September 2016 Archives by subject

Messages sorted by: [ thread ] [ author ] [ date ]
More info on this list...

Starting: Thu Sep 1 01:49:35 EDT 2016
Ending: Fri Sep 30 20:33:22 EDT 2016
Messages: 729

[pypy-commit] cffi default: $ORIGIN is better than . arigo
[pypy-commit] cffi default: - add ffi.compile(debug=flag) arigo
[pypy-commit] cffi default: A direct test arigo
[pypy-commit] cffi default: Backed out changeset 0087e2aec9ef arigo
[pypy-commit] cffi default: Bump version number to 1.8.1 arigo
[pypy-commit] cffi default: bump version number to 1.8.3 arigo
[pypy-commit] cffi default: Docs: remove "char *" from the special casing in the reference arigo
[pypy-commit] cffi default: Document 0ff89f851d2c arigo
[pypy-commit] cffi default: Document inline the various options to compile with embedding arigo
[pypy-commit] cffi default: Document the new warning arigo
[pypy-commit] cffi default: fix arigo
[pypy-commit] cffi default: Fix leak. Remaining issue: if the user close()s the file object in Python, arigo
[pypy-commit] cffi default: Fix test arigo
[pypy-commit] cffi default: hg merge release-1.8 arigo
[pypy-commit] cffi default: hg merge release-1.8 arigo
[pypy-commit] cffi default: If we say Py_LIMITED_API and we're compiling with a debug version of arigo
[pypy-commit] cffi default: Issue #282: probable test and fix arigo
[pypy-commit] cffi default: Issue #283: initializer for nested anonymous structs inside unions arigo
[pypy-commit] cffi default: Kill the ctypes backend. arigo
[pypy-commit] cffi default: Python 3 fix arigo
[pypy-commit] cffi default: skip test arigo
[pypy-commit] cffi default: Skip test if setuptools cannot be imported arigo
[pypy-commit] cffi default: typedef int foo_t[...]; arigo
[pypy-commit] cffi default: Un-document the removal of the ctypes backend arigo
[pypy-commit] cffi default: Update and document v1.8.2 arigo
[pypy-commit] cffi default: update to 1.8.4 and add whatsnew arigo
[pypy-commit] cffi default: Warn when implicitly casting between "char *" and a different pointer arigo
[pypy-commit] cffi default: Windows fun arigo
[pypy-commit] cffi release-1.8: hg merge default arigo
[pypy-commit] cffi release-1.8: md5/sha arigo
[pypy-commit] cffi release-1.8: md5/sha arigo
[pypy-commit] cffi release-1.8: Release branch, starting at 1.8.2 for CPython arigo
[pypy-commit] extradoc extradoc: "import stackless" fails arigo
[pypy-commit] extradoc extradoc: _utf8 DONE in bd9d1dd19c2c arigo
[pypy-commit] extradoc extradoc: Add a task arigo
[pypy-commit] extradoc extradoc: add new done section and move an entry there plan_rich
[pypy-commit] extradoc extradoc: Add references arigo
[pypy-commit] extradoc extradoc: a new job for me plan_rich
[pypy-commit] extradoc extradoc: assign me a new task plan_rich
[pypy-commit] extradoc extradoc: create new empty slide deck for my presentation in pyconza, rmeove test files for reveal.js plan_rich
[pypy-commit] extradoc extradoc: document some progress in the stdlib deadlocks plan_rich
[pypy-commit] extradoc extradoc: done arigo
[pypy-commit] extradoc extradoc: done unwrap_spec(bool) arigo
[pypy-commit] extradoc extradoc: more things to do arigo
[pypy-commit] extradoc extradoc: move compiler-workshop slides into scipyUS2016 dir plan_rich
[pypy-commit] extradoc extradoc: Next crasher arigo
[pypy-commit] extradoc extradoc: Remove DONE entries. Add #2310. arigo
[pypy-commit] extradoc extradoc: Tied this issue to a 3.5 issue arigo
[pypy-commit] extradoc extradoc: tweaks arigo
[pypy-commit] extradoc extradoc: update arigo
[pypy-commit] extradoc extradoc: update arigo
[pypy-commit] extradoc extradoc: updates arigo
[pypy-commit] extradoc extradoc: updates arigo
[pypy-commit] extradoc extradoc: updates arigo
[pypy-commit] pypy.org extradoc: Close the call for donation arigo
[pypy-commit] pypy.org extradoc: Close the STM donation button, as per the January 22 mail on pypy-z. arigo
[pypy-commit] pypy.org extradoc: Expand a bit the update arigo
[pypy-commit] pypy.org extradoc: md5/sha1/sha256 for pypy 5.4.1 arigo
[pypy-commit] pypy.org extradoc: Mention Mozilla at the start of this page arigo
[pypy-commit] pypy.org extradoc: note need to use upstream NumPy HEAD (issue #2403), remove whitepsace at ends of lines mattip
[pypy-commit] pypy.org extradoc: PyPy on FreeBSD is maintained externally on FreshPorts. Thanks DragonSA! arigo
[pypy-commit] pypy.org extradoc: Still link to tmdonate2 arigo
[pypy-commit] pypy.org extradoc: update for 5.4.1 mattip
[pypy-commit] pypy.org extradoc: update the checksums arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy.org extradoc: update the values arigo
[pypy-commit] pypy _py3k-cpyext-A: hack around to get -A tests to run rlamy
[pypy-commit] pypy _py3k-cpyext-A: hack some more: one test passes! rlamy
[pypy-commit] pypy _py3k-cpyext-A: Merge test-cpyext rlamy
[pypy-commit] pypy _warnings: add test that passes untranslated, fails w/pypy -A (TypeError: expected a readable buffer object) mattip
[pypy-commit] pypy _warnings: close branch to be merged mattip
[pypy-commit] pypy _warnings: copy cpython behaviour mattip
[pypy-commit] pypy _warnings: merge default into branch mattip
[pypy-commit] pypy asmmemmgr-for-code-only: Another W^X site. vext01
[pypy-commit] pypy asmmemmgr-for-code-only: Fix translation. vext01
[pypy-commit] pypy asmmemmgr-for-code-only: Mark an argument we send to a C-callback as writable. vext01
[pypy-commit] pypy better-storesink: actually super easy: support CSE on loop-invariant operations (SSA makes this cfbolz
[pypy-commit] pypy better-storesink: add some logging cfbolz
[pypy-commit] pypy better-storesink: add statistics printing after store sinking cfbolz
[pypy-commit] pypy better-storesink: a few corner cases cfbolz
[pypy-commit] pypy better-storesink: a passing test cfbolz
[pypy-commit] pypy better-storesink: a small tweak and two XXXs cfbolz
[pypy-commit] pypy better-storesink: be more careful about merging value, never make the results of find_rep the cfbolz
[pypy-commit] pypy better-storesink: better logging cfbolz
[pypy-commit] pypy better-storesink: compute the merged result more systematically cfbolz
[pypy-commit] pypy better-storesink: constant-fold reads from immutable fields of constants cfbolz
[pypy-commit] pypy better-storesink: don't optimize array fields of virtualizables cfbolz
[pypy-commit] pypy better-storesink: experiment with a more general common subexpression elimination. cfbolz
[pypy-commit] pypy better-storesink: factor out the rewriting to same_as, with an extra safety check cfbolz
[pypy-commit] pypy better-storesink: fix bug cfbolz
[pypy-commit] pypy better-storesink: fix immutable checks cfbolz
[pypy-commit] pypy better-storesink: just call constfold after cse to remove constant operations, instead of doing it cfbolz
[pypy-commit] pypy better-storesink: just use a second union find cfbolz
[pypy-commit] pypy better-storesink: loop_blocks was broken in complex cases cfbolz
[pypy-commit] pypy better-storesink: make cast_pointer introduce a union to be able to track things better cfbolz
[pypy-commit] pypy better-storesink: merge cfbolz
[pypy-commit] pypy better-storesink: merge default cfbolz
[pypy-commit] pypy better-storesink: merge default cfbolz
[pypy-commit] pypy better-storesink: refactor cfbolz
