[pypy-commit] pypy default: Add yet another layer of caching. This is useful when we have to merge

Author: Maciej Fijalkowski Branch: Changeset: r59437:8117b7a66486 Date: 2012-12-15 10:48 +0200 http://bitbucket.org/pypy/pypy/changeset/8117b7a66486/ Log:Add yet another layer of caching. This is useful when we have to merge a few direct calls way too often diff --git a/pypy/transla

[pypy-commit] pypy default: Fix few strange ufuncs on ints

Author: Maciej Fijalkowski Branch: Changeset: r59438:4073ceef4ea7 Date: 2012-12-15 11:17 +0200 http://bitbucket.org/pypy/pypy/changeset/4073ceef4ea7/ Log:Fix few strange ufuncs on ints diff --git a/pypy/module/micronumpy/test/test_ufuncs.py b/pypy/module/micronumpy/test/test_ufuncs.py ---

[pypy-commit] pypy default: Don't cache bogus results in case one of the analyze_direct_call()

Author: Armin Rigo Branch: Changeset: r59439:272e09815758 Date: 2012-12-15 11:57 +0100 http://bitbucket.org/pypy/pypy/changeset/272e09815758/ Log:Don't cache bogus results in case one of the analyze_direct_call() returned a non-cached result (because of 'seen'). diff --git a/pypy/tr

[pypy-commit] pypy default: Backout 6df0aaf4caf2: it seems that the interpreter part is indeed a

Author: Armin Rigo Branch: Changeset: r59440:af2f3e9740fd Date: 2012-12-15 12:21 +0100 http://bitbucket.org/pypy/pypy/changeset/af2f3e9740fd/ Log:Backout 6df0aaf4caf2: it seems that the interpreter part is indeed a tiny bit faster, but the rest of the runtime code (which is also

[pypy-commit] pypy default: improve fcntl.test_flock/test_lockf so they actually test locking

Author: Brian Kearns Branch: Changeset: r59441:aedad230f8b8 Date: 2012-12-13 22:49 -0500 http://bitbucket.org/pypy/pypy/changeset/aedad230f8b8/ Log:improve fcntl.test_flock/test_lockf so they actually test locking diff --git a/pypy/module/fcntl/test/test_fcntl.py b/pypy/module/fcntl/test/t

[pypy-commit] pypy default: cleanup fcntl.flock and fix fcntl.lockf ignoring return value

Author: Brian Kearns Branch: Changeset: r59442:c1024ee7bb9c Date: 2012-12-13 22:50 -0500 http://bitbucket.org/pypy/pypy/changeset/c1024ee7bb9c/ Log:cleanup fcntl.flock and fix fcntl.lockf ignoring return value diff --git a/pypy/module/fcntl/interp_fcntl.py b/pypy/module/fcntl/interp_fcntl.

[pypy-commit] pypy default: attempted fix for fcntl.test_ioctl failure

Author: Brian Kearns Branch: Changeset: r59445:2f2e27f20fed Date: 2012-12-14 12:50 -0500 http://bitbucket.org/pypy/pypy/changeset/2f2e27f20fed/ Log:attempted fix for fcntl.test_ioctl failure diff --git a/pypy/module/fcntl/test/test_fcntl.py b/pypy/module/fcntl/test/test_fcntl.py --- a/pypy

[pypy-commit] pypy default: make sure fcntl.ioctl tests aren't unnecessarily skipped

Author: Brian Kearns Branch: Changeset: r59443:fc6077a8df30 Date: 2012-12-14 04:22 -0500 http://bitbucket.org/pypy/pypy/changeset/fc6077a8df30/ Log:make sure fcntl.ioctl tests aren't unnecessarily skipped diff --git a/pypy/module/fcntl/test/test_fcntl.py b/pypy/module/fcntl/test/test_fcntl

[pypy-commit] pypy default: make fcntl.test_ioctl failure reproducible

Author: Brian Kearns Branch: Changeset: r59444:97d086354044 Date: 2012-12-14 05:57 -0500 http://bitbucket.org/pypy/pypy/changeset/97d086354044/ Log:make fcntl.test_ioctl failure reproducible diff --git a/pypy/module/fcntl/test/test_fcntl.py b/pypy/module/fcntl/test/test_fcntl.py --- a/pypy

[pypy-commit] pypy default: Merged in bdkearns/pypy (pull request #96: fix fcntl.lockf, improve fcntl tests, fix test_ioctl failure)

Author: arigo Branch: Changeset: r59446:92a2d4bc9ea7 Date: 2012-12-15 13:50 +0100 http://bitbucket.org/pypy/pypy/changeset/92a2d4bc9ea7/ Log:Merged in bdkearns/pypy (pull request #96: fix fcntl.lockf, improve fcntl tests, fix test_ioctl failure) diff --git a/pypy/module/fcntl/interp

[pypy-commit] pypy default: Lets try these cflags with the official benchmark. Great results locally (14% better pidigits etc).

Author: Stian Andreassen Branch: Changeset: r59447:3b95bf4a907e Date: 2012-12-15 16:35 +0100 http://bitbucket.org/pypy/pypy/changeset/3b95bf4a907e/ Log:Lets try these cflags with the official benchmark. Great results locally (14% better pidigits etc). diff --git a/pypy/translator/pla

[pypy-commit] pypy default: Backout 3b95bf4a907e. No significans

Author: Stian Andreassen Branch: Changeset: r59448:78309907a0e7 Date: 2012-12-15 19:56 +0100 http://bitbucket.org/pypy/pypy/changeset/78309907a0e7/ Log:Backout 3b95bf4a907e. No significans diff --git a/pypy/translator/platform/linux.py b/pypy/translator/platform/linux.py --- a/pypy/translat

[pypy-commit] pypy default: fix test for untranslated

Author: mattip Branch: Changeset: r59449:c5b6c76270fd Date: 2012-12-13 17:37 -0800 http://bitbucket.org/pypy/pypy/changeset/c5b6c76270fd/ Log:fix test for untranslated diff --git a/pypy/module/select/test/test_select.py b/pypy/module/select/test/test_select.py --- a/pypy/module/select/test

[pypy-commit] pypy default: Add another test for the graph analyzer's handling of recursion.

Author: Armin Rigo Branch: Changeset: r59450:be3ab719fba2 Date: 2012-12-15 23:20 +0100 http://bitbucket.org/pypy/pypy/changeset/be3ab719fba2/ Log:Add another test for the graph analyzer's handling of recursion. diff --git a/pypy/translator/backendopt/test/test_canraise.py b/pypy/translator

[pypy-commit] pypy default: Trying to improve the performance of the GraphAnalyzer in case of recursion: clean up the various caches and consolidate them in a single DependencyTracker class.

Author: Armin Rigo Branch: Changeset: r59451:9f75b268adc1 Date: 2012-12-16 00:45 +0100 http://bitbucket.org/pypy/pypy/changeset/9f75b268adc1/ Log:Trying to improve the performance of the GraphAnalyzer in case of recursion: clean up the various caches and consolidate them in a