April 2016 Archives by author
Starting: Fri Apr 1 02:39:51 EDT 2016
Ending: Sat Apr 30 22:51:13 EDT 2016
Messages: 676
- [pypy-commit] pypy get-heap-stats: (fijal, catalin) - started working on better way to dump heap stats
JohnDoe
- [pypy-commit] pypy numpy_broadcast: Implemented W_Broadcast for numpy.broadcast
Sergey Matyunin
- [pypy-commit] pypy numpy_broadcast: Fixed compiling of W_Broadcast
Sergey Matyunin
- [pypy-commit] pypy numpy_broadcast: In W_Broadcast (micronumpy) added check for number of arguments, added numiter property
Sergey Matyunin
- [pypy-commit] pypy numpy_broadcast_nd: In W_Broadcast (micronumpy) implemented nd attribute
Sergey Matyunin
- [pypy-commit] pypy numpy_broadcast_nd: W_Broadcast (micronumpy) is rewritten using W_FlatIterator for implementation of iters attribute. W_FlatIterator gets optional arguments in constructor.
Sergey Matyunin
- [pypy-commit] pypy numpy_broadcast_nd: Implemented reset for numpy broadcast object.
Sergey Matyunin
- [pypy-commit] pypy numpy_broadcast_nd: fixed compilation error for W_Broadcast
Sergey Matyunin
- [pypy-commit] cffi default: Change get_current_ts to always use _Py_atomic_load_relaxed when available.
Michael McGee
- [pypy-commit] pypy new-jit-log: (diana, richard) added tests for jitlog to encode debug_merge_point and also the implementation
Diana Popa
- [pypy-commit] pypy default: remove dependency on get_rpy_roots in vmprof and maintain a weaklist of
Raemi
- [pypy-commit] stmgc default: prefix some accidentially exposed functions
Raemi
- [pypy-commit] pypy stmgc-c8: remove dependency on get_rpy_roots in vmprof and maintain a weaklist of
Raemi
- [pypy-commit] pypy stmgc-c8: import stmgc
Raemi
- [pypy-commit] pypy default: typo
Raemi
- [pypy-commit] pypy stmgc-c8: enable vmprof and nearly pass the tests
Raemi
- [pypy-commit] pypy default: fix test
Raemi
- [pypy-commit] pypy stmgc-c8: make jit backend test for vmprof pass. however, we are still missing resetting
Raemi
- [pypy-commit] pypy default: try to fix this test that previously used to enumerate all code objs by tracing
Raemi
- [pypy-commit] stmgc default: add mem_reset_on_abort, a mechanism to reset some memory to the value it had on
Raemi
- [pypy-commit] pypy stmgc-c8: import stmgc
Raemi
- [pypy-commit] pypy stmgc-c8: tiny fix
Raemi
- [pypy-commit] pypy stmgc-c8: missing renaming to stm_spinlock
Raemi
- [pypy-commit] pypy stmgc-c8: attempt to fix JIT support for vmprof for STM by resetting the thread-local
Raemi
- [pypy-commit] stmgc default: avoid data race with vmprof where *during* an abort a SIGPROF signal gets
Raemi
- [pypy-commit] pypy stmgc-c8: import stmgc
Raemi
- [pypy-commit] pypy stmgc-c8: ignore signal if we are in the process of aborting a tx in this thread
Raemi
- [pypy-commit] pypy stmgc-c8: make perf map work for "perf top"
Raemi
- [pypy-commit] pypy stmgc-c8: improve perf disassemble script
Raemi
- [pypy-commit] pypy stmgc-c8: improve perf script some more
Raemi
- [pypy-commit] pypy default: cpyext: Move header logic to .h files, and remove the #ifdef guards, these _decl.h files are truly internal.
amauryfa
- [pypy-commit] pypy default: cpyext: Fix the signature of PyMember_GetOne, it takes a const char*
amauryfa
- [pypy-commit] pypy ast-arena: Add an 'Arena' object to the constructor of all AST nodes.
amauryfa
- [pypy-commit] pypy ast-arena: Some progress. The interpret() tests still don't pass.
amauryfa
- [pypy-commit] cffi default: Implement backend.gcp() for the ctypes backend, and remove gc_weakref.
amauryfa
- [pypy-commit] cffi default: Add ffi.gc(ptr, None) which *removes* the destructor in-place on a ffi.gc() object.
amauryfa
- [pypy-commit] cffi default: Add my employer in AUTHORS
amauryfa
- [pypy-commit] pypy py3.5: Hack until the objspace can start and pass some tests.
amauryfa
- [pypy-commit] pypy py3.5: Fix for class definition inside a function.
amauryfa
- [pypy-commit] pypy py3k: Remove unused import, which causes an import loop in py3.5.
amauryfa
- [pypy-commit] pypy default: Add PyDateTimeAPI->TZInfoType. Because it's easy
amauryfa
- [pypy-commit] pypy default: Define PyDateTime_TZInfo struct, and the corresponding Check() functions.
amauryfa
- [pypy-commit] extradoc extradoc: start my talk
antocuni
- [pypy-commit] extradoc extradoc: add some slides
antocuni
- [pypy-commit] extradoc extradoc: more slides
antocuni
- [pypy-commit] extradoc extradoc: more slides
antocuni
- [pypy-commit] extradoc extradoc: fix makefile
antocuni
- [pypy-commit] extradoc extradoc: split into two slides
antocuni
- [pypy-commit] extradoc extradoc: generate talk.pdf
antocuni
- [pypy-commit] extradoc extradoc: write the slides about trace trees
antocuni
- [pypy-commit] extradoc extradoc: regenerate pdf
antocuni
- [pypy-commit] extradoc extradoc: add an XXX
antocuni
- [pypy-commit] pypy resource_warning: move the logic from __del__ to destructor, because we are not allowed to call arbitrary code from the interp-level __del__
antocuni
- [pypy-commit] pypy resource_warning: (antocuni, florinpapa): fix the test, it was because we returned None from space.format_traceback and the following appexec in space.resource_warning got confused
antocuni
- [pypy-commit] pypy resource_warning: (antocuni, florinpapa): make the test better by checking the output with a regexp
antocuni
- [pypy-commit] pypy resource_warning: (antocuni, florinpapa): make the test even better
antocuni
- [pypy-commit] pypy resource_warning: don't import re, it's not needed
antocuni
- [pypy-commit] pypy resource_warning: enable resource-tracking for socket objects
antocuni
- [pypy-commit] pypy resource_warning: improve the test and check that we don't get the warning if we explicitly close the file
antocuni
- [pypy-commit] pypy resource_warning: failing test and corresponding fix
antocuni
- [pypy-commit] pypy resource_warning: don't show the 'anonymous' frame corresponding to the appexec when calling format_traceback
antocuni
- [pypy-commit] pypy resource_warning: don't crash if we enable track_resources at the early startup (in particular, if we pass the -X track-resources option). It is a bit impossible to write an app-level test for it because we always have a bottom frame in that case
antocuni
- [pypy-commit] pypy resource_warning: exit if you don't specify the correct -X option; in theory you should continue and put it in sys._Xoptions, but since it's not implemented it's better to just exit for now
antocuni
- [pypy-commit] pypy default: Add more tests
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: remove unused dict
arigo
- [pypy-commit] pypy default: Add a warning to the docstring
arigo
- [pypy-commit] pypy default: Another translation fix attempt
arigo
- [pypy-commit] pypy jit-constptr-2: x86-32 support
arigo
- [pypy-commit] pypy jit-constptr-2: Kill this test, supersceded by tests in test_rewrite
arigo
- [pypy-commit] pypy jit-constptr-2: Boehm support: the custom tracer won't work there, and the memory may
arigo
- [pypy-commit] pypy jit-constptr-2: tweak
arigo
- [pypy-commit] pypy jit-constptr-2: (untested) ARM support
arigo
- [pypy-commit] pypy jit-constptr-2: fix(?)
