From ahti at ahti.bluemoon.ee Sun Dec 1 10:21:02 2013 From: ahti at ahti.bluemoon.ee (Ahti Heinla) Date: Sun, 01 Dec 2013 11:21:02 +0200 Subject: [pypy-dev] Proposal for sprint in Tallinn, Estonia In-Reply-To: <20131129171720.GA27004@Plop> References: <12834110.ecgp6i7Rsd@karoliine> <20131129171720.GA27004@Plop> Message-ID: <12159165.JKZIkbDSaq@karoliine> Hi, OK, thanks! To make the scheduling easier, I created this Doodle poll. As I understand, the critical thing is that most of the core developers fill out their preferences (or mark all dates as unsuitable, if not interested in the Tallinn sprint at all), since if there are just one or two core developers coming, this is not going to be a successful sprint. http://doodle.com/4uiwehbfed7ryqf8 The sprint lasts for a week. Select the starting dates (Mondays) suitable for you. One of the dates is already in December, and some are very close to the Leysin sprint, but I'm including them just in case. Ahti On Friday 29 November 2013 6:17:20 PM Romain Guillebert wrote: > Hi Ahti > > I'm interested in going if this sprint happens, I'd be able to go after > February 2nd (as long as it's not during PyCon). > > Cheers > Romain > > On 11/26, Ahti Heinla wrote: > > Hi, > > > > I am new to PyPy, but very impressed with what you guys have done. > > Myself I am best known for having been a founding engineer and Chief Technical > > Architect for Skype. Python is my favourite language, and I want to help PyPy > > somehow. > > > > How about I organise a sprint in Tallinn, Estonia (where I live)? I can hopefully > > drum up some interest among local developers, perhaps also get some top > > ex-Skype engineers to join. The requirement to contribute a week full-time is a > > deterrent for many though, so I am not sure yet how many people I can get. > > > > Myself I have no background in interpreters/compilers, but have written > > code for 32 years, often optimised lowlevel code (assembly, C++). I would > > need some hand-holding to get started, but I am sure I'd be productive in > > a day or two. > > > > Ahti > > ahtih on IRC > > _______________________________________________ > > pypy-dev mailing list > > pypy-dev at python.org > > https://mail.python.org/mailman/listinfo/pypy-dev From anto.cuni at gmail.com Mon Dec 2 10:54:12 2013 From: anto.cuni at gmail.com (Antonio Cuni) Date: Mon, 02 Dec 2013 10:54:12 +0100 Subject: [pypy-dev] Proposal for sprint in Tallinn, Estonia In-Reply-To: <12159165.JKZIkbDSaq@karoliine> References: <12834110.ecgp6i7Rsd@karoliine> <20131129171720.GA27004@Plop> <12159165.JKZIkbDSaq@karoliine> Message-ID: <529C58C4.3000900@gmail.com> Hi, On 01/12/13 10:21, Ahti Heinla wrote: > Hi, > > OK, thanks! To make the scheduling easier, I created this Doodle poll. > As I understand, the critical thing is that most of the core developers > fill out their preferences (or mark all dates as unsuitable, if not interested > in the Tallinn sprint at all), since if there are just one or two core developers > coming, this is not going to be a successful sprint. > > http://doodle.com/4uiwehbfed7ryqf8 > > The sprint lasts for a week. Select the starting dates (Mondays) suitable > for you. One of the dates is already in December, and some are very close to > the Leysin sprint, but I'm including them just in case. I checked more carefully my schedule and noticed that I cannot make it in the winter :-( It would work for me only from mid-march. Sorry for the confusion :-( From dimaqq at gmail.com Mon Dec 2 20:29:14 2013 From: dimaqq at gmail.com (Dima Tisnek) Date: Mon, 2 Dec 2013 20:29:14 +0100 Subject: [pypy-dev] clarification cffi vs _ffi_ vs _ffi vs libffi vs _rawffi vs rffi In-Reply-To: References: Message-ID: Thanks, Armin, doc page looks good now! On a related note, I'm thinking to interface jvm/dalvik, similar to jpype (C module) and following pyjnius (cython module), but in cffi. I'd be very grateful if there was a good pypy/cffi project you could point me to as an example, especially of harder bits like reference management. d. On 27 November 2013 18:08, Armin Rigo wrote: > Re-Hi, > > On Wed, Nov 27, 2013 at 5:24 PM, Armin Rigo wrote: >>> http://doc.pypy.org/en/latest/extending.html > > I updated that page. > > > Armin From pjenvey at underboss.org Mon Dec 2 21:08:55 2013 From: pjenvey at underboss.org (Philip Jenvey) Date: Mon, 2 Dec 2013 12:08:55 -0800 Subject: [pypy-dev] PyPy3 release? In-Reply-To: References: Message-ID: <2D814351-5F80-41B5-96AE-EC32A1F47B6F@underboss.org> On Nov 27, 2013, at 9:25 AM, Armin Rigo wrote: > Hi all, > > Any volunteer to handle the PyPy3 release corresponding to PyPy 2.2? I wasn't intending to cut another release until I finished the integer re-optimizations I've been tackling lately, but we could do one anyway. It's a bit weird w/ PyPy3 and PyPy sharing the version numbering scheme, at least for now, since it implies the release schedules are tied together. Maybe they should be though? Calling it PyPy3 w/ the same version scheme seemed to make the most sense vs the other options. A PyPy3 v0.1 could have broken some cases of code like sys.pypy_version_tuple < (1, 5) in the wild. Calling it PyPy 3.0 would have made sense but forced the CPython 2.7 compat. PyPy stick with a 2.x scheme forever. -- Philip Jenvey From rymg19 at gmail.com Mon Dec 2 22:35:19 2013 From: rymg19 at gmail.com (Ryan Gonzalez) Date: Mon, 2 Dec 2013 15:35:19 -0600 Subject: [pypy-dev] clarification cffi vs _ffi_ vs _ffi vs libffi vs _rawffi vs rffi In-Reply-To: References: Message-ID: Check the _tkinter and _sqlite3 modules. On Mon, Dec 2, 2013 at 1:29 PM, Dima Tisnek wrote: > Thanks, Armin, doc page looks good now! > > On a related note, I'm thinking to interface jvm/dalvik, similar to > jpype (C module) and following pyjnius (cython module), but in cffi. > > I'd be very grateful if there was a good pypy/cffi project you could > point me to as an example, especially of harder bits like reference > management. > > d. > > On 27 November 2013 18:08, Armin Rigo wrote: > > Re-Hi, > > > > On Wed, Nov 27, 2013 at 5:24 PM, Armin Rigo wrote: > >>> http://doc.pypy.org/en/latest/extending.html > > > > I updated that page. > > > > > > Armin > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > -- Ryan When your hammer is C++, everything begins to look like a thumb. -------------- next part -------------- An HTML attachment was scrubbed... URL: From anto.cuni at gmail.com Tue Dec 3 00:13:32 2013 From: anto.cuni at gmail.com (Antonio Cuni) Date: Tue, 03 Dec 2013 00:13:32 +0100 Subject: [pypy-dev] PyPy3 release? In-Reply-To: <2D814351-5F80-41B5-96AE-EC32A1F47B6F@underboss.org> References: <2D814351-5F80-41B5-96AE-EC32A1F47B6F@underboss.org> Message-ID: <529D141C.4030107@gmail.com> On 02/12/13 21:08, Philip Jenvey wrote: > > It's a bit weird w/ PyPy3 and PyPy sharing the version numbering scheme, at least for now, since it implies the release schedules are tied together. Maybe they should be though? > > Calling it PyPy3 w/ the same version scheme seemed to make the most sense vs the other options. A PyPy3 v0.1 could have broken some cases of code like sys.pypy_version_tuple < (1, 5) in the wild. Calling it PyPy 3.0 would have made sense but forced the CPython 2.7 compat. PyPy stick with a 2.x scheme forever. another issue is with cpyext: if sys.pypy_version_number is the same, pypy3 extension modules will have the same .pypy-22.so extension as the pypy2 version, causing potentially lots of troubles. I cannot think of a good way to solve the problem though. One possibility is to have pypy_version_number incremented by 3000, so that this would be PyPy 30002.2. Note that this would still break code like pypy_version_number > (2, 2). From pjenvey at underboss.org Tue Dec 3 03:42:57 2013 From: pjenvey at underboss.org (Philip Jenvey) Date: Mon, 2 Dec 2013 18:42:57 -0800 Subject: [pypy-dev] PyPy3 release? In-Reply-To: <529D141C.4030107@gmail.com> References: <2D814351-5F80-41B5-96AE-EC32A1F47B6F@underboss.org> <529D141C.4030107@gmail.com> Message-ID: <54AAD5EA-EA02-4383-917E-C1FFDBC0E662@underboss.org> On Dec 2, 2013, at 3:13 PM, Antonio Cuni wrote: > On 02/12/13 21:08, Philip Jenvey wrote: >> >> It's a bit weird w/ PyPy3 and PyPy sharing the version numbering scheme, at least for now, since it implies the release schedules are tied together. Maybe they should be though? >> >> Calling it PyPy3 w/ the same version scheme seemed to make the most sense vs the other options. A PyPy3 v0.1 could have broken some cases of code like sys.pypy_version_tuple < (1, 5) in the wild. Calling it PyPy 3.0 would have made sense but forced the CPython 2.7 compat. PyPy stick with a 2.x scheme forever. > > > another issue is with cpyext: if sys.pypy_version_number is the same, pypy3 extension modules will have the same .pypy-22.so extension as the pypy2 version, causing potentially lots of troubles. For this particular issue the suffix could be 'pypy3' instead. Though this is not ideal, should 'PyPy3' be reflected in sys.version, sys._mercurial[0], platform.python_implementation(), etc? > > I cannot think of a good way to solve the problem though. One possibility is to have pypy_version_number incremented by 3000, so that this would be PyPy 30002.2. Note that this would still break code like pypy_version_number > (2, 2). This would be weird but I don't think it would break any version number checks, being a higher number. -- Philip Jenvey From fahd.mokaddem at gmail.com Tue Dec 3 11:55:12 2013 From: fahd.mokaddem at gmail.com (Fahd Mokaddem) Date: Tue, 3 Dec 2013 11:55:12 +0100 Subject: [pypy-dev] Hi, do you consider accepting bitcoins? Message-ID: -------------- next part -------------- An HTML attachment was scrubbed... URL: From fijall at gmail.com Tue Dec 3 12:01:18 2013 From: fijall at gmail.com (Maciej Fijalkowski) Date: Tue, 3 Dec 2013 13:01:18 +0200 Subject: [pypy-dev] Hi, do you consider accepting bitcoins? In-Reply-To: References: Message-ID: No, we don't accept bitcoins as of now. You can still donate using paypal though. On Tue, Dec 3, 2013 at 12:55 PM, Fahd Mokaddem wrote: > > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > From fahd.mokaddem at gmail.com Tue Dec 3 12:03:53 2013 From: fahd.mokaddem at gmail.com (Fahd Mokaddem) Date: Tue, 3 Dec 2013 12:03:53 +0100 Subject: [pypy-dev] Hi, do