[pypy-commit] pypy better-storesink: safer mechanism for storing new equivalencies than updating the variable_families cfbolz
[pypy-commit] pypy better-storesink: support for more complex cases in the merge code: cfbolz
[pypy-commit] pypy better-storesink: support for removing loop-invariant heap reads cfbolz
[pypy-commit] pypy better-storesink: two fixes cfbolz
[pypy-commit] pypy better-storesink: used the wrong cache in two places while merging cfbolz
[pypy-commit] pypy better-storesink: use the write analyzer to know when to invalidate the cache cfbolz
[pypy-commit] pypy boehm-rawrefcount: A branch to implement rawrefcount for Boehm (needed by revdb) arigo
[pypy-commit] pypy boehm-rawrefcount: Add missing operation arigo
[pypy-commit] pypy boehm-rawrefcount: For Boehm, we can't pass the callback to rawrefcount.init() and we need arigo
[pypy-commit] pypy boehm-rawrefcount: Import the C code from https://bitbucket.org/arigo/arigo/src/default/hack/pypy-hack/boehm-rawrefcount/ arigo
[pypy-commit] pypy boehm-rawrefcount: Integration in-progress arigo
[pypy-commit] pypy boehm-rawrefcount: Stop trying to use the same test, and wrote a boehm-specific test arigo
[pypy-commit] pypy boehm-rawrefcount: Test and fix arigo
[pypy-commit] pypy boehm-rawrefcount: test and fix arigo
[pypy-commit] pypy boehm-rawrefcount: tweak arigo
[pypy-commit] pypy buffer-interface2: add cpyext-specific __rbuffer__, __wbuffer__ slots for old buffer interface mattip
[pypy-commit] pypy buffer-interface2: add test that passes -A fails untranslated, bf_get* slots are NULL in subclass mattip
[pypy-commit] pypy buffer-interface2: failing test mattip
[pypy-commit] pypy buffer-interface2: fill in tp_as_buffer.* functions that have no dunder slot mattip
[pypy-commit] pypy buffer-interface2: merge default into branch mattip
[pypy-commit] pypy buffer-interface2: merge default into branch mattip
[pypy-commit] pypy buffer-interface2: merge default into branch mattip
[pypy-commit] pypy buffer-interface2: more tests, fixes for buffer interface on app-level classes mattip
[pypy-commit] pypy buffer-interface2: test, add a unicode-specific getreadbuffer mattip
[pypy-commit] pypy buffer-interface2: test, fix whoops for not assigning the buf when readonly mattip
[pypy-commit] pypy buffer-interface: add a flags test from numpy to test_memoryobject.py, implement enough to run it mattip
[pypy-commit] pypy buffer-interface: add all the logic needed to make the tests pass, fixes for translation mattip
[pypy-commit] pypy buffer-interface: close branch to be merged mattip
[pypy-commit] pypy buffer-interface: create fixed arrays of Py_MAX_NDIMS, arbitrarily set to 32, in Py_buffer rather than malloc and leak memory mattip
[pypy-commit] pypy buffer-interface: extend and fix the already-existing test for __array_interfac__ mattip
[pypy-commit] pypy buffer-interface: Fix for test_zjit by explicitly disabling the W_MemoryView case here arigo
[pypy-commit] pypy buffer-interface: fix micronumpy test_zjit by adding _attrs_ to the class mattip
[pypy-commit] pypy buffer-interface: implement view of __array_interface__, fixes issue #52, #53 on pypy/numpy mattip
[pypy-commit] pypy buffer-interface: merge default into branch mattip
[pypy-commit] pypy buffer-interface: modifying ctor.py, might as well add a test for pypy/numpy issues #52, #53 mattip
[pypy-commit] pypy buffer-interface: rewrite _array_from_buffer_3118 without explicit W_MemeoryView (arigato) mattip
[pypy-commit] pypy buffer-interface: this was wrong. if arr.base and dtype='O', simply assume the pointers are gc objects mattip
[pypy-commit] pypy buffer-interface: try to update documentation mattip
[pypy-commit] pypy buffer-interface: use make_ref which has a keepalive_until_here() (arigato) mattip
[pypy-commit] pypy buffer-interface: use make_ref which has a keepalive_until_here() (arigato) mattip
[pypy-commit] pypy buffer-interface: use tobytes if get_raw_address fails, for StringBuffer type buffers mattip
[pypy-commit] pypy cling-support: more detailed benchmarks, used for paper wlav
[pypy-commit] pypy cling-support: optimization as used in the paper wlav
[pypy-commit] pypy closed-branches: Merge closed head ba39bd805638 on branch zarch-simd-support mattip
[pypy-commit] pypy closed-branches: Merge closed head d37191c0da88 on branch cpyext-null-slots mattip
[pypy-commit] pypy closed-branches: Merge closed head faca61b444b4 on branch w-xor-x2 mattip
[pypy-commit] pypy closed-branches: re-close this branch mattip
[pypy-commit] pypy conditional_call_value_2: "Test" and fix. It's hard to really test because the test previously arigo
[pypy-commit] pypy conditional_call_value_2: Add a (passing) test arigo
[pypy-commit] pypy conditional_call_value_2: Add support for COND_CALL_X in the x86 backend arigo
[pypy-commit] pypy conditional_call_value_2: Another attempt at doing a conditional_call that returns a value. arigo
[pypy-commit] pypy conditional_call_value_2: Fix arigo
[pypy-commit] pypy conditional_call_value_2: Goal arigo
[pypy-commit] pypy conditional_call_value_2: in-progress arigo
[pypy-commit] pypy conditional_call_value_2: in-progress arigo
[pypy-commit] pypy conditional_call_value_2: in-progress arigo
[pypy-commit] pypy conditional_call_value_2: in-progress. Issue: there is a lot of logic to handle call_pure already. arigo
[pypy-commit] pypy conditional_call_value_2: more in-progress arigo
[pypy-commit] pypy conditional_call_value_2: Next test, passing arigo
[pypy-commit] pypy conditional_call_value_2: Next test, passing arigo
[pypy-commit] pypy conditional_call_value_2: Test. Improve comment arigo
[pypy-commit] pypy conditional_call_value_2: Tests, progress arigo
[pypy-commit] pypy conditional_call_value_3: Add a flag 'supports_cond_call_value'. E.g. the ARM backend should first arigo
[pypy-commit] pypy conditional_call_value_3: Another case arigo
[pypy-commit] pypy conditional_call_value_3: a runner_test for cond_call_value arigo
[pypy-commit] pypy conditional_call_value_3: close branch, ready to merge: gives 11% improvements on microbenchmarks arigo
[pypy-commit] pypy conditional_call_value_3: Delicate ordering here (shown by test_ll_random) arigo
[pypy-commit] pypy conditional_call_value_3: Extra test arigo
[pypy-commit] pypy conditional_call_value_3: Fixes, more cases arigo
[pypy-commit] pypy conditional_call_value_3: fix tests arigo
[pypy-commit] pypy conditional_call_value_3: I *think* this is not needed right now, and test_ll_random seems to arigo
[pypy-commit] pypy conditional_call_value_3: Implement COND_CALL_VALUE (code copied from conditional_call_value_2) arigo
[pypy-commit] pypy conditional_call_value_3: kill import arigo
[pypy-commit] pypy conditional_call_value_3: merge heads arigo
[pypy-commit] pypy conditional_call_value_3: oops arigo
[pypy-commit] pypy conditional_call_value_3: Random tests for cond_call_value. arigo
[pypy-commit] pypy conditional_call_value_3: Rewrite: expand to cond_call_value arigo
[pypy-commit] pypy conditional_call_value_3: tweak the test arigo
[pypy-commit] pypy conditional_call_value_3: Use @jit.call_shortcut here. arigo
[pypy-commit] pypy cpyext-null-slots: abandon making int(np.string_('abc')) raise a ValueError mattip
[pypy-commit] pypy cpyext-null-slots: add a test that shows incompatibility in method lookup() resolution mattip
