November 2011 Archives by subject
Starting: Tue Nov 1 08:29:58 CET 2011
Ending: Wed Nov 30 22:37:19 CET 2011
Messages: 1443
- [pypy-commit] [COMMENT] Pull request #1 for pypy/benchmarks: Added GZip benchmark
Bitbucket
- [pypy-commit] [COMMENT] Pull request #1 for pypy/benchmarks: Added GZip benchmark
Bitbucket
- [pypy-commit] [OPEN] Pull request #1 for pypy/benchmarks: Added GZip benchmark
Bitbucket
- [pypy-commit] [OPEN] Pull request #14 for pypy/pypy: py3k sys.hash_info
Bitbucket
- [pypy-commit] [OPEN] Pull request #14 for pypy/pypy: py3k fixes - sys.hash_info, more stringobject tests, fixes for stringobject
Bitbucket
- [pypy-commit] [OPEN] Pull request #14 for pypy/pypy: py3k fixes - sys.hash_info, more stringobject tests, fixes for stringobject
Bitbucket
- [pypy-commit] benchmarks default: Essential fix.
arigo
- [pypy-commit] buildbot default: Add PPC32
bivab
- [pypy-commit] buildbot default: Add some documentation on how to run a buildslave
bivab
- [pypy-commit] buildbot default: Add Win64.
arigo
- [pypy-commit] extradoc extradoc: a draft of the sprint report for the blog
antocuni
- [pypy-commit] extradoc extradoc: a few more becnhmarks and some results
hakanardo
- [pypy-commit] extradoc extradoc: add few tasks
fijal
- [pypy-commit] extradoc extradoc: another TODO item
alex_gaynor
- [pypy-commit] extradoc extradoc: First version of the FSCONS talk
arigo
- [pypy-commit] extradoc extradoc: iteration becnhmakrs
hakanardo
- [pypy-commit] extradoc extradoc: latest version
arigo
- [pypy-commit] extradoc extradoc: merge heads
arigo
- [pypy-commit] extradoc extradoc: merge heads
antocuni
- [pypy-commit] extradoc extradoc: merge heads
arigo
- [pypy-commit] extradoc extradoc: minor fixes
hakanardo
- [pypy-commit] extradoc extradoc: note the branches
alex_gaynor
- [pypy-commit] extradoc extradoc: Planning.
arigo
- [pypy-commit] extradoc extradoc: twaks
antocuni
- [pypy-commit] extradoc extradoc: typo
hakanardo
- [pypy-commit] extradoc extradoc: Typo
arigo
- [pypy-commit] extradoc extradoc: update numpy plannign doc
alex_gaynor
- [pypy-commit] lang-scheme default: Added a bush of TypePredicates
boemmels
- [pypy-commit] lang-scheme default: Implement map & reverse
boemmels
- [pypy-commit] lang-scheme default: Implementation of Vectors #(...)
boemmels
- [pypy-commit] lang-scheme default: Implemented external representaions of objects and write function
boemmels
- [pypy-commit] lang-scheme default: naming consistency
boemmels
- [pypy-commit] lang-scheme default: New test: Run and assert tests at scheme level
boemmels
- [pypy-commit] lang-scheme default: Remove dublicated test
boemmels
- [pypy-commit] lang-scheme default: Translation fix
boemmels
- [pypy-commit] lang-scheme default: Use __class__.__name__ in __repr__
boemmels
- [pypy-commit] lang-scheme default: write (and not display) the result of an evaluation
boemmels
- [pypy-commit] Notification: benchmarks
Bitbucket
- [pypy-commit] Notification: pypy
Bitbucket
- [pypy-commit] Notification: pypy
Bitbucket
- [pypy-commit] Notification: pypy
Bitbucket
- [pypy-commit] Notification: pypy
Bitbucket
- [pypy-commit] Notification: pypyjp
Bitbucket
- [pypy-commit] pypy arm-backend-2: (arigo, bivab) remove a reference to boxes
bivab
- [pypy-commit] pypy arm-backend-2: (arigo, bivab) remove gcremovetypeptr for the arm backend for now
bivab
- [pypy-commit] pypy arm-backend-2: Add a method to allocate a scratch register that is managed by the register manager. The register manager keeps a list of temporary boxes that need to freed before before emitting the next operation
bivab
- [pypy-commit] pypy arm-backend-2: add ARM to conftest and test_random
bivab
- [pypy-commit] pypy arm-backend-2: add condition codes for unsigned lower and unsigned higher or same
bivab
- [pypy-commit] pypy arm-backend-2: add some more tests for intergers and guards to cover unary and unsigned cmp ops
bivab
- [pypy-commit] pypy arm-backend-2: Bah, fix.
bivab
- [pypy-commit] pypy arm-backend-2: fix test
bivab
- [pypy-commit] pypy arm-backend-2: fix test
bivab
- [pypy-commit] pypy arm-backend-2: fix tests
bivab
- [pypy-commit] pypy arm-backend-2: improve freeing of boxes in the backend
bivab
- [pypy-commit] pypy arm-backend-2: make sure arm backend tests are only executed when running on ARM
bivab
- [pypy-commit] pypy arm-backend-2: merge disable_merge_different_int_types
bivab
- [pypy-commit] pypy arm-backend-2: merge default
bivab
- [pypy-commit] pypy arm-backend-2: merge default
bivab
- [pypy-commit] pypy arm-backend-2: merge default
bivab
- [pypy-commit] pypy arm-backend-2: merge default
bivab
- [pypy-commit] pypy arm-backend-2: merge default
bivab
- [pypy-commit] pypy arm-backend-2: merge default
bivab
- [pypy-commit] pypy arm-backend-2: merge default (again)
bivab
- [pypy-commit] pypy arm-backend-2: merge heads
bivab
- [pypy-commit] pypy arm-backend-2: more fixes
bivab
- [pypy-commit] pypy arm-backend-2: refactor _ensure_value_is_boxed to use a managed scratch_register and to not return the box for the allocated location anymore
bivab
- [pypy-commit] pypy arm-backend-2: refactor a bit the generation of functions that emit code for cmp operations
bivab
- [pypy-commit] pypy arm-backend-2: refactor check_imm_arg
bivab
- [pypy-commit] pypy arm-backend-2: remove another reference to boxes
bivab
- [pypy-commit] pypy arm-backend-2: remove the inverse argument for the register allocation of cmp operations and use the correct correct condition flags for uint operations
bivab
- [pypy-commit] pypy arm-backend-2: translation fix
bivab
- [pypy-commit] pypy arm-backend-2: typo
bivab
- [pypy-commit] pypy arm-backend-2: Updating the position of the regalloc to soon when merging operations with the following guard leads to some vars being freed to early
bivab
- [pypy-commit] pypy better-jit-hooks: a branch to experiment with some better JIT hooks
fijal
- [pypy-commit] pypy better-jit-hooks: spring cleaning - remove ABORT_BAD_LOOP (unused)
fijal
- [pypy-commit] pypy default: "Fix" this test.
arigo
- [pypy-commit] pypy default: 'merge'
ctismer
- [pypy-commit] pypy default: 'nt.spawnve()', a Windows function
arigo
- [pypy-commit] pypy default: (antocuni, arigo)
arigo
- [pypy-commit] pypy default: (antocuni, hakan, arigo)
arigo
- [pypy-commit] pypy default: (arigato mostly) fix optimizeopt tests that had the arguments to copystrcontent in the wrong order
alex_gaynor
- [pypy-commit] pypy default: (arigo, bivab) merge disable_merge_different_int_types
bivab
- [pypy-commit] pypy default: (arigo, bivab): do not require gcc 4.0 as a default on the mac anymore
bivab
- [pypy-commit] pypy default: (cfbolz, arigo)
arigo
- [pypy-commit] pypy default: (cfbolz, arigo)
arigo
- [pypy-commit] pypy default: (cfbolz, arigo) Clarify the intention.
arigo
- [pypy-commit] pypy default: (fijal and mattip mostly) Implemented support for multidimensional arrays in numpypy, this support extends to most features, such as ufuncs.
alex_gaynor
- [pypy-commit] pypy default: (lukas, carl) Merged list strategies. This allows the contents of a list to be unboxed. See the blog for more details on it.
alex_gaynor
- [pypy-commit] pypy default: (mattip) merge matrixmath, adds arange, transpose, flatiter
fijal
- [pypy-commit] pypy default: (mwp antocuni) make platform work on PowerPC
mwp
- [pypy-commit] pypy default: - add in the backend, for binary instructions, a memo function
arigo
- [pypy-commit] pypy default: __iter__ for flatiter
fijal
- [pypy-commit] pypy default: A check function that allows counting the ops in the peeled loop only. It only works in the simplest case when there is exactly one jump (ie no bridges not ending in finnish)
hakanardo
- [pypy-commit] pypy default: a debug repr for numpy objects
fijal
- [pypy-commit] pypy default: a dot implementation
fijal
- [pypy-commit] pypy default: a failing optimizeopt test
alex_gaynor
- [pypy-commit] pypy default: a failing test
alex_gaynor
- [pypy-commit] pypy default: a failing test for integer pow
alex_gaynor
- [pypy-commit] pypy default: A failing test showing problems with list-comprehension-optimization
fijal
- [pypy-commit] pypy default: a failing test that shows a problem around forcing of virtual immutable recursive structs
cfbolz
- [pypy-commit] pypy default: A sanity check that most probably breaks right now on Windows
arigo
- [pypy-commit] pypy default: A skipped failing test for a case in which the optimizer gets confused
arigo
- [pypy-commit] pypy default: a test and a fix
fijal
- [pypy-commit] pypy default: a test and a fix. test_zjit still fails, has to be investigated deeper
fijal
- [pypy-commit] pypy default: a test that now passes, after the refactorings. however, it fails on cpython
cfbolz
- [pypy-commit] pypy default: Accept py.test.skip()'s exception as also meaning "skip this package" here.
arigo
- [pypy-commit] pypy default: Actually, it seems that the new code replaces any need for the old
arigo
- [pypy-commit] pypy default: add a (probably not too costly) sanity check to make sure that the traces the
cfbolz
- [pypy-commit] pypy default: Add a passing test that rgc.ll_arraycopy cannot raise.
arigo
- [pypy-commit] pypy default: add a potential project idea
fijal
- [pypy-commit] pypy default: add a potential project idea
fijal
- [pypy-commit] pypy default: Add a test for ed83fd7b7ec1.
arigo
- [pypy-commit] pypy default: Add a test here. *Still* passing.
arigo
- [pypy-commit] pypy default: Add a warning message when using stacklet_destroy() with
arigo
- [pypy-commit] pypy default: Add an XXX for this bug.
arigo
- [pypy-commit] pypy default: Add another test that switch() calls are supposed to have an
arigo
- [pypy-commit] pypy default: Add asserts and fix a test. The main point of the asserts is to
arigo
- [pypy-commit] pypy default: Add support for track_allocation=False (the default is True).
arigo
- [pypy-commit] pypy default: Add tests for two special cases of "MOV" in INSN().
arigo
- [pypy-commit] pypy default: Add the requirement that W_XxxObject classes that are different
arigo
- [pypy-commit] pypy default: Added a failing test.
arigo
- [pypy-commit] pypy default: added floatstrategy for lists
l.diekmann
- [pypy-commit] pypy default: added some tests for listobject with floats
l.diekmann
- [pypy-commit] pypy default: Added sqlite3.Connection.enable_load_extension, untested because it requires your libsqlite to have been compiled with the right options. (CPython has no tests for it either). Manually tested by running the GeoDjango test suite.
alex_gaynor
- [pypy-commit] pypy default: added sys.maxint to the compilation hash, to avoid obscure errors on windows
ctismer
- [pypy-commit] pypy default: Added tag release-1.7 for changeset ff4af8f31882
fijal
- [pypy-commit] pypy default: Added tag release-1.7 for changeset ff4af8f31882
fijal
- [pypy-commit] pypy default: adjusted nullpath according to amaury's suggestion
ctismer
- [pypy-commit] pypy default: All these functions PyUnicode_DecodeASCII &co are really similar,
amauryfa
- [pypy-commit] pypy default: Allow at least the front-end to be tested with the "none" gc.
arigo
- [pypy-commit] pypy default: allow inlining into marshal, which is used in cPickle
alex_gaynor
- [pypy-commit] pypy default: allow setarrayitem to update the cache exported from the preamble to the loop the same way setfield does
hakanardo
- [pypy-commit] pypy default: Allow very basic multiple inheritance of app-level types in RPython. Thanks to amaury for the review/suggestions.
alex_gaynor
- [pypy-commit] pypy default: Allows presetup.py to execute setup.py scripts which use
amauryfa
- [pypy-commit] pypy default: also a test where the circular virtual is forced
cfbolz
- [pypy-commit] pypy default: also add a test for indices4
cfbolz
- [pypy-commit] pypy default: Also use the same jitdriver for list(generator).
arigo
- [pypy-commit] pypy default: alternative fix that does not rely on interning ints
hakanardo
- [pypy-commit] pypy default: another attempt to fix translation
fijal
- [pypy-commit] pypy default: Another case: strgetitem on a VStringConcatValue can be resolved
arigo
- [pypy-commit] pypy default: Argh. See comment.
arigo
- [pypy-commit] pypy default: Attempt at producing a Makefile compatible with "nmake lldebug" on Windows.
arigo
- [pypy-commit] pypy default: Attempt to fix generally the issue of VStringPlainValue that can
arigo
- [pypy-commit] pypy default: Attempt to fix this test to check with a valid fd.
arigo
- [pypy-commit] pypy default: Attempt to fix Windows translation.
arigo
- [pypy-commit] pypy default: Attempt to kill 5225def4e322, which contains the broken
arigo
- [pypy-commit] pypy default: Attempting to add a JitDriver to unpackiterable(generator).
arigo
- [pypy-commit] pypy default: Backout 09269d2f8fee. It forces the builder object itself to be escaped.
arigo
- [pypy-commit] pypy default: Backout 5f1a23beccab: the part that changed test_string.py
arigo
- [pypy-commit] pypy default: Backout 6f2534aea5ca. It's more of a mess, because
arigo
- [pypy-commit] pypy default: Backout ad8b93cf993c and 38f173ee998a, which fail on tannit.
arigo
- [pypy-commit] pypy default: bah.
arigo
- [pypy-commit] pypy default: Bah.
arigo
- [pypy-commit] pypy default: Bah. Not really motivated to write a test for this obscure case :-(
arigo
- [pypy-commit] pypy default: Bah. Reorder these two lines as a workaround fix for running:
arigo
- [pypy-commit] pypy default: Better presentation of the docstring
amauryfa
- [pypy-commit] pypy default: break up circular dependencies among short_boxes and give up
hakanardo
- [pypy-commit] pypy default: Chanced upon a "x not in biglist" done in a loop: replace it with
arigo
- [pypy-commit] pypy default: Change VStringPlainValue, refactoring and giving a long
arigo
- [pypy-commit] pypy default: clarrify
alex_gaynor
- [pypy-commit] pypy default: Clean up: min() is now RPython.
arigo
- [pypy-commit] pypy default: Clean-up.