arigo
- [pypy-commit] pypy jit-constptr-2: Test fixes, including moving the two tests about pinned objects to
arigo
- [pypy-commit] extradoc extradoc: Add the current status of the piratepad
arigo
- [pypy-commit] pypy default: (fijal, arigo)
arigo
- [pypy-commit] pypy cleanup-history-rewriting: (fijal, arigo)
arigo
- [pypy-commit] extradoc extradoc: tweaks
arigo
- [pypy-commit] pypy cleanup-history-rewriting: Reorganize do_residual_call() producing
arigo
- [pypy-commit] pypy cleanup-history-rewriting: fix fix
arigo
- [pypy-commit] pypy cleanup-history-rewriting: ready to merge
arigo
- [pypy-commit] pypy default: hg merge cleanup-history-rewriting
arigo
- [pypy-commit] pypy default: Randomly add this here
arigo
- [pypy-commit] pypy vtune: starting
arigo
- [pypy-commit] pypy vtune: very, very, very minimal vtune binding
arigo
- [pypy-commit] pypy vtune: send vtune info only after we patch the generated code's conditional jumps
arigo
- [pypy-commit] extradoc extradoc: misc stuff
arigo
- [pypy-commit] pypy jit-constptr-2: Kill _record_constptrs, no longer used
arigo
- [pypy-commit] pypy jit-constptr-2: ready to merge
arigo
- [pypy-commit] pypy default: hg merge jit-constptr-2
arigo
- [pypy-commit] pypy default: Document branches
arigo
- [pypy-commit] extradoc extradoc: reading the intel optimization manual
arigo
- [pypy-commit] pypy default: Fix: this logic can result in an external call whose stderr is completely
arigo
- [pypy-commit] extradoc extradoc: more
arigo
- [pypy-commit] pypy fix-jitlog: close branch
arigo
- [pypy-commit] pypy default: (edd, palecsandru): Fixed the shown offset values in the jitlog, add a test
arigo
- [pypy-commit] pypy default: mention branch
arigo
- [pypy-commit] pypy default: fix ppc for jit-constptr-2
arigo
- [pypy-commit] pypy default: Print more info, to end up in the captured stdout. For the armhf test failure,
arigo
- [pypy-commit] pypy default: Windows fix: unicode chars are size 2.
arigo
- [pypy-commit] pypy default: Makes the test run, at least
arigo
- [pypy-commit] pypy default: Improve test_seek, and try to make it not fail on windows.
arigo
- [pypy-commit] pypy default: "fix"
arigo
- [pypy-commit] pypy default: Rename argument according to docstring
arigo
- [pypy-commit] pypy default: oops. probable fix. will try to make a unit test
arigo
- [pypy-commit] pypy default: Direct test for load_from_gc_table.
arigo
- [pypy-commit] pypy default: phew, restore showing the assembler
arigo
- [pypy-commit] pypy default: Two tests and two related fixes in typeobject (segfaults the translated
arigo
- [pypy-commit] cffi default: Test and fix for converting empty ffi's in embedding mode
arigo
- [pypy-commit] cffi default: ffi.rawstring(), with a minimal interface
arigo
- [pypy-commit] cffi default: Add a test here
arigo
- [pypy-commit] cffi default: Also accept arrays of int8_t or uint8_t, like ffi.string()
arigo
- [pypy-commit] pypy default: Add ffi.rawstring()
arigo
- [pypy-commit] cffi default: Python3 compat
arigo
- [pypy-commit] pypy default: fix for test_file
arigo
- [pypy-commit] pypy default: fix the detection code (probably) for ARM
arigo
- [pypy-commit] cffi default: Remove again ffi.rawstring(), and implement instead ffi.unpack().
arigo
- [pypy-commit] cffi default: There is no reason to restrict ffi.unpack() to primitives.
arigo
- [pypy-commit] cffi default: Oops. In this case, unpack(p) returns a list of <cdata struct> that
arigo
- [pypy-commit] pypy default: Implement ffi.unpack(); the fast-paths are coming next
arigo
- [pypy-commit] pypy default: Fast paths for unpack()
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy cpyext-ext: A failing test
arigo
- [pypy-commit] pypy cpyext-ext: Raise a SystemError when the situation described in the test happens.
arigo
- [pypy-commit] pypy cpyext-ext: Call make_typedescr() in complexobject.py too. This fixes the existing
arigo
- [pypy-commit] cffi default: update the version number to 1.6
arigo
- [pypy-commit] cffi default: Mention ffi.unpack(), full doc later
arigo
- [pypy-commit] pypy default: update the version number of cffi to 1.6.0 and import the current
arigo
- [pypy-commit] pypy default: was missing: ffi.list_types()
arigo
- [pypy-commit] cffi default: Issue #254: extern "Python+C"
arigo
- [pypy-commit] cffi default: Still delay this a bit longer
arigo
- [pypy-commit] pypy default: import cffi/5d45c8d8df09 and add missing file
arigo
- [pypy-commit] pypy default: Rewrite resource.py to use cffi instead of ctypes_config_cache
arigo
- [pypy-commit] pypy default: port _pypy_wait
arigo
- [pypy-commit] pypy default: Kill ctypes_config_cache
arigo
- [pypy-commit] pypy default: List _resource_cffi here
arigo
- [pypy-commit] pypy cpyext-ext: Failing test
arigo
- [pypy-commit] pypy cpyext-ext: Add another test in cpyext, and fix, including to
arigo
- [pypy-commit] pypy cpyext-ext: Found out that we have space.issequence_w(): move the logic from the
arigo
- [pypy-commit] cffi default: Support help(lib.foo)
arigo
- [pypy-commit] pypy default: Update to cffi/eeef3869b994
arigo
- [pypy-commit] pypy default: import this too
arigo
- [pypy-commit] pypy default: c_forkpty() must have the same "raw" declaration as c_fork().
arigo
- [pypy-commit] pypy default: Mention the issue in the crash message
arigo
- [pypy-commit] pypy default: Uh, the test was not doing anything
arigo
- [pypy-commit] pypy default: PyEval_ThreadsInitialized(): don't return always 1
arigo
- [pypy-commit] cffi default: update
arigo
- [pypy-commit] pypy default: PyPyGILState_Ensure() is not fixed in default. It is only fixed in
arigo
- [pypy-commit] pypy.org extradoc: update the hashes for pypy 5.1.0
arigo
- [pypy-commit] pypy default: Stop early if $branchname or $tagname are invalid
arigo
- [pypy-commit] pypy default: Issue #2277: only special-case two exact lists, not list subclasses,
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy cpyext-ext: Replace empty dict with NULL
arigo
- [pypy-commit] buildbot default: update the buildbot location for bencher4
arigo
- [pypy-commit] pypy default: list libgc-dev (it used to be listed somewhere, but was forgotten)
arigo
- [pypy-commit] cffi default: Move the FFI Interface and Conversion reference sections to their own page.
