ones_like yields incorrect type

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1715>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 21 20:58:37 2014
From: tracker at bugs.pypy.org (Marc Abramowitz)
Date: Fri, 21 Mar 2014 19:58:37 +0000
Subject: [pypy-issue] [issue991] [Feature Request]:
	Implement	PyErr_SetFromErrnoWithFilenameObject
In-Reply-To: <1326099905.08.0.543145843036.issue991@bugs.pypy.org>
Message-ID: <1395431917.95.0.910043271898.issue991@bugs.pypy.org>


Marc Abramowitz <msabramo at gmail.com> added the comment:

It looks like this function is documented now:

http://docs.python.org/2/c-api/exceptions.html?
highlight=pyerr_setfromerrnowithfilenameobject#PyErr_SetFromErrnoWithFilenameObj
ect

http://hg.python.org/cpython/rev/4cc94d30926f

----------
nosy: +msabramo -bra
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue991>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 21 21:00:22 2014
From: tracker at bugs.pypy.org (Marc Abramowitz)
Date: Fri, 21 Mar 2014 20:00:22 +0000
Subject: [pypy-issue] [issue991] [Feature Request]:
	Implement	PyErr_SetFromErrnoWithFilenameObject
In-Reply-To: <1326099905.08.0.543145843036.issue991@bugs.pypy.org>
Message-ID: <1395432022.31.0.79957210798.issue991@bugs.pypy.org>


Marc Abramowitz <msabramo at gmail.com> added the comment:

http://bit.ly/1gjl1jB

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue991>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 21 22:44:56 2014
From: tracker at bugs.pypy.org (Philip Jenvey)
Date: Fri, 21 Mar 2014 21:44:56 +0000
Subject: [pypy-issue] [issue991] [Feature Request]:
	Implement	PyErr_SetFromErrnoWithFilenameObject
In-Reply-To: <1326099905.08.0.543145843036.issue991@bugs.pypy.org>
Message-ID: <1395438296.66.0.50959878304.issue991@bugs.pypy.org>


Philip Jenvey <pjenvey at underboss.org> added the comment:

pypy already provides PyErr_SetFromErrnoWithFilename so writing a 
PyErr_SetFromErrnoWithFilenameObject should be a pretty straightforward task for 
anyone who's interested.

Here its impl:

https://bitbucket.org/pypy/pypy/src/f6f44dc30721299a1005b642653079d736862d65/pyp
y/module/cpyext/pyerrors.py?at=default#cl-145

tests:

https://bitbucket.org/pypy/pypy/src/f6f44dc30721299a1005b642653079d736862d65/pyp
y/module/cpyext/test/test_pyerrors.py?at=default#cl-201

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue991>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 24 15:59:26 2014
From: tracker at bugs.pypy.org (Marc Abramowitz)
Date: Mon, 24 Mar 2014 14:59:26 +0000
Subject: [pypy-issue] [issue991] [Feature Request]:
	Implement	PyErr_SetFromErrnoWithFilenameObject
In-Reply-To: <1326099905.08.0.543145843036.issue991@bugs.pypy.org>
Message-ID: <1395673166.76.0.637753035905.issue991@bugs.pypy.org>


Marc Abramowitz <msabramo at gmail.com> added the comment:

Here's my stab at adding `PyErr_SetFromErrnoWithFilenameObject` to cpyext of 
pypy:

https://bitbucket.org/pypy/pypy/pull-request/211/add-
pyerr_setfromerrnowithfilenameobject/diff

I hope that this helps.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue991>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 24 16:07:29 2014
From: tracker at bugs.pypy.org (Marc Abramowitz)
Date: Mon, 24 Mar 2014 15:07:29 +0000
Subject: [pypy-issue] [issue991] [Feature Request]:
	Implement	PyErr_SetFromErrnoWithFilenameObject
In-Reply-To: <1326099905.08.0.543145843036.issue991@bugs.pypy.org>
Message-ID: <1395673649.5.0.817507723892.issue991@bugs.pypy.org>


Marc Abramowitz <msabramo at gmail.com> added the comment:

Here's my stab at adding `PyErr_SetFromErrnoWithFilenameObject` to cpyext of 
pypy:

- http://bit.ly/1hUD5gD

This was merged last night by Alex_Gaynor and then tweaked a little; see:

- https://bitbucket.org/pypy/pypy/commits/0bc8a37
- https://bitbucket.org/pypy/pypy/commits/a00eb16

I hope that this helps.

----------
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue991>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 24 16:11:26 2014
From: tracker at bugs.pypy.org (Fijal)
Date: Mon, 24 Mar 2014 15:11:26 +0000
Subject: [pypy-issue] [issue991] [Feature Request]:
	Implement	PyErr_SetFromErrnoWithFilenameObject
In-Reply-To: <1326099905.08.0.543145843036.issue991@bugs.pypy.org>
Message-ID: <1395673886.09.0.194144801096.issue991@bugs.pypy.org>


Fijal <fijall at gmail.com> added the comment:

please don't reopen issues to add comments.