arigo
- [pypy-commit] pypy default: Comment this out for now, oopspec doesn't appear to play nice with look_inside_iff
alex_gaynor
- [pypy-commit] pypy default: comparing a unicode char to str char no longer requires allocating a string
alex_gaynor
- [pypy-commit] pypy default: copy for slices
fijal
- [pypy-commit] pypy default: Copy the logic for math.fmod() from CPython 2.7.
arigo
- [pypy-commit] pypy default: copy this function from jitviewer
fijal
- [pypy-commit] pypy default: corner case not handled very well
hakanardo
- [pypy-commit] pypy default: correct the dtype of scalar arrays
alex_gaynor
- [pypy-commit] pypy default: corrected nullpath for windows
ctismer
- [pypy-commit] pypy default: cpyext: Add PyUnicode_*Latin1 functions
amauryfa
- [pypy-commit] pypy default: cpyext: turn some parameters into const char*
amauryfa
- [pypy-commit] pypy default: Disable a half-broken test that fails on Python 2.7 (see comments).
arigo
- [pypy-commit] pypy default: dissable for now, it makes test_convert_from_SmallFunctionSetPBCRepr_to_FunctionsPBCRep fail
hakanardo
- [pypy-commit] pypy default: Do the imports only if the config option is set.
arigo
- [pypy-commit] pypy default: Document slice assignment and deletion for RPython lists.
arigo
- [pypy-commit] pypy default: Don't crash on reading or writing stuff to the history file
arigo
- [pypy-commit] pypy default: Don't use range() here, because it can raise MemoryError.
arigo
- [pypy-commit] pypy default: don't use specialize.ll for these calls, but call_location instead. otherwise
cfbolz
- [pypy-commit] pypy default: don't use the JIT strslice optimization if some of the characters are in an unknown state with regards to whether they're initialized
alex_gaynor
- [pypy-commit] pypy default: don't use w_None here, it breaks cpyext
cfbolz
- [pypy-commit] pypy default: dumb merge
fijal
- [pypy-commit] pypy default: escape '<lambda>' for windows
ctismer
- [pypy-commit] pypy default: Expand the code of unpackiterable() into several versions instead
arigo
- [pypy-commit] pypy default: Export 'setitimer' even if running Python 2.5, which does not have
arigo
- [pypy-commit] pypy default: Export PyDescr_NewMethod and the PyWrapperDescr_Type it returns
amauryfa
- [pypy-commit] pypy default: expose size attribute
fijal
- [pypy-commit] pypy default: failing test
alex_gaynor
- [pypy-commit] pypy default: failing test
hakanardo
- [pypy-commit] pypy default: Finally found out where to put the "assert".
arigo
- [pypy-commit] pypy default: fix
alex_gaynor
- [pypy-commit] pypy default: Fix (probably; tests coming).
arigo
- [pypy-commit] pypy default: fix a crash and translation in micronumpy
alex_gaynor
- [pypy-commit] pypy default: fix an issue in clibffi that is triggered on big endian platforms due to the byte order when casting a larger data type to smaller one to be passed to a function called through ffi
bivab
- [pypy-commit] pypy default: fix an overlow bug
hakanardo
- [pypy-commit] pypy default: Fix annotation issues.
arigo
- [pypy-commit] pypy default: Fix for 612f7784a228.
arigo
- [pypy-commit] pypy default: fix for a typo and cpyext
alex_gaynor
- [pypy-commit] pypy default: fix for invalid inputs in numpy.sqrt
alex_gaynor
- [pypy-commit] pypy default: fix for pow on ints, and translated binops
alex_gaynor
- [pypy-commit] pypy default: Fix for rffi.cast(lltype.Float, r_singlefloat(2))
alex_gaynor
- [pypy-commit] pypy default: Fix for rffi.cast(lltype.SingleFloat, <some r_singlefloat>), if the input type is the same as the restype, just retunr the obj.
alex_gaynor
- [pypy-commit] pypy default: Fix for test_sin_cos.
arigo
- [pypy-commit] pypy default: Fix for test_zrpy_releasegil.
arigo
- [pypy-commit] pypy default: fix for the failing test - StopIteration raised from anywhere kills the generator
alex_gaynor
- [pypy-commit] pypy default: fix for translation and copying scalars
alex_gaynor
- [pypy-commit] pypy default: Fix for various ops on bools.
alex_gaynor
- [pypy-commit] pypy default: Fix syslog.py by following the CPython source, as per
arigo
- [pypy-commit] pypy default: Fix test.
arigo
- [pypy-commit] pypy default: Fix test_bitfields just by relaxing this check here.
arigo
- [pypy-commit] pypy default: fix test_urllib2_localnet - don't be too eager on closing the response, closing
fijal
- [pypy-commit] pypy default: Fix the flow object space for 0bb2bc25fcfc
arigo
- [pypy-commit] pypy default: fix the test
fijal
- [pypy-commit] pypy default: fix the test
fijal
- [pypy-commit] pypy default: Fix the test.
arigo
- [pypy-commit] pypy default: Fix this CPython test, and comment about why I think that PyPy's
arigo
- [pypy-commit] pypy default: fix translation
alex_gaynor
- [pypy-commit] pypy default: fix translation hopefully
alex_gaynor
- [pypy-commit] pypy default: fix translation hopefully
fijal
- [pypy-commit] pypy default: Fix usages of Py_InitModule outside of the module init() function.
amauryfa
- [pypy-commit] pypy default: Fix, maybe temporary, for getting a Windows process to write stuff
arigo
- [pypy-commit] pypy default: Fix.
arigo
- [pypy-commit] pypy default: Fix.
arigo
- [pypy-commit] pypy default: fix.
arigo
- [pypy-commit] pypy default: Fix.
arigo
- [pypy-commit] pypy default: Fix.
arigo
- [pypy-commit] pypy default: Fix.
arigo
- [pypy-commit] pypy default: Fix.
arigo
- [pypy-commit] pypy default: Fix.
arigo
- [pypy-commit] pypy default: Fix.
arigo
- [pypy-commit] pypy default: Fix? the Windows build by using the Windows functions _isnan()
arigo
- [pypy-commit] pypy default: fixes
fijal
- [pypy-commit] pypy default: for completeness sake, bump numbers in the default as well
fijal
- [pypy-commit] pypy default: for concatenation to work, this has to be an empty list (thanks ned)
fijal
- [pypy-commit] pypy default: Found the cause of the failure of test_nongc_attached_to_gc in
arigo
- [pypy-commit] pypy default: Get rid of 'isstartblock'.
arigo
- [pypy-commit] pypy default: Get rid of the test_distutils failure, which (for now) is really
arigo
- [pypy-commit] pypy default: got continuelets ready for windows.
ctismer
- [pypy-commit] pypy default: handle hashing a None rstr, fixes the test I checked in yesterday.
alex_gaynor
- [pypy-commit] pypy default: hg backout 7202b0d9cb70
hakanardo
- [pypy-commit] pypy default: hg backout a27a481ec877
hakanardo
- [pypy-commit] pypy default: hg merge
hakanardo
- [pypy-commit] pypy default: hg merge
hakanardo
- [pypy-commit] pypy default: hg merge
hakanardo
- [pypy-commit] pypy default: hg merge
hakanardo
- [pypy-commit] pypy default: hg merge
hakanardo
- [pypy-commit] pypy default: hopefully make flatiter interoperate nicely with the rest by making it a part
fijal
- [pypy-commit] pypy default: I knew that running a translation first was a good idea
cfbolz
- [pypy-commit] pypy default: imp.find_module() now returns the file object for extension modules.
amauryfa
- [pypy-commit] pypy default: Implement PyUnicode_EncodeUTF8
amauryfa
- [pypy-commit] pypy default: Implement PyWeakref_NewProxy, thanks dcolish!
amauryfa
- [pypy-commit] pypy default: implement setslice in numpy's minilang and add convert a test_zjit, it's failing
alex_gaynor
- [pypy-commit] pypy default: improve the list-comprehension-operation optimization to work for slightly
fijal
- [pypy-commit] pypy default: Improve the tests, testing a fully recursive solution.
arigo
- [pypy-commit] pypy default: interning ints aswell
hakanardo
- [pypy-commit] pypy default: issue887: cpyext: add support for mp_ass_subscript
amauryfa
- [pypy-commit] pypy default: issue944 wontfix
arigo
- [pypy-commit] pypy default: jit.dont_look_inside the rffi functions called with the "win" calling conv.
arigo
- [pypy-commit] pypy default: kill a duplicate test
fijal
- [pypy-commit] pypy default: Kill dependency on "isstartblock".
arigo
- [pypy-commit] pypy default: kill imports
gutworth
- [pypy-commit] pypy default: Kill ovfcheck_lshift(), which was only needed before Python 2.4.
arigo
- [pypy-commit] pypy default: Kill test now that ovfcheck_lshift() is gone.
arigo
- [pypy-commit] pypy default: kill the meeting talk
fijal
- [pypy-commit] pypy default: kill unused nonsensical function
fijal
- [pypy-commit] pypy default: link to planning
fijal
- [pypy-commit] pypy default: made the assembler sources compile so far
ctismer
- [pypy-commit] pypy default: Make tests reflect the reality. It actually works, but a bit by chance
fijal
- [pypy-commit] pypy default: Mallocs of objects with a __del__ are handled differently anyway.
arigo
- [pypy-commit] pypy default: maybe wrong merge or wrong update. Anyway, fixed
ctismer
- [pypy-commit] pypy default: merge
cfbolz
- [pypy-commit] pypy default: merge
cfbolz
- [pypy-commit] pypy default: Merge
ctismer
- [pypy-commit] pypy default: merge
cfbolz
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge
ctismer
- [pypy-commit] pypy default: merge
cfbolz
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: Merge
ctismer
- [pypy-commit] pypy default: merge
l.diekmann
- [pypy-commit] pypy default: merge
fijal
- [pypy-commit] pypy default: merge default
fijal
- [pypy-commit] pypy default: merge default
fijal
- [pypy-commit] pypy default: merge default
fijal
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: Merge heads
amauryfa
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge heads
arigo
- [pypy-commit] pypy default: merge jit-refactor-tests. This replaces check_loops with check_resops which counts every operation including those in the preamble.
hakanardo
- [pypy-commit] pypy default: Merge jitdriver-setparam-all branch that allows setting parameters to all
fijal
- [pypy-commit] pypy default: Merge rgc-mem-pressure. This branch adds memory pressure in some crucial points
fijal
- [pypy-commit] pypy default: Merged jit-dynamic-getarrayitem. Added support for creating custom getarrayitems at jit-compile time. Steals some stuff from anto's ffistruct.
alex_gaynor
- [pypy-commit] pypy default: merged upstream
alex_gaynor
- [pypy-commit] pypy default: merged upstream
alex_gaynor
- [pypy-commit] pypy default: merged upstream
alex_gaynor
- [pypy-commit] pypy default: merged upstream
alex_gaynor
- [pypy-commit] pypy default: minor fixes
fijal
- [pypy-commit] pypy default: Missing f.close().
arigo
- [pypy-commit] pypy default: more places
cfbolz
- [pypy-commit] pypy default: move test_zll_random to llsupport
fijal
- [pypy-commit] pypy default: Move the import globally.
arigo
- [pypy-commit] pypy default: Move the stress test directly to backend/test and rename it
fijal
- [pypy-commit] pypy default: Move these imports to a place where they will only be triggered
arigo
- [pypy-commit] pypy default: Must close the file explicitly; otherwise, on Windows, we cannot unlink it before the GC runs
arigo
- [pypy-commit] pypy default: negative bools??!
cfbolz
- [pypy-commit] pypy default: newline
fijal
- [pypy-commit] pypy default: No-op clarification.
arigo
- [pypy-commit] pypy default: No-op clean-up.
arigo
- [pypy-commit] pypy default: Obscure: mark some builtin modules (but not others) as taking
arigo
- [pypy-commit] pypy default: On Windows, the renamed binary file must end with ".exe".
arigo
- [pypy-commit] pypy default: only use a single counter in xrange iterators (should save a setitem)
hakanardo
- [pypy-commit] pypy default: oops, fix translation
fijal
- [pypy-commit] pypy default: oops, those two are unroll_safe
fijal
- [pypy-commit] pypy default: oups.
arigo
- [pypy-commit] pypy default: Oups.
arigo
- [pypy-commit] pypy default: Oups.
arigo
- [pypy-commit] pypy default: py3k compat. syntax
pjenvey
- [pypy-commit] pypy default: Python 2.5 compatibility.
arigo
- [pypy-commit] pypy default: Python 2.5 support.
arigo
- [pypy-commit] pypy default: raise an exception instead of producing bogus code. Probably would turn
fijal
- [pypy-commit] pypy default: reduce code duplication
cfbolz
- [pypy-commit] pypy default: relevant_environ updated for windows
ctismer
- [pypy-commit] pypy default: Remove implemented functions from stubs.py
amauryfa
- [pypy-commit] pypy default: remove some C-isms
cfbolz
- [pypy-commit] pypy default: remove unused helper function
cfbolz
- [pypy-commit] pypy default: rename numpy -> numpypy
fijal
- [pypy-commit] pypy default: Rename the decorator. Fijal: can you use it on some of
arigo
- [pypy-commit] pypy default: Rename the method 'id' to 'unique_id', which is less prone to random
arigo
- [pypy-commit] pypy default: rename this function to be non-official. it can lead to bugs because the caller
cfbolz
- [pypy-commit] pypy default: reorganize the conditions as armin suggested
alex_gaynor
- [pypy-commit] pypy default: replace other two specialize calls as well
cfbolz
- [pypy-commit] pypy default: reuse the new helper here too
cfbolz
- [pypy-commit] pypy default: Revert most of f47785c0df50, using check_simple_loop() instead.
arigo
- [pypy-commit] pypy default: Rewrite the algorithm of readlines() based on the hypothesis that
arigo
- [pypy-commit] pypy default: rstacklet.py builds ok, but crashes when run.
ctismer
- [pypy-commit] pypy default: setitimer is Unix-only.
arigo
- [pypy-commit] pypy default: similarly simplify some unicode code
cfbolz
- [pypy-commit] pypy default: similarly, no need to call length here
cfbolz
- [pypy-commit] pypy default: simplify _convert_idx_params
cfbolz
- [pypy-commit] pypy default: simplify failing test
alex_gaynor
- [pypy-commit] pypy default: simplify this code a bit.
alex_gaynor
- [pypy-commit] pypy default: Skip an assert in a test that fails on Python 2.5.
arigo
- [pypy-commit] pypy default: Skip if we on't have curses.
arigo
- [pypy-commit] pypy default: skip on win32.
arigo
- [pypy-commit] pypy default: Skip the asmgcc tests on windows.
arigo
- [pypy-commit] pypy default: skip the failing test for now
fijal
- [pypy-commit] pypy default: Skip the fmod test for now.
arigo
- [pypy-commit] pypy default: skip the test in progress
arigo
- [pypy-commit] pypy default: some of the _convert_idx_params implementations are rather nonsensical (and of
cfbolz
- [pypy-commit] pypy default: sqrt ufunc
fijal
- [pypy-commit] pypy default: support some combination of flags, I hope it's good enough
fijal
- [pypy-commit] pypy default: Tentatively rewrite push_arg_as_ffiptr().