arigo
- [pypy-commit] cffi default: ffi.unpack(), extern "Python+C"
arigo
- [pypy-commit] cffi release-1.6: make release branch
arigo
- [pypy-commit] cffi default: kill unused var
arigo
- [pypy-commit] cffi default: Windows fix: 'struct a' is a great name for being predefined and not reusable
arigo
- [pypy-commit] cffi default: Re-add this hack, removed in 51f1337c9b4c; but this time only in tests
arigo
- [pypy-commit] cffi release-1.6: hg merge default
arigo
- [pypy-commit] cffi release-1.6: md5/sha1
arigo
- [pypy-commit] cffi default: hg merge release-1.6
arigo
- [pypy-commit] pypy default: Merged in techtonik/pypy-1/techtonik/introductionrst-simplify-explanation-abo-1460879168046 (pull request #429)
arigo
- [pypy-commit] pypy default: silence test_whatsnew
arigo
- [pypy-commit] pypy default: ooops, thanks amaury for noticing that PyPy was still using the
arigo
- [pypy-commit] pypy default: port cffi/fce0884a0ff6
arigo
- [pypy-commit] pypy.org extradoc: update the values
arigo
- [pypy-commit] pypy default: Issue #2280: syslog(u'foo')
arigo
- [pypy-commit] pypy default: Change the repr of 'lib.func' to include the full signature
arigo
- [pypy-commit] pypy.org extradoc: update the values
arigo
- [pypy-commit] pypy.org extradoc: update the values
arigo
- [pypy-commit] pypy default: Issue #2226
arigo
- [pypy-commit] pypy default: doc tweaks
arigo
- [pypy-commit] cffi default: Clean up
arigo
- [pypy-commit] cffi default: Lies, this didn't print a list
arigo
- [pypy-commit] cffi default: merge heads
arigo
- [pypy-commit] pypy.org extradoc: update the values
arigo
- [pypy-commit] pypy default: typo (thanks LarstiQ)
arigo
- [pypy-commit] pypy remove-raisingops: A branch to get rid of the FAIL_xxx() in 'src/int.h' and keep
arigo
- [pypy-commit] pypy remove-raisingops: Goal: kill this amount of code. Now I have to make things work again :-)
arigo
- [pypy-commit] pypy remove-objspace-options: Remove 'sharesmallstr'
arigo
- [pypy-commit] pypy remove-objspace-options: cleanups
arigo
- [pypy-commit] pypy default: More test fixes after 490058ea54e6
arigo
- [pypy-commit] pypy remove-objspace-options: Now, if an RPython program uses weakrefs when translation.rweakref is
arigo
- [pypy-commit] pypy remove-objspace-options: Hack: the show() method on Blocks and Links now returns the
arigo
- [pypy-commit] pypy remove-objspace-options: Fix for (at least some) ztranslation tests
arigo
- [pypy-commit] pypy remove-objspace-options: missing import
arigo
- [pypy-commit] pypy remove-objspace-options: Class name changed here
arigo
- [pypy-commit] pypy remove-objspace-options: Fix the assert:
arigo
- [pypy-commit] pypy bitstring: A branch to use a bitstring to replace the lists of read or written
arigo
- [pypy-commit] pypy bitstring: The basic implementation
arigo
- [pypy-commit] pypy bitstring: in-progress
arigo
- [pypy-commit] pypy bitstring: Revert the changes here. Must be done more lazily in order to figure
arigo
- [pypy-commit] pypy bitstring: Compute the bitstrings
arigo
- [pypy-commit] pypy bitstring: Port force_from_effectinfo()
arigo
- [pypy-commit] pypy bitstring: More fixes. Remove the CUTOFF in backendopt/writeanalyze.
arigo
- [pypy-commit] pypy bitstring: Fix: this goes outside the other loop
arigo
- [pypy-commit] pypy bitstring: test fix
arigo
- [pypy-commit] pypy bitstring: Comments and tweaks
arigo
- [pypy-commit] pypy remove-raisingops: More kills
arigo
- [pypy-commit] pypy remove-raisingops: progress, and goal
arigo
- [pypy-commit] pypy remove-raisingops: Port the logic to rint.py
arigo
- [pypy-commit] pypy default: Test and fix for unicode_dealloc().
arigo
- [pypy-commit] pypy bitstring: Must capture 'metainterp_sd.all_descrs' really at the end of
arigo
- [pypy-commit] pypy bitstring: Make sure we don't build new EffectInfo instances after
arigo
- [pypy-commit] pypy default: Merged in devin.jeanpierre/pypy-macros (pull request #435)
arigo
- [pypy-commit] pypy bitstring: Dump logs about the bitstring compression
arigo
- [pypy-commit] pypy bitstring: ready to merge
arigo
- [pypy-commit] pypy default: hg merge bitstring
arigo
- [pypy-commit] pypy default: document branch
arigo
- [pypy-commit] pypy default: uh
arigo
- [pypy-commit] pypy default: Document the AttributeError/TypeError difference with CPython about
arigo
- [pypy-commit] pypy remove-raisingops: test fix
arigo
- [pypy-commit] pypy remove-raisingops: a remaining direct usage of int_add_ovf
arigo
- [pypy-commit] pypy remove-raisingops: fix
arigo
- [pypy-commit] pypy remove-raisingops: fix test
arigo
- [pypy-commit] pypy default: Issue #2284: fix urls
arigo
- [pypy-commit] pypy remove-raisingops: Revert the changes in flatten.py, and handle it by jtransform'ing
arigo
- [pypy-commit] pypy default: I'm really often hitting this issue where I get
arigo
- [pypy-commit] pypy default: Add comments
arigo
- [pypy-commit] pypy default: fix: operator.attrgetter("name", not_a_string) would raise a confusing
arigo
- [pypy-commit] stmgc default: mention the status with gcc 6.1
arigo
- [pypy-commit] stmgc default: Note that we *should* not need "-fno-*" but didn't really check so far
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: Share the wrapper logic among all functions with the same signature
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: First try
arigo
- [pypy-commit] stmgc default: Point the default to another demo, because demo2 is failing right now
arigo
- [pypy-commit] pypy.org extradoc: Rephrase
arigo
- [pypy-commit] pypy default: Detail
arigo
- [pypy-commit] pypy.org extradoc: update the values
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: translation fixes (in-progress)
arigo
- [pypy-commit] pypy default: More precise info about why this segfaults on CPython
arigo
- [pypy-commit] pypy default: Simplify stuff
arigo
- [pypy-commit] pypy default: Test and fix for yet another very obscure misfeature of ctypes
arigo
- [pypy-commit] pypy default: The bogus casting is not needed any more after 9f9c409ee27e
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: more translation fighting
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: hg merge default
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: mess in progress
arigo
- [pypy-commit] pypy default: RPython: using "is" to compare functions does not really work
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: Give up on this test, which is also half-not-passing in default
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: More result_is_ll
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: more translation fixes
arigo
- [pypy-commit] pypy share-cpyext-cpython-api: hg merge default
arigo
- [pypy-commit] pypy detect_cpu_count: (catalin_m): Updated CPU detection code in rpython/config/support.py. Now returns the actual number of available cores. Added a non-linux implementation as well. Updated test.
catalin_m
- [pypy-commit] pypy detect_cpu_count: (catalinm) detect_number_of_processors now uses multiprocessing.cpu_count as default behavior, the initial implementation is now just a fallback. Modified test_cpuinfo_linux to properly check its functionality. Modified the cpuinfo string to check regex for more than 10 cpus.