you consider accepting bitcoins? In-Reply-To: References: Message-ID: okay, thanks for your answer. On Tue, Dec 3, 2013 at 12:01 PM, Maciej Fijalkowski wrote: > No, we don't accept bitcoins as of now. You can still donate using > paypal though. > > On Tue, Dec 3, 2013 at 12:55 PM, Fahd Mokaddem > wrote: > > > > > > _______________________________________________ > > pypy-dev mailing list > > pypy-dev at python.org > > https://mail.python.org/mailman/listinfo/pypy-dev > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bokr at oz.net Tue Dec 3 12:13:55 2013 From: bokr at oz.net (Bengt Richter) Date: Tue, 03 Dec 2013 12:13:55 +0100 Subject: [pypy-dev] PyPy3 release? In-Reply-To: <54AAD5EA-EA02-4383-917E-C1FFDBC0E662@underboss.org> References: <2D814351-5F80-41B5-96AE-EC32A1F47B6F@underboss.org> <529D141C.4030107@gmail.com> <54AAD5EA-EA02-4383-917E-C1FFDBC0E662@underboss.org> Message-ID: <529DBCF3.9020703@oz.net> On 12/03/2013 03:42 AM Philip Jenvey wrote: > > On Dec 2, 2013, at 3:13 PM, Antonio Cuni wrote: > >> On 02/12/13 21:08, Philip Jenvey wrote: >>> >>> It's a bit weird w/ PyPy3 and PyPy sharing the version numbering scheme, at least for now, since it implies the release schedules are tied together. Maybe they should be though? >>> >>> Calling it PyPy3 w/ the same version scheme seemed to make the most sense vs the other options. A PyPy3 v0.1 could have broken some cases of code like sys.pypy_version_tuple< (1, 5) in the wild. Calling it PyPy 3.0 would have made sense but forced the CPython 2.7 compat. PyPy stick with a 2.x scheme forever. >> >> >> another issue is with cpyext: if sys.pypy_version_number is the same, pypy3 extension modules will have the same .pypy-22.so extension as the pypy2 version, causing potentially lots of troubles. > > For this particular issue the suffix could be 'pypy3' instead. Though this is not ideal, should 'PyPy3' be reflected in sys.version, sys._mercurial[0], platform.python_implementation(), etc? > >> >> I cannot think of a good way to solve the problem though. One possibility is to have pypy_version_number incremented by 3000, so that this would be PyPy 30002.2. Note that this would still break code like pypy_version_number> (2, 2). > > This would be weird but I don't think it would break any version number checks, being a higher number. > > -- > Philip Jenvey Just a thought: If you consider that pypy is an alternate *implementation* of the python lanaguage, ISTM the implementation should do whatever the version of the python *language* implies (2.7.3 as of now?). The version of the implementation (pypy or other) would seem to be orthogonal. No? Regards, Bengt Richter From amauryfa at gmail.com Tue Dec 3 16:57:14 2013 From: amauryfa at gmail.com (Amaury Forgeot d'Arc) Date: Tue, 3 Dec 2013 16:57:14 +0100 Subject: [pypy-dev] clarification cffi vs _ffi_ vs _ffi vs libffi vs _rawffi vs rffi In-Reply-To: References: Message-ID: 2013/12/2 Ryan Gonzalez > Check the _tkinter and _sqlite3 modules. > But they don't do advanced reference management, even if _tkinter/tclobj.py has some calls to Tcl_IncrRefCount() and Tcl_DecrRefCount(). The lxml module has interesting code to manage cross-references between C and Python. I'm afraid it's a bit difficult to read (it's an almost exact port of the cython version) But I found that ffi.new_handle() and ffi.from_handle() are really handy in this case: https://github.com/amauryfa/lxml/blob/cffi/src/lxml-cffi/proxy.py#L20 > On Mon, Dec 2, 2013 at 1:29 PM, Dima Tisnek wrote: > >> Thanks, Armin, doc page looks good now! >> >> On a related note, I'm thinking to interface jvm/dalvik, similar to >> jpype (C module) and following pyjnius (cython module), but in cffi. >> >> I'd be very grateful if there was a good pypy/cffi project you could >> point me to as an example, especially of harder bits like reference >> management. >> >> d. >> >> On 27 November 2013 18:08, Armin Rigo wrote: >> > Re-Hi, >> > >> > On Wed, Nov 27, 2013 at 5:24 PM, Armin Rigo wrote: >> >>> http://doc.pypy.org/en/latest/extending.html >> > >> > I updated that page. >> > >> > >> > Armin >> _______________________________________________ >> pypy-dev mailing list >> pypy-dev at python.org >> https://mail.python.org/mailman/listinfo/pypy-dev >> > > > > -- > Ryan > When your hammer is C++, everything begins to look like a thumb. > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > > -- Amaury Forgeot d'Arc -------------- next part -------------- An HTML attachment was scrubbed... URL: From dimaqq at gmail.com Tue Dec 3 17:24:28 2013 From: dimaqq at gmail.com (Dima Tisnek) Date: Tue, 3 Dec 2013 17:24:28 +0100 Subject: [pypy-dev] clarification cffi vs _ffi_ vs _ffi vs libffi vs _rawffi vs rffi In-Reply-To: References: Message-ID: Amaury, thanks, this is exactly the kind of example I was looking for! On 3 December 2013 16:57, Amaury Forgeot d'Arc wrote: > 2013/12/2 Ryan Gonzalez >> >> Check the _tkinter and _sqlite3 modules. > > > But they don't do advanced reference management, even if > _tkinter/tclobj.py has some calls to Tcl_IncrRefCount() and > Tcl_DecrRefCount(). > > The lxml module has interesting code to manage cross-references between C > and Python. > I'm afraid it's a bit difficult to read (it's an almost exact port of the > cython version) > But I found that ffi.new_handle() and ffi.from_handle() are really handy in > this case: > https://github.com/amauryfa/lxml/blob/cffi/src/lxml-cffi/proxy.py#L20 > > >> >> On Mon, Dec 2, 2013 at 1:29 PM, Dima Tisnek wrote: >>> >>> Thanks, Armin, doc page looks good now! >>> >>> On a related note, I'm thinking to interface jvm/dalvik, similar to >>> jpype (C module) and following pyjnius (cython module), but in cffi. >>> >>> I'd be very grateful if there was a good pypy/cffi project you could >>> point me to as an example, especially of harder bits like reference >>> management. >>> >>> d. >>> >>> On 27 November 2013 18:08, Armin Rigo wrote: >>> > Re-Hi, >>> > >>> > On Wed, Nov 27, 2013 at 5:24 PM, Armin Rigo wrote: >>> >>> http://doc.pypy.org/en/latest/extending.html >>> > >>> > I updated that page. >>> > >>> > >>> > Armin >>> _______________________________________________ >>> pypy-dev mailing list >>> pypy-dev at python.org >>> https://mail.python.org/mailman/listinfo/pypy-dev >> >> >> >> >> -- >> Ryan >> When your hammer is C++, everything begins to look like a thumb. >> >> _______________________________________________ >> pypy-dev mailing list >> pypy-dev at python.org >> https://mail.python.org/mailman/listinfo/pypy-dev >> > > > > -- > Amaury Forgeot d'Arc From arigo at tunes.org Tue Dec 3 19:47:14 2013 From: arigo at tunes.org (Armin Rigo) Date: Tue, 3 Dec 2013 19:47:14 +0100 Subject: [pypy-dev] Pluggable HTM In-Reply-To: References: Message-ID: Hi Dimitri, On Wed, Nov 27, 2013 at 9:17 AM, Dimitri Vorona wrote: > the original STM proposal spoke of HTM as of a thing of a far future. Now, > Haswells are out and provide built-in HTM support in form of TSX. In the > near future I expect more and more systems to have it. > > Are there plan to make PyPy use HTM if it is available on the system? I don't know yet. I've just started playing with an Intel Haswell, and getting slightly bad results in the form of too many random transaction aborts. This seems so for both "small" transactions that only access some 20KB of data, up to larger transaction of almost 768KB, which is impressively three times the size of the L2 cache; this seems to say that even the L3 cache can dedicate a part of its resources to storing the transaction cache lines. But a naive extrapolation of the single-threaded results shows that, if we had instead 8 threads running with the same results, even on completely independent data, they would still abort too many transactions each. Whenever a transaction needs to be redone without HTM, it really needs to stop all other threads. So "too many" is in this sense: even if it is only 10-20% on each core, it's enough to prevent any scaling beyond just a coupe of cores. It may be that I'm missing something, like a way to learn where conflicts occur. But all in all it is unclear if this is good enough for PyPy (or CPython). The next step, which I might do anyway, would be to extract from the pypy-stm branch the general logic (most notably the numerous conflict-avoiding small changes), and try to run that with HTM. This probably requires writing a different GC, but it should be easy at this point to do, experimentally. A bient?t, Armin. From arigo at tunes.org Tue Dec 3 21:10:13 2013 From: arigo at tunes.org (Armin Rigo) Date: Tue, 3 Dec 2013 21:10:13 +0100 Subject: [pypy-dev] Integrating CPython and PyPy, maybe Message-ID: Hi all, Starting from last night, the nightly builds of PyPy are able to run this kind of code: http://bpaste.net/show/155166/ (You have to adjust the libpython2.x.so path manually.) The idea is to link in the libpython.so, using cffi. Then you can use it much like you would in C. For example, you can import complete CPython packages and call them. This is similar to the earlier experiments done by fijal about calling matplotlib via CPython, except this time done with cffi. The code above supports __add__(), which lets you just write "x3 = x1 + x2" for any two PyObject instances x1 and x2. It's unclear if it's a good idea, but it's at least a cool hack. To be really useful, it needs much more work, of course, starting with a clearer design of what the goal really would be :-) A bient?t, Armin. From ddvento at ucar.edu Tue Dec 3 21:24:29 2013 From: ddvento at ucar.edu (Davide Del Vento) Date: Tue, 03 Dec 2013 13:24:29 -0700 Subject: [pypy-dev] Integrating CPython and PyPy, maybe In-Reply-To: References: Message-ID: <529E3DFD.6020902@ucar.edu> Yes it does sound like a cool hack to me. Let me try to state a possible goal: Allow PyPy to become "exactly" drop-in replacement for CPython not just the current "almost", by using CPython itself for the things that don't work in PyPy Regards, Davide Del Vento, On 12/03/2013 01:10 PM, Armin Rigo wrote: > Hi all, > > Starting from last night, the nightly builds of PyPy are able to run > this kind of code: > > http://bpaste.net/show/155166/ > > (You have to adjust the libpython2.x.so path manually.) > > The idea is to link in the libpython.so, using cffi. Then