arigo
- [pypy-commit] pypy default: Test and check that we don't do stupid copies - it's not only inefficient, it's
fijal
- [pypy-commit] pypy default: Test and fix.
arigo
- [pypy-commit] pypy default: Test and fix: instead of using a custom __setattr__ handler
arigo
- [pypy-commit] pypy default: Test for crashes when we call StringBuilder.build() several times over a growing
arigo
- [pypy-commit] pypy default: Test for the case of really two distinct constant immutable objects
arigo
- [pypy-commit] pypy default: test short preamble and non constant case aswell
hakanardo
- [pypy-commit] pypy default: The hack "y + VERY_LARGE_FLOAT == y" fails to give the correct
arigo
- [pypy-commit] pypy default: The test for ca55b862ead9. Without the fix, it crashes with MemoryError as originally reported by ltratt with the Converge VM.
arigo
- [pypy-commit] pypy default: there was really no optimizeopt test for circular virtuals??!
cfbolz
- [pypy-commit] pypy default: these loops are unrolled anyway, directly access the correct attribute instead
cfbolz
- [pypy-commit] pypy default: This needs to be a list ot be RPython
alex_gaynor
- [pypy-commit] pypy default: This number of quotes suddenly stopped working on Windows, but
arigo
- [pypy-commit] pypy default: This test no longer xfails.
arigo
- [pypy-commit] pypy default: those two functions are unroll_safe
fijal
- [pypy-commit] pypy default: translation fix
arigo
- [pypy-commit] pypy default: Translation fix for x86/test/test_ztranslation.
arigo
- [pypy-commit] pypy default: Translation fix.
arigo
- [pypy-commit] pypy default: Try never to crash when inspect.getsource() fails.
arigo
- [pypy-commit] pypy default: Try to detect conflicting options with the gc.
arigo
- [pypy-commit] pypy default: Try to fix the issue with -lrt being at the beginning and not at the end.
fijal
- [pypy-commit] pypy default: Turn off reporting const-fold errors as WARNINGs. This gives quite a
arigo
- [pypy-commit] pypy default: Tweak for the common case: use isfinite() more often,
arigo
- [pypy-commit] pypy default: Tweak.
arigo
- [pypy-commit] pypy default: Tweaks.
arigo
- [pypy-commit] pypy default: Tweaks. Unsure why but it seems that test_zll_random ends up with
arigo
- [pypy-commit] pypy default: Uh. Fix this function. No clue how it could have worked: it
arigo
- [pypy-commit] pypy default: unpack tuple params, for py3k support
pjenvey
- [pypy-commit] pypy default: Update PyMODINIT_FUNC for C++ extensions.
amauryfa
- [pypy-commit] pypy default: Update the list of irc topics. I had to change details in the script
arigo
- [pypy-commit] pypy default: use autodetect if you don't want to run llgraph
fijal
- [pypy-commit] pypy default: use consistent name
cfbolz
- [pypy-commit] pypy default: use memcpy for numpy.array.copy(), rather than going through the constructor
alex_gaynor
- [pypy-commit] pypy default: Use OneDimIterator if we have one already. Fix it since tests started failing,
fijal
- [pypy-commit] pypy default: use the internal copy, rather than reinstantiate, faster.
alex_gaynor
- [pypy-commit] pypy default: use tuple
fijal
- [pypy-commit] pypy default: what I remember that goes into the release
fijal
- [pypy-commit] pypy default: When hitting a too large value, cleanly give up jit-compilation
arigo
- [pypy-commit] pypy default: Windows fix.
arigo
- [pypy-commit] pypy default: woops, wrong class
cfbolz
- [pypy-commit] pypy default: work out the release announcement
fijal
- [pypy-commit] pypy default: Workaround. See comment.
arigo
- [pypy-commit] pypy default: Workaroundish fix for now: don't use green fields here.
arigo
- [pypy-commit] pypy default: Write a failing test for test_zjit
fijal
- [pypy-commit] pypy default: write a test that _get_interplevel_cls still works if we have multiple string
fijal
- [pypy-commit] pypy default: Write the recursive possibly-looping case of this logic.
arigo
- [pypy-commit] pypy default: Yet another obscure attempt at catching the bug
arigo
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab) close branch
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab) fix test
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab) fix test
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): cast
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): fix
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): fix termios
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): fix test
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): more fixes
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab) allow merging unsigned and non-negative signed values
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab) fix rrandom and _random
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab) get rid of the last traces of ROverflowError
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab) make sure next_instr is unsigned and handlerposition is a non-neg signed O.o
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab) this checks does not make sense
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): add some r_uints here and there
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): add some u's and shuffle operations around
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): Ah, read field as a signed value
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): allow to merge different int types if they have both the same signedness. Generalizing to the larger type
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): enforce sanity when merging inttypes
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): fix an overfix in setobject
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): fix some annotation issues and revert some changes
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): make sure makeipv4addr is called with an unsigned
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (arigo, bivab): more fixes (not really tested)
bivab
- [pypy-commit] pypy disable_merge_different_int_types: (bivab, arigo) Fix.
arigo
- [pypy-commit] pypy disable_merge_different_int_types: (bivab, arigo) Bah, undo this, which was too much.
arigo
- [pypy-commit] pypy disable_merge_different_int_types: (bivab, cfbolz): be more consistent about unsignedness in rsocket
cfbolz
- [pypy-commit] pypy disable_merge_different_int_types: (cfbolz, bivab): unsigned fixes for rbigint
bivab
- [pypy-commit] pypy disable_merge_different_int_types: add missing import
bivab
- [pypy-commit] pypy disable_merge_different_int_types: fix
bivab
- [pypy-commit] pypy disable_merge_different_int_types: Fix for unsigned-ness in hash computation
bivab
- [pypy-commit] pypy disable_merge_different_int_types: make sure calls to ntohs and htons get unsigned values
bivab
- [pypy-commit] pypy disable_merge_different_int_types: merge default
bivab
- [pypy-commit] pypy disable_merge_different_int_types: merge default
bivab
- [pypy-commit] pypy disable_merge_different_int_types: merge default
bivab
- [pypy-commit] pypy disable_merge_different_int_types: merge default
bivab
- [pypy-commit] pypy disable_merge_different_int_types: more fixes
bivab
- [pypy-commit] pypy disable_merge_different_int_types: raise UnionError
bivab
- [pypy-commit] pypy disable_merge_different_int_types: remove a testcase that is now valid
bivab
- [pypy-commit] pypy disable_merge_different_int_types: translation fix for binascii
bivab
- [pypy-commit] pypy disable_merge_different_int_types: translation fix for rmd5 and rsha
bivab
- [pypy-commit] pypy ffistruct: (antocuni, arigo around): correctly truncate all the values to a Signed"
antocuni
- [pypy-commit] pypy ffistruct: (antocuni, arigo): this is probably how the test was meant to be
antocuni
- [pypy-commit] pypy ffistruct: add a new space method to truncate longlongs, similar to space.truncatedint
antocuni
- [pypy-commit] pypy ffistruct: add a test for shorts
antocuni
- [pypy-commit] pypy ffistruct: add an XXX so that I hopefully don't forget this :-)
antocuni
- [pypy-commit] pypy ffistruct: add docstrings for WrapDispatcher methods; raise a proper applevel exception instead of assert False
antocuni
- [pypy-commit] pypy ffistruct: add low-level support to get/set (u)longlong fields in libffi
antocuni
- [pypy-commit] pypy ffistruct: add support and tests for unsigned types
antocuni
- [pypy-commit] pypy ffistruct: add support for char/unichar fields
antocuni
- [pypy-commit] pypy ffistruct: add support for getting/setting signed values other than long
antocuni
- [pypy-commit] pypy ffistruct: add support for longlongs at applevel
antocuni
- [pypy-commit] pypy ffistruct: add unit tests for compute_size_and_alignment; the last ones fails and are commented out for now
antocuni
- [pypy-commit] pypy ffistruct: adjust the total size according to the alignment: this makes more tests passing
antocuni
- [pypy-commit] pypy ffistruct: app level support for pointer fields
antocuni
- [pypy-commit] pypy ffistruct: applevel support for float fields
antocuni
- [pypy-commit] pypy ffistruct: applevel support for single float fields
antocuni
- [pypy-commit] pypy ffistruct: build the TYPE_MAP dictionaries but preserves the list: this is because we want to avoid key-clashing (e.g., on 32bit rffi.UINT: ffi_type_sint is overwritten by rffi.LONG, because _signed_type_for(LONG) returns ffi_type_sint32)
antocuni
- [pypy-commit] pypy ffistruct: copy the logic to cope with field alignment from _rawffi, the failing test now passes
antocuni
- [pypy-commit] pypy ffistruct: crash with a nicer exception if we don't know how to deal with this type
antocuni
- [pypy-commit] pypy ffistruct: hg merge default
antocuni
- [pypy-commit] pypy ffistruct: hg merge default
antocuni
- [pypy-commit] pypy ffistruct: infrastructure to test type_converter.py
antocuni
- [pypy-commit] pypy ffistruct: introduce the dispatcher, whose goal is to convert applevel objtects into low-level values based on the given ffitype. Move there the logic that we used in W_FunctPtr to build the argchain
antocuni
- [pypy-commit] pypy ffistruct: kill duplicate test
antocuni
- [pypy-commit] pypy ffistruct: kill unwrap_truncated_int, and use the nice space.truncatedint_w method
antocuni
- [pypy-commit] pypy ffistruct: low level support for float fields
antocuni
- [pypy-commit] pypy ffistruct: low level support for pointer fields
antocuni
- [pypy-commit] pypy ffistruct: low-level support for single float fields
antocuni
- [pypy-commit] pypy ffistruct: Make rarithmetic.longlongmask() translatable.
arigo
- [pypy-commit] pypy ffistruct: make sure that these two finalizers are lightweight
antocuni
- [pypy-commit] pypy ffistruct: make sure that we correctly handle the app-level-long to interp-level-slong conversion
antocuni
- [pypy-commit] pypy ffistruct: make sure that we properly convert a sbyte >= 128 into a negative value when we set it. This requires to change the TYPE_MAP_INT in clibffi.py, which was wrong before (rffi.CHAR is unsigned, not signed)
antocuni
- [pypy-commit] pypy ffistruct: merge heads
arigo
- [pypy-commit] pypy ffistruct: more tests for pointers and strings
antocuni
- [pypy-commit] pypy ffistruct: more tests for type_converter
antocuni
- [pypy-commit] pypy ffistruct: move the logic to wrap the result of a call in WrapDispatcher. Will write the proper docstrings later, now I have to shutdown the laptop because we are landing :-)
antocuni
- [pypy-commit] pypy ffistruct: move this test at the end, after the ones which directly operate on StructDescr
antocuni
- [pypy-commit] pypy ffistruct: rename dispatchers into the more descriptive {FromAppLevel, ToAppLevel}Converter, and add docstrings
antocuni
- [pypy-commit] pypy ffistruct: rename space.truncatedint into truncatedint_w, and move the corresponding test to test_objspace
antocuni
- [pypy-commit] pypy ffistruct: small refactor, and add a failing test
antocuni
- [pypy-commit] pypy ffistruct: test _as_ffi_pointer_ in type_converter
antocuni
- [pypy-commit] pypy ffistruct: these two functions are not_rpython
antocuni
- [pypy-commit] pypy generator-in-rpython: A first version of the test.
arigo
- [pypy-commit] pypy generator-in-rpython: Generators in the flow space: keep the structure of the graph,
arigo
- [pypy-commit] pypy int-tag-untag-as-operations: a tagged int cannot move
cfbolz
- [pypy-commit] pypy int-tag-untag-as-operations: can only untag odd ints
cfbolz
- [pypy-commit] pypy int-tag-untag-as-operations: merge
cfbolz
- [pypy-commit] pypy int-tag-untag-as-operations: merge default
cfbolz
- [pypy-commit] pypy int-tag-untag-as-operations: merge default (breaks some things, will fix in the next commit)
cfbolz
- [pypy-commit] pypy int-tag-untag-as-operations: refactor the int_tag/untag optimization to fit the new style
cfbolz
- [pypy-commit] pypy int-tag-untag-as-operations: same fix is needed here
cfbolz
- [pypy-commit] pypy int-tag-untag-as-operations: test and fix: make constand-folding of int_tag not crash
cfbolz
- [pypy-commit] pypy jit-dynamic-getarrayitem: added these to base model
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: close for merge
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: don't cache dynamic descrs, progress maybe
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: forgotten test file
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: implement the optimization for set
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: make these not static
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: merged default in
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: merged default in
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: missing attr
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: optimize {get, set}interiorfield_{raw, gc} for itemsizes that can be addressed using special x86 addressing
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: progress, gets are now compiled and executed properly
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: put tests in a sane subclass and only run the new ones in the x86 backend
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: remove the dupe
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: show the annotator this case isn't posssible
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: start working on the JIT optimization
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: started working on support for creating array get/set item at compile time
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: support in teh x86 backend
alex_gaynor
- [pypy-commit] pypy jit-dynamic-getarrayitem: translation-ish fix, except it still breaks
alex_gaynor
- [pypy-commit] pypy jit-refactor-tests: close merged branch
hakanardo
- [pypy-commit] pypy jit-refactor-tests: conversion in progress
hakanardo
- [pypy-commit] pypy jit-refactor-tests: convert test
hakanardo
- [pypy-commit] pypy jit-refactor-tests: convert test
hakanardo
- [pypy-commit] pypy jit-refactor-tests: convert test
hakanardo
- [pypy-commit] pypy jit-refactor-tests: convert tests
hakanardo
- [pypy-commit] pypy jit-refactor-tests: converted test
hakanardo
- [pypy-commit] pypy jit-refactor-tests: converted test
hakanardo
- [pypy-commit] pypy jit-refactor-tests: converted test
hakanardo
- [pypy-commit] pypy jit-refactor-tests: converted test
hakanardo
- [pypy-commit] pypy jit-refactor-tests: converted tests
hakanardo
- [pypy-commit] pypy jit-refactor-tests: convreted tests
hakanardo
- [pypy-commit] pypy jit-refactor-tests: count all operations in each test
hakanardo
- [pypy-commit] pypy jit-refactor-tests: fix indentation
hakanardo
- [pypy-commit] pypy jit-refactor-tests: hg merge default
hakanardo
- [pypy-commit] pypy jit-refactor-tests: hg merge default
hakanardo
- [pypy-commit] pypy jit-refactor-tests: hg merge default
hakanardo
- [pypy-commit] pypy jit-refactor-tests: hg merge default ; hg revert test_zjit.py -r default
hakanardo
- [pypy-commit] pypy jit-refactor-tests: kill check_loops
hakanardo
- [pypy-commit] pypy jit-refactor-tests: kill check_loops, it is now replaced with check_resops
hakanardo
- [pypy-commit] pypy jit-refactor-tests: merge
hakanardo
- [pypy-commit] pypy jit-refactor-tests: Resurrect check_loops
hakanardo
- [pypy-commit] pypy jit-targets: A check function that allows counting the ops in the peeled loop only. It only works in the simplest case when there is exactly one jump (ie no bridges not ending in finnish)
hakanardo
- [pypy-commit] pypy jit-targets: a first failed atempt to support retrace, we need to redesign...