catalin_m
- [pypy-commit] pypy detect_cpu_count: (catalin_m) Renamed support.detect_number_of_processors to support.detect_number_of_usable_processors to (somewhat) avoid confusion. Put back the usable cpu counting logic (cpu_count / 2 for >= 4 cpus with a minimum of 3, cpu_count otherwise). Now applies to BSD and MACOS as well. Moved the logic to support.usable_processors_from_total_processors(total) and added tests for it. Changed tests to reflect changes in the support module
catalin_m
- [pypy-commit] pypy default: try to reduce the size of concrete syntax tree nodes
cfbolz
- [pypy-commit] pypy default: don't store line and column on Nonterminal, introduce a special one-child
cfbolz
- [pypy-commit] pypy default: don't suggest the hybrid GC
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove objspace.std.withrangelist option, just use objspace.std.withliststrategies
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove the withprebuiltchar option and make it part of sharesmallstr
cfbolz
- [pypy-commit] pypy remove-objspace-options: if we have no rweakref, make classes uncollectable as opposed to not supporting
cfbolz
- [pypy-commit] pypy remove-objspace-options: forgot to fix this
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove the withtypeversion option and turn it on by default
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove withmethodcache option and have it always be on
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove option docs
cfbolz
- [pypy-commit] pypy remove-objspace-options: again, weakrefs no longer needed
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove withidentitydict option and make that always on
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove some vestiges of withsmalldicts(!) in mocking code
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove getattributeshortcut option and turn it on by default
cfbolz
- [pypy-commit] pypy remove-objspace-options: move the tests for now on-by-default identity tracking to test_typeobject
cfbolz
- [pypy-commit] pypy remove-objspace-options: remove the withmapdict option and turn it on by default
cfbolz
- [pypy-commit] pypy remove-objspace-options: use mapdict for all the subclassing
cfbolz
- [pypy-commit] pypy remove-objspace-options: problems that the enabling mapdict by default found:
cfbolz
- [pypy-commit] pypy remove-objspace-options: the assert was too strong
cfbolz
- [pypy-commit] pypy remove-objspace-options: various fixes for translation and tests
cfbolz
- [pypy-commit] pypy remove-objspace-options: another withmethodcache
cfbolz
- [pypy-commit] pypy remove-objspace-options: (cfbolz, arigo advising) test the cutoff and make it less random
cfbolz
- [pypy-commit] pypy remove-objspace-options: (cfbolz, arigo advising) somewhat randomly raise the cutoff to 3000 to make
cfbolz
- [pypy-commit] pypy remove-objspace-options: even more different (comes from a different module now)
cfbolz
- [pypy-commit] pypy remove-objspace-options: document branch
cfbolz
- [pypy-commit] pypy remove-objspace-options: close to-be-merged branch
cfbolz
- [pypy-commit] pypy default: merge remove-objspace-options
cfbolz
- [pypy-commit] pypy default: merge heads
cfbolz
- [pypy-commit] pypy share-mapdict-methods: don't put anything into __init__.py
cfbolz
- [pypy-commit] pypy share-mapdict-methods: - try to reduce duplication of methods by putting a lot of the methods that
cfbolz
- [pypy-commit] pypy py3k-update: merge e35996d1c1b6
cfbolz
- [pypy-commit] pypy share-mapdict-methods: fix problems with __del__: only make a single RPython subclass if the base
cfbolz
- [pypy-commit] pypy share-mapdict-methods: seems it's wrong to share deldictvalue, for somewhat involved reasons:
cfbolz
- [pypy-commit] pypy default: remove the weird _become method
cfbolz
- [pypy-commit] pypy default: be more consistent in naming
cfbolz
- [pypy-commit] pypy share-mapdict-methods-2: new attempt: try to share methods by simply sticking the same function objects
cfbolz
- [pypy-commit] pypy share-mapdict-methods-2: don't inline 5 fields into everything, just into old-style classes and
cfbolz
- [pypy-commit] pypy share-mapdict-methods-2: fix problems with __del__: only make a single RPython subclass if
cfbolz
- [pypy-commit] pypy default: merge default
cfbolz
- [pypy-commit] pypy share-mapdict-methods: close abandoned branch
cfbolz
- [pypy-commit] pypy share-mapdict-methods: tighten test
cfbolz
- [pypy-commit] pypy default: merge share-mapdict-methods-2
cfbolz
- [pypy-commit] pypy default: Make attrgetter a single type -- this brings PyPy in line with CPython.
devin.jeanpierre
- [pypy-commit] pypy default: Correctly leak replaced object in PyList_SET_ITEM.
devin.jeanpierre
- [pypy-commit] pypy default: Move decref somewhere less worrisome.
devin.jeanpierre
- [pypy-commit] pypy default: Restore PyList_GET_ITEM. Oops!
devin.jeanpierre
- [pypy-commit] pypy default: Correct docstring for PyList_SET_ITEM.
devin.jeanpierre
- [pypy-commit] pypy default: Correct the return type and documentation of PyList_SET_ITEM.
devin.jeanpierre
- [pypy-commit] pypy default: Compile with -Werror in cpyext tests.
devin.jeanpierre
- [pypy-commit] pypy default: merge
devin.jeanpierre
- [pypy-commit] pypy default: Undo 0f0ab7f / e648ab7 I think/hope.
devin.jeanpierre
- [pypy-commit] pypy cpyext-werror: Compile with -Werror in cpyext tests.
devin.jeanpierre
- [pypy-commit] pypy default: an attempt to fix translation
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: backout the merge of the branch we need to think more about
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] benchmarks default: add the ability to number the counts
fijal
- [pypy-commit] benchmarks default: improve
fijal
- [pypy-commit] extradoc extradoc: write a blog post
fijal
- [pypy-commit] extradoc extradoc: merge
fijal
- [pypy-commit] pypy default: we use enum34 not enum
fijal
- [pypy-commit] pypy default: fix a strange case
fijal
- [pypy-commit] pypy default: fix strange annotation errors for small targets
fijal
- [pypy-commit] extradoc extradoc: start working on tooling stability doc
fijal
- [pypy-commit] extradoc extradoc: merge
fijal
- [pypy-commit] pypy default: kill the depthmap calculation - it's unused by anything and takes memory. Leave the frame size calculations, so we can assert things (why not)
fijal
- [pypy-commit] pypy default: Merged in vasanthaganeshk/pypy (pull request #433)
fijal
- [pypy-commit] pypy resource_warning: (florin, antocuni) Add resource warning test for file descriptors
florinpapa
- [pypy-commit] pypy resource_warning: (florin, antocuni) IN-PROGRESS: Attach and show the creation traceback. Test fails and we don't know why
florinpapa
- [pypy-commit] pypy resource_warning: (florin, antocuni) Rename resource_warning to track_resources
florinpapa
- [pypy-commit] pypy resource_warning: (florin, antocuni) Add get_track_resources to retrieve flag value
florinpapa
- [pypy-commit] pypy resource_warning: (florin, antocuni) Add -X track-resources runtime flag to the interpreter
florinpapa
- [pypy-commit] pypy resource_warning: Merge heads
florinpapa
- [pypy-commit] pypy py3.3-hashfix: Use intobject hash function for specialisedtuple
kunalgrover05
- [pypy-commit] pypy py3.3-hashfix: Raise NotImplemented when value not an integer or float in specialisedtuple
kunalgrover05
- [pypy-commit] pypy py3.3-hashfix: Added test
kunalgrover05
- [pypy-commit] pypy py3.3-hashfix: Minor fix
kunalgrover05
- [pypy-commit] pypy 33_fix_itertools: Work on adding pickling support for the itertools classes. (Doesn't work yet for product)
marky1991
- [pypy-commit] pypy 33_fix_itertools: Fixed it in pyinteractive (and fixed a correctness issue), but translation is still broken.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Fixed translation.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Fix itertools failures.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Get rid of some unnecessary changes.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Get rid of some unnecessary changes.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Oops. Didn't mean to commit those changes to test_itertools. Backing that out.
marky1991
- [pypy-commit] pypy 33_fix_itertools: That 'unnecessary' change was required to make translation not fail. Add it back.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Add newlines to match PEP8 and as requested.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Undo unnecessary name change.
marky1991
- [pypy-commit] pypy py3k: Merged in marky1991/pypy_new/33_fix_itertools (pull request #437)
marky1991
- [pypy-commit] pypy 33_fix_itertools: Fix the islice test.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Sort of fix the remaining itertools tests. Translation is currently broken and I had to use a lot of isinstance assertions to make the annotator happy. Committing to ask about it.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Fix translation. (Thanks ronan)
marky1991
- [pypy-commit] pypy 33_fix_itertools: Got rid of unnecessary space.wrap calls and was then able to get rid of two of the three type assertions.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Get rid of all of the changes in test_itertools I never meant to actually commit.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Fix infinite loop bug in reduce.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Fix final two failing tests.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Get rid of out-of-date comment.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Respond to review.