[pypy-commit] pypy cpyext-null-slots: cpython compatible slot lookup does not traverse mro, implement and ensure slots are filled at PyType_Ready mattip
[pypy-commit] pypy default: Aaargh again. Borked the locking here. Shows up as a rare crash arigo
[pypy-commit] pypy default: Aargh, missing header meant that the "char *" was truncated to 32 bits arigo
[pypy-commit] pypy default: Actually, we can just move that block of code outside the unrolled loop. arigo
[pypy-commit] pypy default: Add a passing test that direct running and annotation both crash arigo
[pypy-commit] pypy default: add comment arigo
[pypy-commit] pypy default: Added tag release-pypy2.7-v5.4.1 for changeset 050d84dd7899 mattip
[pypy-commit] pypy default: add release note for 5.4.1 mattip
[pypy-commit] pypy default: add try_inline, like always_inline and dont_inline cfbolz
[pypy-commit] pypy default: Align wrapped method compile_module() API with compile_module() function rlamy
[pypy-commit] pypy default: Another attempt at fixing the original problem arigo
[pypy-commit] pypy default: Another passing test arigo
[pypy-commit] pypy default: backout d1382b9d843b: doesn't work, dladdr() doesn't return the symlink's arigo
[pypy-commit] pypy default: Better fix for 47fa78e23bfb (in obscure cases like the 'return' trace arigo
[pypy-commit] pypy default: Better override _compute_value() than get_w_value(): the latter is arigo
[pypy-commit] pypy default: build the hashmap only once, not *every time* we call _has_reflected_op cfbolz
[pypy-commit] pypy default: Bundle all the static bits of compilation info into a SystemCompilationInfo object rlamy
[pypy-commit] pypy default: Check in extradoc/talk/bucharest2016/jit-backend-8vhY1ArTsh.txt in the arigo
[pypy-commit] pypy default: Clean up some variants of 'conditional_call_*' that are rarely or never arigo
[pypy-commit] pypy default: Document 'search-path-from-libpypy', and write down what it simplifies arigo
[pypy-commit] pypy default: documentation mattip
[pypy-commit] pypy default: Document branch arigo
[pypy-commit] pypy default: document branches mattip
[pypy-commit] pypy default: document branches and update release 5.4.1 mattip
[pypy-commit] pypy default: Document merged branch rlamy
[pypy-commit] pypy default: Document sys.getsizeof() arigo
[pypy-commit] pypy default: don't crash in merge_if_blocks if the values are symbolics cfbolz
[pypy-commit] pypy default: Don't put an __init__ in a Test class, it makes py.test skip(!?!!) that class arigo
[pypy-commit] pypy default: Explicitly detect that we found OpenSSL 1.1, and abort with a message arigo
[pypy-commit] pypy default: factor common code between compile_extension_module() and compile_extension_module_applevel() rlamy
[pypy-commit] pypy default: factor out some invariant code from the unrolling_iterable loop. no sense in cfbolz
[pypy-commit] pypy default: fix arigo
[pypy-commit] pypy default: Fix -A segfault in test_tp_new_in_subclass_type(), but now it fails in regular testing rlamy
[pypy-commit] pypy default: Fix: this is needed to JIT code that uses ll_ullong_py_mod_zer arigo
[pypy-commit] pypy default: fix __all__ code cfbolz
[pypy-commit] pypy default: fix b127faf95f86 which updated contributors but removed the rest of LICENSE mattip
[pypy-commit] pypy default: fix bug in codewriter about passing the exitswitch variable to a call cfbolz
[pypy-commit] pypy default: Fixes arigo
[pypy-commit] pypy default: fix formatting mattip
[pypy-commit] pypy default: Fix for sandbox tests arigo
[pypy-commit] pypy default: Fix PyLong_FromVoidPtr to return int/long like CPython. arigo
[pypy-commit] pypy default: Fix test arigo
[pypy-commit] pypy default: Fix tests (test_throw1 and test_throw2 were identical at this point) arigo
[pypy-commit] pypy default: Fix the initial value returned by sys.getrecursionlimit(). Used to be arigo
[pypy-commit] pypy default: fix translation mattip
[pypy-commit] pypy default: Fix translation and add a warning in annmodel.unionof rlamy
[pypy-commit] pypy default: fix translation with --no-allworkingmodules arigo
[pypy-commit] pypy default: fix unintentional skip of untranslated tests mattip
[pypy-commit] pypy default: Fix what occurs near the limit MAX_STACK_DEPTH: I *think* it would arigo
[pypy-commit] pypy default: For now, forbid --profopt if we're building a shared pypy arigo
[pypy-commit] pypy default: Future py3k compat arigo
[pypy-commit] pypy default: Hack at our included py.test to fail, not quietly skip, when it sees a arigo
[pypy-commit] pypy default: hg merge conditional_call_value_3 arigo
[pypy-commit] pypy default: hg merge search-path-from-libpypy arigo
[pypy-commit] pypy default: If libpypy-c.so is a symlink, look first around the non-deferred path to arigo
[pypy-commit] pypy default: If the expected exception was raised, the SSLContext can't be shut down yet stefanor
[pypy-commit] pypy default: import cffi/b81ca61b6de6 arigo
[pypy-commit] pypy default: import cffi/cb6147abe7af arigo
[pypy-commit] pypy default: Improve error message when attempting to modify annotations in a fixed graph rlamy
[pypy-commit] pypy default: improve the trace of import statements to not contain calls cfbolz
[pypy-commit] pypy default: Issue #2325/2361: __class__ assignment between two classes with the same arigo
[pypy-commit] pypy default: Issue #2388: the problem is obscure interaction with a different call (I arigo
[pypy-commit] pypy default: Issue #2389: the custom error handler may return a 'pos' that is smaller arigo
[pypy-commit] pypy default: Kill FakeSpace.include_dir rlamy
[pypy-commit] pypy default: LibreSSL uses a higher value of OPENSSL_VERSION_NUMBER just because arigo
[pypy-commit] pypy default: make sure exception is raised if call returns error value mattip
[pypy-commit] pypy default: make sure test of PyObject_GetBuffer runs to completion, fix implementation mattip
[pypy-commit] pypy default: merge cfbolz
[pypy-commit] pypy default: Merge branch 'test-cpyext' rlamy
[pypy-commit] pypy default: merge branch to convert message to str in warn(message, ...) like cpython mattip
[pypy-commit] pypy default: merge branch to reject 'a'.strip(buffer(' ')) like cpython mattip
[pypy-commit] pypy default: merge buffer-interface, which implements buffer-protocol parts of cpyext and numpypy mattip
[pypy-commit] pypy default: Merge compile_extension_module() and compile_extension_module_applevel() into a single version that takes a sys_info instead of a space rlamy
[pypy-commit] pypy default: Merged in force-virtual-state (pull request #478) sbauman
[pypy-commit] pypy default: merge heads arigo
[pypy-commit] pypy default: Merge in force-virtual-state sbauman
[pypy-commit] pypy default: Minor test and fix arigo
[pypy-commit] pypy default: move all Py_buffer to memoryobject, create helper function mattip
[pypy-commit] pypy default: Moved tag release-pypy2.7-v5.4.1 to changeset 0e2d9a73f5a1 (from changeset 050d84dd7899) mattip
[pypy-commit] pypy default: Move the bit checking inside helpers, share it from the two places arigo
[pypy-commit] pypy default: Oops! rlamy
[pypy-commit] pypy default: Oops, bug: if a throw() propagates the error sent in, then the 'return' arigo
[pypy-commit] pypy default: oops, revert f6eabcc3367c: found the difference between test_throw1 and arigo
[pypy-commit] pypy default: os.access() and float arguments is a translation crash with sandboxing arigo