hakanardo
- [pypy-commit] pypy jit-targets: abort unsupported case
hakanardo
- [pypy-commit] pypy jit-targets: adapt test framework to new interface
hakanardo
- [pypy-commit] pypy jit-targets: add test and comment
hakanardo
- [pypy-commit] pypy jit-targets: aliasing vrituals are pointless except that it generates a crash
hakanardo
- [pypy-commit] pypy jit-targets: allow orignial jump_args to be used in the peeled loop
hakanardo
- [pypy-commit] pypy jit-targets: Bah. Fix.
arigo
- [pypy-commit] pypy jit-targets: better way of checking that the retracecount is not exceeded
hakanardo
- [pypy-commit] pypy jit-targets: bridge support
hakanardo
- [pypy-commit] pypy jit-targets: Cannot attach the LoopToken by default on a jump(), because we
arigo
- [pypy-commit] pypy jit-targets: centralize target token counter
hakanardo
- [pypy-commit] pypy jit-targets: centralize the unrolling call to optimizeopt
hakanardo
- [pypy-commit] pypy jit-targets: count all operations in each test
hakanardo
- [pypy-commit] pypy jit-targets: dont crash if not inlining the same short preamble as is beeing produced
hakanardo
- [pypy-commit] pypy jit-targets: dont inline short preamble when falling back to the full preamble
hakanardo
- [pypy-commit] pypy jit-targets: dont replace the the JitCellToken when retracing
hakanardo
- [pypy-commit] pypy jit-targets: dont use invalidated loops
hakanardo
- [pypy-commit] pypy jit-targets: elements assigned in different order now
hakanardo
- [pypy-commit] pypy jit-targets: ensure loops are freed
hakanardo
- [pypy-commit] pypy jit-targets: exception support
hakanardo
- [pypy-commit] pypy jit-targets: extract the peeled loop from the combined preamble and peeled loop trace
hakanardo
- [pypy-commit] pypy jit-targets: fallback to preamble if inlining short preamble fails
hakanardo
- [pypy-commit] pypy jit-targets: first simple loop now passed along all the way in the new format using labels
hakanardo
- [pypy-commit] pypy jit-targets: first test with a brigde passing
hakanardo
- [pypy-commit] pypy jit-targets: fix
hakanardo
- [pypy-commit] pypy jit-targets: Fix runner_test.
arigo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test
hakanardo
- [pypy-commit] pypy jit-targets: fix test to actually count the number of specialized versions of the loop
hakanardo
- [pypy-commit] pypy jit-targets: fix test to not retrace when the guard is created and only to count the number of int_mul which is what the test is about
hakanardo
- [pypy-commit] pypy jit-targets: Fix test_random.
arigo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: fix tests
hakanardo
- [pypy-commit] pypy jit-targets: Fix tests.
arigo
- [pypy-commit] pypy jit-targets: Fix the test.
arigo
- [pypy-commit] pypy jit-targets: Fix the XXX.
arigo
- [pypy-commit] pypy jit-targets: Fix.
arigo
- [pypy-commit] pypy jit-targets: Fix.
arigo
- [pypy-commit] pypy jit-targets: Fix.
arigo
- [pypy-commit] pypy jit-targets: fixed test
hakanardo
- [pypy-commit] pypy jit-targets: fixed test
hakanardo
- [pypy-commit] pypy jit-targets: Fixes.
arigo
- [pypy-commit] pypy jit-targets: hack to support aliased virtuals
hakanardo
- [pypy-commit] pypy jit-targets: hg merge
hakanardo
- [pypy-commit] pypy jit-targets: hg merge
hakanardo
- [pypy-commit] pypy jit-targets: hg merge
hakanardo
- [pypy-commit] pypy jit-targets: hg merge
hakanardo
- [pypy-commit] pypy jit-targets: hg merge default
hakanardo
- [pypy-commit] pypy jit-targets: hg merge default
hakanardo
- [pypy-commit] pypy jit-targets: hg merge default
hakanardo
- [pypy-commit] pypy jit-targets: hg merge default
hakanardo
- [pypy-commit] pypy jit-targets: hg merge default
hakanardo
- [pypy-commit] pypy jit-targets: hg merge default (a bit painfully)
arigo
- [pypy-commit] pypy jit-targets: hg merge jit-refactor-tests
hakanardo
- [pypy-commit] pypy jit-targets: hg merge jit-refactor-tests
hakanardo
- [pypy-commit] pypy jit-targets: hg merge jit-refactor-tests
hakanardo
- [pypy-commit] pypy jit-targets: hg merge jit-refactor-tests
hakanardo
- [pypy-commit] pypy jit-targets: hg merge jit-refactor-tests
hakanardo
- [pypy-commit] pypy jit-targets: hg merge jit-refactor-tests
hakanardo
- [pypy-commit] pypy jit-targets: ignore for now
hakanardo
- [pypy-commit] pypy jit-targets: Improve randomness: generates a LABEL that is not always the very first
arigo
- [pypy-commit] pypy jit-targets: Improve the checking: kills values that are not explicitly given
arigo
- [pypy-commit] pypy jit-targets: indent
hakanardo
- [pypy-commit] pypy jit-targets: inject debug counter after initial label if present
hakanardo
- [pypy-commit] pypy jit-targets: introduce targets that can be placed somewhere in a trace that can be used jump targets
hakanardo
- [pypy-commit] pypy jit-targets: jumps are allowed to jump to a TargetToken beloning to another procedure
hakanardo
- [pypy-commit] pypy jit-targets: kill insert_loop_token
hakanardo
- [pypy-commit] pypy jit-targets: kill optimizer.bridge
hakanardo
- [pypy-commit] pypy jit-targets: kill OptInlineShortPreamble
hakanardo
- [pypy-commit] pypy jit-targets: make OptSimplify follow and test_optimizebasic follow the new model
hakanardo
- [pypy-commit] pypy jit-targets: make sure all jitcell tokens and traces are actually counted
hakanardo
- [pypy-commit] pypy jit-targets: make sure history.Stats dont keep things alive
hakanardo
- [pypy-commit] pypy jit-targets: merge heads
arigo
- [pypy-commit] pypy jit-targets: merge messup?
hakanardo
- [pypy-commit] pypy jit-targets: more consisten naming, fixed test
hakanardo
- [pypy-commit] pypy jit-targets: move the decition wheter to unroll or not back into optimizeopt
hakanardo
- [pypy-commit] pypy jit-targets: no need to export jump_args
hakanardo
- [pypy-commit] pypy jit-targets: no need to export values
hakanardo
- [pypy-commit] pypy jit-targets: optimize_trace might be forced to insert sameas operations infront of the label
hakanardo
- [pypy-commit] pypy jit-targets: place the virtual_state on the target_token
hakanardo
- [pypy-commit] pypy jit-targets: propagate quasi_immutable_deps
hakanardo
- [pypy-commit] pypy jit-targets: recursion support
hakanardo
- [pypy-commit] pypy jit-targets: reenable autoconstruction of jump descr in oparser
hakanardo
- [pypy-commit] pypy jit-targets: refactor unrolling to use the new target resop
hakanardo
- [pypy-commit] pypy jit-targets: refactoring in progress
hakanardo
- [pypy-commit] pypy jit-targets: reintroduce inputargs on loops
hakanardo
- [pypy-commit] pypy jit-targets: reintroduce logging of loops
hakanardo
- [pypy-commit] pypy jit-targets: rename
hakanardo
- [pypy-commit] pypy jit-targets: Rename ProcedureToken to JitCellToken. It now refers to all compiled traces starting from a specific JitCell and it is used as a decsr of jumps produced by the frontend to indicate the target of the jump. The optimizer will the convert this to a jump to a TargetToken (which refers to a specific label in an already compiled trace). If that is not yet possible it will be converted into a label resop with a new TargetTocken
hakanardo
- [pypy-commit] pypy jit-targets: rename TARGET to LABEL
hakanardo
- [pypy-commit] pypy jit-targets: rename TargetToken.cell_token into TargetToke.targeting_jitcell_token
hakanardo
- [pypy-commit] pypy jit-targets: renames
hakanardo
- [pypy-commit] pypy jit-targets: renaming, redefining and reeanbling the "loop" counters (in progress)
hakanardo
- [pypy-commit] pypy jit-targets: respect retrace limit
hakanardo
- [pypy-commit] pypy jit-targets: restrict match_by_id to ops in the peeled loop
hakanardo
- [pypy-commit] pypy jit-targets: retrace support
hakanardo
- [pypy-commit] pypy jit-targets: retraces ending with a virtual state matching a previously compiled trace
hakanardo
- [pypy-commit] pypy jit-targets: Start fixing graphpage.py.
arigo
- [pypy-commit] pypy jit-targets: started to fix tests (in progress)
hakanardo
- [pypy-commit] pypy jit-targets: store the short preamble on the TargetToken instead
hakanardo
- [pypy-commit] pypy jit-targets: Subtle fix. Used to show up in test_loop_unroll occasionally.
arigo
- [pypy-commit] pypy jit-targets: support CALL_ASSEMBLER
hakanardo
- [pypy-commit] pypy jit-targets: support for bridges in progress
hakanardo
- [pypy-commit] pypy jit-targets: Support LABEL in the x86 backend. Probably not RPython yet.
arigo
- [pypy-commit] pypy jit-targets: support max_retrace_guards
hakanardo
- [pypy-commit] pypy jit-targets: Test and fix, laboriously.
arigo
- [pypy-commit] pypy jit-targets: test_ajit.test_basic now passing
hakanardo
- [pypy-commit] pypy jit-targets: test_loop_1 passing
hakanardo
- [pypy-commit] pypy jit-targets: test_target_loop_kept_alive_or_not relies on send_bridge_to_backend() NOT increasing the generation of its original preamble (its current action on default is to increas the generation of the peeled loop, which is a bit of a no-op). Is this realy the desiered behaviour?
hakanardo
- [pypy-commit] pypy jit-targets: traces from interpreter now working again
hakanardo
- [pypy-commit] pypy jit-targets: translation fix
hakanardo
- [pypy-commit] pypy jit-targets: translation fix
hakanardo
- [pypy-commit] pypy jit-targets: translation fix
hakanardo
- [pypy-commit] pypy jit-targets: translation fixes
arigo
- [pypy-commit] pypy jit-targets: use TargetToken to refere to a target
hakanardo
- [pypy-commit] pypy jit-targets: we now need inputargs again...
hakanardo
- [pypy-commit] pypy jit-targets: we still need it here
hakanardo
- [pypy-commit] pypy jit-targets: XXX a reminder, for later
arigo
- [pypy-commit] pypy jitdriver-setparam-all: Change set_param API. Now we run set_param(driver-or-None, 'name', value)
fijal
- [pypy-commit] pypy jitdriver-setparam-all: close about to be merged branch
fijal
- [pypy-commit] pypy jitdriver-setparam-all: Re-add an assert that was lost.
arigo
- [pypy-commit] pypy list-strategies: added more tests for so far uncovered code in listobject.setslice
l.diekmann
- [pypy-commit] pypy list-strategies: Closing branch to be merged.
alex_gaynor
- [pypy-commit] pypy list-strategies: Create 2 versions of GeneratorIterator.unpack_into, one which takes a W_ListObject and one which takes an RPython list. This is to allow unwrapped lists to be built from generators with intermediaries (and to fix a translation error).
alex_gaynor
- [pypy-commit] pypy list-strategies: don't encode exact offsets
cfbolz
- [pypy-commit] pypy list-strategies: don't use the JIT strslice optimization if some of the characters are in an unknown state with regards to whether they're initialized
alex_gaynor
- [pypy-commit] pypy list-strategies: fix from merge
alex_gaynor
- [pypy-commit] pypy list-strategies: implement a fast path for list.pop() (without arguments)
cfbolz
- [pypy-commit] pypy list-strategies: merge default
fijal
- [pypy-commit] pypy list-strategies: merged default in
alex_gaynor
- [pypy-commit] pypy list-strategies: merged default in
alex_gaynor
- [pypy-commit] pypy list-strategies: Merged default in, resolved merge conflicts (involved removing an optimization that had been done in a different way on default).
alex_gaynor
- [pypy-commit] pypy list-strategies: remove dead/commented out code
fijal
- [pypy-commit] pypy list-strategies: untested paths
fijal
- [pypy-commit] pypy list-strategies: When using a string list-strategy have the same behavior on str.join with one element lists.
alex_gaynor
- [pypy-commit] pypy matrixmath-reshape: create branch
mattip
- [pypy-commit] pypy matrixmath-reshape: fix shape, reshape. Ready for review
mattip
- [pypy-commit] pypy matrixmath: add a REVIEW file, just because IRC logs might be lost
fijal
- [pypy-commit] pypy matrixmath: add explicit test for arange float
mattip
- [pypy-commit] pypy matrixmath: Add responses to REVIEW
mattip
- [pypy-commit] pypy matrixmath: add test for numpy.flatiter, more tests for transpose
mattip
- [pypy-commit] pypy matrixmath: adding set_shape, some tests pass, some fail
mattip
- [pypy-commit] pypy matrixmath: anothe review note
alex_gaynor
- [pypy-commit] pypy matrixmath: close merged branch
fijal
- [pypy-commit] pypy matrixmath: find_size no longer raises ValueError
fijal
- [pypy-commit] pypy matrixmath: fix find_size for Scalar
mattip
- [pypy-commit] pypy matrixmath: leave branch-unrelated comment
fijal
- [pypy-commit] pypy matrixmath: make 0-dimension arrays (scalars) numpy-compatible
mattip
- [pypy-commit] pypy matrixmath: maybe found a problem, add comment instead of fixing it
mattip
- [pypy-commit] pypy matrixmath: merge
fijal
- [pypy-commit] pypy matrixmath: merge (one)
fijal
- [pypy-commit] pypy matrixmath: Merge to avoid creating new head, not ready for review
mattip
- [pypy-commit] pypy matrixmath: Merge to avoid creating new head, not ready for review
mattip
- [pypy-commit] pypy matrixmath: Merge transpose, reshape/shape
mattip
- [pypy-commit] pypy matrixmath: Merge work in progress
mattip
- [pypy-commit] pypy matrixmath: Merge: arange, transpose ready for review
mattip
- [pypy-commit] pypy matrixmath: merged default in
alex_gaynor
- [pypy-commit] pypy matrixmath: oops, did not see arange is applevel. Commit a test and remove a REVIEW item
fijal
- [pypy-commit] pypy matrixmath: Remove the notion of reshape/set_shape and clean up the branch
fijal
- [pypy-commit] pypy matrixmath: remove, all items are done
fijal
- [pypy-commit] pypy matrixmath: Simply kill descr_new_flatiter
fijal
- [pypy-commit] pypy matrixmath: Some cleanup: remove parameters that's never used, rename class, delete some more setshape code, fix a case of arange with bools.
alex_gaynor
- [pypy-commit] pypy matrixmath: some code-prettyness changes
mattip
- [pypy-commit] pypy matrixmath: split tests - work in progress
mattip
- [pypy-commit] pypy matrixmath: test, implelent arange
mattip
- [pypy-commit] pypy matrixmath: tests, implementation of flat
mattip
- [pypy-commit] pypy matrixmath: tests, implementation of transpose
mattip
- [pypy-commit] pypy micronumpy-resync: close old abandoned branch
fijal
- [pypy-commit] pypy nedbat-sandbox: An implementation of do_ll_os__ll_os_fstat, no tests, not sure if it's right.
ned
- [pypy-commit] pypy nedbat-sandbox: Avoid a compile at startup by using our own marshalling of stat results, instead of calling into rmarshal to do it.
ned
- [pypy-commit] pypy numpy NDimArray: close abandoned branch, work happen on numpy-multidim one
fijal
- [pypy-commit] pypy numpy-dtype-refactor-complex: initial work on complex values
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: Added a decorator for specialize:call_location
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: added files I forgot
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: all tests pass! now it just needs to be made RPyhon
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: allowisntantiating numpy boxes from applevel
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: begin refactoring everything. nothing works.