marky1991
- [pypy-commit] pypy 33_fix_itertools: Use support.gc_collect instead of regular gc.collect
marky1991
- [pypy-commit] pypy cpyext-ext: test some list methods
mattip
- [pypy-commit] pypy cpyext-ext: refactor error checking to make sure the strategy is correct
mattip
- [pypy-commit] pypy cpyext-ext: add missing methods, randomly revert some to ObjectStrategy when called, raise on the rest
mattip
- [pypy-commit] pypy cpyext-ext: add missing methods, test
mattip
- [pypy-commit] pypy cpyext-ext: fix strategy for zip()
mattip
- [pypy-commit] pypy cpyext-ext: test, implement more strategy methods, remove methods implemented in base class
mattip
- [pypy-commit] pypy cpyext-ext: finish all CPyListStrategy methods
mattip
- [pypy-commit] pypy cpyext-ext: typo
mattip
- [pypy-commit] pypy cpyext-ext: guess a bit why test_tp_new_in_subclass_of_type recurses?
mattip
- [pypy-commit] pypy stdlib-2.7.11: merge default into branch
mattip
- [pypy-commit] pypy cleanup-includes: refactor creation of macro definitions and remove numpy includes
mattip
- [pypy-commit] pypy cleanup-includes: remove outdated headers
mattip
- [pypy-commit] pypy cleanup-includes: use constants from micronumpy rather than redefining them
mattip
- [pypy-commit] pypy cleanup-includes: add parts of headers removed in 4518b83c9ee2, fixes for testing
mattip
- [pypy-commit] pypy cleanup-includes: if a header is specified, put the declarations and macros in that header
mattip
- [pypy-commit] pypy cleanup-includes: move ndarray declarations to seperate header, adjust header creation
mattip
- [pypy-commit] pypy cleanup-includes: typo
mattip
- [pypy-commit] pypy cleanup-includes: copy all generated headers
mattip
- [pypy-commit] pypy cleanup-includes: merge default into branch
mattip
- [pypy-commit] pypy cleanup-includes: fix merge
mattip
- [pypy-commit] pypy cpyext-ext: actually call the new fcn (fijal)
mattip
- [pypy-commit] pypy release-5.x: merge default into release
mattip
- [pypy-commit] pypy release-5.x: update version to 5.1
mattip
- [pypy-commit] pypy default: start release 5.1
mattip
- [pypy-commit] pypy default: restart whatsnew
mattip
- [pypy-commit] pypy cleanup-includes: close branch to be merged
mattip
- [pypy-commit] pypy default: merge cleanup-includes which improves micronumpy header status
mattip
- [pypy-commit] pypy release-5.x: merge default into branch
mattip
- [pypy-commit] pypy default: document merged branch
mattip
- [pypy-commit] pypy default: restart whatsnew-head
mattip
- [pypy-commit] pypy release-5.x: merge default into branch
mattip
- [pypy-commit] pypy default: update binary version
mattip
- [pypy-commit] pypy default: document changes between 5.0 and this release
mattip
- [pypy-commit] pypy cpyext-ext: remove debug cruft
mattip
- [pypy-commit] pypy cpyext-ext: make fooc3 more like numpy's PyCDoubleArrType_Type
mattip
- [pypy-commit] pypy cpyext-ext: cleanup
mattip
- [pypy-commit] pypy cpyext-ext: avoid segfault in -A tests, still lookiong for why this happens
mattip
- [pypy-commit] pypy cpyext-ext: remove printing
mattip
- [pypy-commit] pypy cpyext-ext: improve debug output when WARN_ABOUT_MISSING_SLOT_FUNCTIONS is True
mattip
- [pypy-commit] pypy numpy_broadcast: test, implement overflow checking; small cleanups
mattip
- [pypy-commit] pypy numpy_broadcast: merge default into branch
mattip
- [pypy-commit] pypy numpy_broadcast: close branch to be merged
mattip
- [pypy-commit] pypy default: merge branch which provides numpy.broadcast
mattip
- [pypy-commit] pypy default: document merged branches
mattip
- [pypy-commit] pypy default: update contributors
mattip
- [pypy-commit] pypy cpyext-ext: merge default into branch
mattip
- [pypy-commit] pypy py3k: fix for win32
mattip
- [pypy-commit] pypy default: untabbify
mattip
- [pypy-commit] pypy cpyext-ext: merge default into branch
mattip
- [pypy-commit] pypy cpyext-ext: fixes from merge
mattip
- [pypy-commit] buildbot default: ensure cp step will not fail
mattip
- [pypy-commit] pypy cpyext-ext: fix tests for -A
mattip
- [pypy-commit] pypy default: add
mattip
- [pypy-commit] pypy default: update for 5.1 release
mattip
- [pypy-commit] pypy release-5.x: merge default into branch
mattip
- [pypy-commit] pypy cpyext-ext: merge default into branch
mattip
- [pypy-commit] pypy release-5.x: fix merge
mattip
- [pypy-commit] buildbot default: try to fix 11de30c674f6
mattip
- [pypy-commit] pypy default: update documentation for 5.1 release
mattip
- [pypy-commit] pypy cpyext-ext: try harder to remove ndarray macros and decls by adding the header name after # for all GLOBALS
mattip
- [pypy-commit] pypy default: test, fix to not print mandlebrot to log files
mattip
- [pypy-commit] pypy default: add s390x to repackage script
mattip
- [pypy-commit] pypy default: Added tag release-5.1 for changeset 3260adbeba4a
mattip
- [pypy-commit] pypy default: expand header of release note
mattip
- [pypy-commit] pypy.org extradoc: update website download page for 5.1
mattip
- [pypy-commit] buildbot default: deactivate nightly jit-benchmark-linux-x86-64-single-run runs
mattip
- [pypy-commit] pypy cpyext-ext: add defines to post_include_bits when translating and name not in pypy_macros.h
mattip
- [pypy-commit] pypy cpyext-ext: remove debug cruft
mattip
- [pypy-commit] pypy cpyext-ext: typo
mattip
- [pypy-commit] pypy cpyext-ext: fix test, PyThreadState_Get() cannot be called after PyEval_SaveThread since the state is set to NULL
mattip
- [pypy-commit] pypy cpyext-ext: add tests for bytearray, pass on cpython with -A
mattip
- [pypy-commit] pypy cpyext-ext: add PyBytesArrayObject, based on PyStringObject. Still not clear how to make it mutable
mattip
- [pypy-commit] pypy cpyext-ext: remove unused imports
mattip
- [pypy-commit] pypy cpyext-ext: try to summarize IRC conversation
mattip
- [pypy-commit] pypy cpyext-for-merge: prepare cpyext-ext for merging back to default
mattip
- [pypy-commit] pypy cpyext-for-merge: merge default into branch
mattip
- [pypy-commit] pypy cpyext-for-merge: update TODO
mattip
- [pypy-commit] pypy cpyext-ext: update TODO
mattip
- [pypy-commit] pypy cpyext-for-merge: skip 'wishlist' tests
mattip
- [pypy-commit] pypy release-5.x: revert 8d781f7a74f7, fix issue #2282
mattip
- [pypy-commit] pypy release-5.x: update version to 5.1.1
mattip
- [pypy-commit] pypy release-5.x: fix test (revert part of a03329def3ec)
mattip
- [pypy-commit] pypy cpyext-for-merge: merge default into branch
mattip
- [pypy-commit] pypy cpyext-for-merge: skip 'wishlist' test
mattip
- [pypy-commit] pypy release-5.x: tweak headers for new pypy_numpy.h
mattip
- [pypy-commit] pypy cpyext-for-merge: fix or skip runappdirect variations of tests
mattip
- [pypy-commit] pypy cpyext-for-merge: skip runappdirect tests for merge
mattip
- [pypy-commit] pypy cpyext-for-merge: merge default into branch
mattip
- [pypy-commit] pypy cpyext-for-merge: document branch
mattip
- [pypy-commit] pypy default: reference rawrefcount, remove dead link
mattip
- [pypy-commit] pypy cpyext-for-merge: revert 8d781f7a74f7, fix issue #2282
mattip
- [pypy-commit] pypy cpyext-for-merge: fix test (revert part of a03329def3ec)
mattip
- [pypy-commit] pypy cpyext-for-merge: tweak headers for new pypy_numpy.h
mattip
- [pypy-commit] pypy cpyext-for-merge: tweak includes to reduce warnings
mattip
- [pypy-commit] pypy cpyext-for-merge: done in 5c74afbd7d74
mattip
- [pypy-commit] pypy cpyext-for-merge: document the header directory
mattip
- [pypy-commit] pypy cpyext-ext: merge cpyext-for-merge back into branch
mattip
- [pypy-commit] pypy cpyext-ext: revert skipped tests from merge
mattip
- [pypy-commit] pypy cpyext-for-merge: fix for win32 and c90
mattip
- [pypy-commit] pypy cpyext-for-merge: close branch to be merged
mattip
- [pypy-commit] pypy default: merge a major refactoring of parts of cpyext to default
mattip
- [pypy-commit] extradoc extradoc: add a draft blog post
mattip
- [pypy-commit] extradoc extradoc: add a talk
mattip
- [pypy-commit] extradoc extradoc: add pdf
mattip
- [pypy-commit] extradoc extradoc: update
mattip
- [pypy-commit] pypy default: update TODO
mattip
- [pypy-commit] pypy default: document merged branch
mattip
- [pypy-commit] pypy numpy_broadcast_nd: merge default into branch
mattip
- [pypy-commit] pypy numpy_broadcast_nd: add failing test - W_Broadcast.__init__
mattip
- [pypy-commit] pypy numpy_broadcast_nd: try to treat W_Broadcast as true W_NumpyObjects
mattip
- [pypy-commit] pypy cpyext-werror: merge default into branch
mattip
- [pypy-commit] extradoc extradoc: shorten
mattip
- [pypy-commit] pypy py3k: Get rid of tab in Python source file.
