[pypy-commit] pypy default: There's now an extra variable here, since we're not reading it out of it's object.

Author: Alex Gaynor Branch: Changeset: r45067:d6ae0a63a316 Date: 2011-06-22 17:55 -0700 http://bitbucket.org/pypy/pypy/changeset/d6ae0a63a316/ Log:There's now an extra variable here, since we're not reading it out of it's object. diff --git a/pypy/module/pypyjit/test_pypy_c/test_ins

[pypy-commit] pypy default: Emit strgetiteim from the string optimization in a way that allows it to be seen by other optimizers.

Author: Alex Gaynor Branch: Changeset: r45066:fe17cc28746d Date: 2011-06-22 16:52 -0700 http://bitbucket.org/pypy/pypy/changeset/fe17cc28746d/ Log:Emit strgetiteim from the string optimization in a way that allows it to be seen by other optimizers. diff --git a/pypy/jit/metainterp/o

[pypy-commit] pypy default: unicodegetitem can't return negative numbers.

Author: Alex Gaynor Branch: Changeset: r45065:cca2ff5601ef Date: 2011-06-22 16:29 -0700 http://bitbucket.org/pypy/pypy/changeset/cca2ff5601ef/ Log:unicodegetitem can't return negative numbers. diff --git a/pypy/jit/metainterp/optimizeopt/intbounds.py b/pypy/jit/metainterp/optimizeopt/intbo

[pypy-commit] pypy default: Merged app level builder.

Author: Alex Gaynor Branch: Changeset: r45064:5b62f71347c8 Date: 2011-06-22 13:28 -0700 http://bitbucket.org/pypy/pypy/changeset/5b62f71347c8/ Log:Merged app level builder. diff --git a/pypy/jit/codewriter/policy.py b/pypy/jit/codewriter/policy.py --- a/pypy/jit/codewriter/policy.py +++ b/p

[pypy-commit] pypy applevel-builder: Close branch in prep for merging.

Author: Alex Gaynor Branch: applevel-builder Changeset: r45063:bebe52b55c79 Date: 2011-06-22 13:28 -0700 http://bitbucket.org/pypy/pypy/changeset/bebe52b55c79/ Log:Close branch in prep for merging. ___ pypy-commit mailing list [email protected]

[pypy-commit] pypy applevel-builder: Merged default.

Author: Alex Gaynor Branch: applevel-builder Changeset: r45062:fa5c17a11ab9 Date: 2011-06-22 13:27 -0700 http://bitbucket.org/pypy/pypy/changeset/fa5c17a11ab9/ Log:Merged default. diff --git a/pypy/jit/metainterp/history.py b/pypy/jit/metainterp/history.py --- a/pypy/jit/metainterp/history.p

[pypy-commit] pypy applevel-builder: Allow inlining into the builders

Author: Alex Gaynor Branch: applevel-builder Changeset: r45061:8e72dc12e190 Date: 2011-06-22 12:27 -0700 http://bitbucket.org/pypy/pypy/changeset/8e72dc12e190/ Log:Allow inlining into the builders diff --git a/pypy/jit/codewriter/policy.py b/pypy/jit/codewriter/policy.py --- a/pypy/jit/codew

[pypy-commit] Notification: Re: Pull request

You have received a notification from Philip Jenvey. This fix was already pulled, thanks -- Change your notification settings at https://bitbucket.org/account/notifications/ ___ pypy-commit mailing list [email protected] http://mail.python.org/mai

[pypy-commit] Notification: pypy

You have received a notification from landtuna. Hi, I forked pypy. My fork is at https://bitbucket.org/landtuna/pypy. -- Change your notification settings at https://bitbucket.org/account/notifications/ ___ pypy-commit mailing list [email protected]

[pypy-commit] pypy inline-dict-ops: better I think

Author: Maciej Fijalkowski Branch: inline-dict-ops Changeset: r45060:ddb0c8eda26c Date: 2011-06-22 19:59 +0200 http://bitbucket.org/pypy/pypy/changeset/ddb0c8eda26c/ Log:better I think diff --git a/pypy/jit/codewriter/jtransform.py b/pypy/jit/codewriter/jtransform.py --- a/pypy/jit/codewrit

[pypy-commit] pypy inline-dict-ops: make the test pass, unsure if it"s complete

Author: Maciej Fijalkowski Branch: inline-dict-ops Changeset: r45059:6fbbe0d349ae Date: 2011-06-22 19:56 +0200 http://bitbucket.org/pypy/pypy/changeset/6fbbe0d349ae/ Log:make the test pass, unsure if it"s complete diff --git a/pypy/jit/codewriter/jtransform.py b/pypy/jit/codewriter/jtransfo

[pypy-commit] extradoc extradoc: print the address of the function

Author: Antonio Cuni Branch: extradoc Changeset: r3773:323338533f37 Date: 2011-06-22 18:34 +0200 http://bitbucket.org/pypy/extradoc/changeset/323338533f37/ Log:print the address of the function diff --git a/talk/ep2011/talk/ctypesbench.py b/talk/ep2011/talk/ctypesbench.py --- a/talk/ep2011/t