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: bit of cleanup
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: comment these out, they're in the wrong place in the MRO with the mixins
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: consistancy
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: convert to use the new libffi support
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: don't try to do arithmatic on small types
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: expose another thing at app level
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: expose more stuff at app level
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: expose some more stuff at app-levle
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: first few translation fixes
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: fix
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: fix a bunch of tests, not sure how I ddi'nt catch this before
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: fix tests for 32-bit
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: fix the bool tests
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: fix translation
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: fix-ish, need to talk to fijal about why these mehods have different signatures than expected
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: get these tests passing again
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: major progress towards translating
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: many more tests passing
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default in
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default in
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default in
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default in
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default in
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: merged default in
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: Merged default in.
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: Merged default in.
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: Merged default in.
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: more progress towards translating
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: more translation fixes
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: more updates
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: more updates to code and tests
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: progress, we now inherit from int
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: put a ton of ufuncs back, more methods on numpy.generic objs, a few more fixes
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: recognize scalars in array's constrcutor when guessing types
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: remove some dead code, doesn't fix translation as I hoped
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: running all tests together now passes
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: Specialize these properly
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: start re-adding many ops
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: start to re-flesh out the dtype interface. now we get to the fun part of exposing the boxes at app level
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: stuff seems to translate!
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: subclassing types, failing tests for bools (why are they so special :()
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: Teach llimpl about {get, set}interiorfield_raw with floats.
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: tons more box types, and some methods on them
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: translation fix
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: translation fix
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: translation fixes
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: update for new module name
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: update the meta data on these funcs
alex_gaynor
- [pypy-commit] pypy numpy-dtype-refactor: Update these tests.
alex_gaynor
- [pypy-commit] pypy numpy-monkeyaround: abandon tests, implementation of numpy arange, transpose
mattip
- [pypy-commit] pypy numpy-monkeyaround: add test and implementation for transpose, add test for set_shape
mattip
- [pypy-commit] pypy numpy-monkeyaround: Add test for arange
mattip
- [pypy-commit] pypy numpy-monkeyaround: branch for experimental implementations
mattip
- [pypy-commit] pypy numpy-monkeyaround: test, implement arange
mattip
- [pypy-commit] pypy numpy-multidim-exp: close abandoned branch, multidim support in trunk
fijal
- [pypy-commit] pypy numpy-multidim-shards: A branch to use shards instead of just raw indexes that have to be computed
fijal
- [pypy-commit] pypy numpy-multidim-shards: a fix and a comment
fijal
- [pypy-commit] pypy numpy-multidim-shards: add a passing test, just because
fijal
- [pypy-commit] pypy numpy-multidim-shards: add shape_agreement broadcast test, fix for it
mattip
- [pypy-commit] pypy numpy-multidim-shards: broadcast passes tests, code needs review
mattip
- [pypy-commit] pypy numpy-multidim-shards: clean up tests since broadcast, iterator changes broke them
mattip
- [pypy-commit] pypy numpy-multidim-shards: close merged branch
fijal
- [pypy-commit] pypy numpy-multidim-shards: code cleanup
mattip
- [pypy-commit] pypy numpy-multidim-shards: don't copy a list any more - the cause was fixed in 0bdcf803c0fd
fijal
- [pypy-commit] pypy numpy-multidim-shards: Fight a bit with test_zjit. I'm still not 100% happy, but getting there
fijal
- [pypy-commit] pypy numpy-multidim-shards: fix compile tests
fijal
- [pypy-commit] pypy numpy-multidim-shards: fix multidim tests
fijal
- [pypy-commit] pypy numpy-multidim-shards: fix signatures
fijal
- [pypy-commit] pypy numpy-multidim-shards: fix those tests - downside: they're for fortran layout
fijal
- [pypy-commit] pypy numpy-multidim-shards: Fix until some tests start passing. Introduce more iterators
fijal
- [pypy-commit] pypy numpy-multidim-shards: Implement shards correctly. Allow fortran and C order. Temporarily replace repr
fijal
- [pypy-commit] pypy numpy-multidim-shards: in-progress. Get this into some shape so we can run tests
fijal
- [pypy-commit] pypy numpy-multidim-shards: Make array iterators a once-off immutable things.
fijal
- [pypy-commit] pypy numpy-multidim-shards: make create_slice have a better interface
fijal
- [pypy-commit] pypy numpy-multidim-shards: make more tests pass
fijal
- [pypy-commit] pypy numpy-multidim-shards: make ones support multidim arrays. Also write a passing test
fijal
- [pypy-commit] pypy numpy-multidim-shards: merge default
fijal
- [pypy-commit] pypy numpy-multidim-shards: merge default in
fijal
- [pypy-commit] pypy numpy-multidim-shards: module name change
mattip
- [pypy-commit] pypy numpy-multidim-shards: more rpythonization (and potential crash avoidance)
fijal
- [pypy-commit] pypy numpy-multidim-shards: move to_str to the base class
fijal
- [pypy-commit] pypy numpy-multidim-shards: one more test
fijal
- [pypy-commit] pypy numpy-multidim-shards: one more test and make the other test failing without problems with repr
fijal
- [pypy-commit] pypy numpy-multidim-shards: one more test and start working on repr
fijal
- [pypy-commit] pypy numpy-multidim-shards: pass few more tests. everything has shards and backshards
fijal
- [pypy-commit] pypy numpy-multidim-shards: progress
fijal
- [pypy-commit] pypy numpy-multidim-shards: promote shape length
fijal
- [pypy-commit] pypy numpy-multidim-shards: provide some sort of descr_repr (a broken one) and a fix
fijal
- [pypy-commit] pypy numpy-multidim-shards: remove unused parameter
fijal
- [pypy-commit] pypy numpy-multidim-shards: rename shards to strides
fijal
- [pypy-commit] pypy numpy-multidim-shards: repr and str pass tests
mattip
- [pypy-commit] pypy numpy-multidim-shards: rpython fixes (?)
fijal
- [pypy-commit] pypy numpy-multidim-shards: rpythonize a bit. Disable _immutable_fields_ for now
fijal
- [pypy-commit] pypy numpy-multidim-shards: some partial fixes, will continue later
fijal
- [pypy-commit] pypy numpy-multidim-shards: split tests, test_repr passes
mattip
- [pypy-commit] pypy numpy-multidim-shards: start to fix descr_repr
mattip
- [pypy-commit] pypy numpy-multidim-shards: start working on broadcasting - a helper function
fijal
- [pypy-commit] pypy numpy-multidim: (antocuni, mwp)merge heads, wanted to checkin on default, did it on branch by mistake
mwp
- [pypy-commit] pypy numpy-multidim: (mwp antocuni) make platform work on PowerPC
mwp
- [pypy-commit] pypy numpy-multidim: a nicely passing test and a test we want to work on
fijal
- [pypy-commit] pypy numpy-multidim: a skipped test
fijal
- [pypy-commit] pypy numpy-multidim: Add two tests. One breaks, but it's not exactly a merge blocker
fijal
- [pypy-commit] pypy numpy-multidim: argmax works for multidim now
alex_gaynor
- [pypy-commit] pypy numpy-multidim: avoid exposing random parameters to python
alex_gaynor
- [pypy-commit] pypy numpy-multidim: better comment
alex_gaynor
- [pypy-commit] pypy numpy-multidim: close branch for merge
alex_gaynor
- [pypy-commit] pypy numpy-multidim: Few missing tests, code improvements
fijal
- [pypy-commit] pypy numpy-multidim: fix test_zjit until test_slice
fijal
- [pypy-commit] pypy numpy-multidim: fixes for translation
mattip
- [pypy-commit] pypy numpy-multidim: For now multidimarray.argmax is a TypeError. Wait until we implement flatiter
fijal
- [pypy-commit] pypy numpy-multidim: kill a function that was only used by one test. a bit pep-8ify, not too
fijal
- [pypy-commit] pypy numpy-multidim: leave a comment
fijal
- [pypy-commit] pypy numpy-multidim: leave some comments
fijal
- [pypy-commit] pypy numpy-multidim: make compile use a real tokenizer - breaks test_zjit for now
fijal
- [pypy-commit] pypy numpy-multidim: make parser not rpython (we'll think about it later) and use the same trick
fijal
- [pypy-commit] pypy numpy-multidim: match the percise error message of numpy
alex_gaynor
- [pypy-commit] pypy numpy-multidim: merge
fijal
- [pypy-commit] pypy numpy-multidim: merge numpy-multidim-shards
fijal
- [pypy-commit] pypy numpy-multidim: Merge str and repr cleanup
mattip
- [pypy-commit] pypy numpy-multidim: merged default in
alex_gaynor
- [pypy-commit] pypy numpy-multidim: pep-8, use StringBuilder
mattip
- [pypy-commit] pypy numpy-multidim: pep8
fijal
- [pypy-commit] pypy numpy-multidim: progress on test_zjit. Now test_slice is almost ready, barring few arraylen_gc
fijal
- [pypy-commit] pypy numpy-multidim: raise NotImplementedError in case we call reduce directly on a multidim array
fijal
- [pypy-commit] pypy numpy-multidim: repr/str: add tests for numpy compliance, code cleanup
mattip
- [pypy-commit] pypy numpy-multidim: shuffle code around. Now get_code lives outside of tests
fijal
- [pypy-commit] pypy numpy-multidim: simplification, a new (and unused) class and some comments
fijal
- [pypy-commit] pypy numpy-multidim: some code-prettyness changes
alex_gaynor
- [pypy-commit] pypy numpy-multidim: some work on broadcast iterator. test_zjit requires some support from
fijal
- [pypy-commit] pypy op_malloc_gc: A branch to simplify the backend interface: instead of
arigo
- [pypy-commit] pypy op_malloc_gc: Attempt to collapse several NEW-like operations into a single
arigo
- [pypy-commit] pypy op_malloc_gc: Clean up the tests.
arigo
- [pypy-commit] pypy op_malloc_gc: Clean-ups, comments, reorderings.
arigo
- [pypy-commit] pypy op_malloc_gc: Fix some tests.
arigo
- [pypy-commit] pypy op_malloc_gc: hg merge default
arigo
- [pypy-commit] pypy op_malloc_gc: In-progress.
arigo
- [pypy-commit] pypy op_malloc_gc: Kill unused attribute.
arigo
- [pypy-commit] pypy op_malloc_gc: Missing rounding up: tests and fix.
arigo
- [pypy-commit] pypy op_malloc_gc: Move the new tests to their own file.
arigo
- [pypy-commit] pypy op_malloc_gc: newstr, newunicode.
arigo
- [pypy-commit] pypy op_malloc_gc: No-op.
arigo
- [pypy-commit] pypy op_malloc_gc: Plan which tests are still needed (and the corresponding code).
arigo
- [pypy-commit] pypy op_malloc_gc: Progress.
arigo
- [pypy-commit] pypy op_malloc_gc: Progress.
arigo
- [pypy-commit] pypy op_malloc_gc: Refactor more. Test_rewrite passes.
arigo
- [pypy-commit] pypy op_malloc_gc: Refactor this logic into its own class, shared by the two GC strategies.
arigo
- [pypy-commit] pypy op_malloc_gc: Refactoring, in-progress.
arigo
- [pypy-commit] pypy op_malloc_gc: Test and fix for allocating tiny structures --- they are smaller
arigo
- [pypy-commit] pypy op_malloc_gc: The interface that I really want.
arigo
- [pypy-commit] pypy ppc-jit-backend: (bivab, hager): Fixed bug in decode64
hager
- [pypy-commit] pypy ppc-jit-backend: (bivab, hager): Make first tests on PPC64 pass
hager
- [pypy-commit] pypy ppc-jit-backend: (bivab, arigo, edelsohn, hager): Fix strange bug by removing r13 from the MANAGED_REGS and adding a line of long forgotten code.
hager
- [pypy-commit] pypy ppc-jit-backend: (bivab, edelsohn, hager): change size of allocated stack space at function calls
hager
- [pypy-commit] pypy ppc-jit-backend: (bivab, hager):
hager
- [pypy-commit] pypy ppc-jit-backend: (bivab, hager): Code refactorings and debugging in int operations
hager
- [pypy-commit] pypy ppc-jit-backend: (bivab, hager): Make function descriptor in case of 64 bit for the generated machine code.
hager
- [pypy-commit] pypy ppc-jit-backend: (bivab, hager): Read function address out of function descriptor in case of 64 bit.
hager
- [pypy-commit] pypy ppc-jit-backend: (edelsohn, hager): Some experiments in _gen_exit_path
hager
- [pypy-commit] pypy ppc-jit-backend: _ppc_direct_bootstrap_code must be function descriptor on PPC64
edelsohn
- [pypy-commit] pypy ppc-jit-backend: add a helper method to the register manager to allocate a scratch register
bivab
- [pypy-commit] pypy ppc-jit-backend: Add PPC64 ovf operations
edelsohn
- [pypy-commit] pypy ppc-jit-backend: add test for direct call of assembler code
hager
- [pypy-commit] pypy ppc-jit-backend: Add ugly way to avoid problems with holes in MANAGED_REGS
hager
- [pypy-commit] pypy ppc-jit-backend: Added a few methods which will be needed later on
hager
- [pypy-commit] pypy ppc-jit-backend: Added code for call to C functions again :(
hager
- [pypy-commit] pypy ppc-jit-backend: Added code that verifies sane use of scratch register.