you can use > it much like you would in C. For example, you can import complete > CPython packages and call them. This is similar to the earlier > experiments done by fijal about calling matplotlib via CPython, except > this time done with cffi. > > The code above supports __add__(), which lets you just write "x3 = x1 > + x2" for any two PyObject instances x1 and x2. It's unclear if it's > a good idea, but it's at least a cool hack. To be really useful, it > needs much more work, of course, starting with a clearer design of > what the goal really would be :-) > > > A bient?t, > > Armin. > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > From arigo at tunes.org Tue Dec 3 22:50:50 2013 From: arigo at tunes.org (Armin Rigo) Date: Tue, 3 Dec 2013 22:50:50 +0100 Subject: [pypy-dev] Integrating CPython and PyPy, maybe In-Reply-To: <529E3DFD.6020902@ucar.edu> References: <529E3DFD.6020902@ucar.edu> Message-ID: Hi Davide, On Tue, Dec 3, 2013 at 9:24 PM, Davide Del Vento wrote: > Let me try to state a possible goal: > > Allow PyPy to become "exactly" drop-in replacement for CPython not just the > current "almost", by using CPython itself for the things that don't work in > PyPy Thank you for stating this ambitious goal that I could not have thought about myself. Sadly, however, I have to tell you, it doesn't work this way. :-) Armin From ddvento at ucar.edu Tue Dec 3 23:54:19 2013 From: ddvento at ucar.edu (Davide Del Vento) Date: Tue, 03 Dec 2013 15:54:19 -0700 Subject: [pypy-dev] Integrating CPython and PyPy, maybe In-Reply-To: References: <529E3DFD.6020902@ucar.edu> Message-ID: Hi Armin, >> Let me try to state a possible goal: >> >> Allow PyPy to become "exactly" drop-in replacement for CPython not >>just the >> current "almost", by using CPython itself for the things that don't >>work in >> PyPy > > Thank you for stating this ambitious goal that I could not have > thought about myself. Sadly, however, I have to tell you, it > doesn't work this way. :-) I know it doesn't. However, if my understanding is correct, the only incompatibilities between code that works on CPython and doesn't in PyPy is something that uses some non-pure-python libraries. Now, if that is correct, you can use your technique to access those libraries from CPython where they work by definition. Would these libs work fine in PyPy with this hack? If so, then we're done, at least in theory. In practice, the hack is a lot of dirty code (including the always-changing path), but maybe it could be scripted under something like a "native import" command which could do all the plumbing work under the cover (and set the path once and forever at install time in some way). Am I missing something or is it simply too much work to make it right? Cheers, Davide From andrea.jeradi at studenti.univr.it Fri Dec 6 11:24:44 2013 From: andrea.jeradi at studenti.univr.it (andrea.jeradi at studenti.univr.it) Date: Fri, 06 Dec 2013 11:24:44 +0100 Subject: [pypy-dev] Questions about pypy log Message-ID: Hello, I'm Andrea Jeradi, a computer science student of university of Verona (Italy). I'm doing a project and I'm using pypy. On Internet I have not found explanation about the meaning of some entry of the log. [b6faf360f285] {jit-summary Tracing:????? ??? 2??? 0.006712 Backend:????? ??? 2??? 0.000884 TOTAL:????? ??? ??? 0.955633 ops:???????????? ??? 705 recorded ops:??? ??? 168 ? calls:???????? ??? 3 guards:????????? ??? 58 opt ops:???????? ??? 263 opt guards:????? ??? 58 forcings:??????? ??? 0 abort: trace too long:??? 0 abort: compiling:??? 0 abort: vable escape:??? 0 abort: bad loop: ??? 0 abort: force quasi-immut:??? 0 nvirtuals:?????? ??? 0 nvholes:???????? ??? 0 nvreused:??????? ??? 0 Total # of loops:??? 1 Total # of bridges:??? 1 Freed # of loops:??? 0 Freed # of bridges:??? 0 [b6faf3623ec5] jit-summary} Exist a site, a paper or something else that explain it? Particularly what means 'abort: vable escape', 'abort: bad loop', 'abort: force quasi-immut', 'nvirtuals', 'nvholes' and 'nvreused'? Thanks for the help, best regards Andrea Jeradi -------------- next part -------------- An HTML attachment was scrubbed... URL: From rymg19 at gmail.com Fri Dec 6 15:29:58 2013 From: rymg19 at gmail.com (Ryan) Date: Fri, 06 Dec 2013 08:29:58 -0600 Subject: [pypy-dev] Questions about pypy log In-Reply-To: References: Message-ID: This may seem simple, but have you tried using jitviewer? andrea.jeradi at studenti.univr.it wrote: >Hello, > >I'm Andrea Jeradi, a computer science student of university of Verona >(Italy). I'm doing a project and I'm using pypy. On Internet I have not > found explanation about the meaning of some entry of the log. > > [b6faf360f285] {jit-summary > Tracing:????? ??? 2??? 0.006712 > Backend:????? ??? 2??? 0.000884 > TOTAL:????? ??? ??? 0.955633 > ops:???????????? ??? 705 > recorded ops:??? ??? 168 > ? calls:???????? ??? 3 > guards:????????? ??? 58 > opt ops:???????? ??? 263 > opt guards:????? ??? 58 > forcings:??????? ??? 0 > abort: trace too long:??? 0 > abort: compiling:??? 0 > abort: vable escape:??? 0 > abort: bad loop: ??? 0 > abort: force quasi-immut:??? 0 > nvirtuals:?????? ??? 0 > nvholes:???????? ??? 0 > nvreused:??????? ??? 0 > Total # of loops:??? 1 > Total # of bridges:??? 1 > Freed # of loops:??? 0 > Freed # of bridges:??? 0 > [b6faf3623ec5] jit-summary} > > Exist a site, a paper or something else that explain it? >Particularly what means 'abort: vable escape', 'abort: bad loop', >'abort: force quasi-immut', 'nvirtuals', 'nvholes' and 'nvreused'? > > Thanks for the help, > best regards > Andrea Jeradi > > > >------------------------------------------------------------------------ > >_______________________________________________ >pypy-dev mailing list >pypy-dev at python.org >https://mail.python.org/mailman/listinfo/pypy-dev -- Sent from my Android device with K-9 Mail. Please excuse my brevity. -------------- next part -------------- An HTML attachment was scrubbed... URL: From arigo at tunes.org Sat Dec 7 15:01:54 2013 From: arigo at tunes.org (Armin Rigo) Date: Sat, 7 Dec 2013 15:01:54 +0100 Subject: [pypy-dev] Questions about pypy log In-Reply-To: References: Message-ID: Hi Andrea, On Fri, Dec 6, 2013 at 11:24 AM, wrote: > I'm Andrea Jeradi, a computer science student of university of Verona > (Italy). I'm doing a project and I'm using pypy. On Internet I have not > found explanation about the meaning of some entry of the log. The whole log's format and what it prints are internal to PyPy and changes from time to time. If you find obscure entries that you cannot find documented anywhere, it's because they are not recognized as very useful so they were never documented. To find more information about them, the source code of PyPy is the reference. A bient?t, Armin. From taavi.burns at gmail.com Mon Dec 9 01:17:12 2013 From: taavi.burns at gmail.com (Taavi Burns) Date: Sun, 8 Dec 2013 19:17:12 -0500 Subject: [pypy-dev] Pluggable HTM In-Reply-To: References: Message-ID: My gut says that HTM would be something used to help make STM commits faster (if that's possible), not to replace PyPy's STM machinery entirely. And maybe with something like CPython, one could replace the GIL entirely with HTM, but you'd probably want to make the "ticks between releases" a lot shorter to reduce the chance of conflicts. Things to try in my Copious Amounts of Free Time. :/ On Tue, Dec 3, 2013 at 1:47 PM, Armin Rigo wrote: > Hi Dimitri, > > On Wed, Nov 27, 2013 at 9:17 AM, Dimitri Vorona wrote: >> the original STM proposal spoke of HTM as of a thing of a far future. Now, >> Haswells are out and provide built-in HTM support in form of TSX. In the >> near future I expect more and more systems to have it. >> >> Are there plan to make PyPy use HTM if it is available on the system? > > I don't know yet. I've just started playing with an Intel Haswell, > and getting slightly bad results in the form of too many random > transaction aborts. > > This seems so for both "small" transactions that only access some 20KB > of data, up to larger transaction of almost 768KB, which is > impressively three times the size of the L2 cache; this seems to say > that even the L3 cache can dedicate a part of its resources to storing > the transaction cache lines. > > But a naive extrapolation of the single-threaded results shows that, > if we had instead 8 threads running with the same results, even on > completely independent data, they would still abort too many > transactions each. Whenever a transaction needs to be redone without > HTM, it really needs to stop all other threads. So "too many" is in > this sense: even if it is only 10-20% on each core, it's enough to > prevent any scaling beyond just a coupe of cores. > > It may be that I'm missing something, like a way to learn where > conflicts occur. But all in all it is unclear if this is good enough > for PyPy (or CPython). The next step, which I might do anyway, would > be to extract from the pypy-stm branch the general logic (most notably > the numerous conflict-avoiding small changes), and try to run that > with HTM. This probably requires writing a different GC, but it > should be easy at this point to do, experimentally. > > > A bient?t, > > Armin. > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev -- taa /*eof*/ From andrea.jeradi at studenti.univr.it Mon Dec 9 10:06:40 2013 From: andrea.jeradi at studenti.univr.it (Andrea Jeradi) Date: Mon, 09 Dec 2013 10:06:40 +0100 Subject: [pypy-dev] Questions about pypy log In-Reply-To: References: Message-ID: <0C2C4D3E-EB92-4C8C-92AD-318E734841A1@studenti.univr.it> Hello, I have just tried to use jitviewer but it shows me the traces, the opcode and the assembly. For me it is easier to use the log file directly, and so i need to understand the meaning of the jit-summary, and for completeness what the entry of the summary refers to. For example the ?abort: trace too long? entry shows me the number of traces that have been aborted because they were too long (for standard option of pypy trace_limit=6000). But i don?t understand the meaning of the other entrys. Il giorno 07/dic/2013, alle ore 15:01, Armin Rigo ha scritto: > Hi Andrea, > > On Fri, Dec 6, 2013 at 11:24 AM, wrote: >> I'm Andrea Jeradi, a computer science student of university of Verona >> (Italy). I'm