----------
nosy: +fijal
status: chatting -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue991>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 25 03:59:28 2014
From: tracker at bugs.pypy.org (YAMAMOTO Takashi)
Date: Tue, 25 Mar 2014 02:59:28 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395716368.76.0.46671741939.issue1514@bugs.pypy.org>


YAMAMOTO Takashi <yamt at mwd.biglobe.ne.jp> added the comment:

"./pypy-c ./pytest.py -A -v pypy/module/zipimport" passed for me. ("43 passed")
can you explain?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 25 08:30:48 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Tue, 25 Mar 2014 07:30:48 +0000
Subject: [pypy-issue] [issue991] [Feature Request]:
	Implement	PyErr_SetFromErrnoWithFilenameObject
In-Reply-To: <1326099905.08.0.543145843036.issue991@bugs.pypy.org>
Message-ID: <1395732648.7.0.274599293451.issue991@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

just want to chime in: you are welcome to add comments to issues.  I think that
what fijal is complaining about is really the issue system, in which adding a
comment *always* re-opens the issue.  The commenter needs to be aware of this
and then re-close the issue afterwards.

----------
nosy: +arigo
status: resolved -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue991>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 25 10:48:59 2014
From: tracker at bugs.pypy.org (Dmitrey)
Date: Tue, 25 Mar 2014 09:48:59 +0000
Subject: [pypy-issue] [issue1716] jitviewer: tcl-dev,
	tk-dev should be added to jitviewer dependencies
In-Reply-To: <1395740939.64.0.378646962826.issue1716@bugs.pypy.org>
Message-ID: <1395740939.64.0.378646962826.issue1716@bugs.pypy.org>


New submission from Dmitrey <dmitrey15 at ukr.net>:

jitviewer: tcl-dev, tk-dev should be added to jitviewer dependencies (it prints
"tcl.h not found", "tk.h not found")

----------
messages: 6628
nosy: Dmitrey, pypy-issue
priority: bug
status: unread
title: jitviewer: tcl-dev, tk-dev should be added to jitviewer dependencies

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1716>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 25 15:29:32 2014
From: tracker at bugs.pypy.org (Dmitrey)
Date: Tue, 25 Mar 2014 14:29:32 +0000
Subject: [pypy-issue] [issue1717] numpy: bug with boolean indexation
In-Reply-To: <1395757772.75.0.157729165741.issue1717@bugs.pypy.org>
Message-ID: <1395757772.75.0.157729165741.issue1717@bugs.pypy.org>


New submission from Dmitrey <dmitrey15 at ukr.net>:

>>> from numpy import *
>>> a = array((1,5)).reshape(2,1)
>>> a[0,atleast_1d(True)]=100
>>> a
array([[100],
       [  5]])

>>>> from numpy import *
>>>> a = array((1,5)).reshape(2,1)
>>>> a[0,atleast_1d(True)]=100
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
IndexError: index 1 is out of bounds for axis 1 with size 1

----------
messages: 6629
nosy: Dmitrey, pypy-issue
priority: bug
status: unread
title: numpy: bug with boolean indexation

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1717>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 25 15:31:28 2014
From: tracker at bugs.pypy.org (Dmitrey)
Date: Tue, 25 Mar 2014 14:31:28 +0000
Subject: [pypy-issue] [issue1718] numpy: bug with linalg.norm
In-Reply-To: <1395757888.89.0.725383810343.issue1718@bugs.pypy.org>
Message-ID: <1395757888.89.0.725383810343.issue1718@bugs.pypy.org>


New submission from Dmitrey <dmitrey15 at ukr.net>:

>>> from numpy import inf
>>> from numpy.linalg import norm
>>> norm((1,2,3),inf)
3

>>>> from numpy import inf
>>>> from numpy.linalg import norm
>>>> norm((1,2,3),inf)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File
"/home/dmitrey/Install/pypy-c-jit-70035-049d8d5677f9-linux/site-packages/numpy/linalg/linalg.py",
line 2067, in norm
    return abs(x).max(axis=axis)
TypeError: expected integer, got tuple object

----------
messages: 6630
nosy: Dmitrey, pypy-issue
priority: bug
status: unread
title: numpy: bug with linalg.norm

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1718>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 25 15:48:49 2014
From: tracker at bugs.pypy.org (Dmitrey)
Date: Tue, 25 Mar 2014 14:48:49 +0000
Subject: [pypy-issue] [issue1719] numpy: bug with array indexation
In-Reply-To: <1395758929.38.0.0933997146753.issue1719@bugs.pypy.org>
Message-ID: <1395758929.38.0.0933997146753.issue1719@bugs.pypy.org>


New submission from Dmitrey <dmitrey15 at ukr.net>:

>>>> from numpy import *
>>>> r=ones(3)
>>>> ind = array([],int32)
>>>> tmp = array([],float64)
>>>> r[ind]=tmp
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
IndexError: index 0 is out of bounds for axis 0 with size 0

(in cpython it works)

----------
messages: 6631
nosy: Dmitrey, pypy-issue
priority: bug
status: unread
title: numpy: bug with array indexation