[pypy-commit] extradoc extradoc: more tweaks

Author: Antonio Cuni Branch: extradoc Changeset: r3772:5a1477ddb233 Date: 2011-06-22 18:07 +0200 http://bitbucket.org/pypy/extradoc/changeset/5a1477ddb233/ Log:more tweaks diff --git a/talk/ep2011/talk/talk.rst b/talk/ep2011/talk/talk.rst --- a/talk/ep2011/talk/talk.rst +++ b/talk/ep2011/tal

[pypy-commit] extradoc extradoc: more slides

Author: Antonio Cuni Branch: extradoc Changeset: r3771:40aef6687ef6 Date: 2011-06-22 17:57 +0200 http://bitbucket.org/pypy/extradoc/changeset/40aef6687ef6/ Log:more slides diff --git a/talk/ep2011/talk/talk.rst b/talk/ep2011/talk/talk.rst --- a/talk/ep2011/talk/talk.rst +++ b/talk/ep2011/tal

[pypy-commit] pypy default: remove some more unnecesary imports

Author: Maciej Fijalkowski Branch: Changeset: r45058:c9072cc74109 Date: 2011-06-22 16:21 +0200 http://bitbucket.org/pypy/pypy/changeset/c9072cc74109/ Log:remove some more unnecesary imports diff --git a/pypy/jit/metainterp/history.py b/pypy/jit/metainterp/history.py --- a/pypy/jit/metainter

Re: [pypy-commit] pypy default: When a virtual is forced, and then subsequenly an immutable field is read out of it, the value is known if it was seen in a setfield, because it can't be set again by a

On 06/22/2011 03:50 PM, Alex Gaynor wrote: Given that a virtual is forced only right before it escapes, and the external call would flush the heap cache, what's the value there? Not all external values flush the heap cache. A lot of external calls just read things, don't write. Also, being u

Re: [pypy-commit] pypy default: When a virtual is forced, and then subsequenly an immutable field is read out of it, the value is known if it was seen in a setfield, because it can't be set again by a

Given that a virtual is forced only right before it escapes, and the external call would flush the heap cache, what's the value there? Alex On Wed, Jun 22, 2011 at 2:34 AM, Carl Friedrich Bolz wrote: > Hi Alex, > > This is part of a more general problem: If a virtual is forced the heap > cache

[pypy-commit] pypy dict-strategies: fix hint that was broken in the strategy refactoring

Author: Carl Friedrich Bolz Branch: dict-strategies Changeset: r45057:21484a3e3772 Date: 2011-06-22 15:46 +0200 http://bitbucket.org/pypy/pypy/changeset/21484a3e3772/ Log:fix hint that was broken in the strategy refactoring diff --git a/pypy/objspace/std/celldict.py b/pypy/objspace/std/celld

[pypy-commit] pypy default: merge

Author: Maciej Fijalkowski Branch: Changeset: r45055:58447690c064 Date: 2011-06-22 15:33 +0200 http://bitbucket.org/pypy/pypy/changeset/58447690c064/ Log:merge diff --git a/pypy/module/pypyjit/test_pypy_c/test_instance.py b/pypy/module/pypyjit/test_pypy_c/test_instance.py --- a/pypy/module

[pypy-commit] pypy default: fun with pyflakes

Author: Maciej Fijalkowski Branch: Changeset: r45054:441d0965244a Date: 2011-06-22 15:24 +0200 http://bitbucket.org/pypy/pypy/changeset/441d0965244a/ Log:fun with pyflakes diff --git a/pypy/jit/metainterp/pyjitpl.py b/pypy/jit/metainterp/pyjitpl.py --- a/pypy/jit/metainterp/pyjitpl.py +++ b

[pypy-commit] pypy inline-dict-ops: start working on inlining of simple (non-looping) dict ops into assembler

Author: Maciej Fijalkowski Branch: inline-dict-ops Changeset: r45053:04c0bfcc0891 Date: 2011-06-22 15:06 +0200 http://bitbucket.org/pypy/pypy/changeset/04c0bfcc0891/ Log:start working on inlining of simple (non-looping) dict ops into assembler diff --git a/pypy/jit/codewriter/jtransf

[pypy-commit] Notification: jitviewer

You have received a notification from nikuda. Hi, I forked jitviewer. My fork is at https://bitbucket.org/nikuda/jitviewer. -- Change your notification settings at https://bitbucket.org/account/notifications/ ___ pypy-commit mailing list pypy-commit@py

Re: [pypy-commit] pypy default: When a virtual is forced, and then subsequenly an immutable field is read out of it, the value is known if it was seen in a setfield, because it can't be set again by a

Hi Alex, This is part of a more general problem: If a virtual is forced the heap cache is not informed of the values that are written into the newly allocated object. This is useful also for fields that are not immutable. Do you maybe feel like generalizing this? Cheers, Carl Friedrich On 06/2