doing a project and I'm using pypy. On Internet I have not >> found explanation about the meaning of some entry of the log. > > The whole log's format and what it prints are internal to PyPy and > changes from time to time. If you find obscure entries that you > cannot find documented anywhere, it's because they are not recognized > as very useful so they were never documented. To find more > information about them, the source code of PyPy is the reference. > > > A bient?t, > > Armin. -------------- next part -------------- An HTML attachment was scrubbed... URL: From jpcaissy at piji.ca Sun Dec 8 20:46:53 2013 From: jpcaissy at piji.ca (Jean-Philippe Caissy) Date: Sun, 8 Dec 2013 14:46:53 -0500 Subject: [pypy-dev] Compiling stmgc-c4 fails on 32bit Message-ID: Hi, I'm trying to compile stmgc-c4 on a fresh Ubuntu 12.04 32bit machine (lxc containers if that matters), and it fails. However, a fresh 12.04 64bit installation, it compiles without any problem. Both the jit and no-jit fail (-Ojit and -O2). The stack trace for the jit is really long, and can be found here : https://gist.github.com/jpcaissy/7862899 and the stack trace for the no-jit can be found here : https://gist.github.com/jpcaissy/7862913 Could that be a bug, or am I doing something wrong? Thanks, JP -------------- next part -------------- An HTML attachment was scrubbed... URL: From arigo at tunes.org Mon Dec 9 11:32:14 2013 From: arigo at tunes.org (Armin Rigo) Date: Mon, 9 Dec 2013 11:32:14 +0100 Subject: [pypy-dev] Questions about pypy log In-Reply-To: <0C2C4D3E-EB92-4C8C-92AD-318E734841A1@studenti.univr.it> References: <0C2C4D3E-EB92-4C8C-92AD-318E734841A1@studenti.univr.it> Message-ID: Hi again, On Mon, Dec 9, 2013 at 10:06 AM, Andrea Jeradi wrote: > (...) i need to > understand the meaning of the jit-summary, and for completeness what the > entry of the summary refers to. As I said, half the entries are very internal. I could try to explain them to you but that would take a long explanation and not give you anything useful. If you're really interested in that level of detail, start by browsing the source code of PyPy's JIT. If you're not really interested in it, please consider that they are just random numbers. A bient?t, Armin. From fijall at gmail.com Mon Dec 9 11:42:03 2013 From: fijall at gmail.com (Maciej Fijalkowski) Date: Mon, 9 Dec 2013 12:42:03 +0200 Subject: [pypy-dev] Questions about pypy log In-Reply-To: References: <0C2C4D3E-EB92-4C8C-92AD-318E734841A1@studenti.univr.it> Message-ID: On Mon, Dec 9, 2013 at 12:32 PM, Armin Rigo wrote: > Hi again, > > On Mon, Dec 9, 2013 at 10:06 AM, Andrea Jeradi > wrote: >> (...) i need to >> understand the meaning of the jit-summary, and for completeness what the >> entry of the summary refers to. > > As I said, half the entries are very internal. I could try to explain > them to you but that would take a long explanation and not give you > anything useful. If you're really interested in that level of detail, > start by browsing the source code of PyPy's JIT. If you're not really > interested in it, please consider that they are just random numbers. Also, if you're *really* interested, come to #pypy on IRC and that's a better place for an interactive discussion. From arigo at tunes.org Mon Dec 9 11:58:38 2013 From: arigo at tunes.org (Armin Rigo) Date: Mon, 9 Dec 2013 11:58:38 +0100 Subject: [pypy-dev] Compiling stmgc-c4 fails on 32bit In-Reply-To: References: Message-ID: Hi Jean-Philippe, On Sun, Dec 8, 2013 at 8:46 PM, Jean-Philippe Caissy wrote: > I'm trying to compile stmgc-c4 on a fresh Ubuntu 12.04 32bit machine (lxc > containers if that matters), and it fails. However, a fresh 12.04 64bit > installation, it compiles without any problem. For now, it's known to only compile on 64bit Linux. It's all very experimental still, so there was no effort to port the fixes we did to other platforms. I guess the non-jit version should mostly work on 32bit Linux, btw, and fixing it would be easy; but the JIT requires a (little) bit more work. A bient?t, Armin. From mario.pernici at gmail.com Thu Dec 12 22:40:09 2013 From: mario.pernici at gmail.com (Mario Pernici) Date: Thu, 12 Dec 2013 22:40:09 +0100 Subject: [pypy-dev] faster rbigint multiplication Message-ID: Hello, I would like to submit a patch to rbigint.py (50 lines of code in _x_mul_), speeding up multiplication; on my computer (x86_64) it seems to be at least as fast as CPython for numbers with few words, about 2x faster for numbers with more than 20 words. Where should I post it? Mario -------------- next part -------------- An HTML attachment was scrubbed... URL: From alex.gaynor at gmail.com Thu Dec 12 22:42:34 2013 From: alex.gaynor at gmail.com (Alex Gaynor) Date: Thu, 12 Dec 2013 13:42:34 -0800 Subject: [pypy-dev] faster rbigint multiplication In-Reply-To: References: Message-ID: Hi Mario, You can either file a ticket and upload a patch at bugs.pypy.org, or send a pull request on bitbucket. Alex On Thu, Dec 12, 2013 at 1:40 PM, Mario Pernici wrote: > > Hello, > I would like to submit a patch to rbigint.py (50 lines of code in > _x_mul_), > speeding up multiplication; on my computer (x86_64) > it seems to be at least as fast as CPython for numbers with few words, > about 2x faster for numbers with more than 20 words. > Where should I post it? > > Mario > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > > -- "I disapprove of what you say, but I will defend to the death your right to say it." -- Evelyn Beatrice Hall (summarizing Voltaire) "The people's good is the highest law." -- Cicero GPG Key fingerprint: 125F 5C67 DFE9 4084 -------------- next part -------------- An HTML attachment was scrubbed... URL: From naylor.b.david at gmail.com Mon Dec 16 10:40:50 2013 From: naylor.b.david at gmail.com (David Naylor) Date: Mon, 16 Dec 2013 12:40:50 +0300 Subject: [pypy-dev] FreeBSD 9.2 buildslave: first run In-Reply-To: References: <634914A010D0B943A035D226786325D4446A18BDD4@EXVMBX020-12.exch020.serverdata.net> Message-ID: <3898390.kQhPxa5noV@dragon.dg> On Sunday, 17 November 2013 19:09:00 Maciej Fijalkowski wrote: > On Sun, Nov 17, 2013 at 6:22 PM, Tobias Oberstein > > wrote: > > Hi, > > > > the new FreeBSD buildslave has run the first time: > > > > http://buildbot.pypy.org/builders/pypy-c-jit-freebsd-9-x86-64/builds/6 > > > > PyPy was built successfully, but there are a couple of issues: > > > > https://bugs.pypy.org/issue1637 > > https://bugs.pypy.org/issue1638 > > https://bugs.pypy.org/issue1639 > > > > I'll address those. > > > > There is another issue that bugs me .. look at the built times: > > > > [Timer] Timings: > > [Timer] annotate --- 576.0 s > > [Timer] rtype_lltype --- 1323.7 s > > [Timer] pyjitpl_lltype --- 1098.6 s > > [Timer] backendopt_lltype --- 314.2 s > > [Timer] stackcheckinsertion_lltype --- 161.5 s > > [Timer] database_c --- 451.3 s > > [Timer] source_c --- 668.5 s > > [Timer] compile_c --- 2403.8 s > > [Timer] =========================================== > > [Timer] Total: --- 6997.6 s > > > > and compare with "Linux 64 on allegro64": > > > > [Timer] Timings: > > [Timer] annotate --- 641.5 s > > [Timer] rtype_lltype --- 1528.5 s > > [Timer] pyjitpl_lltype --- 938.7 s > > [Timer] backendopt_lltype --- 242.2 s > > [Timer] stackcheckinsertion_lltype --- 206.8 s > > [Timer] database_c --- 325.3 s > > [Timer] source_c --- 357.7 s > > [Timer] compile_c --- 246.4 s > > [Timer] =========================================== > > [Timer] Total: --- 4487.1 s > > > > ***** > > > > Why the heck does "compile_c" take 10x the time on FreeBSD? > > > > Note that FreeBSD builder uses Clang .. could that be a reason? > > > > Any hints on that are welcome! > > > > Cheers, > > Tobias > > > > > > > > _______________________________________________ > > pypy-dev mailing list > > pypy-dev at python.org > > https://mail.python.org/mailman/listinfo/pypy-dev > > I would guess because we can't guess the number of processors we don't > parallelize the build correctly. Hi, Please see attached for a patch that fixes this. It should be trivial to add support for the other bsd systems, however I am not sure where the other bsds put their sysctl. Tobias: can you please confirm if this patch works (please make sure there is no MAKEFLAGS specified). The Ports Collection invokes the Makefile directly so never needed this. Regards -------------- next part -------------- A non-text attachment was scrubbed... Name: patch-rpython__config__support.py Type: text/x-python Size: 1186 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 196 bytes Desc: This is a digitally signed message part. URL: From tobias.oberstein at tavendo.de Mon Dec 16 14:12:03 2013 From: tobias.oberstein at tavendo.de (Tobias Oberstein) Date: Mon, 16 Dec 2013 05:12:03 -0800 Subject: [pypy-dev] FreeBSD 9.2 buildslave: first run In-Reply-To: <3898390.kQhPxa5noV@dragon.dg> References: <634914A010D0B943A035D226786325D4446A18BDD4@EXVMBX020-12.exch020.serverdata.net> <3898390.kQhPxa5noV@dragon.dg> Message-ID: <634914A010D0B943A035D226786325D4446BAC77E0@EXVMBX020-12.exch020.serverdata.net> Hi David, the FreeBSD buildslave now has appropriate env flags (MAKEFLAGS=-j8) which seems to solve the issue without patching .. /Tobias > -----Urspr?ngliche Nachricht----- > Von: David Naylor [mailto:naylor.b.david at gmail.com] > Gesendet: Montag, 16. Dezember 2013 10:41 > An: Maciej Fijalkowski > Cc: Tobias Oberstein; pypy-dev at python.org > Betreff: Re: [pypy-dev] FreeBSD 9.2 buildslave: first run > > On Sunday, 17 November 2013 19:09:00 Maciej Fijalkowski wrote: > > On Sun, Nov 17, 2013 at 6:22 PM, Tobias Oberstein > > > > wrote: > > > Hi, > > > > > > the new FreeBSD buildslave has run the first time: > > > > > > http://buildbot.pypy.org/builders/pypy-c-jit-freebsd-9-x86-64/builds > > > /6 > > > > > > PyPy was built successfully, but there are a couple of issues: > > > > > > https://bugs.pypy.org/issue1637 > > > https://bugs.pypy.org/issue1638 > > > https://bugs.pypy.org/issue1639 > > > > > > I'll address those. > > > > > > There is another issue that bugs me .. look at the built times: > > > > > > [Timer] Timings: > > > [Timer] annotate --- 576.0 s > > > [Timer] rtype_lltype --- 1323.7 s > > > [Timer] pyjitpl_lltype --- 1098.6 s > > > [Timer] backendopt_lltype --- 314.2 s > > > [Timer] stackcheckinsertion_lltype --- 161.5 s > > > [Timer] database_c --- 451.3 s > > > [Timer] source_c --- 668.5 s > > > [Timer] compile_c --- 2403.8 s > > > [Timer] =========================================== > > > [Timer] Total: --- 6997.6 s > > > > > > and compare with "Linux 64 on allegro64": > > > > > > [Timer] Timings: > > > [Timer] annotate --- 641.5 s > > > [Timer] rtype_lltype --- 1528.5 s > > > [Timer] pyjitpl_lltype --- 938.7 s > > > [Timer] backendopt_lltype --- 242.2 s > > > [Timer] stackcheckinsertion_lltype --- 206.8 s > > > [Timer] database_c --- 325.3 s > > > [Timer] source_c --- 357.7 s > > > [Timer] compile_c --- 246.4 s > > > [Timer] =========================================== > > > [Timer] Total: --- 4487.1 s > > > > > > ***** > > > > > > Why the heck does "compile_c" take 10x the time on FreeBSD? > > > > > > Note that FreeBSD builder uses Clang .. could that be a reason? > > > > > > Any hints on that are welcome! > > > > > > Cheers, > > > Tobias > > > > > > > > > > > > _______________________________________________ > > > pypy-dev mailing list > > > pypy-dev at python.org > > > https://mail.python.org/mailman/listinfo/pypy-dev > > > > I would guess because we can't guess the number of processors we don't > > parallelize the build correctly. > > Hi, > > Please see attached for a patch that fixes this. It should be trivial to add > support for the other bsd systems, however I am not sure where the other > bsds put their sysctl. > > Tobias: can you please confirm if this patch works (please make sure there is > no MAKEFLAGS specified). The Ports Collection invokes the Makefile directly > so never needed this. > > Regards From r.voigtlaender at gmail.com Thu Dec 19 08:57:47 2013 From: r.voigtlaender at gmail.com (=?ISO-8859-1?Q?Robert_Voigtl=E4nder?=) Date: Thu, 19 Dec 2013 08:57:47 +0100 Subject: [pypy-dev] gevent Message-ID: Hi, I am relatively new to python and more or less completely new to pypy. So far I have pypy running on my raspberry pi. I don?t use virtualenv. Modules for pypy I simply installed via easy_installer from the pypy bin dir. I now try to get gevent installed. I found this remark: ?stackless support - eventlet just works and gevent requires pypycore and pypy-hacks branch of gevent (which mostly disables cython-based modules)? On the github site there is a small instruction on how to install gevent with pip in the virtualenv. $ virtualenv -p /path/to/bin/pypy venv $ source venv/bin/activate (venv)$ pip install git+git://github.com/schmir/gevent at pypy-hacks (venv)$ pip install cffi (venv)$ pip install git+git://github.com/gevent-on-pypy/pypycore (venv)$ export GEVENT_LOOP=pypycore.loop But I am not able to translate this into an installation with easy_install for my environment (without venv) Can someone give me a tip for getting started? Thanks Robert -------------- next part -------------- An HTML attachment was scrubbed... URL: From piotr.skamruk at gmail.com Thu Dec 19 09:58:38 2013 From: piotr.skamruk at gmail.com (Piotr Skamruk) Date: Thu, 19 Dec 2013 09:58:38 +0100 Subject: [pypy-dev] gevent In-Reply-To: References: Message-ID: Try: easy_install cffi git clone git://github.com/schmir/gevent && cd gevent && git checkout pypy-hacks && pypy setup.py install git clone git://github.com/gevent-on-pypy/pypycore && cd pypycore && pypy setup.py install After that - You have to remember to export GEVENT_LOOP=pypycore.loop before starting any of your program which would use this setup. 2013/12/19 Robert Voigtl?nder : > Hi, > > > > I am relatively new to python and more or less completely new to pypy. So > far I have pypy running on my raspberry pi. I don?t use virtualenv. Modules > for pypy I simply installed via easy_installer from the pypy bin dir. > > I now try to get gevent installed. > > > I found this remark: ?stackless support - eventlet just works and gevent > requires pypycore and pypy-hacks branch of gevent (which mostly disables > cython-based modules)? > > > > On the github site there is a small instruction on how to install gevent > with pip in the virtualenv. > > > > $ virtualenv -p /path/to/bin/pypy venv > > $ source venv/bin/activate > > (venv)$ pip install git+git://github.com/schmir/gevent at pypy-hacks > > (venv)$ pip install cffi > > (venv)$ pip install git+git://github.com/gevent-on-pypy/pypycore > > (venv)$ export GEVENT_LOOP=pypycore.loop > > > > But I am not able to translate this into an installation with easy_install > for my environment (without venv) > > > > Can someone give me a tip for getting started? > > > > Thanks > > Robert > > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > From r.voigtlaender at gmail.com Thu Dec 19 15:56:22 2013 From: r.voigtlaender at gmail.com (=?ISO-8859-1?Q?Robert_Voigtl=E4nder?=) Date: Thu, 19 Dec 2013 15:56:22 +0100 Subject: [pypy-dev] AttributeError: 'socket' object has no attribute '_reuse' Message-ID: Thanks to Piotr I now got gevent installed. But now I get a new error when trying to run my python program. The Webserver (with SSE) runs but when I try to load a site I get the error below. I need to get it running .. pypy is so awesomely faster than cpython. Robert Traceback (most recent call last): File "build/bdist.linux-armv6l/egg/gevent/greenlet.py", line 328, in run result = self._run(*self.args, **self.kwargs) File "build/bdist.linux-armv6l/egg/gevent/pywsgi.py", line 653, in handle handler = self.handler_class(socket, address, self) File "build/bdist.linux-armv6l/egg/gevent/pywsgi.py", line 175, in __init__ self.rfile = socket.makefile('rb', -1) File "build/bdist.linux-armv6l/egg/gevent/socket.py", line 379, in makefile return _fileobject(type(self)(_sock=self), mode, bufsize) File "/usr/lib/pypy-upstream/lib-python/2.7/socket.py", line 300, in __init__ sock._reuse() AttributeError: 'socket' object has no attribute '_reuse' >( failed with AttributeError -------------- next part -------------- An HTML attachment was scrubbed... URL: From piotr.skamruk at gmail.com Thu Dec 19 16:59:23 2013 From: piotr.skamruk at gmail.com (Piotr Skamruk) Date: Thu, 19 Dec 2013 16:59:23 +0100 Subject: [pypy-dev] AttributeError: 'socket' object has no attribute '_reuse' In-Reply-To: References: Message-ID: It's because lib-python/2.7/socket.py has reference to _reuse (around 185 line - look at Antonios comment in https://mail.python.org/pipermail/pypy-commit/2013-August/076563.html ), but this method isn't defined in rpython/rlib/rsocket.py You should fill issue in https://bugs.pypy.org Probably it will be connected in some way with https://bugs.pypy.org/issue1568 2013/12/19 Robert Voigtl?nder : > Thanks to Piotr I now got gevent installed. > But now I get a new error when trying to run my python program. The > Webserver (with SSE) runs but when I try to load a site I get the error > below. > > I need to get it running .. pypy is so awesomely faster than cpython. > > > Robert > > > Traceback (most recent call last): > File "build/bdist.linux-armv6l/egg/gevent/greenlet.py", line 328, in run > result = self._run(*self.args, **self.kwargs) > File "build/bdist.linux-armv6l/egg/gevent/pywsgi.py", line 653, in handle > handler = self.handler_class(socket, address, self) > File "build/bdist.linux-armv6l/egg/gevent/pywsgi.py", line 175, in > __init__ > self.rfile = socket.makefile('rb', -1) > File "build/bdist.linux-armv6l/egg/gevent/socket.py", line 379, in > makefile > return _fileobject(type(self)(_sock=self), mode, bufsize) > File "/usr/lib/pypy-upstream/lib-python/2.7/socket.py", line 300, in > __init__ > sock._reuse() > AttributeError: 'socket' object has no attribute '_reuse' > 0x465d630L fileno=6 address=0.0.0.0:9999>>( sock=192.168.1.199:, ('84.63.102.205', 52661))> failed with AttributeError > > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > From alex.gaynor at gmail.com Thu Dec 19 17:01:11 2013 From: alex.gaynor at gmail.com (Alex Gaynor) Date: Thu, 19 Dec 2013 08:01:11 -0800 Subject: [pypy-dev] AttributeError: 'socket' object has no attribute '_reuse' In-Reply-To: References: Message-ID: No, this isn't a bug in PyPy. If gevent wants to use the internal details of the socket module in way's that aren't defined, they need to pass somethign which matches the required interface. Alex On Thu, Dec 19, 2013 at 7:59 AM, Piotr Skamruk wrote: > It's because lib-python/2.7/socket.py has reference to _reuse (around > 185 line - look at Antonios comment in > https://mail.python.org/pipermail/pypy-commit/2013-August/076563.html > ), but this method isn't defined in rpython/rlib/rsocket.py > > You should fill issue in https://bugs.pypy.org > Probably it will be connected in some way with > https://bugs.pypy.org/issue1568 > > 2013/12/19 Robert Voigtl?nder : > > Thanks to Piotr I now got gevent installed. > > But now I get a new error when trying to run my python program. The > > Webserver (with SSE) runs but when I try to load a site I get the error > > below. > > > > I need to get it running .. pypy is so awesomely faster than cpython. > > > > > > Robert > > > > > > Traceback (most recent call last): > > File "build/bdist.linux-armv6l/egg/gevent/greenlet.py", line 328, in > run > > result = self._run(*self.args, **self.kwargs) > > File "build/bdist.linux-armv6l/egg/gevent/pywsgi.py", line 653, in > handle > > handler = self.handler_class(socket, address, self) > > File "build/bdist.linux-armv6l/egg/gevent/pywsgi.py", line 175, in > > __init__ > > self.rfile = socket.makefile('rb', -1) > > File "build/bdist.linux-armv6l/egg/gevent/socket.py", line 379, in > > makefile > > return _fileobject(type(self)(_sock=self), mode, bufsize) > > File "/usr/lib/pypy-upstream/lib-python/2.7/socket.py", line 300, in > > __init__ > > sock._reuse() > > AttributeError: 'socket' object has no attribute '_reuse' > > at > > 0x465d630L fileno=6 address=0.0.0.0:9999>>( fileno=8 > > sock=192.168.1.199:, ('84.63.102.205', 52661))> failed with > AttributeError > > > > > > _______________________________________________ > > pypy-dev mailing list > > pypy-dev at python.org > > https://mail.python.org/mailman/listinfo/pypy-dev > > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > -- "I disapprove of what you say, but I will defend to the death your right to say it." -- Evelyn Beatrice Hall (summarizing Voltaire) "The people's good is the highest law." -- Cicero GPG Key fingerprint: 125F 5C67 DFE9 4084 -------------- next part -------------- An HTML attachment was scrubbed... URL: From piotr.skamruk at gmail.com Thu Dec 19 17:14:15 2013 From: piotr.skamruk at gmail.com (Piotr Skamruk) Date: Thu, 19 Dec 2013 17:14:15 +0100 Subject: [pypy-dev] AttributeError: 'socket' object has no attribute '_reuse' In-Reply-To: <52B3197E.2040506@gmail.com> References: <52B3197E.2040506@gmail.com> Message-ID: Thanx for correction ;) From anto.cuni at gmail.com Thu Dec 19 17:06:22 2013 From: anto.cuni at gmail.com (Antonio Cuni) Date: Thu, 19 Dec 2013 17:06:22 +0100 Subject: [pypy-dev] AttributeError: 'socket' object has no attribute '_reuse' In-Reply-To: References: Message-ID: <52B3197E.2040506@gmail.com> On 19/12/13 17:01, Alex Gaynor wrote: > No, this isn't a bug in PyPy. If gevent wants to use the internal details of > the socket module in way's that aren't defined, they need to pass somethign > which matches the required interface. it's worth noting that eventlet had the same issue, and it was fixed here (thanks to... Alex :)): https://github.com/eventlet/eventlet/commit/2633322d6581beacd39d832284f17d461eb25098 From r.voigtlaender at gmail.com Thu Dec 19 19:31:24 2013 From: r.voigtlaender at gmail.com (=?ISO-8859-1?Q?Robert_Voigtl=E4nder?