mjacob
- [pypy-commit] pypy py3k: 2to3
mjacob
- [pypy-commit] pypy py3k: 2to3
mjacob
- [pypy-commit] pypy py3k: Raise NotImplementedError only when /dev/urandom is not present. Fixes app-level test.
mjacob
- [pypy-commit] pypy default: Add more details on how to update the stdlib files while properly tracking renames.
mjacob
- [pypy-commit] pypy default: (cfbolz, palecsandru): Fixed what's new that was backed out. The test did not know about the backouts.
palecsandru
- [pypy-commit] pypy default: (cfbolz, palecsandru): Fix failing test that was not adapted when the implementation changed.
palecsandru
- [pypy-commit] pypy fix-jitlog: (edd, palecsandru): Fixed the shown offset values in the jitlog
palecsandru
- [pypy-commit] pypy default: fix range check thinko; calculate w/ bits not bytes
pjenvey
- [pypy-commit] pypy py3k: fix range check thinko; calculate w/ bits not bytes
pjenvey
- [pypy-commit] pypy py3k: kill this assert now that 3.3 supports Q (longlong) conversions, for 32bit
pjenvey
- [pypy-commit] pypy py3k: prefer oefmt for new code
pjenvey
- [pypy-commit] pypy default: clarify UINT handling
pjenvey
- [pypy-commit] pypy py3k: clarify UINT handling
pjenvey
- [pypy-commit] pypy py3k: fix for py3k's new convert scheme
pjenvey
- [pypy-commit] pypy py3k: merge default
pjenvey
- [pypy-commit] pypy py3k: reapply test_hash skips from ce0cdf69b07a
pjenvey
- [pypy-commit] pypy py3k: merge kunalgrover05/pypy/py3.3-hashfix (pull request #402)
pjenvey
- [pypy-commit] pypy py3.3-hashfix: close branch
pjenvey
- [pypy-commit] pypy py3k: merge default
pjenvey
- [pypy-commit] pypy py3k: refactor and allow specifying flags
pjenvey
- [pypy-commit] pypy default: fix isinstance(deque(), Hashable) on the pure python deque
pjenvey
- [pypy-commit] pypy py3k: relax some assertions
pjenvey
- [pypy-commit] pypy py3k: translation fixes for osx
pjenvey
- [pypy-commit] pypy py3k: -v/PYTHONVERBOSE & PYTHONCASEOK are now supported thanks to importlib
pjenvey
- [pypy-commit] pypy default: simplify sys_exc_info to return None for a cleared exception and reuse some of
pjenvey
- [pypy-commit] pypy py3k: fix translation on osx, specialize utime when called w/ None
pjenvey
- [pypy-commit] pypy py3k: simplify sys_exc_info to return None for a cleared exception and reuse some of
pjenvey
- [pypy-commit] pypy py3k: fix
pjenvey
- [pypy-commit] pypy py3k: issue1903: remove the lazy __context__ recording potentially getting the wrong
pjenvey
- [pypy-commit] pypy py3k: add a failing test from test_contextlib: we lack WHY_SILENCED
pjenvey
- [pypy-commit] pypy py3k: fix exec from @hidden_applevel functions
pjenvey
- [pypy-commit] pypy py3k: hide run_command_line now that exec works
pjenvey
- [pypy-commit] pypy py3k: globals deserves module=True
pjenvey
- [pypy-commit] pypy default: don't prepend '<builtin>' more than once (not exactly sure how this happens but
pjenvey
- [pypy-commit] pypy py3k: don't prepend '<builtin>' more than once (not exactly sure how this happens but
pjenvey
- [pypy-commit] pypy py3k: adjust per PyCode's changes to frozen modules and app_name='__import__'
pjenvey
- [pypy-commit] pypy default: fix reraise in hidden functions
pjenvey
- [pypy-commit] pypy py3k: fix reraise in hidden functions
pjenvey
- [pypy-commit] pypy py3k: readd for_hidden from default
pjenvey
- [pypy-commit] pypy py3k: minor cleanup
pjenvey
- [pypy-commit] pypy py3k: minor cleanup
pjenvey
- [pypy-commit] pypy default: is_jit_debug takes the integer opnum as argument
plan_rich
- [pypy-commit] pypy new-jit-log: merged default, translation issue
plan_rich
- [pypy-commit] pypy new-jit-log: refactoring the jitlog, inlines encode methods, finish_once of cpu now prints enter count of bridges, entries and loops
plan_rich
- [pypy-commit] pypy new-jit-log: catchup default
plan_rich
- [pypy-commit] pypy new-jit-log: do not release gil for two logger functions. they are called in @rgc.no_release_gil
plan_rich
- [pypy-commit] pypy new-jit-log: another releasegil=False
plan_rich
- [pypy-commit] pypy new-jit-log: try init from env variable rewritten, there is currently no filtering mechanism. filtering does not make sense anymore
plan_rich
- [pypy-commit] pypy new-jit-log: exposed enable_jitlog in module _vmprof, JITLOG env variable now controls the jitlog output file (bugfix), setup_once is now called while initializing the metainterp_sd
plan_rich
- [pypy-commit] pypy new-jit-log: rvmprof.enable_jitlog now correctly writes the header, teardown must still be called correctly
plan_rich
- [pypy-commit] pypy new-jit-log: finish is now called in the testing case
plan_rich
- [pypy-commit] pypy default: catching up with the backend changes load_from_gc_table. stuffed the gc references before the constant pool, which makes it easy to access them using r13
plan_rich
- [pypy-commit] pypy default: wrong parameter to LG command and wrong jump target to bridge
plan_rich
- [pypy-commit] pypy new-jit-log: catchup with default
plan_rich
- [pypy-commit] pypy new-jit-log: use tmpdir instead of tempfile, added a version to the jitlog
plan_rich
- [pypy-commit] pypy new-jit-log: added test to ensure version number is correctly written
plan_rich
- [pypy-commit] pypy new-jit-log: (work in progress) adding a new API to the jit driver to return a rather "generic" list of primitive types that describe the execution state of any meta interpreter
plan_rich
- [pypy-commit] pypy default: (s390x) execute patch_stack_checks before materialize loop, translation issue fixed
plan_rich
- [pypy-commit] pypy default: (s390x) jumping to the wrong address
plan_rich
- [pypy-commit] pypy default: (s390x) store_force_index didnt use the new descr index
plan_rich
- [pypy-commit] pypy new-jit-log: extended the return value of get_location
plan_rich
- [pypy-commit] pypy default: tested & impl pair allocation. there was an edge case which caused the allocation to fail.