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1719>
________________________________________

From tracker at bugs.pypy.org  Tue Mar 25 18:59:13 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Tue, 25 Mar 2014 17:59:13 +0000
Subject: [pypy-issue] [issue1719] numpy: bug with array indexation
In-Reply-To: <1395758929.38.0.0933997146753.issue1719@bugs.pypy.org>
Message-ID: <1395770353.04.0.656427643383.issue1719@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

fixed in 26e0a68bb6dd

----------
nosy: +bdk
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1719>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 03:54:58 2014
From: tracker at bugs.pypy.org (coli)
Date: Wed, 26 Mar 2014 02:54:58 +0000
Subject: [pypy-issue] [issue1720] Can not compile win32api with pypy
In-Reply-To: <1395802498.9.0.0604541194408.issue1720@bugs.pypy.org>
Message-ID: <1395802498.9.0.0604541194408.issue1720@bugs.pypy.org>


New submission from coli <coli at rocketmail.com>:

PythonService.cpp
win32\src\PythonService.cpp(627) : error C3861: 'Py_SetProgramName': identifier
not found
win32\src\PythonService.cpp(632) : error C3861: 'Py_Initialize': identifier not
found
win32\src\PythonService.cpp(644) : error C3861: 'PySys_SetArgv': identifier not
found
error: command 'C:\Program Files (x86)\Microsoft Visual Studio 9.0\VC\BIN\cl.exe
' failed with exit status 2

It does not appear pypy has these 3 functions?

----------
messages: 6633
nosy: coli, pypy-issue
priority: feature
status: unread
title: Can not compile win32api with pypy

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1720>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 07:09:26 2014
From: tracker at bugs.pypy.org (INADA Naoki)
Date: Wed, 26 Mar 2014 06:09:26 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>


New submission from INADA Naoki <songofacandy at gmail.com>:

When source code contains uncalled function produce large constant, It seems PyPy does constant folding for it.
Please run attached sample script.

----------
messages: 6634
nosy: methane, pypy-issue
priority: bug
status: unread
title: Too aggressive constant folding?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 07:15:09 2014
From: tracker at bugs.pypy.org (INADA Naoki)
Date: Wed, 26 Mar 2014 06:15:09 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395814509.94.0.195016369623.issue1721@bugs.pypy.org>


INADA Naoki <songofacandy at gmail.com> added the comment:

I use PyPy-2.2.1 on Mac OS X.

----------
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 07:27:14 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Wed, 26 Mar 2014 06:27:14 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395815234.1.0.616686500658.issue1721@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

You forgot to attach the example.  (Note also that CPython likely does the same
thing; please tell us if it does not.)

----------
nosy: +arigo

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 07:32:32 2014
From: tracker at bugs.pypy.org (INADA Naoki)
Date: Wed, 26 Mar 2014 06:32:32 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395815552.33.0.239427764404.issue1721@bugs.pypy.org>


INADA Naoki <songofacandy at gmail.com> added the comment:

I'm sorry. I attached sample file.

And here is results:

$ /usr/bin/time -l python pypy_constant_folding.py
Hello
        0.02 real         0.01 user         0.00 sys
   4337664  maximum resident set size
...

$ /usr/bin/time -l pypy pypy_constant_folding.py
Hello
        1.86 real         0.62 user         0.89 sys
2163933184  maximum resident set size

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 17:27:05 2014
From: tracker at bugs.pypy.org (Johan Rade)
Date: Wed, 26 Mar 2014 16:27:05 +0000
Subject: [pypy-issue] [issue1720] Can not compile win32api with pypy
In-Reply-To: <1395802498.9.0.0604541194408.issue1720@bugs.pypy.org>
Message-ID: <1395851225.1.0.758230005094.issue1720@bugs.pypy.org>


Johan Rade <johan.rade at gmail.com> added the comment:

Are you trying to embed PyPy in a C or C++ program?
If so, take a look at http://pypy.readthedocs.org/en/latest/embedding.html

----------
nosy: +jrade
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1720>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 18:57:27 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Wed, 26 Mar 2014 17:57:27 +0000
Subject: [pypy-issue] [issue1715] numpy: ones_like yields incorrect type
In-Reply-To: <1395430843.6.0.596739030349.issue1715@bugs.pypy.org>
Message-ID: <1395856647.34.0.212677977536.issue1715@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

fixed in bf6bb7bf1035

----------
nosy: +bdk
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1715>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 19:08:13 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Wed, 26 Mar 2014 18:08:13 +0000
Subject: [pypy-issue] [issue1711] numpy.copy: ValueError: Unknown order: K
In-Reply-To: <1395138535.24.0.0759444635711.issue1711@bugs.pypy.org>
Message-ID: <1395857293.02.0.537740744668.issue1711@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

fixed in db6f95b7c608

----------
nosy: +bdk
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1711>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 19:30:03 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Wed, 26 Mar 2014 18:30:03 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395858603.04.0.0364277645454.issue1514@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