[pypy-commit] pypy default: os.stat(): support full-precision nanosecond times in translated RPython arigo
[pypy-commit] pypy default: pfff another hack on top of this pile of hacks arigo
[pypy-commit] pypy default: profopt was accidentally silently disabled on all platforms, by arigo
[pypy-commit] pypy default: Put the explanation in the docstring and in the error message arigo
[pypy-commit] pypy default: PyPy creates a SOABI entry in confvars, Numpy needs only the correct SO entry mattip
[pypy-commit] pypy default: refactor for py3k pjenvey
[pypy-commit] pypy default: Remove again unused import arigo
[pypy-commit] pypy default: removed my hack, use filter instead (thx armin) plan_rich
[pypy-commit] pypy default: remove one more pyobj force, rework documentation mattip
[pypy-commit] pypy default: remove unecessary check (arigato) mattip
[pypy-commit] pypy default: remove unsupported operation in the stress tests (call_cond_value) plan_rich
[pypy-commit] pypy default: Rename the exception that we call TestException to some class name that arigo
[pypy-commit] pypy default: replace a number of _annspecialcase_ by @specialize in rlib/ cfbolz
[pypy-commit] pypy default: replace some _annspecialcase_ with @specialize in the interpreter cfbolz
[pypy-commit] pypy default: revert 2b244ba62cf4, redo the distutils part to be cpython compliant (no SOABI) mattip
[pypy-commit] pypy default: rposix.sync() arigo
[pypy-commit] pypy default: Search first around the pypy executable (like previously), arigo
[pypy-commit] pypy default: some inlining decorators (instead of function attributes) cfbolz
[pypy-commit] pypy default: Specialize str_decode_utf_8_impl on allow_surrogates, to resolve a translation failure in pycket rlamy
[pypy-commit] pypy default: test, fix for raising non-TypeError exception when conversion fails mattip
[pypy-commit] pypy default: test, fix for returning a W_LongObject mattip
[pypy-commit] pypy default: test, fix issue #2395 - do not force pyobj mattip
[pypy-commit] pypy default: test, fix Py_buffer format, which can be a string (issue #2396 and IRC discussion) mattip
[pypy-commit] pypy default: Test and fix arigo
[pypy-commit] pypy default: Test that loops automatically unroll if they are written anywhere in the arigo
[pypy-commit] pypy default: These specialize.arg() are not valid, because callers inside runicode arigo
[pypy-commit] pypy default: This test is meant to test for virtualize.optimize_COND_CALL(). arigo
[pypy-commit] pypy default: translation fix arigo
[pypy-commit] pypy default: Tweaks: don't use 'int(float-value-of-st.st_Xtime)': because precision arigo
[pypy-commit] pypy default: Tweak the build_cffi_imports output: clarify the status in case of arigo
[pypy-commit] pypy default: update for 5.4.1 mattip
[pypy-commit] pypy default: Update to cffi/5fc2d2cf2f5d arigo
[pypy-commit] pypy default: update to cffi/b518a1326d6c arigo
[pypy-commit] pypy default: update to cffi/c61b49c67c83 arigo
[pypy-commit] pypy default: update version minor number, make test run untranslated mattip
[pypy-commit] pypy default: Update whatsnew-head for the force-virtual-state sbauman
[pypy-commit] pypy default: Using PyType_GenericNew is wrong for a metaclass (effectively reverts 6ea4b78fa64b) rlamy
[pypy-commit] pypy default: Warning arigo
[pypy-commit] pypy default: Why final=False here? arigo
[pypy-commit] pypy default: Write a note arigo
[pypy-commit] pypy default: Write down a comment about profopt in JIT-enabled binaries arigo
[pypy-commit] pypy force-virtual-state: Add some explanation for the new test sbauman
[pypy-commit] pypy force-virtual-state: Add test case for generating the proper guard on a raw pointer during short preamble creation sbauman
[pypy-commit] pypy force-virtual-state: Add test for edge case when forcing virtual args (currently failing) sbauman
[pypy-commit] pypy force-virtual-state: Emit proper instruction sequence for checking raw pointers sbauman
[pypy-commit] pypy force-virtual-state: Ensure make_inputargs fails properly when given arguments with type information sbauman
[pypy-commit] pypy force-virtual-state: Explicitly pass force_boxes arg in all cases (suggested by cfbolz) sbauman
[pypy-commit] pypy force-virtual-state: Merge sbauman
[pypy-commit] pypy force-virtual-state: Merge sbauman
[pypy-commit] pypy force-virtual-state: Merge with default sbauman
[pypy-commit] pypy force-virtual-state: Merge with default sbauman
[pypy-commit] pypy force-virtual-state: Merge with known working version sbauman
[pypy-commit] pypy force-virtual-state: Missing comma sbauman
[pypy-commit] pypy force-virtual-state: Remove more junk sbauman
[pypy-commit] pypy force-virtual-state: Remove useless line sbauman
[pypy-commit] pypy force-virtual-state: Tentative solution to having aliased objects in virtual state when forcing sbauman
[pypy-commit] pypy guard-compatible: add debug prints to at least have a chance to figure out what is going on cfbolz
[pypy-commit] pypy guard-compatible: fix fast isinstance check with guard_compatible cfbolz
[pypy-commit] pypy guard-compatible: fix translation cfbolz
[pypy-commit] pypy guard-compatible: fix translation cfbolz
[pypy-commit] pypy guard-compatible: just print the pointer constant cfbolz
[pypy-commit] pypy guard-compatible: merge default cfbolz
[pypy-commit] pypy guard-compatible: merge default cfbolz
[pypy-commit] pypy guard-compatible: somewhat obscure: if there is a condition on the "version" (ie second argument cfbolz
[pypy-commit] pypy ppc-vsx-support: apply translation fixes for ppc plan_rich
[pypy-commit] pypy ppc-vsx-support: catchup with default plan_rich
[pypy-commit] pypy ppc-vsx-support: document branches plan_rich
[pypy-commit] pypy ppc-vsx-support: do not force alloc variable of trace op the bridge is attached to. use a scratch reg for x86 plan_rich
[pypy-commit] pypy ppc-vsx-support: drafting new implementation plan_rich
[pypy-commit] pypy ppc-vsx-support: expand gpr/fpr to a vector reg plan_rich
[pypy-commit] pypy ppc-vsx-support: failargs can have None entries, consider this while iterating plan_rich
[pypy-commit] pypy ppc-vsx-support: fixed vecopt tests with the new delayed execution plan_rich
[pypy-commit] pypy ppc-vsx-support: fix ppc accum issue plan_rich
[pypy-commit] pypy ppc-vsx-support: killed a bug in unpacking multiple plan_rich
[pypy-commit] pypy ppc-vsx-support: merge default plan_rich
[pypy-commit] pypy ppc-vsx-support: merge default plan_rich
[pypy-commit] pypy ppc-vsx-support: merge default plan_rich
[pypy-commit] pypy ppc-vsx-support: merge default plan_rich
[pypy-commit] pypy ppc-vsx-support: merge default plan_rich
[pypy-commit] pypy ppc-vsx-support: merge default plan_rich
[pypy-commit] pypy ppc-vsx-support: merge default plan_rich
[pypy-commit] pypy ppc-vsx-support: new test to ensure that operations that only the necessary index calculations are emitted plan_rich
[pypy-commit] pypy ppc-vsx-support: not_implemented takes only one argument, remove pxor(result, result) which contained valid data plan_rich
[pypy-commit] pypy ppc-vsx-support: one parameter too much plan_rich
[pypy-commit] pypy ppc-vsx-support: operations now can be delayed (if they are pure), operation arguments can pull them just before the op itself is emitted plan_rich