plan_rich
- [pypy-commit] pypy default: translation issue resolved
plan_rich
- [pypy-commit] pypy new-jit-log: instead of many, only one syscall to write is executed while initializing the jitlog (even slows down the test execution)
plan_rich
- [pypy-commit] pypy new-jit-log: use _get_vmprof to get a handle on cintf in the logger, this should resolve the translation issue
plan_rich
- [pypy-commit] pypy new-jit-log: solved translation issue
plan_rich
- [pypy-commit] pypy default: (s390x) prevent setting the key none in reg_bindings
plan_rich
- [pypy-commit] pypy default: (s390x) marked two tests to be brittle, they some times fail only on one machine I do not have access to...
plan_rich
- [pypy-commit] pypy new-jit-log: merged default
plan_rich
- [pypy-commit] pypy new-jit-log: renamed variable
plan_rich
- [pypy-commit] pypy default: rewritten test, sleeping up to ~25 second to wait for the initial thread to start
plan_rich
- [pypy-commit] pypy default: (s390x) overwrote already defined variable, lead to wrong assignment
plan_rich
- [pypy-commit] pypy new-jit-log: renamed jitdriver_sd property only in one place
plan_rich
- [pypy-commit] pypy default: use TempInt instead of TempVar, the latter has not type property
plan_rich
- [pypy-commit] buildbot default: added a new builder that I have access to (s390x) own-linux
plan_rich
- [pypy-commit] buildbot default: added own-linux-s390x-2 as a nightly
plan_rich
- [pypy-commit] pypy new-jit-log: forgot to set get_location to pypy's main jitdriver
plan_rich
- [pypy-commit] pypy default: print the progress uppon test failure (test_gil), increased the time it sleeps while checking the progress
plan_rich
- [pypy-commit] pypy new-jit-log: modified ztranslation test to check the get_location and ensure it can be translated
plan_rich
- [pypy-commit] pypy new-jit-log: refactoring simplifies the logging of a trace. now (just for the purpose of the jitlog) there is a unique id for each traces which makes parsing also easier
plan_rich
- [pypy-commit] pypy default: typo in release notes
plan_rich
- [pypy-commit] pypy new-jit-log: working on the API for get_location (nearly done)
plan_rich
- [pypy-commit] pypy new-jit-log: moved .../metainterp/jitlog.py to rlib (since the vm implementer uses the jitlog directly)
plan_rich
- [pypy-commit] pypy new-jit-log: hurrayy! the new get_location function API is now extensible and pypy translates (this took me quite a while)
plan_rich
- [pypy-commit] pypy new-jit-log: reenabled the prefix compression and added a new test to check commonprefix
plan_rich
- [pypy-commit] pypy new-jit-log: more tests stressing the prefix compression in the log
plan_rich
- [pypy-commit] pypy new-jit-log: fixed up the test that integrates the whole jitlog on resoperations
plan_rich
- [pypy-commit] pypy new-jit-log: renamed old module
plan_rich
- [pypy-commit] pypy new-jit-log: fixed up tests
plan_rich
- [pypy-commit] pypy new-jit-log: creating jit logger for base optimization tests
plan_rich
- [pypy-commit] pypy new-jit-log: removed old files, adapted main pypy jit driver
plan_rich
- [pypy-commit] pypy new-jit-log: fixed translation issues
plan_rich
- [pypy-commit] pypy new-jit-log: start_new_trace was called too late, it must be called before the optimizer is invoked. added an ABORT_TRACE marker and encoding it into the jitlog
plan_rich
- [pypy-commit] pypy new-jit-log: setting the metainterp_sd earlier to be able to resolve constptr even for the trace that has not been optimized
plan_rich
- [pypy-commit] pypy new-jit-log: init merge point needs a counter before encoding the types,
plan_rich
- [pypy-commit] pypy new-jit-log: added jitlog_disable to the _vmprof module
plan_rich
- [pypy-commit] pypy new-jit-log: stitching was not written to the log, but it is now
plan_rich
- [pypy-commit] pypy new-jit-log: copy machine code byte by byte instead of converting it to HEX
plan_rich
- [pypy-commit] pypy new-jit-log: fixed tests and added missing method in rvmprof/__init__.py (for disable_jitlog)
plan_rich
- [pypy-commit] pypy.org extradoc: updated the architectures on features.html and regenerated them
plan_rich
- [pypy-commit] pypy.org extradoc: some clarification and added operating systems on which they run (regenerated the page as well)
plan_rich
- [pypy-commit] pypy default: updated the FAQ entry (cpu architectures supported)
plan_rich
- [pypy-commit] pypy.org extradoc: x86 on one line, regenerated source
plan_rich
- [pypy-commit] pypy py3.5-raffael_t: Add @ (matmul) to Grammar and base methodtable, Switch used lib to 3.5
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Add complete Grammar 3.5
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Add simple mmult (@) test
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: AST um @ erweitert
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Add @= token, Set Grammar3.5
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Write matmul method as binary function in ast
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Change to right opcode for matmul
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: (patch plan_rich) Set Python 3.5 as library for module, Comment unfinished ops cmds in astcompiler
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Add missing opcodes to 3.5 lib, Fix error mat_mul -> matrix_multiply
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Define matmul method in module, Add to dict in objspace
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Clean up unused 3.5 opcodes, Define inplace_matmul operator
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Remove undefined Grammar again, so that only @ is new (fix Assertionerror)
raff... at gmail.com
- [pypy-commit] pypy py3.5-raffael_t: Uncomment opscode stacks
raff... at gmail.com
- [pypy-commit] pypy follow_symlinks: hg merge rposix-for-3
rlamy
- [pypy-commit] pypy follow_symlinks: Update execve() docstring
rlamy
- [pypy-commit] pypy follow_symlinks: Separate implementations of execv() and execve() since their signature and validation logic differ
rlamy
- [pypy-commit] pypy follow_symlinks: Allow path as file descriptor in execve()
rlamy
- [pypy-commit] pypy follow_symlinks: Allow specifying path as file descriptor in os.listdir()
rlamy
- [pypy-commit] pypy rposix-for-3: Don't attempt to define posix-only constants on Windows
rlamy
- [pypy-commit] pypy py3k: hg merge default
rlamy
- [pypy-commit] pypy py3k: port test to py3
rlamy
- [pypy-commit] pypy default: Merge branch 'rposix-for-3'
rlamy
- [pypy-commit] pypy default: Update whatsnew
rlamy
- [pypy-commit] pypy py3k: hg merge default
rlamy
- [pypy-commit] pypy py3k: Merge branch 'follow_symlinks'
rlamy
- [pypy-commit] pypy follow_symlinks: Add missing test file
rlamy
- [pypy-commit] pypy py3k: hg merge follow_symlinks
rlamy
- [pypy-commit] pypy rposix-for-3: hg merge default
rlamy
- [pypy-commit] pypy follow_symlinks: hg merge py3k
rlamy
- [pypy-commit] pypy follow_symlinks: Allow specifying path as a file descriptor in statvfs()
rlamy
- [pypy-commit] pypy rposix-for-3: Reuse rposix definition of TIMESPEC in rposix_stat