./pytest.py pypy/module/zipimport

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Wed Mar 26 23:06:39 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Wed, 26 Mar 2014 22:06:39 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395871599.61.0.751266314846.issue1721@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Are you sure you're comparing CPython on 64-bit with PyPy on 64-bit?  From your
results, my guess is that your CPython is on 32-bit.  On Linux, I get no
explosion on 32-bit (both with CPython and PyPy), and explosion on 64-bit (both
with CPython and PyPy).  So it's a bug that should also be reported on CPython.
 I agree that PyPy could fix it though.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 00:36:33 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Wed, 26 Mar 2014 23:36:33 +0000
Subject: [pypy-issue] [issue1590] numpy: NaN valued numpypy.float64 can be
	converted to int
In-Reply-To: <1377504676.46.0.56280241899.issue1590@bugs.pypy.org>
Message-ID: <1395876993.52.0.325312487407.issue1590@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

fixed in 297f0a5cd941

----------
nosy: +bdk
status: unread -> resolved

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1590>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 03:14:12 2014
From: tracker at bugs.pypy.org (INADA Naoki)
Date: Thu, 27 Mar 2014 02:14:12 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395886452.11.0.911264378933.issue1721@bugs.pypy.org>


INADA Naoki <songofacandy at gmail.com> added the comment:

Yes.
The problem is large foldable constant (e.g. b'x' * (2**32)) in function not called is calculated.
This is nasty when I skip tests requires large memory.

@pytest.mark.skipif(True, reason="This test requires very larege memory")
def test_largebynary():
    x = b'x' * (2**32)
    do_something(x)

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 03:20:08 2014
From: tracker at bugs.pypy.org (INADA Naoki)
Date: Thu, 27 Mar 2014 02:20:08 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395886808.86.0.0359820533108.issue1721@bugs.pypy.org>


INADA Naoki <songofacandy at gmail.com> added the comment:

$ /usr/local/bin/python -c "import platform; print platform.architecture()"
('64bit', '')
$ /usr/bin/time -l /usr/local/bin/python pypy_constant_folding.py
Hello
        0.02 real         0.01 user         0.00 sys
   4337664  maximum resident set size

/usr/local/bin/python is built with Homebrew.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 03:22:43 2014
From: tracker at bugs.pypy.org (INADA Naoki)
Date: Thu, 27 Mar 2014 02:22:43 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395886963.69.0.116284986773.issue1721@bugs.pypy.org>


INADA Naoki <songofacandy at gmail.com> added the comment:

I'm sorry, I can reproduce this behavior on Python 3.4.
I'll report this to CPython.

$ /usr/bin/time  -l /usr/local/bin/python3 pypy_constant_folding.py
Hello
        2.77 real         1.36 user         1.41 sys
4300234752  maximum resident set size

$ /usr/bin/time -l pypy pypy_constant_folding.py
Hello
        3.14 real         1.17 user         1.64 sys
4311416832  maximum resident set size

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 07:52:20 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Thu, 27 Mar 2014 06:52:20 +0000
Subject: [pypy-issue] [issue1720] Can not compile win32api with pypy
In-Reply-To: <1395802498.9.0.0604541194408.issue1720@bugs.pypy.org>
Message-ID: <1395903140.32.0.484458695244.issue1720@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

No, it seems to be simply compiling with cpyext the "win32api" module.  As we
are low on people with Windows machines, we can't test all the Windows-only
extension modules, even though I agree that "win32api" seems central.  I don't
know it well enough to know what the best path forward would be: either hack it
to not call these three functions; or add support for the three functions in
pypy/module/cpyext/, either as stubs or as a real implementation; or reimplement
win32api or a part of it by rewriting it as a cffi module (which looks like a
big job, even if it also looks like all APIs should be very easy to map).

----------
nosy: +arigo

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1720>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 07:56:42 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Thu, 27 Mar 2014 06:56:42 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395903402.07.0.181022324888.issue1721@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

It could be the case that `platform.architecture()` reports a bogus result.  Try
to print `sys.maxint`.  If you get 2147483647 --- i.e. a number smaller than
2**31 --- then you have a 32-bit CPython and it will not try to constant-fold
`b"x"*(2**31)`.  Another way to see the problem is to try instead to
constant-fold `b"x"*(2**30)`, which should explode the memory everywhere (but
only to 1GB).

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 08:12:20 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Thu, 27 Mar 2014 07:12:20 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395904340.56.0.604664364155.issue1514@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

bdk: I cannot reproduce this.  test_zipimport_deflated seems to pass for me
since 4e0378f68161, which is just after 6002c93c0cc0.  The backout in
c402f888f629 doesn't seem to have fixed (or broken) anything because no test
from there seemed to fail at that time.  Can you double-check?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 08:15:59 2014
From: tracker at bugs.pypy.org (Brian Kearns)
Date: Thu, 27 Mar 2014 07:15:59 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395904559.86.0.962756228406.issue1514@bugs.pypy.org>