=) Date: Thu, 19 Dec 2013 19:31:24 +0100 Subject: [pypy-dev] AttributeError: 'socket' object has no attribute '_reuse' In-Reply-To: <52B3197E.2040506@gmail.com> References: <52B3197E.2040506@gmail.com> Message-ID: Thanks for the feedback. I would like to file an issue for this. Which would be the correct repository for this? https://github.com/schmir/gevent/tree/pypy-hacks or https://github.com/gevent-on-pypy/pypycore ? Robert On 19 December 2013 17:06, Antonio Cuni wrote: > On 19/12/13 17:01, Alex Gaynor wrote: > >> No, this isn't a bug in PyPy. If gevent wants to use the internal details >> of >> the socket module in way's that aren't defined, they need to pass >> somethign >> which matches the required interface. >> > > it's worth noting that eventlet had the same issue, and it was fixed here > (thanks to... Alex :)): > > https://github.com/eventlet/eventlet/commit/2633322d6581beacd39d832284f17d > 461eb25098 > > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev > -------------- next part -------------- An HTML attachment was scrubbed... URL: From matti.picus at gmail.com Mon Dec 23 00:15:32 2013 From: matti.picus at gmail.com (Matti Picus) Date: Mon, 23 Dec 2013 01:15:32 +0200 Subject: [pypy-dev] numpy fails trigonometry with complex numbers, what to do? In-Reply-To: References: <5046AB18.8020202@gmail.com> Message-ID: <52B77294.7030007@gmail.com> After more than a year, I am trying to engage with numpy about their non-compliant C99 complex math routines. FWIW, the trigger for this was that we still see failures on our micronumpy -A tests with numpy 1.8. Matti On 09/06/2012 10:12 AM, Armin Rigo wrote: > Hi, > > On Wed, Sep 5, 2012 at 3:30 AM, Matti Picus wrote: >>>>> numpy.arccos(complex(0.,-0.)) >> (1.5707963267948966-0j) >>>>> cmath.acos(complex(0.,-0.)) >> (1.5707963267948966+0j) >>>>> cmath.acos(complex(float('inf'),2.3)) >> -infj >>>>> numpy.arccos(complex(float('inf'),2.3)) >> (0.78539816339744828-inf*j) > According to the C99 standard Annex G (draft, > http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1124.pdf), the cmath > answer is the correct one in both cases. I don't know if that really > means that numpy didn't care about the details. It sounds a bit > strange given that it has tests for it; I fear it rather means that > numpy implemented a different standard. But maybe that's me being too > optimistic/pessimistic (depending on the point of view). I would > indeed ask on numpy mailing lists or submit a bug entry and see their > reaction. > > > A bient?t, > > Armin. From prasadjoshi124 at gmail.com Tue Dec 24 07:41:41 2013 From: prasadjoshi124 at gmail.com (Prasad Joshi) Date: Tue, 24 Dec 2013 12:11:41 +0530 Subject: [pypy-dev] Interested in GSoC 2014 Message-ID: Hello All, I am interested in participating GSoC 2014. I have went through last year's eligibility criterion, I think I am qualified to participate. I know GSoC 2014 still has more than 2-3 months, however I would like to start early. Please let me know suggestions. After going through GSoC 2013 page (https://wiki.python.org/moin/SummerOfCode/2013), I think I would be interested in Core Python or PyPy. Please let me know how should I start. Thanks a lot for your inputs. Thanks and Regards, Prasad From mail2shine at qq.com Tue Dec 24 09:15:52 2013 From: mail2shine at qq.com (=?ISO-8859-1?B?S2FTaGluaW5n?=) Date: Tue, 24 Dec 2013 16:15:52 +0800 Subject: [pypy-dev] why logging in pypy not need acquire lock? Message-ID: why logging in pypy not need acquire lock? left from pypy whle right from cpython -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 56A93EDB at 4F4D555A.B842B952.jpg Type: image/jpeg Size: 266338 bytes Desc: not available URL: From mail2shine at qq.com Tue Dec 24 10:00:13 2013 From: mail2shine at qq.com (=?ISO-8859-1?B?S2FTaGluaW5n?=) Date: Tue, 24 Dec 2013 17:00:13 +0800 Subject: [pypy-dev] why logging in pypy not need acquire lock? In-Reply-To: References: Message-ID: I take this question because I use the mlogging under pypy have something wrong like that: [10458] 4127 eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee [10464] 426 eeeeeeeeeeeeeeeeee[10461] 1670 iiiiiiiiiiiiiiii the right log should be: [10458] 4127 eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee [10464] 426 eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee [10461] 1670 iiiiiiiiiiiiiiii about mlogging https://pypi.python.org/pypi/mlogging ------------------ Original ------------------ From: "KaShining";; Date: Tue, Dec 24, 2013 04:15 PM To: "pypy-dev"; Subject: why logging in pypy not need acquire lock? why logging in pypy not need acquire lock? left from pypy whle right from cpython -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 105090A3 at 95DAA86C.1D4DB952.jpg Type: image/jpeg Size: 266338 bytes Desc: not available URL: From mail2shine at qq.com Tue Dec 24 10:35:46 2013 From: mail2shine at qq.com (=?ISO-8859-1?B?S2FTaGluaW5n?=) Date: Tue, 24 Dec 2013 17:35:46 +0800 Subject: [pypy-dev] why logging in pypy not need acquire lock? In-Reply-To: References: Message-ID: Please skip this question. I had found the reason! It's the mlogging bug! ------------------ Original ------------------ From: "KaShining";; Date: Tue, Dec 24, 2013 05:00 PM To: "pypy-dev"; Subject: Re:why logging in pypy not need acquire lock? I take this question because I use the mlogging under pypy have something wrong like that: [10458] 4127 eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee [10464] 426 eeeeeeeeeeeeeeeeee[10461] 1670 iiiiiiiiiiiiiiii the right log should be: [10458] 4127 eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee [10464] 426 eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee [10461] 1670 iiiiiiiiiiiiiiii about mlogging https://pypi.python.org/pypi/mlogging ------------------ Original ------------------ From: "KaShining";; Date: Tue, Dec 24, 2013 04:15 PM To: "pypy-dev"; Subject: why logging in pypy not need acquire lock? why logging in pypy not need acquire lock? left from pypy whle right from cpython -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: 6D8CA44F at CF570723.7255B952.jpg Type: image/jpeg Size: 266338 bytes Desc: not available URL: From arigo at tunes.org Tue Dec 24 11:08:55 2013 From: arigo at tunes.org (Armin Rigo) Date: Tue, 24 Dec 2013 11:08:55 +0100 Subject: [pypy-dev] why logging in pypy not need acquire lock? In-Reply-To: References: Message-ID: Hi KaShining, On Tue, Dec 24, 2013 at 9:15 AM, KaShining wrote: > why logging in pypy not need acquire lock? > left from pypy whle right from cpython PyPy's stdlib is from CPython 2.7.3, where it doesn't acquire a lock either. It seems it was some bug that was discovered and fixed in a more recent 2.7.x. We'll get it too when we upgrade the CPython stdlib version in PyPy. A bient?t, Armin. From arigo at tunes.org Tue Dec 24 11:13:25 2013 From: arigo at tunes.org (Armin Rigo) Date: Tue, 24 Dec 2013 11:13:25 +0100 Subject: [pypy-dev] Interested in GSoC 2014 In-Reply-To: References: Message-ID: Hi Prasad, On Tue, Dec 24, 2013 at 7:41 AM, Prasad Joshi wrote: > After going through GSoC 2013 page > (https://wiki.python.org/moin/SummerOfCode/2013), I think I would be > interested in Core Python or PyPy. Please let me know how should I > start. PyPy is most active on IRC (although right now it's Christmas, so maybe less than usual). You should come around on #pypy on irc.freenode.net and ask. A bient?t, Armin. From johan.rade at gmail.com Tue Dec 24 10:23:56 2013 From: johan.rade at gmail.com (=?ISO-8859-1?Q?Johan_R=E5de?=) Date: Tue, 24 Dec 2013 10:23:56 +0100 Subject: [pypy-dev] Leysin Message-ID: Hi, I'm a software developer from Sweden. I would like to attend the Leysin Winter sprint. How do I register and reserve accommodation? Best regards, Johan R?de From mail2shine at qq.com Tue Dec 24 13:49:55 2013 From: mail2shine at qq.com (=?utf-8?B?S2FTaGluaW5n?=) Date: Tue, 24 Dec 2013 20:49:55 +0800 Subject: [pypy-dev] why logging in pypy not need acquire lock? Message-ID: ok ------------------ ???? ------------------ ???: "Armin Rigo" ; ????: 2013?12?24?(???) 