hager
- [pypy-commit] pypy ppc-jit-backend: Added regalloc_push and regalloc_pop => test_jump passes
hager
- [pypy-commit] pypy ppc-jit-backend: Added some comments in ppcgen/ppc_assembler.py
hager
- [pypy-commit] pypy ppc-jit-backend: Adjust emit_call for PPC64 ABI.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Adjust frame layout and store TOC pointer.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Alias TOC as r2.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Align PPC64 stack
edelsohn
- [pypy-commit] pypy ppc-jit-backend: allocate and use a scratch register for get/set arrayitem in case the scale > 0
bivab
- [pypy-commit] pypy ppc-jit-backend: Allow allocation without setting the scratch register value
hager
- [pypy-commit] pypy ppc-jit-backend: another fix for the ppc tests
bivab
- [pypy-commit] pypy ppc-jit-backend: asm_helper_adr is function descriptor on PPC64
edelsohn
- [pypy-commit] pypy ppc-jit-backend: changed BACKCHAIN_SIZE to 3 * WORD
hager
- [pypy-commit] pypy ppc-jit-backend: Correct indentation of prepare_setarrayitem_raw.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Deallocate stack in emit_call on PPC64 path.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Define getarrayitem_raw, getfield_raw, etc. Sign extend getfield result.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Different backchain sizes depending on the architecture
hager
- [pypy-commit] pypy ppc-jit-backend: Do sanity check
hager
- [pypy-commit] pypy ppc-jit-backend: enable direct bootstrap code
hager
- [pypy-commit] pypy ppc-jit-backend: Ensure exclusive usage of r0
hager
- [pypy-commit] pypy ppc-jit-backend: Ensure exlusive usage of r0 in b_abs
hager
- [pypy-commit] pypy ppc-jit-backend: First set SP when saving managed registers
hager
- [pypy-commit] pypy ppc-jit-backend: fix an issue in clibffi that is triggered on big endian platforms due to the byte order when casting a larger data type to smaller one to be passed to a function called through ffi
bivab
- [pypy-commit] pypy ppc-jit-backend: fix for tests
bivab
- [pypy-commit] pypy ppc-jit-backend: Fixed assertion that made multiple tests fail due to side effect
hager
- [pypy-commit] pypy ppc-jit-backend: fixed offset bug in loading parameters from stack
hager
- [pypy-commit] pypy ppc-jit-backend: Handle stack alignment in emit_call
hager
- [pypy-commit] pypy ppc-jit-backend: If functions with many (> 8) arguments are called,
hager
- [pypy-commit] pypy ppc-jit-backend: Implemented calls to C functions.
hager
- [pypy-commit] pypy ppc-jit-backend: Implemented COPYSTRCONTENT
hager
- [pypy-commit] pypy ppc-jit-backend: Implemented COPYUNICODECONTENT
hager
- [pypy-commit] pypy ppc-jit-backend: Implemented NEW_ARRAY
hager
- [pypy-commit] pypy ppc-jit-backend: Implemented NEW_WITH_VTABLE
hager
- [pypy-commit] pypy ppc-jit-backend: Implemented NEWSTR and NEWUNICODE
hager
- [pypy-commit] pypy ppc-jit-backend: Make PPC64 _emit_call abi offset clearer.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: merge
hager
- [pypy-commit] pypy ppc-jit-backend: merge
hager
- [pypy-commit] pypy ppc-jit-backend: merge
hager
- [pypy-commit] pypy ppc-jit-backend: merge
hager
- [pypy-commit] pypy ppc-jit-backend: merge
hager
- [pypy-commit] pypy ppc-jit-backend: merge upstream
bivab
- [pypy-commit] pypy ppc-jit-backend: more test fixes
bivab
- [pypy-commit] pypy ppc-jit-backend: Optimize zero-extend and sign-extend in _ensure_result_bit_extension
edelsohn
- [pypy-commit] pypy ppc-jit-backend: PPC64 guard compares
edelsohn
- [pypy-commit] pypy ppc-jit-backend: PPC64 support for _save_managed_regs
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Re-adding David's changes, which I killed on merge :(
hager
- [pypy-commit] pypy ppc-jit-backend: refactor load_from_addr. clarify register usage in store_reg.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Refactored code in opassembler.py, divided the code in class OpAssembler
hager
- [pypy-commit] pypy ppc-jit-backend: Remove GPR 1 from those registers which are preserved across function calls
hager
- [pypy-commit] pypy ppc-jit-backend: Remove some comments and blank lines
hager
- [pypy-commit] pypy ppc-jit-backend: Remove syntax error
hager
- [pypy-commit] pypy ppc-jit-backend: Remove typo from _gen_exit_path
hager
- [pypy-commit] pypy ppc-jit-backend: Replace cmpi with cmpwi and cmpdi
hager
- [pypy-commit] pypy ppc-jit-backend: saved_registers must allocate and deallocate a full stack frame
edelsohn
- [pypy-commit] pypy ppc-jit-backend: setarrayitem and getarrayitem cannot add offset with addi.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: setarrayitem and getarrayitem offsets are immediate values.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Started implementation of CALL_ASSEMBLER
hager
- [pypy-commit] pypy ppc-jit-backend: Started implementation of NEW, first test passes
hager
- [pypy-commit] pypy ppc-jit-backend: Store PPC64 LR at frame_depth + WORD in prologue. Load R2 in
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Tentatively rewrite push_arg_as_ffiptr().
arigo
- [pypy-commit] pypy ppc-jit-backend: Use a more clear way to compute offsets
hager
- [pypy-commit] pypy ppc-jit-backend: use force index field as scratch field
hager
- [pypy-commit] pypy ppc-jit-backend: Use PPC64 instructions in _emit_call stack adjustment.
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Use r0 as one-element stack
hager
- [pypy-commit] pypy ppc-jit-backend: Use rlwinm for PPC32 zero-extend
edelsohn
- [pypy-commit] pypy ppc-jit-backend: Use safe way for relative jumps
hager
- [pypy-commit] pypy ppc-jit-backend: Use sld for PPC64 shifts in getitem and unicode.
edelsohn
- [pypy-commit] pypy py3k: (chronitis) Add a couple more string test cases, and fix contains and expandtabs appropriately.
amauryfa
- [pypy-commit] pypy py3k: (chronitis) Add bytes.maketrans, and allow buffer-compatible objects to
amauryfa
- [pypy-commit] pypy py3k: (chronitis) Implement bytes.fromhex().
amauryfa
- [pypy-commit] pypy py3k: (chronitis) String tests should test (bytes) strings.
amauryfa
- [pypy-commit] pypy py3k: (chronitis) update std.__doc__ and bytes.__doc__.
amauryfa
- [pypy-commit] pypy py3k: 2to3
pjenvey
- [pypy-commit] pypy py3k: _warnings is always required by test.regrtest
amauryfa
- [pypy-commit] pypy py3k: a bunch of fixes for complex, includes removing floordiv, divmod, and mod
alex_gaynor
- [pypy-commit] pypy py3k: A non quadratic implementation of random.getrandbits(),
amauryfa
- [pypy-commit] pypy py3k: added cmath.isfinite
alex_gaynor
- [pypy-commit] pypy py3k: Allow bytes source code in compile()
amauryfa
- [pypy-commit] pypy py3k: backout 21b2914fdb96 pending type.name switching to unicode
pjenvey
- [pypy-commit] pypy py3k: Buffered IO should catch EINTR and retry.
amauryfa
- [pypy-commit] pypy py3k: builtins.ascii
pjenvey
- [pypy-commit] pypy py3k: Convert from __nonzero__ to __bool__.
alex_gaynor
- [pypy-commit] pypy py3k: Disallow implicit concatenation of bytes and strings
amauryfa
- [pypy-commit] pypy py3k: Fix _random.get_randbits()
amauryfa
- [pypy-commit] pypy py3k: fix bytes' repr
pjenvey
- [pypy-commit] pypy py3k: Fix for py.test running CPython test suite
amauryfa
- [pypy-commit] pypy py3k: Fix most tests in __builtin__ module,
amauryfa
- [pypy-commit] pypy py3k: Fix object.__reduce_ex__
amauryfa
- [pypy-commit] pypy py3k: Fix pickle of builtin types.
amauryfa
- [pypy-commit] pypy py3k: fix str() on bytes, reenable the -b cmd line opt
pjenvey
- [pypy-commit] pypy py3k: fix test_escape_encode (thanks amaury)
pjenvey
- [pypy-commit] pypy py3k: Fix tests in module/rctime
amauryfa
- [pypy-commit] pypy py3k: Fix translation
amauryfa
- [pypy-commit] pypy py3k: fix translation
pjenvey
- [pypy-commit] pypy py3k: fixes for app_main
alex_gaynor
- [pypy-commit] pypy py3k: get ascii translating
pjenvey
- [pypy-commit] pypy py3k: gzip needs zlib of course
amauryfa
- [pypy-commit] pypy py3k: handle dict views in dir()
pjenvey
- [pypy-commit] pypy py3k: hg merge default
amauryfa
- [pypy-commit] pypy py3k: hg merge default
amauryfa
- [pypy-commit] pypy py3k: hg merge default
amauryfa
- [pypy-commit] pypy py3k: imp.get_magic() returns bytes
amauryfa
- [pypy-commit] pypy py3k: Implement imp.cache_from_source()
amauryfa
- [pypy-commit] pypy py3k: Implement IOBase._checkClosed(), will maybe fix test_fptlib and test_poplib.
amauryfa
- [pypy-commit] pypy py3k: Implement sys.hash_info. Initial patch by chronitis, thanks!
amauryfa
- [pypy-commit] pypy py3k: improve pep3120 support
pjenvey
- [pypy-commit] pypy py3k: Make sure sys.stdout or sys.stderr are initialized when app_main prints
amauryfa
- [pypy-commit] pypy py3k: merge default
pjenvey
- [pypy-commit] pypy py3k: merge default
pjenvey
- [pypy-commit] pypy py3k: merge heads
amauryfa
- [pypy-commit] pypy py3k: merge heads
amauryfa
- [pypy-commit] pypy py3k: merged upstream
alex_gaynor
- [pypy-commit] pypy py3k: merged upstream
alex_gaynor
- [pypy-commit] pypy py3k: no longer needed in py3k
pjenvey
- [pypy-commit] pypy py3k: partially revert 3e341a64f85c, conftest needs to stay py
pjenvey
- [pypy-commit] pypy py3k: pass through source as bytes to the compiler
pjenvey
- [pypy-commit] pypy py3k: reapply our sysconfig modifications to 3.2
pjenvey
- [pypy-commit] pypy py3k: Remove dict.has_key, and switch dict.{keys, values, items} to return views and remove the view* methods.
alex_gaynor
- [pypy-commit] pypy py3k: remove remnants of the getslice operations
amauryfa
- [pypy-commit] pypy py3k: Rename: sys.long_info -> sys.int_info
amauryfa
- [pypy-commit] pypy py3k: Share code between bytes and bytearray constructors.
amauryfa
- [pypy-commit] pypy py3k: split up these tests
pjenvey
- [pypy-commit] pypy py3k: support the TextIOWrapper write_through option
pjenvey
- [pypy-commit] pypy py3k: test for rev a94eb72cdc3f
pjenvey
- [pypy-commit] pypy py3k: unicode string should not join bytes items
amauryfa
- [pypy-commit] pypy py3k: Update the list of test files with the content of 3.2/test/test_*.py.
amauryfa
- [pypy-commit] pypy py3k: update type's repr
pjenvey
- [pypy-commit] pypy py3k: zlib only deals with bytes, not str
amauryfa
- [pypy-commit] pypy release-1.7.x: Added tag release-1.7 for changeset ff4af8f31882
fijal
- [pypy-commit] pypy release-1.7.x: Create a 1.7 branch. Bump version numbers
fijal
- [pypy-commit] pypy release-1.7.x: fix the test
fijal
- [pypy-commit] pypy release-1.7.x: got continuelets ready for windows.
ctismer
- [pypy-commit] pypy release-1.7.x: made the assembler sources compile so far
ctismer
- [pypy-commit] pypy release-1.7.x: merge in default
fijal
- [pypy-commit] pypy release-1.7.x: relevant_environ updated for windows
ctismer
- [pypy-commit] pypy release-1.7.x: rstacklet.py builds ok, but crashes when run.
ctismer
- [pypy-commit] pypy rgc-mem-pressure: closed merged branch
fijal
- [pypy-commit] pypy rgc-mem-pressure: optimize it slightly. not look up the dictionary each time we see _digest_size
fijal
- [pypy-commit] pypy separate-applevel-numpy: merge default
fijal
- [pypy-commit] pypy set-strategies: _is_sane_hash was renamed to _never_equal_to_string
l.diekmann
- [pypy-commit] pypy set-strategies: _never_equal_to_string makes no sense here
l.diekmann
- [pypy-commit] pypy set-strategies: _newobj moved to W_SetObject and W_FrozenSetObject
l.diekmann
- [pypy-commit] pypy set-strategies: added _is_sane_hash to delitem; fixed _is_sane_hash in has_key; added strategy tests
l.diekmann
- [pypy-commit] pypy set-strategies: added different method for symmetric_difference_update when strategies match
l.diekmann
- [pypy-commit] pypy set-strategies: added docstrings
l.diekmann
- [pypy-commit] pypy set-strategies: added EmptySetStrategy + tests
l.diekmann
- [pypy-commit] pypy set-strategies: added fastpath for issubset and isdisjoint
l.diekmann
- [pypy-commit] pypy set-strategies: added fastpath for not comparable sets (starting with difference)
l.diekmann
- [pypy-commit] pypy set-strategies: added fastpath to intersection and fixed not_contain_equal_elements
l.diekmann
- [pypy-commit] pypy set-strategies: added fix and tests for clear and __sub__
l.diekmann
- [pypy-commit] pypy set-strategies: added fixes and tests for symmetric_difference[_update]
l.diekmann
- [pypy-commit] pypy set-strategies: added from_storage_and_strategy function
l.diekmann
- [pypy-commit] pypy set-strategies: added intelligent way to treat the different strategies in W_SetObject.difference
l.diekmann
- [pypy-commit] pypy set-strategies: added methods raising NotImplemented error
l.diekmann
- [pypy-commit] pypy set-strategies: added strategy test for union
l.diekmann
- [pypy-commit] pypy set-strategies: added StringStrategy for sets
l.diekmann
- [pypy-commit] pypy set-strategies: added test and fix for inplace_and
l.diekmann
- [pypy-commit] pypy set-strategies: added test and fix for set(<generator>)
l.diekmann
- [pypy-commit] pypy set-strategies: added test and fix for inplace sub
l.diekmann
- [pypy-commit] pypy set-strategies: added test and fix for inplace_or
l.diekmann
- [pypy-commit] pypy set-strategies: added test and fix for issubset and issuperset
l.diekmann
- [pypy-commit] pypy set-strategies: added test and fix for update on empty sets
l.diekmann
- [pypy-commit] pypy set-strategies: added test for user generated subclass of setobject
l.diekmann
- [pypy-commit] pypy set-strategies: added tests and fix for unhashable items in combination with EmptySetStrategy
l.diekmann
- [pypy-commit] pypy set-strategies: added tests for intersection_multiple order
l.diekmann
- [pypy-commit] pypy set-strategies: added tests for setstrategies
l.diekmann
- [pypy-commit] pypy set-strategies: All tests for setobject are working (but there is still untested code)
l.diekmann
- [pypy-commit] pypy set-strategies: also copy storage of frozenset to avoid changing frozenset in methods like intersection, difference, etc
l.diekmann
- [pypy-commit] pypy set-strategies: Altough the if-part will never be executed in IntegerSetStrategy, the annotator doesn't know what type d is. It could be an int-dict and then d[w_key], where w_key is always a wrapped object because of the getkeys()-method, would degenerate this object to an integer.