Brian Kearns <bdkearns at gmail.com> added the comment:

with yamt patch -- zipimport tests fail. at some point i tried modified/simplified 
patches but never got something that both translated and passed all tests, so 
everything was backed out to previous state. at some point may have had something 
that translated and passed zipimport, but failed elsewhere.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 08:45:08 2014
From: tracker at bugs.pypy.org (YAMAMOTO Takashi)
Date: Thu, 27 Mar 2014 07:45:08 +0000
Subject: [pypy-issue] [issue1514] module behaviour incompatibility which
	makes eventlet fail
In-Reply-To: <1370934852.01.0.927470601666.issue1514@bugs.pypy.org>
Message-ID: <1395906308.79.0.42395475697.issue1514@bugs.pypy.org>


YAMAMOTO Takashi <yamt at mwd.biglobe.ne.jp> added the comment:

the problem seems that "unlazy" part of my patch ends up to loading "struct" module's
app level code, which imports "sturct" module again and causes recursions.
we should add sys.modules before initializing the module.
new version of a patch, pypy-import.diff3, passes at least zipimport tests.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1514>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 09:33:44 2014
From: tracker at bugs.pypy.org (mattip)
Date: Thu, 27 Mar 2014 08:33:44 +0000
Subject: [pypy-issue] [issue1722] tcl tests failing for tcl version 8.5.2
In-Reply-To: <1395909224.42.0.029978604364.issue1722@bugs.pypy.org>
Message-ID: <1395909224.42.0.029978604364.issue1722@bugs.pypy.org>


New submission from mattip <matti.picus at gmail.com>:

win32 is failing test_tcl in stdlib. We should add the fix to cpython issue 19320
http://bugs.python.org/issue19320

----------
assignedto: mattip
messages: 6652
nosy: mattip, pypy-issue
priority: bug
status: unread
title: tcl tests failing for tcl version 8.5.2

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1722>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 09:38:25 2014
From: tracker at bugs.pypy.org (mattip)
Date: Thu, 27 Mar 2014 08:38:25 +0000
Subject: [pypy-issue] [issue1723] test runner filling temp directory with
	_testcapi modules
In-Reply-To: <1395909505.08.0.09145133023.issue1723@bugs.pypy.org>
Message-ID: <1395909505.08.0.09145133023.issue1723@bugs.pypy.org>


New submission from mattip <matti.picus at gmail.com>:

Every test run is recompiling _testcapimodule.c, filling up temp directory with 
many subdirs for each nightly buildbot run.
We should fix lib_pypy/_testcapi.py to use a hash like cffi does to build once 
for each version of pypy and c-files
We should also add a cleanup step to buildbot?

----------
messages: 6653
nosy: mattip, pypy-issue
priority: feature
status: unread
title: test runner filling temp directory with _testcapi modules

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1723>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 10:13:39 2014
From: tracker at bugs.pypy.org (mattip)
Date: Thu, 27 Mar 2014 09:13:39 +0000
Subject: [pypy-issue] [issue1709] check _cffi_backend on Windows
In-Reply-To: <1395090984.0.0.602689979075.issue1709@bugs.pypy.org>
Message-ID: <1395911619.2.0.53533381273.issue1709@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

is there a test that exhibits this behaviour?

----------
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1709>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 11:05:59 2014
From: tracker at bugs.pypy.org (INADA Naoki)
Date: Thu, 27 Mar 2014 10:05:59 +0000
Subject: [pypy-issue] [issue1721] Too aggressive constant folding?
In-Reply-To: <1395814166.43.0.41821499984.issue1721@bugs.pypy.org>
Message-ID: <1395914759.13.0.478413518763.issue1721@bugs.pypy.org>


INADA Naoki <songofacandy at gmail.com> added the comment:

I've filed this to bugs.python.org
http://bugs.python.org/issue21074

Difference between Python 2 and 3 is described at there.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1721>
________________________________________

From tracker at bugs.pypy.org  Thu Mar 27 14:16:13 2014
From: tracker at bugs.pypy.org (coli)
Date: Thu, 27 Mar 2014 13:16:13 +0000
Subject: [pypy-issue] [issue1720] Can not compile win32api with pypy
In-Reply-To: <1395802498.9.0.0604541194408.issue1720@bugs.pypy.org>
Message-ID: <1395926173.85.0.0677938801672.issue1720@bugs.pypy.org>


coli <coli at rocketmail.com> added the comment:

I've seen an implementation of those function at 
https://bitbucket.org/pypy/pypy/branch/embedded-pypy , but they are not merged in 
for some reason.

And yeah, without win32api nothing useful really works.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1720>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 08:12:59 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 28 Mar 2014 07:12:59 +0000
Subject: [pypy-issue] [issue1709] check _cffi_backend on Windows
In-Reply-To: <1395090984.0.0.602689979075.issue1709@bugs.pypy.org>
Message-ID: <1395990779.98.0.252545716785.issue1709@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Sorry, I don't really know at the moment.  This is more a "review needed" report.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1709>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 08:23:32 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 28 Mar 2014 07:23:32 +0000
Subject: [pypy-issue] [issue1723] test runner filling temp directory with
	_testcapi modules
In-Reply-To: <1395909505.08.0.09145133023.issue1723@bugs.pypy.org>
Message-ID: <1395991412.55.0.577734882875.issue1723@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Looking at the files in /tmp on tannit, dated say from the night of March 27, I
see indeed a large quantity of subdirectories "tmpXXXXXXX" with a
_testcapi.pypy-23.so in them.  Fixing that would be the first step.  But there
is a more stuff around, too.  Is there any way we could instead convince the
builds to use some other general directory than /tmp, like a local directory in
the build, maybe with Linux-level hackery to redirect /tmp?