[pypy-commit] pypy ppc-vsx-support: removed issue that modified list. if the vectorization was bailed this left behind modifications that should not have happened plan_rich
[pypy-commit] pypy ppc-vsx-support: remove some unused options, change some tests in micronumpy plan_rich
[pypy-commit] pypy ppc-vsx-support: remove unsupported operation in the stress tests (call_cond_value) plan_rich
[pypy-commit] pypy ppc-vsx-support: renamed costmodel class to generic one (it is used also for ppc, not only for x86) plan_rich
[pypy-commit] pypy ppc-vsx-support: resolve issues in test_schedule introduced by vec_load/store plan_rich
[pypy-commit] pypy ppc-vsx-support: resolve some test issues plan_rich
[pypy-commit] pypy ppc-vsx-support: return register immediately if it is the right one (base case) plan_rich
[pypy-commit] pypy ppc-vsx-support: same patch for x86 plan_rich
[pypy-commit] pypy ppc-vsx-support: syntax error plan_rich
[pypy-commit] pypy ppc-vsx-support: test used old vec_raw_store/load plan_rich
[pypy-commit] pypy ppc-vsx-support: translation fix plan_rich
[pypy-commit] pypy ppc-vsx-support: translation issue, no common base class plan_rich
[pypy-commit] pypy ppc-vsx-support: translation issues plan_rich
[pypy-commit] pypy ppc-vsx-support: translation issues plan_rich
[pypy-commit] pypy ppc-vsx-support: try to align the loop (should improve ppc and s390x load/store performance plan_rich
[pypy-commit] pypy ppc-vsx-support: update the llgraph runner to exec vec_load/store plan_rich
[pypy-commit] pypy ppc-vsx-support: use detect_vsx in ppc, not detect_simd_z plan_rich
[pypy-commit] pypy ppc-vsx-support: use os.open instead of open() to read from /proc files at runtime plan_rich
[pypy-commit] pypy ppc-vsx-support: use x86 scratch regs instead plan_rich
[pypy-commit] pypy ppc-vsx-support: vector operations (load/store) where not considered in dep. construction (was not needed up to now) plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: adding properties and new parameters in py3.5 (work in progress) plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: both bz2 and lzma seem to be in shape for py3.5 plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: count diff too often plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: issue in bz2, data not appended plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: pass all lzma tests plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: progress on the lzma changes to allow max_length parameter plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: py3.5 modify test, read accepts None as first argument plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: simplifications for bz2 decompress plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: tracking the rest of the buffer that has not been read plan_rich
[pypy-commit] pypy py3.5-bz2-lzma: unused data is now added again for decompression plan_rich
[pypy-commit] pypy py3.5-corowrapper: 'yield from' starts to work again, a bit arigo
[pypy-commit] pypy py3.5-corowrapper: A branch to add the coroutine_wrapper type from CPython arigo
[pypy-commit] pypy py3.5-corowrapper: A direct test for "async for" (I have still no idea how we're supposed arigo
[pypy-commit] pypy py3.5-corowrapper: aiter_wrapper arigo
[pypy-commit] pypy py3.5-corowrapper: Another test and fix arigo
[pypy-commit] pypy py3.5-corowrapper: CO_FUTURE_GENERATOR_STOP, not tested so far arigo
[pypy-commit] pypy py3.5-corowrapper: comment arigo
[pypy-commit] pypy py3.5-corowrapper: comment arigo
[pypy-commit] pypy py3.5-corowrapper: Comment, test about StopAsyncIteration arigo
[pypy-commit] pypy py3.5-corowrapper: Complain more clearly arigo
[pypy-commit] pypy py3.5-corowrapper: CoroutineWrapper arigo
[pypy-commit] pypy py3.5-corowrapper: Direct test for 'await' arigo
[pypy-commit] pypy py3.5-corowrapper: First test: space.next() must not accept coroutines (and not by *name* arigo
[pypy-commit] pypy py3.5-corowrapper: fix arigo
[pypy-commit] pypy py3.5-corowrapper: Fixes arigo
[pypy-commit] pypy py3.5-corowrapper: fix interp_jit arigo
[pypy-commit] pypy py3.5-corowrapper: fix the stack depth computation arigo
[pypy-commit] pypy py3.5-corowrapper: fix the test arigo
[pypy-commit] pypy py3.5-corowrapper: in-progress arigo
[pypy-commit] pypy py3.5-corowrapper: in-progress arigo
[pypy-commit] pypy py3.5-corowrapper: in-progress arigo
[pypy-commit] pypy py3.5-corowrapper: in-progress (I think that basic generators are working again) arigo
[pypy-commit] pypy py3.5-corowrapper: Next failing test arigo
[pypy-commit] pypy py3.5-corowrapper: next fix arigo
[pypy-commit] pypy py3.5-corowrapper: one translation fix arigo
[pypy-commit] pypy py3.5-corowrapper: pass test_generator arigo
[pypy-commit] pypy py3.5-corowrapper: Refactor/cleanup/complete the opcodes arigo
[pypy-commit] pypy py3.5-corowrapper: reverse the order of these two values on the valuestack, as needed for arigo
[pypy-commit] pypy py3.5-corowrapper: rewrite rewrite rewrite in progress arigo
[pypy-commit] pypy py3.5-corowrapper: Setting __name__ on generator-iterator objects (uh?) arigo
[pypy-commit] pypy py3.5-corowrapper: sys.set_coroutine_wrapper() arigo
[pypy-commit] pypy py3.5-corowrapper: Test and fix arigo
[pypy-commit] pypy py3.5-corowrapper: Test and fix arigo
[pypy-commit] pypy py3.5-corowrapper: Test and fix: must reset w_yielded_from to None when running arigo
[pypy-commit] pypy py3.5-corowrapper: Test for gi_yieldfrom arigo
[pypy-commit] pypy py3.5-corowrapper: translation fixes arigo
[pypy-commit] pypy py3.5-corowrapper: Two tests, one passing arigo
[pypy-commit] pypy py3.5-set-sentinel: extended doc, renamed function, starting to implement the function fully plan_rich
[pypy-commit] pypy py3.5-set-sentinel: release the sentinel lock before the thread dies plan_rich
[pypy-commit] pypy py3.5: Add a minimal _ssl.MemoryBIO to allow "import ssl" amauryfa
[pypy-commit] pypy py3.5: Add memoryview.nbytes. amauryfa
[pypy-commit] pypy py3.5: Another one arigo
[pypy-commit] pypy py3.5: audioop.byteswap() arigo
[pypy-commit] pypy py3.5: avoid sys.version during bootstrapping as its setup lazily (3.6 uses pjenvey
[pypy-commit] pypy py3.5: Back out ce4d4b3c159f, i.e. redo 26dfa2f67afd, in the py3.5 branch where arigo
[pypy-commit] pypy py3.5: Fix for argument clinic's use of the "bool" specifier, which now maps arigo
[pypy-commit] pypy py3.5: Fix for test_interpreter.test_build_map_unpack arigo
[pypy-commit] pypy py3.5: Fix many tests in module/_ssl amauryfa
[pypy-commit] pypy py3.5: Fix some pickling tests arigo
[pypy-commit] pypy py3.5: Fix tests for audioop.byteswap(), and test_wave passes. amauryfa
[pypy-commit] pypy py3.5: Fix translation rlamy
[pypy-commit] pypy py3.5: Fix translation amauryfa
[pypy-commit] pypy py3.5: formatting arigo
[pypy-commit] pypy py3.5: hg merge py3k rlamy
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k arigo
[pypy-commit] pypy py3.5: hg merge py3k rlamy
[pypy-commit] pypy py3.5: Implement default keyword on min & max with tests remarkablerocket
[pypy-commit] pypy py3.5: Implement round ndigits=None behaviour as per CPython 3.5 docs remarkablerocket
[pypy-commit] pypy py3.5: Implement the -I option to pypy arigo
[pypy-commit] pypy py3.5: Issue #2400: Rename the interp-level module to '_operator' arigo
[pypy-commit] pypy py3.5: Make error messages for dict and other non-numeric types consistent with CPython 3.5 remarkablerocket