rlamy
- [pypy-commit] pypy rposix-for-3: Add wrapper for fstatat()
rlamy
- [pypy-commit] pypy follow_symlinks: hg merge rposix-for-3
rlamy
- [pypy-commit] pypy follow_symlinks: Add support for dir_fd and follow_symlinks in posix.stat()
rlamy
- [pypy-commit] pypy default: fix test
rlamy
- [pypy-commit] pypy default: Skip some tests when the function under test is unavailable (e.g. on Windows)
rlamy
- [pypy-commit] pypy follow_symlinks: Create path_or_fd unwrapper
rlamy
- [pypy-commit] pypy follow_symlinks: Share a bit more code between the different cases in interp_posix.utime()
rlamy
- [pypy-commit] pypy follow_symlinks: Use path_or_fd unwrapper in utime()
rlamy
- [pypy-commit] pypy follow_symlinks: Enable fd support in utime()
rlamy
- [pypy-commit] pypy py3k: Merged in marky1991/pypy_new/33_fix_itertools (pull request #424)
rlamy
- [pypy-commit] pypy py3k: fix code formatting
rlamy
- [pypy-commit] pypy py3k: Merged in marky1991/pypy_new/33_fix_itertools (pull request #425)
rlamy
- [pypy-commit] pypy follow_symlinks: Factor out the pattern for calling an rposix function on a Path object
rlamy
- [pypy-commit] pypy follow_symlinks: Simplify stat(), fix error messages, add missing lstat() functionality
rlamy
- [pypy-commit] pypy follow_symlinks: Fall back to the buffer protocol in space.bytes_w()
rlamy
- [pypy-commit] pypy follow_symlinks: match CPython error messages better
rlamy
- [pypy-commit] pypy follow_symlinks: Allow fd in os.pathconf()
rlamy
- [pypy-commit] pypy follow_symlinks: Corner case: don't crash if path == -1
rlamy
- [pypy-commit] pypy rposix-for-3: Close branch rposix-for-3
rlamy
- [pypy-commit] pypy default: Merged in rposix-for-3 (pull request #427)
rlamy
- [pypy-commit] pypy py3k: hg merge default
rlamy
- [pypy-commit] pypy default: Try to fix OS X translation
rlamy
- [pypy-commit] pypy follow_symlinks: DRY: path unwrappers
rlamy
- [pypy-commit] pypy follow_symlinks: Use path_or_fd in readlink()
rlamy
- [pypy-commit] pypy stat_ns: implement the st_xtime_ns fields in stat_result()
rlamy
- [pypy-commit] pypy default: fix annotation error reporting for stat_result.__getitem__
rlamy
- [pypy-commit] pypy py3k: winreg must be a builtin module on Windows
rlamy
- [pypy-commit] pypy follow_symlinks: hg merge py3k
rlamy
- [pypy-commit] pypy follow_symlinks: (pjenvey, ronan) Query the buffer protocol in space.fsencode_w, not
rlamy
- [pypy-commit] pypy follow_symlinks: Add comment about the use of wrap_oserror() in utime()
rlamy
- [pypy-commit] pypy follow_symlinks: Close branch follow_symlinks
rlamy
- [pypy-commit] pypy py3k: Merged in follow_symlinks (pull request #428)
rlamy
- [pypy-commit] pypy stat_ns: hg merge py3k
rlamy
- [pypy-commit] pypy stat_ns: Close branch stat_ns
rlamy
- [pypy-commit] pypy py3k: Merged in stat_ns (pull request #430)
rlamy
- [pypy-commit] pypy gcheader-decl: GC headers don't need to be in the database
rlamy
- [pypy-commit] pypy gcheader-decl: document branch
rlamy
- [pypy-commit] pypy gcheader-decl: Close branch gcheader-decl
rlamy
- [pypy-commit] pypy default: Merged in gcheader-decl (pull request #432)
rlamy
- [pypy-commit] pypy vendor/stdlib-3.5.1: Branch for importing the 3.5.1 stdlib from CPython
rlamy
- [pypy-commit] pypy py3.5: Open branch for 3.5
rlamy
- [pypy-commit] pypy py3.5-raffael_t: Start branch py3.5-raffael_t
rlamy
- [pypy-commit] pypy py3.5-raffael_t: Fix location of stdlib after rebasing
rlamy
- [pypy-commit] pypy py3.5-raffael_t: Reapply Raffael's changes to opcode.py
rlamy
- [pypy-commit] pypy py3.5-raffael_t: fix typo
rlamy
- [pypy-commit] pypy default: Remove unnecessary import, which may trigger an unwanted early import of 'types'
rlamy
- [pypy-commit] pypy py3k-update: hg merge 73a49ec9edc3
rlamy
- [pypy-commit] pypy py3k-update: Catch SyntaxError on test_resource.py so that the test is skipped when trying to run py3 code on py2
rlamy
- [pypy-commit] pypy py3k-update: hg merge default (before cpyext-for-merge merge)
rlamy
- [pypy-commit] pypy release-5.x: GC headers don't need to be in the database (grafted from b05e778c895c)
rlamy
- [pypy-commit] pypy py3k-update: hg merge default (broken)
rlamy
- [pypy-commit] pypy py3k-update: fix some tests
rlamy
- [pypy-commit] pypy py3k-update: rm stray pypy2-ism
rlamy
- [pypy-commit] pypy py3k-update: rm more stray pypy2-isms
rlamy
- [pypy-commit] pypy default: Do 'import os' less hackishly in test_cpyext.py
rlamy
- [pypy-commit] pypy py3k-update: hg merge default
rlamy
- [pypy-commit] pypy py3k-update: fix
rlamy
- [pypy-commit] pypy py3k-update: Fix some tests
rlamy
- [pypy-commit] pypy py3k-update: 2to3fy two tests
rlamy
- [pypy-commit] pypy py3k-update: Create PyDictProxy_Check{,Exact}
rlamy
- [pypy-commit] pypy py3k-update: more test fixes
rlamy
- [pypy-commit] pypy py3k-update: more test fixes
rlamy
- [pypy-commit] pypy py3k-update: fix fix fix
rlamy
- [pypy-commit] pypy py3k-update: Fix pointer types in test module creation
rlamy
- [pypy-commit] pypy remove-getarrayitem-pure: Start removing getarrayitem_gc_pure ops
sbauman
- [pypy-commit] pypy remove-getarrayitem-pure: Make some of the tests work
sbauman
- [pypy-commit] pypy default: Skip PTY tests that hang forever on kFreeBSD
stefanor
- [pypy-commit] pypy default: Pretend to be using a terminal that supports clear
stefanor
- [pypy-commit] pypy default: The JIT detect_cpu function throws exceptions on some CPUs (that don't support JIT), so it can't be used in non-JIT contexts
stefanor
- [pypy-commit] pypy default: Avoid hanging forever on amd64 Debian/kFreeBSD (where it appears this timeout doesn't work)
stefanor
- [pypy-commit] pypy default: That was a bug-fix to the NEON bug-fix
stefanor
- [pypy-commit] pypy default: Sandbox didn't work in 5.0
stefanor
- [pypy-commit] pypy techtonik/introductionrst-simplify-explanation-abo-1460879168046: introduction.rst simplify explanation about PyPy
techtonik
- [pypy-commit] pypy techtonik/introductionrst-simplify-explanation-abo-1460879168046: introduction.rst attempt to explain it a little better
techtonik
- [pypy-commit] pypy techtonik/introductionrst-simplify-explanation-abo-1460879168046: introduction.rst add comments be Armin Rigo
techtonik
- [pypy-commit] pypy default: changed spelling of Fedora, changed yum to dnf
vasanthaganeshk
- [pypy-commit] pypy default: added Debian equivalent packages for Fedora
vasanthaganeshk
- [pypy-commit] pypy fix-jitlog: Test the JIT logger deals with forwarding pointers.
vext01
- [pypy-commit] pypy verbose-imports: Log things if verbose flags are supplied
william_ml_leslie
- [pypy-commit] pypy verbose-imports: Log nothing if sys hasn't been imported yet
william_ml_leslie
- [pypy-commit] pypy verbose-imports: Tests for verbose flag functionality
william_ml_leslie
- [pypy-commit] pypy verbose-imports: Test verbose = 0, too
william_ml_leslie
- [pypy-commit] pypy verbose-imports: Update pypy's usage string now that we have -v support
william_ml_leslie
Last message date:
Sat Apr 30 22:51:13 EDT 2016
Archived on: Sat Apr 30 22:51:17 EDT 2016
This archive was generated by
Pipermail 0.09 (Mailman edition).