----------
nosy: +arigo
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1723>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 08:31:46 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 28 Mar 2014 07:31:46 +0000
Subject: [pypy-issue] [issue1720] Can not compile win32api with pypy
In-Reply-To: <1395802498.9.0.0604541194408.issue1720@bugs.pypy.org>
Message-ID: <1395991906.29.0.587135955728.issue1720@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Just to be clear, what I don't know is: among projects out there that depend on
win32api, which proportion don't care about the performance of calls to win32api
functions, and which proportion do care?  If it's often the case that Python
projects tend to call many short-running functions of win32api, then simply
attempting to recompile it with cpyext is going to give terribly bad
performance.  In that case, we'd need to redo win32api with cffi.  But I would
like someone that knows a bit of all of Python-on-Windows, win32api and PyPy to
chime in.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1720>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 09:01:06 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 28 Mar 2014 08:01:06 +0000
Subject: [pypy-issue] [issue1716] jitviewer: tcl-dev,
	tk-dev should be added to jitviewer dependencies
In-Reply-To: <1395740939.64.0.378646962826.issue1716@bugs.pypy.org>
Message-ID: <1395993666.5.0.139221342415.issue1716@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

If jitviewer needs tcl/tk, this looks like a bug to me.

----------
nosy: +arigo
status: unread -> chatting

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1716>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 09:04:28 2014
From: tracker at bugs.pypy.org (Dmitrey)
Date: Fri, 28 Mar 2014 08:04:28 +0000
Subject: [pypy-issue] [issue1716] jitviewer: tcl-dev,
	tk-dev should be added to jitviewer dependencies
In-Reply-To: <1395740939.64.0.378646962826.issue1716@bugs.pypy.org>
Message-ID: <1395993868.02.0.628704622922.issue1716@bugs.pypy.org>


Dmitrey <dmitrey15 at ukr.net> added the comment:

I got those messages until I installed tcl-dev, tk-dev from ubuntu aptitude

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1716>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 09:49:20 2014
From: tracker at bugs.pypy.org (mattip)
Date: Fri, 28 Mar 2014 08:49:20 +0000
Subject: [pypy-issue] [issue1723] test runner filling temp directory with
	_testcapi modules
In-Reply-To: <1395909505.08.0.09145133023.issue1723@bugs.pypy.org>
Message-ID: <1395996560.26.0.922083907216.issue1723@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

What do you think of the hash idea to build once, use many times?

fwiw, we support various non-documented environmental variables to control things 
like where to put files (PYPY_USESSION_DIR), compiler (CC) and probably others, 
"someone" should document them in pypy/doc/config/index.rst

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1723>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 09:58:47 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 28 Mar 2014 08:58:47 +0000
Subject: [pypy-issue] [issue1723] test runner filling temp directory with
	_testcapi modules
In-Reply-To: <1395909505.08.0.09145133023.issue1723@bugs.pypy.org>
Message-ID: <1395997127.69.0.709229614699.issue1723@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Sorry, I'm fine with the hash idea.  Please do it.  I was already looking for a
general solution that would fix all the places that write stuff in /tmp, instead
of fixing "only" the most obvious place, but that may not be very useful after all.

PYPY_USESSION_DIR is not a configuration option (as pypy/doc/config/index.rst
documents), it is an environment variable.  (And this particular variable only
controls where /tmp/usession-NUM is created, which is only one directory among
the large number created in /tmp in a test run.)  I don't really think they are
essential either; they are just hacks to change various directories.  Again, if
someone like a Linux distribution wants to ensure that things stay in a
controlled environment, they would be better served by chroot for example.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1723>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 10:06:09 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 28 Mar 2014 09:06:09 +0000
Subject: [pypy-issue] [issue1716] jitviewer: tcl-dev,
	tk-dev should be added to jitviewer dependencies
In-Reply-To: <1395740939.64.0.378646962826.issue1716@bugs.pypy.org>
Message-ID: <1395997569.66.0.883346309649.issue1716@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Sorry if there is some context somewhere else, but this bug report is not enough
(at least for me).  Please explain exactly what you where trying to do, and give
what you got.

Grepping in the complete jitviewer source code, it does not seem to contain the
word "tk" at all.  I can only guess that some package needed by jitviewer
indirectly needs tk.  This is not a problem of jitviewer at all...