[pypy-commit] pypy py3.5: Make int() called with invalid argument type raise error message consistent with CPython remarkablerocket
[pypy-commit] pypy py3.5: math.gcd() arigo
[pypy-commit] pypy py3.5: math.isclose() arigo
[pypy-commit] pypy py3.5: Meant to be turned off by default arigo
[pypy-commit] pypy py3.5: merge branch plan_rich
[pypy-commit] pypy py3.5: merge bz2 and lzma changes plan_rich
[pypy-commit] pypy py3.5: Merged in remarkablerocket/pypy/py3.5 (pull request #480) arigo
[pypy-commit] pypy py3.5: merged py3.5-bz2-lzma plan_rich
[pypy-commit] pypy py3.5: Merged pypy/pypy/py3.5 into py3.5 remarkablerocket
[pypy-commit] pypy py3.5: merge py3k pjenvey
[pypy-commit] pypy py3.5: Move this method as a global function arigo
[pypy-commit] pypy py3.5: os.sync() arigo
[pypy-commit] pypy py3.5: Pff, took me too long to figure out this was causing an obscure error at arigo
[pypy-commit] pypy py3.5: removed some skips for modules that work now! plan_rich
[pypy-commit] pypy py3.5: Remove dunder aliases from _operator, as on CPython rlamy
[pypy-commit] pypy py3.5: Remove excess spaces from error message (commit f1684db) remarkablerocket
[pypy-commit] pypy py3.5: Renaming a built-in module does not change the name of the corresponding usemodules.* option rlamy
[pypy-commit] pypy py3.5: Replace checks for None and space.w_None with space.is_none remarkablerocket
[pypy-commit] pypy py3.5: Replace invalid RPython in floatobject.py (commit 0bf6794, pull request #481) remarkablerocket
[pypy-commit] pypy py3.5: Revert change to rpython/rlib/rfloat.py in commit 0bf6794 (pull request #481); amend error message in pypy/objspace/std/floatobject.py instead. remarkablerocket
[pypy-commit] pypy py3.5: simplify pjenvey
[pypy-commit] pypy py3.5: Skip the micronumpy tests on py3.5. Micronumpy is anyway not compiled arigo
[pypy-commit] pypy py3.5: socket.py now depends on the 'select' module. amauryfa
[pypy-commit] pypy py3.5: SOMAXCONN, socket.listen(no argument) arigo
[pypy-commit] pypy py3.5: sys.hash_info.algorithm = 'rpython' amauryfa
[pypy-commit] pypy py3.5: Tweak tweak, now it seems to translate arigo
[pypy-commit] pypy py3.5: Uh arigo
[pypy-commit] pypy py3.5: untabify arigo
[pypy-commit] pypy py3.5: Update int(x, base=10) behaviour for Python 3.4. remarkablerocket
[pypy-commit] pypy py3.5: Use ast.NameConstant instead of ast.Name for None/True/False arigo
[pypy-commit] pypy py3.5: Use built-in module _operator to define built-in functions bin(), oct(), hex(). rlamy
[pypy-commit] pypy py3.5: We can't really do anything without the '_struct' module arigo
[pypy-commit] pypy py3.5: Write down a message to avoid being hit by 0419bc15bcf5 again arigo
[pypy-commit] pypy py35-getbuiltin: Don't 'inject' the .pyc magic number from the outside, but import it from __pypy__. amauryfa
[pypy-commit] pypy py35-getbuiltin: Experiment with space.getbuiltinmodule(): it should not interact with sys.modules. amauryfa
[pypy-commit] pypy py35-getbuiltin: Progress: space.getbuiltinmodule() does not use or update sys.modules amauryfa
[pypy-commit] pypy py3k-faulthandler: A branch to really implement faulthandler arigo
[pypy-commit] pypy py3k-faulthandler: add explicit dependency arigo
[pypy-commit] pypy py3k-faulthandler: Add some more of the test functions arigo
[pypy-commit] pypy py3k-faulthandler: Change the API in rvmprof to support more directly what we need, arigo
[pypy-commit] pypy py3k-faulthandler: De-duplicate traceback entries after a LOC_JITTED arigo
[pypy-commit] pypy py3k-faulthandler: fix arigo
[pypy-commit] pypy py3k-faulthandler: fix arigo
[pypy-commit] pypy py3k-faulthandler: fixes arigo
[pypy-commit] pypy py3k-faulthandler: fixes fixes arigo
[pypy-commit] pypy py3k-faulthandler: fix the test arigo
[pypy-commit] pypy py3k-faulthandler: in-progress arigo
[pypy-commit] pypy py3k-faulthandler: in-progress arigo
[pypy-commit] pypy py3k-faulthandler: in-progress arigo
[pypy-commit] pypy py3k-faulthandler: in-progress arigo
[pypy-commit] pypy py3k-faulthandler: Port pypy to the new interface arigo
[pypy-commit] pypy py3k-faulthandler: re-add _read_null() arigo
[pypy-commit] pypy py3k-faulthandler: translation fixes arigo
[pypy-commit] pypy py3k-faulthandler: translation fixes arigo
[pypy-commit] pypy py3k-faulthandler: Translation test arigo
[pypy-commit] pypy py3k-faulthandler: tweak arigo
[pypy-commit] pypy py3k-faulthandler: tweaks, in-progress arigo
[pypy-commit] pypy py3k-faulthandler: tweaks, in-progress arigo
[pypy-commit] pypy py3k: 2to3 rlamy
[pypy-commit] pypy py3k: 2to3fy cpyext's test_memoryobject.py rlamy
[pypy-commit] pypy py3k: 2to3fy test rlamy
[pypy-commit] pypy py3k: Aaaargh arigo
[pypy-commit] pypy py3k: Add explicitly @jit.call_shortcut here arigo
[pypy-commit] pypy py3k: Add merged branches arigo
[pypy-commit] pypy py3k: Add tests for st.st_Xtime_ns arigo
[pypy-commit] pypy py3k: attempt to adapt missing parts of 29d14733e007 to cffi pjenvey
[pypy-commit] pypy py3k: backout bd9d1dd19c2c, per a8a4ffe56b74 rlamy
[pypy-commit] pypy py3k: backport 146f531d6c86 arigo
[pypy-commit] pypy py3k: cpyext is required to actually test _imp.load_dynamic(). rlamy
[pypy-commit] pypy py3k: Delete RPython impls of PyObject_CheckBuffer and PyObject_GetBuffer once again rlamy
[pypy-commit] pypy py3k: Duplicate arigo
[pypy-commit] pypy py3k: Fix: _utf8 cannot be a quasi-immutable. Will document why and do a arigo
[pypy-commit] pypy py3k: fix bytebuffer test rlamy
[pypy-commit] pypy py3k: fixes arigo
[pypy-commit] pypy py3k: Fix for test_hash arigo
[pypy-commit] pypy py3k: Fix NullImporter test and revert unnecessary stdlib change rlamy
[pypy-commit] pypy py3k: Fix skipping of AppTestApi rlamy
[pypy-commit] pypy py3k: Fix test_load_dynamic() to match CPython behaviour rlamy
[pypy-commit] pypy py3k: fix translation rlamy
[pypy-commit] pypy py3k: Fix translation rlamy
[pypy-commit] pypy py3k: For now, we should get most of the intended effect with @jit.elidable arigo
[pypy-commit] pypy py3k: Functions and methods that are actually "built-in" on CPython might end arigo
[pypy-commit] pypy py3k: graft eb88834f82b2 arigo
[pypy-commit] pypy py3k: hg merge default rlamy
[pypy-commit] pypy py3k: hg merge default rlamy
[pypy-commit] pypy py3k: hg merge default arigo
[pypy-commit] pypy py3k: hg merge default arigo
[pypy-commit] pypy py3k: hg merge default arigo
[pypy-commit] pypy py3k: hg merge default arigo
[pypy-commit] pypy py3k: hg merge default arigo
[pypy-commit] pypy py3k: hg merge default arigo
[pypy-commit] pypy py3k: hg merge default arigo
[pypy-commit] pypy py3k: hg merge default arigo
[pypy-commit] pypy py3k: hg merge default rlamy
[pypy-commit] pypy py3k: hg merge default rlamy
[pypy-commit] pypy py3k: In the py3k branch, use 'python3.3' to run -A tests, not 'python3'. arigo
[pypy-commit] pypy py3k: kill _handle, subprocess.Handle basically replaces it pjenvey
[pypy-commit] pypy py3k: Kill unused MODULES_THAT_ALWAYS_SHADOW and associated tests (importlib handles that logic now) rlamy
[pypy-commit] pypy py3k: Likely fix arigo
[pypy-commit] pypy py3k: Manual merge of a translation fix I've done only in the py3.5 branch (sorry) arigo
[pypy-commit] pypy py3k: memoryview indexing returns integers, not bytes arigo
[pypy-commit] pypy py3k: merge default pjenvey