l.diekmann
- [pypy-commit] pypy set-strategies: always use issubset instead of issuperset
l.diekmann
- [pypy-commit] pypy set-strategies: another test for discard; cleaned up discard code
l.diekmann
- [pypy-commit] pypy set-strategies: another way of creating a frozen set
l.diekmann
- [pypy-commit] pypy set-strategies: argument must be None to create a new empty set
l.diekmann
- [pypy-commit] pypy set-strategies: be sure that w_obj is setobject
l.diekmann
- [pypy-commit] pypy set-strategies: checking for string makes no sense here
l.diekmann
- [pypy-commit] pypy set-strategies: Cleaned up setobject.py
l.diekmann
- [pypy-commit] pypy set-strategies: create set from iterable to check length and use fastpath
l.diekmann
- [pypy-commit] pypy set-strategies: delegated not_contain_equal_elements method to strategies
l.diekmann
- [pypy-commit] pypy set-strategies: difference always expects w_other to be a set
l.diekmann
- [pypy-commit] pypy set-strategies: differentiation between set types happens in W_SetObject and W_FrozenSetObject (more OO)
l.diekmann
- [pypy-commit] pypy set-strategies: discard is deprecated. instead we use remove
l.diekmann
- [pypy-commit] pypy set-strategies: discard is not needed anymore
l.diekmann
- [pypy-commit] pypy set-strategies: do not use getkeys as this is not very efficient
l.diekmann
- [pypy-commit] pypy set-strategies: EmptySet.add() switches to correct strategy now
l.diekmann
- [pypy-commit] pypy set-strategies: erasing bug in _intersection_wrapped. added test and fix
l.diekmann
- [pypy-commit] pypy set-strategies: FakeInt is needed for this test class but setup_class is overwritten
l.diekmann
- [pypy-commit] pypy set-strategies: First basic implementation of strategies for SetObjects
l.diekmann
- [pypy-commit] pypy set-strategies: fix and test for fakeobject in has_key
l.diekmann
- [pypy-commit] pypy set-strategies: fix and tests for fakeints in instrategy
l.diekmann
- [pypy-commit] pypy set-strategies: fix in EmptySetStrategy.issuperset
l.diekmann
- [pypy-commit] pypy set-strategies: fix needed for translation
l.diekmann
- [pypy-commit] pypy set-strategies: fix needed for translation
l.diekmann
- [pypy-commit] pypy set-strategies: fixed _mixin_
l.diekmann
- [pypy-commit] pypy set-strategies: fixed bug in determination of strategy
l.diekmann
- [pypy-commit] pypy set-strategies: fixed bug in difference method for objectsets and added tests
l.diekmann
- [pypy-commit] pypy set-strategies: fixed bug in issuperset, more tests, some optimization
l.diekmann
- [pypy-commit] pypy set-strategies: fixed copy and paste error. SetStrategy needs one more argument
l.diekmann
- [pypy-commit] pypy set-strategies: fixed creating new set based on another set (needs to be copied)
l.diekmann
- [pypy-commit] pypy set-strategies: fixed EmptySetStrategy.issuperset
l.diekmann
- [pypy-commit] pypy set-strategies: fixed eq__Set_settypedef
l.diekmann
- [pypy-commit] pypy set-strategies: fixed ne__Set_settypedef
l.diekmann
- [pypy-commit] pypy set-strategies: fixed recent popitem changes
l.diekmann
- [pypy-commit] pypy set-strategies: forgot argument for abstract method copy
l.diekmann
- [pypy-commit] pypy set-strategies: forgot self in method _isdisjoint_wrapped
l.diekmann
- [pypy-commit] pypy set-strategies: frozenset does not need to be copied
l.diekmann
- [pypy-commit] pypy set-strategies: give variables some meaningful names
l.diekmann
- [pypy-commit] pypy set-strategies: implemented new iteratorimplementation (similar to dictmultiobject)
l.diekmann
- [pypy-commit] pypy set-strategies: implemented popitem on W_SetObject
l.diekmann
- [pypy-commit] pypy set-strategies: in intersection_multiple start with the smallest to avoid unnecessary comparisons
l.diekmann
- [pypy-commit] pypy set-strategies: just check for unhashable objects here
l.diekmann
- [pypy-commit] pypy set-strategies: keep the same order for similar methods
l.diekmann
- [pypy-commit] pypy set-strategies: maintain invariant that first argument is always self
l.diekmann
- [pypy-commit] pypy set-strategies: make sure the annotator sees this as set or a subclass of set
l.diekmann
- [pypy-commit] pypy set-strategies: make_setdata_from_w_iterable is not needed anymore
l.diekmann
- [pypy-commit] pypy set-strategies: merge
cfbolz
- [pypy-commit] pypy set-strategies: merge default
cfbolz
- [pypy-commit] pypy set-strategies: merge default
cfbolz
- [pypy-commit] pypy set-strategies: merge with default
l.diekmann
- [pypy-commit] pypy set-strategies: merge with default
l.diekmann
- [pypy-commit] pypy set-strategies: merged default into set-strategies
l.diekmann
- [pypy-commit] pypy set-strategies: more test coverage
l.diekmann
- [pypy-commit] pypy set-strategies: move objectstrategy case to ObjectSetStrategy
l.diekmann
- [pypy-commit] pypy set-strategies: need to use r_dict when storing wrapped objects
l.diekmann
- [pypy-commit] pypy set-strategies: need to use StopItertion to check for last element in list
l.diekmann
- [pypy-commit] pypy set-strategies: no need to check since w_other is always a set here
l.diekmann
- [pypy-commit] pypy set-strategies: not needed anymore
l.diekmann
- [pypy-commit] pypy set-strategies: now we dont reference i before assignment anymore
l.diekmann
- [pypy-commit] pypy set-strategies: obviuosly d_obj still could be an int-dict
l.diekmann
- [pypy-commit] pypy set-strategies: one more abstract method fix
l.diekmann
- [pypy-commit] pypy set-strategies: only iterate over keys
l.diekmann
- [pypy-commit] pypy set-strategies: optimized intersection_multiple some more
l.diekmann
- [pypy-commit] pypy set-strategies: other_w can't be resized
l.diekmann
- [pypy-commit] pypy set-strategies: possible fix for translation
l.diekmann
- [pypy-commit] pypy set-strategies: refactored difference of setobjects
l.diekmann
- [pypy-commit] pypy set-strategies: refactored discard/delitem and wrote some more tests
l.diekmann
- [pypy-commit] pypy set-strategies: refactored initialisation of W_SetObject
l.diekmann
- [pypy-commit] pypy set-strategies: refactored intersection for sets
l.diekmann
- [pypy-commit] pypy set-strategies: refactored isdisjoint
l.diekmann
- [pypy-commit] pypy set-strategies: refactored issuperset (no wrapping when strategies are equal)
l.diekmann
- [pypy-commit] pypy set-strategies: refactored symmetric_difference
l.diekmann
- [pypy-commit] pypy set-strategies: refactored symmetric_difference for sets
l.diekmann
- [pypy-commit] pypy set-strategies: refactoring: replaced issubset by issuperset
l.diekmann
- [pypy-commit] pypy set-strategies: referenced i before assignment if others_w is None/empty
l.diekmann
- [pypy-commit] pypy set-strategies: removed old comment
l.diekmann
- [pypy-commit] pypy set-strategies: removed unused methods
l.diekmann
- [pypy-commit] pypy set-strategies: removed/chnaged old comments
l.diekmann
- [pypy-commit] pypy set-strategies: renamed cast_to/from_void_star to (un)erase
l.diekmann
- [pypy-commit] pypy set-strategies: renamed delitem to remove
l.diekmann
- [pypy-commit] pypy set-strategies: renamed not_comparable to more convenient not_contain_equal_elements
l.diekmann
- [pypy-commit] pypy set-strategies: replaced getkeys by using iterator
l.diekmann
- [pypy-commit] pypy set-strategies: replaced getkeys in hash_FrozenSet with iterator
l.diekmann
- [pypy-commit] pypy set-strategies: replaced more make_setdata_from_w_iterbale by _newobj() and set_strategy_from_w_iterable()
l.diekmann
- [pypy-commit] pypy set-strategies: replaced space.iterator with iterator implementation for sets
l.diekmann
- [pypy-commit] pypy set-strategies: replaced w_left.strategy with self where possible
l.diekmann
- [pypy-commit] pypy set-strategies: reuse set_difference_update__Set
l.diekmann
- [pypy-commit] pypy set-strategies: review code. add plenty of XXXs
cfbolz
- [pypy-commit] pypy set-strategies: satisfying the annotator
l.diekmann
- [pypy-commit] pypy set-strategies: skip currently not supported tests
l.diekmann
- [pypy-commit] pypy set-strategies: some more optimization
l.diekmann
- [pypy-commit] pypy set-strategies: space not necessary here?
l.diekmann
- [pypy-commit] pypy set-strategies: switch back to empty strategy on remove and clear
l.diekmann
- [pypy-commit] pypy set-strategies: tell annotator that this obj must be a set
l.diekmann
- [pypy-commit] pypy set-strategies: test and fix for W_SetObject.pop()
l.diekmann
- [pypy-commit] pypy set-strategies: this comment won't be needed anymore
l.diekmann
- [pypy-commit] pypy set-strategies: this is already checked in _difference_base
l.diekmann
- [pypy-commit] pypy set-strategies: this is done with not_contain_equal_elements
l.diekmann
- [pypy-commit] pypy set-strategies: this is the same but hopefully it will satisfy the annotator
l.diekmann
- [pypy-commit] pypy set-strategies: to be consistent create a set and call difference_update here too
l.diekmann
- [pypy-commit] pypy set-strategies: uh? this test is clearly dict order dependent
cfbolz
- [pypy-commit] pypy set-strategies: unnecessary code
l.diekmann
- [pypy-commit] pypy set-strategies: use copy and intersect_update
l.diekmann
- [pypy-commit] pypy set-strategies: use string strategy when appending string to empty set
l.diekmann
- [pypy-commit] pypy set-strategies: use the correct dict (here: r_dict for wrapped items)
l.diekmann
- [pypy-commit] pypy set-strategies: using a for loop is much simpler here
l.diekmann
- [pypy-commit] pypy set-strategies: w_iterable must never be a list
l.diekmann
- [pypy-commit] pypy set-strategies: W_SetObject not takes w_iterable as init value instead of r_dict
l.diekmann
- [pypy-commit] pypy set-strategies: was not rpython
l.diekmann
- [pypy-commit] pypy set-strategies: we do not enforce EmptySetStrategy for empty sets
l.diekmann
- [pypy-commit] pypy set-strategies: what the hell did we do here!?
l.diekmann
- [pypy-commit] pypy set-strategies: when updating empty list simply copy storage and strategy from the other set
l.diekmann
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) app-level slice tests pass
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) app-level [not]test_specialisedtuple pass
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) app-level eq and hash test pass
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) check eq delegates when necessary
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) create specialisedtuple class dynamically
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) fix repr in tool/pytest/appsupport.py in case an exception is raised
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) improve eq test and kill commented code
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) isspecialisedtupleobjectintint passes
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) make sure that tuple in test_eq does not delegate
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) make sure that tuple in test_getitem does not delegate
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) make sure that tuple in test_len does not delegate
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) move instantiaton code into specialisedtupleobject.py
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) not interested in 1-tuples really, kill the code
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) spelling error
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) starting to implement TupleIntInt
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) test length of specialised tuples
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) test_hash_against_normal_tuple passes
mwp
- [pypy-commit] pypy SpecialisedTuples: (antocuni, mwp) use new magic for defining helper method
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) add code for ordering of specialised 2-tuples
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) add test for creating float-float-tuples
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) add tests and code for some specialised 3-tuples + add slice multimethod
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) add tests and code to generate name of each specialised class from its element types
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) add tests for non-delegated neq and ordering
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) also support specialised tuples with 'any' type
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) create Classes for float-float and str-str specialisations
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) equality and order tests now check w_other is same specialisation to avoid mixed type comparisons
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) extend hash test to check floats which happen to be integers
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) fix hash so it deals with flaots that are ints properly
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) make_specialised_class take a tuple, not a list - rename and assert
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) move try_specialisation to be a class method of specialised class, and unroll specialisation loop
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) pass new class name as parameter to creator and tidy locals
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) refactor test for correct hashes and extend create and eq tests
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) reinstate inherited tuple tests, and add mul__SpecialisedTuple_ANY to fix identity test failure
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) replace specific code to create SpecialisedTupleObjects with generic
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) store ANY elements wrapped, and fix bug in hash test
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) use a tuple of types as parameter to make_specialised_class
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp) use unrolling_iterable to generate access to tuple elements
mwp
- [pypy-commit] pypy SpecialisedTuples: (mwp, antocuni) Create a branch with tuples specialised by type
mwp
- [pypy-commit] pypy step-one-xrange: special case xrange without any step specified
hakanardo
- [pypy-commit] pypy step-one-xrange: test ensuring xrange iterator only produces a single setitem
hakanardo
- [pypy-commit] pypy stm: A few extra operations that are always allowed.
arigo
- [pypy-commit] pypy stm: A poor man's lock: just use a regular counter and check it every second.
arigo
- [pypy-commit] pypy stm: Added a file in that branch too to memorize things to do.
arigo
- [pypy-commit] pypy stm: Baaaaah. setjmp() cannot be called on a jmp_buf that belongs to a parent
arigo
- [pypy-commit] pypy stm: Break a line that is definitely too long in the log.
arigo
- [pypy-commit] pypy stm: Extend the test.
arigo
- [pypy-commit] pypy stm: Fix the test.
arigo
- [pypy-commit] pypy stm: fix.
arigo
- [pypy-commit] pypy stm: Fix.
arigo
- [pypy-commit] pypy stm: getarrayitem in funcgen.py. Refactor test_funcgen to actually
arigo
- [pypy-commit] pypy stm: getarrayitem. Disabled for now because it's missing the C impl.
arigo
- [pypy-commit] pypy stm: Hack. It may stay around if no solution to the problem is found,
arigo
- [pypy-commit] pypy stm: Hard-code the STM logic here for now.
arigo
- [pypy-commit] pypy stm: hg merge default
arigo
- [pypy-commit] pypy stm: hg merge default
arigo
- [pypy-commit] pypy stm: hum.
arigo
- [pypy-commit] pypy stm: Improve targetdemo.
arigo
- [pypy-commit] pypy stm: In-progress
arigo
- [pypy-commit] pypy stm: interiorfield operations.
arigo
- [pypy-commit] pypy stm: kill these two C functions.
arigo
- [pypy-commit] pypy stm: malloc_varsize.
arigo
- [pypy-commit] pypy stm: More tests, fixes.
arigo
- [pypy-commit] pypy stm: One more.
arigo
- [pypy-commit] pypy stm: setarrayitem
arigo
- [pypy-commit] pypy stm: setarrayitem.
arigo
- [pypy-commit] pypy stm: Start a test for the complicated logic in funcgen.py,
arigo
- [pypy-commit] pypy stm: Test and fix.
arigo
- [pypy-commit] pypy stm: Tweaks.
arigo
- [pypy-commit] pypy stm: update
arigo
- [pypy-commit] pypy stm: update
arigo
- [pypy-commit] pypy stm: update
arigo
- [pypy-commit] pypy stm: Use default=False, and enable it only in -O2/O3/Ojit, like the
arigo
- [pypy-commit] pypy stm: Yay! targetdemo is fixed and seems to be working.