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1716>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 10:12:46 2014
From: tracker at bugs.pypy.org (Dmitrey)
Date: Fri, 28 Mar 2014 09:12:46 +0000
Subject: [pypy-issue] [issue1716] jitviewer: tcl-dev,
	tk-dev should be added to jitviewer dependencies
In-Reply-To: <1395740939.64.0.378646962826.issue1716@bugs.pypy.org>
Message-ID: <1395997966.54.0.745567621396.issue1716@bugs.pypy.org>


Dmitrey <dmitrey15 at ukr.net> added the comment:

while executing 

pypy-c jitviewer.py -q -c my_file.py

I got those messages (tcl.h not found, tk.h not found) in ubuntu 13.10 until I
had installed tcl-dev, tk-dev.
Well, it really seems to be optional dependencies for 3rd party packages, but
the output is somewhat annoying.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1716>
________________________________________

From tracker at bugs.pypy.org  Fri Mar 28 12:45:22 2014
From: tracker at bugs.pypy.org (Armin Rigo)
Date: Fri, 28 Mar 2014 11:45:22 +0000
Subject: [pypy-issue] [issue1716] jitviewer: tcl-dev,
	tk-dev should be added to jitviewer dependencies
In-Reply-To: <1395740939.64.0.378646962826.issue1716@bugs.pypy.org>
Message-ID: <1396007122.59.0.987469197246.issue1716@bugs.pypy.org>


Armin Rigo <armin.rigo at gmail.com> added the comment:

Ok, so the problem is that running "try: import tkinter; except ImportError:..."
will print an error message to stderr because cffi doesn't gobble them.  I
suppose the best place to report this problem (even if solving it cleanly might
be difficult) is the cffi issue tracker.  Please post it there, as it is
unrelated to jitviewer and to pypy: https://bitbucket.org/cffi/cffi/issues

----------
status: chatting -> invalid

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1716>
________________________________________

From tracker at bugs.pypy.org  Sat Mar 29 22:32:27 2014
From: tracker at bugs.pypy.org (mattip)
Date: Sat, 29 Mar 2014 21:32:27 +0000
Subject: [pypy-issue] [issue1723] test runner filling temp directory with
	_testcapi modules
In-Reply-To: <1395909505.08.0.09145133023.issue1723@bugs.pypy.org>
Message-ID: <1396128747.49.0.492597275418.issue1723@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

fix commited to build testcapi once, needs confirmation after nightly builds
changeset ac3fb021b26a

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1723>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 30 10:03:37 2014
From: tracker at bugs.pypy.org (mattip)
Date: Sun, 30 Mar 2014 08:03:37 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1396166617.04.0.683462057289.issue1696@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

coli, please try again with the latest nightly from
http://buildbot.pypy.org/nightly/trunk/pypy-c-jit-latest-win32.zip
It seems to install all the pieces for me, but then fails.

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 30 15:10:08 2014
From: tracker at bugs.pypy.org (coli)
Date: Sun, 30 Mar 2014 13:10:08 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1396185008.05.0.794362217922.issue1696@bugs.pypy.org>


coli <coli at rocketmail.com> added the comment:

Yup, getting a different error now:


  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/basecommand.py", line 122, in main
    status = self.run(options, args)
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/commands/install.py", line 279, in run
    requirement_set.install(install_options, global_options, root=options.root_path)
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/req.py", line 1380, in install
    requirement.install(install_options, global_options, *args, **kwargs)
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/req.py", line 664, in install
    self.move_wheel_files(self.source_dir, root=root)
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/req.py", line 894, in move_wheel_files
    pycompile=self.pycompile,
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/wheel.py", line 308, in move_wheel_files
    generated.extend(maker.make(spec))
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/_vendor/distlib/scripts.py", line 305, in make
    self._make_script(entry, filenames, options=options)
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/_vendor/distlib/scripts.py", line 209, in _make_script
    self._write_script(scriptnames, shebang, script, filenames, ext)
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/_vendor/distlib/scripts.py", line 151, in _write_script
    launcher = self._get_launcher('t')
  File "c:\users\owl\appdata\local\temp\tmpxqt7hj\pip.zip\pip/_vendor/distlib/scripts.py", line 284, in _get_launcher
    result = finder(distlib_package).find(name).bytes
AttributeError: 'NoneType' object has no attribute 'bytes'

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 30 21:07:43 2014
From: tracker at bugs.pypy.org (mattip)
Date: Sun, 30 Mar 2014 19:07:43 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1396206463.14.0.777097134344.issue1696@bugs.pypy.org>


mattip <matti.picus at gmail.com> added the comment:

result = finder(distlib_package).find(name)
the wheel installer is looking for name=t32.exe deep in a zipfile (pip.zip)
\pip.zip\pip\_vendor\distlib\t32.exe
and not finding it.

If I expand the pip.zip, intsallation proceeds flawlessly. I'm not sure where to 
look next, perhaps something is wrong with zip directories and pypy win32?