[pypy-commit] pypy py3k: merge default pjenvey
[pypy-commit] pypy py3k: merge heads arigo
[pypy-commit] pypy py3k: move all Py_buffer to memoryobject, create helper function rlamy
[pypy-commit] pypy py3k: Move the docstrings to the proper place arigo
[pypy-commit] pypy py3k: next attempt arigo
[pypy-commit] pypy py3k: outdated comment pjenvey
[pypy-commit] pypy py3k: quick untested fix arigo
[pypy-commit] pypy py3k: reapply 995f8366f830 to cffi: accept py3 strs via CreateProcessW pjenvey
[pypy-commit] pypy py3k: regenerate pjenvey
[pypy-commit] pypy py3k: Remove obsolete code rlamy
[pypy-commit] pypy py3k: Replace marshal with _md5 in test, so that it works on CPython (no idea why this is needed, but whatevs) rlamy
[pypy-commit] pypy py3k: Restore missing import arigo
[pypy-commit] pypy py3k: Set up ImportError attributes properly in _imp.load_dynamic() rlamy
[pypy-commit] pypy py3k: Skip in test_app_main instead of failing if there is no python3.3 rlamy
[pypy-commit] pypy py3k: Skip PyPy-specific test on CPython rlamy
[pypy-commit] pypy py3k: Skip test_random_attr.py on pypy3 rlamy
[pypy-commit] pypy py3k: Skip this import test which I don't understand at all, and don't really arigo
[pypy-commit] pypy py3k: test, fix Py_buffer format, which can be a string (issue #2396 and IRC discussion) rlamy
[pypy-commit] pypy py3k: Test and fix arigo
[pypy-commit] pypy py3k: Test and fix arigo
[pypy-commit] pypy py3k: Test and fix: correctly count the number of kwonly argument given, for arigo
[pypy-commit] pypy py3k: Test and fix: don't give a unicode co_filename from interactive.py arigo
[pypy-commit] pypy py3k: Test and fix for allowing __len__ to return any index-like rlamy
[pypy-commit] pypy py3k: translation fix arigo
[pypy-commit] pypy py3k: Try to blindly fix some casting issues inside interp_time, showing up arigo
[pypy-commit] pypy py3k: utf-8 encoding before translation accepts lone surrogates, because it is arigo
[pypy-commit] pypy quad-color-gc: Add teardown method to codegen ntruessel
[pypy-commit] pypy quad-color-gc: Claim QCGC supports rweakref to make pypy executable useable ntruessel
[pypy-commit] pypy quad-color-gc: Fix compilation rule ntruessel
[pypy-commit] pypy quad-color-gc: Fix some broken testcases due to wrong typeids for immortal objects ntruessel
[pypy-commit] pypy quad-color-gc: Fix weakref test ntruessel
[pypy-commit] pypy quad-color-gc: Performance improvements ntruessel
[pypy-commit] pypy quad-color-gc: Probably last optimization import from qcgc repo ntruessel
[pypy-commit] pypy quad-color-gc: QCGC codebase update ntruessel
[pypy-commit] pypy quad-color-gc: QCGC update ntruessel
[pypy-commit] pypy quad-color-gc: QCGC update (performance) ntruessel
[pypy-commit] pypy quad-color-gc: Register weakrefs (test failing due to not collected targets) ntruessel
[pypy-commit] pypy quad-color-gc: Remove object layout hack, update qcgc codebase ntruessel
[pypy-commit] pypy quad-color-gc: Skip correct test, disable checks in qcgc and prepare weakref registration ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc codebase ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc codebase ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc codebase ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc codebase ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc codebase ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc codebase (bugfix) ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc codebase (performance optimizations) ntruessel
[pypy-commit] pypy quad-color-gc: Update qcgc codebase again ntruessel
[pypy-commit] pypy redirect-assembler-jitlog: merge default plan_rich
[pypy-commit] pypy redirect-assembler-jitlog: new tag tmp_callback to correctly make the connection between call_assembler <-> trace plan_rich
[pypy-commit] pypy redirect-assembler-jitlog: translation issue, assigning unique number to tmp_callback jit cell token plan_rich
[pypy-commit] pypy redirect-assembler-jitlog: translation issues plan_rich
[pypy-commit] pypy release-5.x: Add a failing (skipped) whitebox test and a test about ctypes that fails arigo
[pypy-commit] pypy release-5.x: Added tag release-pypy2.7-v5.4.1 for changeset 050d84dd7899 mattip
[pypy-commit] pypy release-5.x: Avoid blowing up with results that are ~0 but <0 due to floating point imprecision stefanor
[pypy-commit] pypy release-5.x: Bump recursionlimit, for translating with cpython stefanor
[pypy-commit] pypy release-5.x: Explicitly detect that we found OpenSSL 1.1, and abort with a message mattip
[pypy-commit] pypy release-5.x: fix b127faf95f86 which updated contributors but removed the rest of LICENSE mattip
[pypy-commit] pypy release-5.x: haaaack on top of previous hacks arigo
[pypy-commit] pypy release-5.x: import cffi/b81ca61b6de6 mattip
[pypy-commit] pypy release-5.x: import cffi/cb6147abe7af mattip
[pypy-commit] pypy release-5.x: Issue #2388: the problem is obscure interaction with a different call (I mattip
[pypy-commit] pypy release-5.x: Issue #2389: the custom error handler may return a 'pos' that is smaller mattip
[pypy-commit] pypy release-5.x: Moved tag release-pypy2.7-v5.4.1 to changeset 0e2d9a73f5a1 (from changeset 050d84dd7899) mattip
[pypy-commit] pypy release-5.x: Move the bit checking inside helpers, share it from the two places mattip
[pypy-commit] pypy release-5.x: name can be None stefanor
[pypy-commit] pypy release-5.x: pfff another hack on top of this pile of hacks mattip
[pypy-commit] pypy release-5.x: Probable fix for issue #2383: have 'list(S())' call 'S.__getitem__' if S arigo
[pypy-commit] pypy release-5.x: test, fix issue #2395 - do not force pyobj mattip
[pypy-commit] pypy release-5.x: update for 5.4.1 mattip
[pypy-commit] pypy release-5.x: update version numbers, make test run untranslated mattip
[pypy-commit] pypy remove-getarrayitem-pure: Move code sbauman
[pypy-commit] pypy remove-getarrayitem-pure: Sanity checks sbauman
[pypy-commit] pypy reverse-debugger: Added tag RevDB-pypy2.7-v5.4.1 for changeset 4909c06daf41 arigo
[pypy-commit] pypy reverse-debugger: Added tag RevDB-pypy2.7-v5.4.1 for changeset d7724c0a5700 arigo
[pypy-commit] pypy reverse-debugger: Add test file (not working at all so far) arigo
[pypy-commit] pypy reverse-debugger: Comments. Disable a code path that (by mistake) casts between a GC arigo
[pypy-commit] pypy reverse-debugger: Disable micronumpy as well, for now arigo
[pypy-commit] pypy reverse-debugger: emit the byte "release gil" before actually releasing the GIL arigo
[pypy-commit] pypy reverse-debugger: fix, basic test passes arigo
[pypy-commit] pypy reverse-debugger: Fix for Linux 32-bit arigo
[pypy-commit] pypy reverse-debugger: hg merge boehm-rawrefcount arigo
[pypy-commit] pypy reverse-debugger: hg merge boehm-rawrefcount arigo
[pypy-commit] pypy reverse-debugger: hg merge default arigo
[pypy-commit] pypy reverse-debugger: in-progress arigo
[pypy-commit] pypy reverse-debugger: Move the tests to the external repo arigo
[pypy-commit] pypy reverse-debugger: Print full precision doubles arigo
[pypy-commit] pypy reverse-debugger: Remove the runtime part, move it to http://bitbucket.org/pypy/revdb/ arigo
[pypy-commit] pypy reverse-debugger: Rename the env var REVDB arigo
[pypy-commit] pypy rmod-radd-slots: fix translation mattip
[pypy-commit] pypy rmod-radd-slots: test, add functioning __rmod__ and dummy __radd__ to W_AbstractBytesObject mattip
