Mailman 3 June 2016 - pypy-dev

fix libpypy-c.so needing certein versions
by Katelyn McNair 22 Jun '16

22 Jun '16

I have been trying to install PyPy locally for the last 3 days. I didn't have Openssl, which install fine. But then PyPy started not liking the new version (libssl.so.1.0.1), so I got the archive for 1.0.0, and installed. But then PyPy didn't like that the .so files didn't have to correct version name in them. [katelyn@anthill bin]$ ./pypy ./pypy: /home/katelyn/lib/libssl.so.1.0.0: version `OPENSSL_1.0.1' not found (required by /home/katelyn/opt/pypy2-v5.3.0-linux64/bin/libpypy-c.so) ./pypy: /home/katelyn/lib/libssl.so.1.0.0: version `OPENSSL_1.0.0' not found (required by /home/katelyn/opt/pypy2-v5.3.0-linux64/bin/libpypy-c.so) ./pypy: /home/katelyn/lib/libcrypto.so.1.0.0: version `OPENSSL_1.0.0' not found (required by /home/katelyn/opt/pypy2-v5.3.0-linux64/bin/libpypy-c.so) ./pypy: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by /home/katelyn/opt/pypy2-v5.3.0-linux64/bin/libpypy-c.so) ./pypy: /lib64/libc.so.6: version `GLIBC_2.15' not found (required by /home/katelyn/opt/pypy2-v5.3.0-linux64/bin/libpypy-c.so) It took me a day to find the hack here: http://stackoverflow.com/questions/18390833/no-version-information-available Although I had to put both OPENSSL_1.0.0 and OPENSSL_1.0.1in the file. Now I have to go build my own libc.so since libpypy.so doesn't like the one I have. [katelyn@anthill openssl-1.0.0s]$ pypy /home3/katelyn/opt/pypy2-v5.3.0-linux64/bin/pypy: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by /home/katelyn/opt/pypy2-v5.3.0-linux64/bin/libpypy-c.so) /home3/katelyn/opt/pypy2-v5.3.0-linux64/bin/pypy: /lib64/libc.so.6: version `GLIBC_2.15' not found (required by /home/katelyn/opt/pypy2-v5.3.0-linux64/bin/libpypy-c.so) -- -katelyn

4 3

PyParallel-style threads
by Omer Katz 21 Jun '16

21 Jun '16

Hi all, There was an experiment based on CPython's code called PyParallel <https://github.com/pyparallel/pyparallel> that allows running threads in parallel without STM and modifying source code of both Python and C extensions. The only limitation is that they disallow mutation of global state in parallel context. I briefly mentioned it before on PyPy's freenode channel. I'd like to discuss why the approach is useful, how it can benefit PyPy users and how can it be implemented. Allowing to run in parallel without mutating global state can help servers use each thread to handle a request. It can also allow to log in parallel or send an HTTP request (or an AMQP message) without sharing the response with the main thread. This is useful in some cases and since PyParallel managed to keep the same semantics it (shouldn't) break CPyExt. If we keep to the following rules: 1. No global state mutation is allowed 2. No new keywords or code modifications required 3. No CPyExt code is allowed (for now) I believe that users can somewhat benefit from this implementation if done correctly. As for implementation, if we can trace the code running in the thread and ensure it's not mutating global state and that CPyExt is never used during the thread's course we can simply release the GIL when such a thread is run. That requires less knowledge than using STM and less code modifications. However I think that attempting to do so will introduce the same issue with caching traces (Armin am I correct here?). As for CPyExt, we could copy the same code modifications that PyParallels did but I suspect that it will be so slow that the benefit of running in parallel will be completely lost for all cases but very long threads. Is what I'm suggesting even possible? How challenging will it be? Thanks, Omer Katz.

4 11

bugfix PyPy 2.7 v5.3.1 ready for release
by Matti Picus 16 Jun '16

16 Jun '16

We fixed a few bugs and would like to release, please help us make sure everything is ready to go. The downloads are here https://bitbucket.org/pypy/pypy/downloads the donwload page has been updated http://pypy.org/download.html and the release notice is ready to be posted to the blog http://doc.pypy.org/en/latest/release-pypy2.7-v5.3.1.html http://doc.pypy.org/en/latest/whatsnew-pypy2-5.3.1.html Let me know if something needs fixing... Thanks Matti

1 0

PyPy2 v5.3.0 released, help us get the word out
by Matti Picus 09 Jun '16

09 Jun '16

5 7

W^X in the PyPy JIT?
by Edd Barrett 09 Jun '16

09 Jun '16

Hi! Recently OpenBSD has started enforcing W^X [1] for all programs, unless the program is on a file system with a special 'wxallowed' flag. Regardless of the flag, a message is emitted to the dmesg buffer. Starting PyPy will emit such a message: pypy(62301): mmap W^X violation I suppose the memory the JIT is using for traces has been mapped W+X? Assuming this is the case, the security of PyPy (on all platforms) could be improved by mapping the memory W during trace compilation, and then re-mapping the memory X once compilation is complete. Of course, it might not be the JIT at all... Cheers [1] https://en.wikipedia.org/wiki/W%5EX -- Best Regards Edd Barrett http://www.theunixzoo.co.uk

2 1

v5.3.0 numpy support (specifically numpy/core/src/multiarray/multiarray.c.src)
by Eli Stevens (Gmail) 09 Jun '16

09 Jun '16

I asked about the following about a month ago: ====================================================================== ERROR: Failure: ImportError (No module named numpy.core.multiarray_tests) ---------------------------------------------------------------------- Traceback (most recent call last): File "/home/elis/venv/ppdb/site-packages/nose/loader.py", line 418, in loadTestsFromName addr.filename, addr.module) File "/home/elis/venv/ppdb/site-packages/nose/importer.py", line 47, in importFromPath return self.importFromDir(dir_path, fqname) File "/home/elis/venv/ppdb/site-packages/nose/importer.py", line 94, in importFromDir mod = load_module(part_fqname, fh, filename, desc) File "/home/elis/venv/ppdb/site-packages/numpy/core/tests/test_multiarray.py", line 22, in <module> from numpy.core.multiarray_tests import ( ImportError: No module named numpy.core.multiarray_tests And was told by Matti: > multiarray_tests comes from numpy/core/src/multiarray/multiarray.c.src which is compiled to a C-API module. > We skip building it as we have quite a way to go before we can support that level of C-API compatibility. Is there any change in behavior expected with the new release regarding these kinds of errors? Thanks, Eli

2 1

PyPy2.7 v5.3 release candidates available for download
by Matti Picus 07 Jun '16

07 Jun '16

Please check the binaries and source packages available at https://bitbucket.org/pypy/pypy/downloads I uses a slightly different naming scheme to differentiate from pypy3.3, so the files are now https://bitbucket.org/pypy/pypy/downloads/pypy2-v5.3.0-* Matti

1 0

pypy3.3-v5.2-alpha1 release
by Matti Picus 05 Jun '16

05 Jun '16

We are almost ready to release pypy3.3-v5.2-alpha1, but need some help to make sure all is OK. Please try the download packages here https://bitbucket.org/pypy/pypy/downloads also, if someone could check that the links related to pypy3.3 all work on this page that would be great http://pypy.org/download.html If the page is still showing PyPy3 2.4.0, wait 30 minutes or so from now and then hit refresh on your browser. Thanks, and kudos to the pypy3 team. Matti

4 4