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Sun Mar 30 21:47:41 2014
From: tracker at bugs.pypy.org (coli)
Date: Sun, 30 Mar 2014 19:47:41 +0000
Subject: [pypy-issue] [issue1696] Can not install pip with get-pip.py
In-Reply-To: <1393856361.2.0.130806515172.issue1696@bugs.pypy.org>
Message-ID: <1396208861.64.0.946215243274.issue1696@bugs.pypy.org>


coli <coli at rocketmail.com> added the comment:

This may be a bug at distlib, I found 
https://bitbucket.org/pypa/distlib/issue/40/using-distlibscripts-from-within-a-
zipfile

it was closed as fixed though

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1696>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 31 16:30:17 2014
From: tracker at bugs.pypy.org (lesshaste)
Date: Mon, 31 Mar 2014 14:30:17 +0000
Subject: [pypy-issue] [issue1724] numpy: numpy.random.binomial fails
In-Reply-To: <1396276217.61.0.245768287371.issue1724@bugs.pypy.org>
Message-ID: <1396276217.61.0.245768287371.issue1724@bugs.pypy.org>


New submission from lesshaste <drraph at gmail.com>:

Using Mar 15 2014 daily snapshot.

>>>> import numpy as np
>>>> n=10
>>>> v = np.random.binomial(n,1/4,n)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'NoneType' object is not callable

----------
messages: 6672
nosy: lesshaste, pypy-issue
priority: bug
release: 2.2
status: unread
title: numpy: numpy.random.binomial fails

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1724>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 31 17:58:21 2014
From: tracker at bugs.pypy.org (wolfganglechner)
Date: Mon, 31 Mar 2014 15:58:21 +0000
Subject: [pypy-issue] [issue1725] numpy randint interval
In-Reply-To: <1396281501.7.0.56634307888.issue1725@bugs.pypy.org>
Message-ID: <1396281501.7.0.56634307888.issue1725@bugs.pypy.org>


New submission from wolfganglechner <wolfgang.lechner at gmail.com>:

In python, numpy.random.randint(N) produces a number between 0 and N-1, in pypy 
a number between 0 and N including N. 

expceted bahavior:
>>> import numpy as np
>>> for i in range(100):
...     print np.random.randint(2)
1
0
1
1
1
0
1
1
...

actual behavior:
>>> import numpy as np
>>> for i in range(100):
...     print np.random.randint(2)
0
1
2
...

----------
messages: 6673
nosy: pypy-issue, wolfganglechner
priority: bug
status: unread
title: numpy randint interval

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1725>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 31 18:32:07 2014
From: tracker at bugs.pypy.org (lesshaste)
Date: Mon, 31 Mar 2014 16:32:07 +0000
Subject: [pypy-issue] [issue1726] numpy :numpy.unique reports
	NotImplementedError: sorting	of non-numeric types 'void' is
	not implemented
In-Reply-To: <1396283527.28.0.961256505937.issue1726@bugs.pypy.org>
Message-ID: <1396283527.28.0.961256505937.issue1726@bugs.pypy.org>


New submission from lesshaste <drraph at gmail.com>:

March 15 pypy daily snapshot.

#!/usr/bin/python

import numpy as np
def asvoid(arr):
    arr = np.ascontiguousarray(arr)
    return arr.view(np.dtype((np.void, arr.dtype.itemsize * arr.shape[-1])))


def nodistinctcols(M):
    MT = asvoid(M.T)
    uniqs = np.unique(MT)
    return len(uniqs)

n= 16

X = np.array([np.random.randint(2, size = n) for i in xrange(2**n)])
print "nodistinctcols(X.T)", nodistinctcols(X.T)

Output:
nodistinctcols(X.T)Traceback (most recent call last):
  File "app_main.py", line 72, in run_toplevel
  File "./test.py", line 17, in <module>
    print "nodistinctcols(X.T)", nodistinctcols(X.T)
  File "./test.py", line 11, in nodistinctcols
    uniqs = np.unique(MT)
  File
"/home/raph/python/pypy-2.3-alpha-20140316-linux_x86_64-portable/site-packages/numpy/lib/arraysetops.py",
line 196, in unique
    ar.sort()
NotImplementedError: sorting of non-numeric types 'void' is not implemented


Expected output:

nodistinctcols(X.T) 41465

----------
messages: 6674
nosy: lesshaste, pypy-issue
priority: bug
status: unread
title: numpy :numpy.unique reports NotImplementedError: sorting of non-numeric types 'void' is not implemented

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1726>
________________________________________

From tracker at bugs.pypy.org  Mon Mar 31 20:01:55 2014
From: tracker at bugs.pypy.org (lesshaste)
Date: Mon, 31 Mar 2014 18:01:55 +0000
Subject: [pypy-issue] [issue1727] numpy: numpy.random.choice not implemented
In-Reply-To: <1396288915.29.0.9511952625.issue1727@bugs.pypy.org>
Message-ID: <1396288915.29.0.9511952625.issue1727@bugs.pypy.org>


New submission from lesshaste <drraph at gmail.com>:

March 15 daily snapshot

import numpy as np
np.random.choice([-1,0,0,1], size = 10)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: 'module' object has no attribute 'choice'

----------
messages: 6675
nosy: lesshaste, pypy-issue
priority: bug
status: unread
title: numpy: numpy.random.choice not implemented

________________________________________
PyPy bug tracker <tracker at bugs.pypy.org>
<https://bugs.pypy.org/issue1727>
________________________________________