[pypy-commit] pypy search-path-from-libpypy: Fix/skip parts of this test (could in theory play tricks, but that's arigo
[pypy-commit] pypy search-path-from-libpypy: ready to merge arigo
[pypy-commit] pypy search-path-from-libpypy: Search for the lib-python/lib_pypy directores starting from the arigo
[pypy-commit] pypy stdlib-2.7.11: merge default pjenvey
[pypy-commit] pypy stricter-strip: close branch to be merged mattip
[pypy-commit] pypy stricter-strip: test, fix for argument to strip() - must be None, str, unicode (cpython compatibility) mattip
[pypy-commit] pypy sys-getsizeof: A branch to "implement" sys.getsizeof(), or at least give some number. arigo
[pypy-commit] pypy test-cpyext: Create sys_info in setup_class() instead of setup_method() rlamy
[pypy-commit] pypy test-cpyext: Extract make_source() function from .import_module() rlamy
[pypy-commit] pypy test-cpyext: fix rlamy
[pypy-commit] pypy test-cpyext: Fix test_recursive_package_import to actually check what it claims to rlamy
[pypy-commit] pypy test-cpyext: hg merge default rlamy
[pypy-commit] pypy test-cpyext: hg merge default rlamy
[pypy-commit] pypy test-cpyext: Inline create_so() rlamy
[pypy-commit] pypy test-cpyext: kill w_record_imported_module and unskip test_recursive_package_import() on -A rlamy
[pypy-commit] pypy test-cpyext: Make compile_extension_module() a method of sys_info rlamy
[pypy-commit] pypy test-cpyext: Move implementation of load_module to SystemCompilationInfo subclasses rlamy
[pypy-commit] pypy test-cpyext: Move import_extension impl to sys_info rlamy
[pypy-commit] pypy test-cpyext: Move import_module impl to sys_info rlamy
[pypy-commit] pypy test-cpyext: rm debugging leftover rlamy
[pypy-commit] pypy test-cpyext: Simplify import_extension() by extracting make_methods() rlamy
[pypy-commit] pypy test-cpyext: Simplify import_module(): kill load_it param and remove duplication with reimport_module() rlamy
[pypy-commit] pypy test-cpyext: Simplify load_it=False logic in import_module() rlamy
[pypy-commit] pypy test-cpyext: Stop using StdObjSpace and FakeSpace for -A tests rlamy
[pypy-commit] pypy union-side-effects: Add an elementary (but failing) hypothesis test for union() rlamy
[pypy-commit] pypy union-side-effects: Add test for constants rlamy
[pypy-commit] pypy union-side-effects: Create union() as a side-effect-free binary function to replace unionof() wherever possible rlamy
[pypy-commit] pypy union-side-effects: Disallow unions of char and unichar, since they make the annotator inconsistent rlamy
[pypy-commit] pypy union-side-effects: Expand hypothesis testing of union() until it fails rlamy
[pypy-commit] pypy union-side-effects: Explicitly forbid comparisons between str and unicode rlamy
[pypy-commit] pypy union-side-effects: Fix str vs unicode correctness issues rlamy
[pypy-commit] pypy union-side-effects: Fix union() and give a correct explanation for the s1==s2 case rlamy
[pypy-commit] pypy union-side-effects: hg merge default rlamy
[pypy-commit] pypy union-side-effects: Revert change to run_formatter() and fix unicode vs str issues. rlamy
[pypy-commit] pypy union-side-effects: Use more general strategy in 2 tests rlamy
[pypy-commit] pypy union-side-effects: Use union() instead of unionof() in a few places rlamy
[pypy-commit] pypy value-classes: _value_class_=True implies _immutable_=True for now sbauman
[pypy-commit] pypy value-classes: Add some missing _immutable_fields_ annotations for BuiltinCode objects sbauman
[pypy-commit] pypy value-classes: Add test case where a virtualized value class is forced sbauman
[pypy-commit] pypy value-classes: Add test for case of Virtual/ConstPtr (failing currently) sbauman
[pypy-commit] pypy value-classes: Add value_class information to SizeDescr sbauman
[pypy-commit] pypy value-classes: Attempted support for unpacking constant pointers sbauman
[pypy-commit] pypy value-classes: Catch edge case sbauman
[pypy-commit] pypy value-classes: Change is_immutable to is_value_class sbauman
[pypy-commit] pypy value-classes: Cleanup sbauman
[pypy-commit] pypy value-classes: Clean up for virtualstate.py sbauman
[pypy-commit] pypy value-classes: Convert _immutable_ to _value_class_ sbauman
[pypy-commit] pypy value-classes: Fixes for unpacking const pointers sbauman
[pypy-commit] pypy value-classes: Fix handling of intbounds in virtuastate sbauman
[pypy-commit] pypy value-classes: Fix some JIT tests sbauman
[pypy-commit] pypy value-classes: Fix some translation errors sbauman
[pypy-commit] pypy value-classes: Fix test sbauman
[pypy-commit] pypy value-classes: Fix tests sbauman
[pypy-commit] pypy value-classes: GcStruct support for value_class hint sbauman
[pypy-commit] pypy value-classes: Generate fieldboxes and faux virtual state sbauman
[pypy-commit] pypy value-classes: Get simple test working sbauman
[pypy-commit] pypy value-classes: Guard recursion in generate_guards sbauman
[pypy-commit] pypy value-classes: Kill dead rop function sbauman
[pypy-commit] pypy value-classes: Make a note and undo some changes to tests sbauman
[pypy-commit] pypy value-classes: Prospective test sbauman
[pypy-commit] pypy value-classes: Remove dead code sbauman
[pypy-commit] pypy value-classes: Remove old comments sbauman
[pypy-commit] pypy value-classes: Replace all _immutable_=True annotations with _immutable_fields_=[...] sbauman
[pypy-commit] pypy value-classes: Start working on _value_class_ annotation sbauman
[pypy-commit] pypy value-classes: Stop gap for dealing with Virtual/ConstPtr case sbauman
[pypy-commit] pypy value-classes: Work on merging virtual and non-virtual states sbauman
[pypy-commit] pypy zarch-simd-support: add new files to detect features on zarch + basic vector test plan_rich
[pypy-commit] pypy zarch-simd-support: add runtime check to detect vector facility plan_rich
[pypy-commit] pypy zarch-simd-support: a nearly complete vector backend for s390x plan_rich
[pypy-commit] pypy zarch-simd-support: close branch plan_rich
[pypy-commit] pypy zarch-simd-support: copied over vector_ext file, some minor modifications plan_rich
[pypy-commit] pypy zarch-simd-support: do not generate a new operation for index calc if it can be expressed as a single operation plan_rich
[pypy-commit] pypy zarch-simd-support: implemented all vectro resops for s390x plan_rich
[pypy-commit] pypy zarch-simd-support: int_is_true fixed plan_rich
[pypy-commit] pypy zarch-simd-support: load, store, add, subtract vector versions plan_rich
[pypy-commit] pypy zarch-simd-support: more changes to get first tests to fail at assembly plan_rich
[pypy-commit] pypy zarch-simd-support: more vec op backend impl. missing, pack/unpack for i/f plan_rich
[pypy-commit] pypy zarch-simd-support: more vector ops, add, subtract, logic or, and, xor, float mul, float div plan_rich
[pypy-commit] pypy zarch-simd-support: more vector ops, int->float, comparison float plan_rich
[pypy-commit] pypy zarch-simd-support: some more pack/unpack cases implemented plan_rich
[pypy-commit] pypy zarch-simd-support: start new branch for zarch simd plan_rich

Last message date: Fri Sep 30 20:33:22 EDT 2016
Archived on: Fri Sep 30 20:33:30 EDT 2016

Messages sorted by: [ thread ] [ author ] [ date ]
More info on this list...

This archive was generated by Pipermail 0.09 (Mailman edition).