arigo
- [pypy-commit] pypy stm: Yay, test_ztranslation passes. The issue was direct memcpy on STRs,
arigo
- [pypy-commit] pypy stm: Yay, the first example of RPython program that runs successfully
arigo
- [pypy-commit] pypy type-specialized-instances: (l.diekmann, cfbolz): a branch to play with the idea of type-specializing instances
l.diekmann
- [pypy-commit] pypy type-specialized-instances: added tests and fixes for overwriting attributes with another type and for untaggable ints
l.diekmann
- [pypy-commit] pypy type-specialized-instances: do not use index anymore to read attributes. in future the attributes manage (un)erasing and (un)wrapping of their values themselves
l.diekmann
- [pypy-commit] pypy type-specialized-instances: fallback to PlainAttribute on AttributeError. this fix is needed for modules, since they do not have a typedef
l.diekmann
- [pypy-commit] pypy type-specialized-instances: fix: former PlainAttribute is now AbstractStoredAttribute
l.diekmann
- [pypy-commit] pypy type-specialized-instances: fixed read with new selector (still not sure if this is the right fix)
l.diekmann
- [pypy-commit] pypy type-specialized-instances: fixed some tests where storage items are erased. there is no general erasing pair anymore so we need to use the methods from the correct AbstractAttribute to (un)erase
l.diekmann
- [pypy-commit] pypy type-specialized-instances: introduced IntAttribute
l.diekmann
- [pypy-commit] pypy type-specialized-instances: None should be erased by PlainAttribute
l.diekmann
- [pypy-commit] pypy type-specialized-instances: read attributes only through Attribute class. fixed tests
l.diekmann
- [pypy-commit] pypy type-specialized-instances: removed some old debug prints
l.diekmann
- [pypy-commit] pypy type-specialized-instances: removed some old prints
l.diekmann
- [pypy-commit] pypy type-specialized-instances: reverted last fix, found a new (and better) one
l.diekmann
- [pypy-commit] pypy type-specialized-instances: started fixing tests to work with new selector
l.diekmann
- [pypy-commit] pypy win64 test: buildbot problem partially solved: sys.maxint is now also hacked when called by the buildbot (which uses pytest, not the test_all
ctismer
- [pypy-commit] pypy win64 test: close old branch
ctismer
- [pypy-commit] pypy win64 test: closing badly named old branch. I guess that the changes are still visible from win64_gborg
ctismer
- [pypy-commit] pypy win64 test: merge default
ctismer
- [pypy-commit] pypy win64-stage1: _continuelet on win64 _works_ !
ctismer
- [pypy-commit] pypy win64-stage1: a few more int/long unifications.
ctismer
- [pypy-commit] pypy win64-stage1: a few more int/long unifications.
ctismer
- [pypy-commit] pypy win64-stage1: Added a bit to commondefs.h about the work in progress.
ctismer
- [pypy-commit] pypy win64-stage1: Added a check for is_emulated_long.
ctismer
- [pypy-commit] pypy win64-stage1: Added a line to an outdated doc about boehm on windows
ctismer
- [pypy-commit] pypy win64-stage1: added explicit _WIN32/_WIN64 defns for nmake
ctismer
- [pypy-commit] pypy win64-stage1: added snakepit to predefined libdirs
ctismer
- [pypy-commit] pypy win64-stage1: added the win64.asm source from cpython
ctismer
- [pypy-commit] pypy win64-stage1: added win64.asm
ctismer
- [pypy-commit] pypy win64-stage1: Backed out changeset: 0bb2bc25fcfc
ctismer
- [pypy-commit] pypy win64-stage1: Backed out changeset: b78a605fb452
ctismer
- [pypy-commit] pypy win64-stage1: bad bad
ctismer
- [pypy-commit] pypy win64-stage1: blaming myself about bad ideas. is_valid_int should never raise an exception, basta!
ctismer
- [pypy-commit] pypy win64-stage1: boehm on win64 without __asm, of course :-)
ctismer
- [pypy-commit] pypy win64-stage1: boehm-gc support for win64
ctismer
- [pypy-commit] pypy win64-stage1: buildbot problem partially solved: sys.maxint is now also hacked when called by the buildbot (which uses pytest, not the test_all
ctismer
- [pypy-commit] pypy win64-stage1: changed all checks for (int, long) to is_valid_int.
ctismer
- [pypy-commit] pypy win64-stage1: Check in a test that fails (and works on "default")
arigo
- [pypy-commit] pypy win64-stage1: cleanup of merge conflicts
ctismer
- [pypy-commit] pypy win64-stage1: continuelet support on win64, - something doesn't work, checked in anyway
ctismer
- [pypy-commit] pypy win64-stage1: enabled support won win64 standalone builds
ctismer
- [pypy-commit] pypy win64-stage1: fix arg to is_valid_int in fakearenaaddress.__add__
ctismer
- [pypy-commit] pypy win64-stage1: fix of is_valid_int call
ctismer
- [pypy-commit] pypy win64-stage1: fixed clibffi.py, but win32.c must be replaced, and I don't know yes by what.
ctismer
- [pypy-commit] pypy win64-stage1: fixed int tests to use is_valid_int
ctismer
- [pypy-commit] pypy win64-stage1: fixed mmap.resize for win64
ctismer
- [pypy-commit] pypy win64-stage1: fixed test_intobject.
ctismer
- [pypy-commit] pypy win64-stage1: forgot this patch for sys.maxint, which has bad consequences.
ctismer
- [pypy-commit] pypy win64-stage1: hacked a bit on ll2ctypes/do_allocation_in_far_regions
ctismer
- [pypy-commit] pypy win64-stage1: hacked a bit on ll2ctypes/do_allocation_in_far_regions
ctismer
- [pypy-commit] pypy win64-stage1: hacked a bit on ll2ctypes/do_allocation_in_far_regions
ctismer
- [pypy-commit] pypy win64-stage1: ignoring L formatting, unification of int and long
ctismer
- [pypy-commit] pypy win64-stage1: manual big review of changes
ctismer
- [pypy-commit] pypy win64-stage1: manual merge
ctismer
- [pypy-commit] pypy win64-stage1: merge
ctismer
- [pypy-commit] pypy win64-stage1: merge
ctismer
- [pypy-commit] pypy win64-stage1: merge
ctismer
- [pypy-commit] pypy win64-stage1: merge
ctismer
- [pypy-commit] pypy win64-stage1: merge
ctismer
- [pypy-commit] pypy win64-stage1: Merge
ctismer
- [pypy-commit] pypy win64-stage1: Merge
ctismer
- [pypy-commit] pypy win64-stage1: Merge
ctismer
- [pypy-commit] pypy win64-stage1: Merge
ctismer
- [pypy-commit] pypy win64-stage1: Merge
ctismer
- [pypy-commit] pypy win64-stage1: Merge
ctismer
- [pypy-commit] pypy win64-stage1: merge defauli
ctismer
- [pypy-commit] pypy win64-stage1: Merge Default
ctismer
- [pypy-commit] pypy win64-stage1: merge glitch
ctismer
- [pypy-commit] pypy win64-stage1: merge heads
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: Merge with default
ctismer
- [pypy-commit] pypy win64-stage1: merged with default
ctismer
- [pypy-commit] pypy win64-stage1: mertsch
ctismer
- [pypy-commit] pypy win64-stage1: Mission Accomplished
ctismer
- [pypy-commit] pypy win64-stage1: more unification of int and long
ctismer
- [pypy-commit] pypy win64-stage1: new name for the branch
ctismer
- [pypy-commit] pypy win64-stage1: oupps
ctismer
- [pypy-commit] pypy win64-stage1: relevant_environ updated for windows
ctismer
- [pypy-commit] pypy win64-stage1: removed a check in r_bigint that creates more problems now than it solves
ctismer
- [pypy-commit] pypy win64-stage1: removed a merge glitch
ctismer
- [pypy-commit] pypy win64-stage1: removed win32 blocker 'XXX'
ctismer
- [pypy-commit] pypy win64-stage1: reverted a bad import of pypy into app_operator
ctismer
- [pypy-commit] pypy win64-stage1: reverted merge with default
ctismer
- [pypy-commit] pypy win64-stage1: simple but ugly hack to make 'test_mod_ovf_zer' pass.
ctismer
- [pypy-commit] pypy win64-stage1: small merge glitch with big effect
ctismer
- [pypy-commit] pypy win64-stage1: Some corrections to ll2ctypes. Actually, they did not solve my problem:
ctismer
- [pypy-commit] pypy win64-stage1: Starting stage2 a bit by defining maxint in rarithmetic
ctismer
- [pypy-commit] pypy win64-stage1: starting to replace sys.maxint
ctismer
- [pypy-commit] pypy win64-stage1: temp hack. something makes this cast necessary. Find the reason!
ctismer
- [pypy-commit] pypy win64-stage1: test_zpy works on win64
ctismer
- [pypy-commit] pypy win64-stage1: the first glitch, expected
ctismer
- [pypy-commit] pypy win64-stage1: this patch makes at least the setup part of test_newgc work.
ctismer
- [pypy-commit] pypy win64-stage1: three more typos
ctismer
- [pypy-commit] pypy win64-stage1: typo-typopo
ctismer
- [pypy-commit] pypy win64-stage1: unification of int/long.
ctismer
- [pypy-commit] pypy win64_gborg: a little bit of comments for me ; -)
ctismer
- [pypy-commit] pypy win64_gborg: added file to memorize things to do.
ctismer
- [pypy-commit] pypy win64_gborg: Adjusted 'long' in most c/src files, but tried carefully not to mix up things where CPython is involved.
ctismer
- [pypy-commit] pypy win64_gborg: adjusted the overflow checks
ctismer
- [pypy-commit] pypy win64_gborg: all errors are gone from test_typed.py.
ctismer
- [pypy-commit] pypy win64_gborg: all of test_ll_os works now (more than before is started win64 ; -)
ctismer
- [pypy-commit] pypy win64_gborg: close old branch
ctismer
- [pypy-commit] pypy win64_gborg: coming closer, 49 tests passed, 16 failed in test_typed.py
ctismer
- [pypy-commit] pypy win64_gborg: continuing win64
ctismer
- [pypy-commit] pypy win64_gborg: corrected the formatting of constants.
ctismer
- [pypy-commit] pypy win64_gborg: disabled errno check on win64 and py 2.7.2 - ctypes bug
ctismer
- [pypy-commit] pypy win64_gborg: disabled errno check on win64 and py 2.7.2 - ctypes bug
ctismer
- [pypy-commit] pypy win64_gborg: fixed a lot of GC and JIT/assembler word sizes (argh, testing now ; -)
ctismer
- [pypy-commit] pypy win64_gborg: fixed memorylayout of the GC for win64, format characters
ctismer
- [pypy-commit] pypy win64_gborg: fixed ovfcheck, which needs to skip symbolics. This caused 50 or more gs tests to fail
ctismer
- [pypy-commit] pypy win64_gborg: fixed rbigint
ctismer
- [pypy-commit] pypy win64_gborg: fixed test_compute_hash
ctismer
- [pypy-commit] pypy win64_gborg: fixed test_rerased to show only the single error which it apparently had before ; -)
ctismer
- [pypy-commit] pypy win64_gborg: gc inspector works
ctismer
- [pypy-commit] pypy win64_gborg: got one of four things in test_typed.py to run.
ctismer
- [pypy-commit] pypy win64_gborg: ll_os.times() works now
ctismer
- [pypy-commit] pypy win64_gborg: ll_os.utimes works, too
ctismer
- [pypy-commit] pypy win64_gborg: merge
ctismer
- [pypy-commit] pypy win64_gborg: merge
ctismer
- [pypy-commit] pypy win64_gborg: merge
ctismer
- [pypy-commit] pypy win64_gborg: merge
ctismer
- [pypy-commit] pypy win64_gborg: merge
ctismer
- [pypy-commit] pypy win64_gborg: merge
ctismer
- [pypy-commit] pypy win64_gborg: merge
ctismer
- [pypy-commit] pypy win64_gborg: merge default
ctismer
- [pypy-commit] pypy win64_gborg: merge default
ctismer
- [pypy-commit] pypy win64_gborg: Merge with default
ctismer
- [pypy-commit] pypy win64_gborg: Merge with default
ctismer
- [pypy-commit] pypy win64_gborg: Merge with default
ctismer
- [pypy-commit] pypy win64_gborg: Merge with default
ctismer
- [pypy-commit] pypy win64_gborg: Merge with default
ctismer
- [pypy-commit] pypy win64_gborg: modulo 4 tests (flot/unicode conversion), it all works.
ctismer
- [pypy-commit] pypy win64_gborg: re-enabled rwin32
ctismer
- [pypy-commit] pypy win64_gborg: re-enabled rwin32
ctismer
- [pypy-commit] pypy win64_gborg: removed the last bug from test_typed.py ehich is not related to rwin32.py buggyness
ctismer
- [pypy-commit] pypy win64_gborg: renamed to win64-stage1
ctismer
- [pypy-commit] pypy win64_gborg: reverted a few changes which cannot take external macros
ctismer
- [pypy-commit] pypy win64_gborg: simplified primitive.py by using the types 'Signed' and 'Unsigned' which are defined in g_prerequisites.h
ctismer
- [pypy-commit] pypy win64_gborg: temporarily disabled import of rwin32
ctismer
- [pypy-commit] pypy win64_gborg: test_chdir is fixed now for win32
ctismer
- [pypy-commit] pypy win64_gborg: test_ll_os_stat works now as well
ctismer
- [pypy-commit] pypy win64_gborg: test_ll_os_stat works now, too. Hint: never assume 'c:\temp' exists. Use the environ!
ctismer
- [pypy-commit] pypy win64_gborg: test_ll_time: test_time_sleep works
ctismer
- [pypy-commit] pypy win64_gborg: test_posix: test_isatty works
ctismer
- [pypy-commit] pypy win64_gborg: test_posix: test_open works
ctismer
- [pypy-commit] pypy win64_gborg: This fix looks wrong. The JMP target is still only 4 bytes even
arigo
- [pypy-commit] pypy win64_gborg: typo correction
ctismer
- [pypy-commit] pypy.org extradoc: All *binary* versions
arigo
- [pypy-commit] pypy.org extradoc: bump the release number
fijal
- [pypy-commit] pypy.org extradoc: checksums
fijal
- [pypy-commit] pypy.org extradoc: Link win32.
arigo
- [pypy-commit] pypy.org extradoc: merge
fijal
- [pypy-commit] pypy.org extradoc: regen html
fijal
- [pypy-commit] pypy.org extradoc: update numbers
fijal
- [pypy-commit] pyrepl default: adapt encopyright to hg
RonnyPfannschmidt
- [pypy-commit] pyrepl default: cherry pick over tox.ini and hgignore from py3ksupport branch
RonnyPfannschmidt
- [pypy-commit] pyrepl default: Port 5e062fe507c3 from pypy.
arigo
- [pypy-commit] pyrepl py3ksupport: add unicode alias to unix_console
RonnyPfannschmidt
- [pypy-commit] pyrepl py3ksupport: fix up keymap creation
RonnyPfannschmidt
- [pypy-commit] pyrepl py3ksupport: merge default
RonnyPfannschmidt
- [pypy-commit] pyrepl py3ksupport: merge from default
RonnyPfannschmidt
- [pypy-commit] pyrepl py3ksupport: skip functional tests in case of python3 :(
RonnyPfannschmidt
Last message date:
Wed Nov 30 22:37:19 CET 2011
Archived on: Wed Nov 30 22:37:22 CET 2011
This archive was generated by
Pipermail 0.09 (Mailman edition).