18:08 ???: "KaShining" ; ??: "pypy-dev" ; ??: Re: [pypy-dev] why logging in pypy not need acquire lock? Hi KaShining, On Tue, Dec 24, 2013 at 9:15 AM, KaShining wrote: > why logging in pypy not need acquire lock? > left from pypy whle right from cpython PyPy's stdlib is from CPython 2.7.3, where it doesn't acquire a lock either. It seems it was some bug that was discovered and fixed in a more recent 2.7.x. We'll get it too when we upgrade the CPython stdlib version in PyPy. A bient??t, Armin. -------------- next part -------------- An HTML attachment was scrubbed... URL: From romain.py at gmail.com Tue Dec 24 15:59:07 2013 From: romain.py at gmail.com (Romain Guillebert) Date: Tue, 24 Dec 2013 15:59:07 +0100 Subject: [pypy-dev] Leysin In-Reply-To: References: Message-ID: <20131224145907.GA11633@Plop> Hi Johan You can register there : https://bitbucket.org/pypy/extradoc/src/d0ccfcf501f92072e856faf1fa1b0de06a54d703/sprintinfo/leysin-winter-2014/people.txt?at=extradoc (send a pull request), I guess Armin will take care of the booking if you choose Ermina, otherwise, you will have to book your accommodation yourself. Cheers Romain On 12/24, Johan R?de wrote: > Hi, > > I'm a software developer from Sweden. > I would like to attend the Leysin Winter sprint. > How do I register and reserve accommodation? > > Best regards, > Johan R?de > > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev From jashan.newton at gmail.com Wed Dec 25 09:20:05 2013 From: jashan.newton at gmail.com (Jashan Goyal) Date: Wed, 25 Dec 2013 13:50:05 +0530 Subject: [pypy-dev] Kindly guide Message-ID: hi, I am jashan goyal.I am interested in contributing to pypy please guide me how to begin. Thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: From fijall at gmail.com Wed Dec 25 09:57:31 2013 From: fijall at gmail.com (Maciej Fijalkowski) Date: Wed, 25 Dec 2013 10:57:31 +0200 Subject: [pypy-dev] [Python-Dev] Interested in GSoC 2014 In-Reply-To: References: Message-ID: On Tue, Dec 24, 2013 at 8:41 AM, Prasad Joshi wrote: > Hello All, > > I am interested in participating GSoC 2014. I have went through last > year's eligibility criterion, I think I am qualified to participate. I > know GSoC 2014 still has more than 2-3 months, however I would like to > start early. Please let me know suggestions. > > After going through GSoC 2013 page > (https://wiki.python.org/moin/SummerOfCode/2013), I think I would be > interested in Core Python or PyPy. Please let me know how should I > start. > > Thanks a lot for your inputs. > > Thanks and Regards, > Prasad > _______________________________________________ > Python-Dev mailing list > Python-Dev at python.org > https://mail.python.org/mailman/listinfo/python-dev > Unsubscribe: https://mail.python.org/mailman/options/python-dev/fijall%40gmail.com Hi Prasad. I've removed the python-dev from this thread, please don't cross-post messages like that. If you're interested in PyPy, a good start would be to tell you what more you're interested in. A list of potential medium-sized projects is available here: http://doc.pypy.org/en/latest/project-ideas.html To answer specific questions, we're very much IRC based. #pypy on freenode Cheers, fijal From arigo at tunes.org Wed Dec 25 18:54:43 2013 From: arigo at tunes.org (Armin Rigo) Date: Wed, 25 Dec 2013 18:54:43 +0100 Subject: [pypy-dev] Leysin In-Reply-To: <20131224145907.GA11633@Plop> References: <20131224145907.GA11633@Plop> Message-ID: Hi Johan! On Tue, Dec 24, 2013 at 3:59 PM, Romain Guillebert wrote: >> I'm a software developer from Sweden. >> I would like to attend the Leysin Winter sprint. Sure, you are welcome! And great: another person --- the sprint was looking thin so far :-) If you didn't do it so far, you should show up on IRC and say hi, and maybe discuss a bit what topic you're interested in more precisely, if any. It's on #pypy on irc.freenode.net. >> How do I register and reserve accommodation? > > You can register there : > https://bitbucket.org/pypy/extradoc/src/(...) We can also enter your name directly here :-) Just tell us your precise dates. I am arranging the group's booking, in rooms of 2 or 3 people (if this suits you --- otherwise you can also get your own room). A bient?t, Armin. From arigo at tunes.org Thu Dec 26 22:20:52 2013 From: arigo at tunes.org (Armin Rigo) Date: Thu, 26 Dec 2013 22:20:52 +0100 Subject: [pypy-dev] Kindly guide In-Reply-To: References: Message-ID: Hi Jashan, On Wed, Dec 25, 2013 at 9:20 AM, Jashan Goyal wrote: > I am jashan goyal.I am interested in contributing to pypy please guide me > how to begin. Have a look at our getting-started guides: http://doc.pypy.org/en/latest/getting-started.html http://doc.pypy.org/en/latest/getting-started-python.html http://doc.pypy.org/en/latest/getting-started-dev.html Then show up on IRC, in #pypy at irc.freenode.net. This is the best way to find some of us and to explain what precisely you're interested in more precisely (PyPy is a big project with many different aspects). A bient?t, Armin. From johan.rade at gmail.com Sat Dec 28 11:41:00 2013 From: johan.rade at gmail.com (=?ISO-8859-1?Q?Johan_R=E5de?=) Date: Sat, 28 Dec 2013 11:41:00 +0100 Subject: [pypy-dev] Leysin In-Reply-To: References: <20131224145907.GA11633@Plop> Message-ID: <52BEAABC.2050704@gmail.com> Hi everyone, Let me just quickly introduce myself. I did a PhD in mathematics at University of Texas and a postdoc at Stanford University. After that I have done academic research in mathematics and commercial software development, mainly in C++. I am one of the founders of and former CTO at Qlucore, a bioinformatics software company. My main interest is analysis and visualization of large scientific data sets. The best tool I know for that purpose is CPython combined with C++. I think PyPy has the potential to become an even more powerful tool. I plan to attend the Leysin Sprint. Here are two suggestions what I might work on: * Help with the effort to port NumPy to PyPy * See if it feasible to port Qt (PySide) to PyPy using CFFI. See you in Leysin, Johan On 2013-12-25 18:54, Armin Rigo wrote: > Hi Johan! > > On Tue, Dec 24, 2013 at 3:59 PM, Romain Guillebert wrote: >>> I'm a software developer from Sweden. >>> I would like to attend the Leysin Winter sprint. > > Sure, you are welcome! And great: another person --- the sprint was > looking thin so far :-) If you didn't do it so far, you should show > up on IRC and say hi, and maybe discuss a bit what topic you're > interested in more precisely, if any. It's on #pypy on > irc.freenode.net. From johan.rade at gmail.com Sat Dec 28 11:41:00 2013 From: johan.rade at gmail.com (=?ISO-8859-1?Q?Johan_R=E5de?=) Date: Sat, 28 Dec 2013 11:41:00 +0100 Subject: [pypy-dev] Leysin In-Reply-To: References: <20131224145907.GA11633@Plop> Message-ID: <52BEAABC.2050704@gmail.com> Hi everyone, Let me just quickly introduce myself. I did a PhD in mathematics at University of Texas and a postdoc at Stanford University. After that I have done academic research in mathematics and commercial software development, mainly in C++. I am one of the founders of and former CTO at Qlucore, a bioinformatics software company. My main interest is analysis and visualization of large scientific data sets. The best tool I know for that purpose is CPython combined with C++. I think PyPy has the potential to become an even more powerful tool. I plan to attend the Leysin Sprint. Here are two suggestions what I might work on: * Help with the effort to port NumPy to PyPy * See if it feasible to port Qt (PySide) to PyPy using CFFI. See you in Leysin, Johan On 2013-12-25 18:54, Armin Rigo wrote: > Hi Johan! > > On Tue, Dec 24, 2013 at 3:59 PM, Romain Guillebert wrote: >>> I'm a software developer from Sweden. >>> I would like to attend the Leysin Winter sprint. > > Sure, you are welcome! And great: another person --- the sprint was > looking thin so far :-) If you didn't do it so far, you should show > up on IRC and say hi, and maybe discuss a bit what topic you're > interested in more precisely, if any. It's on #pypy on > irc.freenode.net. From matti.picus at gmail.com Sat Dec 28 20:39:08 2013 From: matti.picus at gmail.com (Matti Picus) Date: Sat, 28 Dec 2013 21:39:08 +0200 Subject: [pypy-dev] Leysin In-Reply-To: <52BEAABC.2050704@gmail.com> References: <20131224145907.GA11633@Plop> <52BEAABC.2050704@gmail.com> Message-ID: <52BF28DC.3050805@gmail.com> It would be nice to get a modern gui widget set working with pypy. Personally, I prefer a more liberal license, which leads to wx rather than qt, but I imagine, given the name of your company, that you do not share my preference :). In any case, you might want to take a look at the wxPython-cffi Google Summer of Code project for the amount of work involved in porting a code generator to cffi. WxPython uses sip, pyside uses shiboken. blog posts http://waedt.blogspot.co.il/ code https://bitbucket.org/waedt/wxpython_cffi Matti On 12/28/2013 12:41 PM, Johan R?de wrote: > Hi everyone, > > Let me just quickly introduce myself. > > I did a PhD in mathematics at University of Texas > and a postdoc at Stanford University. > After that I have done academic research in mathematics > and commercial software development, mainly in C++. > I am one of the founders of and former CTO at Qlucore, > a bioinformatics software company. > > My main interest is analysis and visualization of large scientific > data sets. > The best tool I know for that purpose is CPython combined with C++. > I think PyPy has the potential to become an even more powerful tool. > > I plan to attend the Leysin Sprint. > Here are two suggestions what I might work on: > * Help with the effort to port NumPy to PyPy > * See if it feasible to port Qt (PySide) to PyPy using CFFI. > > See you in Leysin, > Johan > > On 2013-12-25 18:54, Armin Rigo wrote: >> Hi Johan! >> >> On Tue, Dec 24, 2013 at 3:59 PM, Romain Guillebert >> wrote: >>>> I'm a software developer from Sweden. >>>> I would like to attend the Leysin Winter sprint. >> >> Sure, you are welcome! And great: another person --- the sprint was >> looking thin so far :-) If you didn't do it so far, you should show >> up on IRC and say hi, and maybe discuss a bit what topic you're >> interested in more precisely, if any. It's on #pypy on >> irc.freenode.net. > > > > > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev From prasadjoshi124 at gmail.com Sun Dec 29 07:55:54 2013 From: prasadjoshi124 at gmail.com (Prasad Joshi) Date: Sun, 29 Dec 2013 12:25:54 +0530 Subject: [pypy-dev] [refactor-str-types] Translating the PyPy Python interpreter failed. Message-ID: prasad at pjoshi:~/python/pypy/pypy/goal$ python ../../rpython/bin/rpython --opt=jit targetpypystandalone.py ... ... ... ... [Timer] Timings: [Timer] annotate --- 138.9 s [Timer] ========================================== [Timer] Total: --- 138.9 s [translation:info] Error: [translation:info] File "/home/prasad/python/pypy/rpython/translator/goal/translate.py", line 318, in main [translation:info] drv.proceed(goals) [translation:info] File "/home/prasad/python/pypy/rpython/translator/driver.py", line 534, in proceed [translation:info] return self._execute(goals, task_skip = self._maybe_skip()) [translation:info] File "/home/prasad/python/pypy/rpython/translator/tool/taskengine.py", line 114, in _execute [translation:info] res = self._do(goal, taskcallable, *args, **kwds) [translation:info] File "/home/prasad/python/pypy/rpython/translator/driver.py", line 283, in _do [translation:info] res = func() [translation:info] File "/home/prasad/python/pypy/rpython/translator/driver.py", line 320, in task_annotate [translation:info] s = annotator.build_types(self.entry_point, self.inputtypes) [translation:info] File "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 91, in build_types [translation:info] return self.build_graph_types(flowgraph, inputcells, complete_now=complete_now) [translation:info] File "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 145, in build_graph_types [translation:info] self.complete() [translation:info] File "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 199, in complete [translation:info] self.complete_pending_blocks() [translation:info] File "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 194, in complete_pending_blocks [translation:info] self.processblock(graph, block) [translation:info] File "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 344, in processblock [translation:info] self.flowin(graph, block) [translation:info] File "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 408, in flowin [translation:info] self.consider_op(block, i) [translation:info] File "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 602, in consider_op [translation:info] resultcell = consider_meth(*argcells) [translation:info] File "<3933-codegen /home/prasad/python/pypy/rpython/annotator/annrpython.py:641>", line 3, in consider_op_getattr [translation:info] return arg.getattr(*args) [translation:info] File "/home/prasad/python/pypy/rpython/annotator/unaryop.py", line 144, in getattr [translation:info] raise AnnotatorError("Cannot find attribute %r on %r" % (attr, obj)) [translation:ERROR] AnnotatorError: [translation:ERROR] [translation:ERROR] Cannot find attribute 'isspace' on SomeUnicodeCodePoint() [translation:ERROR] [translation:ERROR] [translation:ERROR] v324 = getattr(v323, ('isspace')) [translation:ERROR] [translation:ERROR] In : [translation:ERROR] Happened at file /home/prasad/python/pypy/rpython/rlib/rstring.py line 98 [translation:ERROR] [translation:ERROR] ==> if not value[i].isspace(): [translation:ERROR] [translation:ERROR] Known variable annotations: [translation:ERROR] v323 = SomeUnicodeCodePoint() [translation:ERROR] [translation:ERROR] Processing block: [translation:ERROR] block at 64 is a [translation:ERROR] in (rpython.rlib.rstring:89)rsplit__unicode [translation:ERROR] containing the following operations: [translation:ERROR] v323 = getitem(value_0, i_0) [translation:ERROR] v324 = getattr(v323, ('isspace')) [translation:ERROR] v325 = simple_call(v324) [translation:ERROR] v326 = bool(v325) [translation:ERROR] --end-- [translation] start debugger... From fijall at gmail.com Sun Dec 29 09:08:26 2013 From: fijall at gmail.com (Maciej Fijalkowski) Date: Sun, 29 Dec 2013 10:08:26 +0200 Subject: [pypy-dev] [refactor-str-types] Translating the PyPy Python interpreter failed. In-Reply-To: References: Message-ID: merge the default on this branch On Sun, Dec 29, 2013 at 8:55 AM, Prasad Joshi wrote: > prasad at pjoshi:~/python/pypy/pypy/goal$ python > ../../rpython/bin/rpython --opt=jit targetpypystandalone.py > ... > ... > ... > ... > [Timer] Timings: > [Timer] annotate --- 138.9 s > [Timer] ========================================== > [Timer] Total: --- 138.9 s > [translation:info] Error: > [translation:info] File > "/home/prasad/python/pypy/rpython/translator/goal/translate.py", line > 318, in main > [translation:info] drv.proceed(goals) > [translation:info] File > "/home/prasad/python/pypy/rpython/translator/driver.py", line 534, in > proceed > [translation:info] return self._execute(goals, task_skip = > self._maybe_skip()) > [translation:info] File > "/home/prasad/python/pypy/rpython/translator/tool/taskengine.py", line > 114, in _execute > [translation:info] res = self._do(goal, taskcallable, *args, **kwds) > [translation:info] File > "/home/prasad/python/pypy/rpython/translator/driver.py", line 283, in > _do > [translation:info] res = func() > [translation:info] File > "/home/prasad/python/pypy/rpython/translator/driver.py", line 320, in > task_annotate > [translation:info] s = annotator.build_types(self.entry_point, > self.inputtypes) > [translation:info] File > "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 91, > in build_types > [translation:info] return self.build_graph_types(flowgraph, > inputcells, complete_now=complete_now) > [translation:info] File > "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 145, > in build_graph_types > [translation:info] self.complete() > [translation:info] File > "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 199, > in complete > [translation:info] self.complete_pending_blocks() > [translation:info] File > "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 194, > in complete_pending_blocks > [translation:info] self.processblock(graph, block) > [translation:info] File > "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 344, > in processblock > [translation:info] self.flowin(graph, block) > [translation:info] File > "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 408, > in flowin > [translation:info] self.consider_op(block, i) > [translation:info] File > "/home/prasad/python/pypy/rpython/annotator/annrpython.py", line 602, > in consider_op > [translation:info] resultcell = consider_meth(*argcells) > [translation:info] File "<3933-codegen > /home/prasad/python/pypy/rpython/annotator/annrpython.py:641>", line > 3, in consider_op_getattr > [translation:info] return arg.getattr(*args) > [translation:info] File > "/home/prasad/python/pypy/rpython/annotator/unaryop.py", line 144, in > getattr > [translation:info] raise AnnotatorError("Cannot find attribute %r > on %r" % (attr, obj)) > [translation:ERROR] AnnotatorError: > [translation:ERROR] > [translation:ERROR] Cannot find attribute 'isspace' on SomeUnicodeCodePoint() > [translation:ERROR] > [translation:ERROR] > [translation:ERROR] v324 = getattr(v323, ('isspace')) > [translation:ERROR] > [translation:ERROR] In (rpython.rlib.rstring:89)rsplit__unicode at 0x1e5d0bd0>: > [translation:ERROR] Happened at file > /home/prasad/python/pypy/rpython/rlib/rstring.py line 98 > [translation:ERROR] > [translation:ERROR] ==> if not value[i].isspace(): > [translation:ERROR] > [translation:ERROR] Known variable annotations: > [translation:ERROR] v323 = SomeUnicodeCodePoint() > [translation:ERROR] > [translation:ERROR] Processing block: > [translation:ERROR] block at 64 is a 'rpython.flowspace.flowcontext.SpamBlock'> > [translation:ERROR] in (rpython.rlib.rstring:89)rsplit__unicode > [translation:ERROR] containing the following operations: > [translation:ERROR] v323 = getitem(value_0, i_0) > [translation:ERROR] v324 = getattr(v323, ('isspace')) > [translation:ERROR] v325 = simple_call(v324) > [translation:ERROR] v326 = bool(v325) > [translation:ERROR] --end-- > [translation] start debugger... > _______________________________________________ > pypy-dev mailing list > pypy-dev at python.org > https://mail.python.org/mailman/listinfo/pypy-dev From johan.rade at gmail.com Sun Dec 29 09:38:26 2013 From: johan.rade at gmail.com (=?ISO-8859-1?Q?Johan_R=E5de?=) Date: Sun, 29 Dec 2013 09:38:26 +0100 Subject: [pypy-dev] Leysin In-Reply-To: <52BF28DC.3050805@gmail.com> References: <20131224145907.GA11633@Plop> <52BEAABC.2050704@gmail.com> <52BF28DC.3050805@gmail.com> Message-ID: <52BFDF82.1000906@gmail.com> Hi Matti, Thanks for the links to the GSoC project. They should be useful. I have never used wxWidgets. That is my only reason for choosing Qt. --Johan On 2013-12-28 20:39, Matti Picus wrote: > It would be nice to get a modern gui widget set working with pypy. > Personally, I prefer a more liberal license, which leads to wx rather > than qt, but I imagine, given the name of your company, that you do not > share my preference :). > In any case, you might want to take a look at the wxPython-cffi Google > Summer of Code project for the amount of work involved in porting a code > generator to cffi. WxPython uses sip, pyside uses shiboken. > blog posts http://waedt.blogspot.co.il/ > code https://bitbucket.org/waedt/wxpython_cffi > Matti > > On 12/28/2013 12:41 PM, Johan R?de wrote: >> I plan to attend the Leysin Sprint. >> Here are two suggestions what I might work on: >> * Help with the effort to port NumPy to PyPy >> * See if it feasible to port Qt (PySide) to PyPy using CFFI. From johan.rade at gmail.com Sun Dec 29 09:38:26 2013 From: johan.rade at gmail.com (=?ISO-8859-1?Q?Johan_R=E5de?=) Date: Sun, 29 Dec 2013 09:38:26 +0100 Subject: [pypy-dev] Leysin In-Reply-To: <52BF28DC.3050805@gmail.com> References: <20131224145907.GA11633@Plop> <52BEAABC.2050704@gmail.com> <52BF28DC.3050805@gmail.com> Message-ID: <52BFDF82.1000906@gmail.com> Hi Matti, Thanks for the links to the GSoC project. They should be useful. I have never used wxWidgets. That is my only reason for choosing Qt. --Johan On 2013-12-28 20:39, Matti Picus wrote: > It would be nice to get a modern gui widget set working with pypy. > Personally, I prefer a more liberal license, which leads to wx rather > than qt, but I imagine, given the name of your company, that you do not > share my preference :). > In any case, you might want to take a look at the wxPython-cffi Google > Summer of Code project for the amount of work involved in porting a code > generator to cffi. WxPython uses sip, pyside uses shiboken. > blog posts http://waedt.blogspot.co.il/ > code https://bitbucket.org/waedt/wxpython_cffi > Matti > > On 12/28/2013 12:41 PM, Johan R?de wrote: >> I plan to attend the Leysin Sprint. >> Here are two suggestions what I might work on: >> * Help with the effort to port NumPy to PyPy >> * See if it feasible to port Qt (PySide) to PyPy using CFFI. From matti.picus at gmail.com Mon Dec 30 20:42:14 2013 From: matti.picus at gmail.com (Matti Picus) Date: Mon, 30 Dec 2013 21:42:14 +0200 Subject: [pypy-dev] arm-v7 buildbot down for the next two weeks Message-ID: <52C1CC96.9090907@gmail.com> An HTML attachment was scrubbed... URL: