[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Jeffrey Yasskin added the comment: Merged to 3.1 in r67611, 3.0.x in r67721, and 2.6.x in r67722. -- keywords: -needs review status: open -> closed versions: +Python 2.6 ___ Python tracker

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Jeffrey Yasskin <[EMAIL PROTECTED]> added the comment: Thanks for the review, Raymond. Committed to the trunk in r67494. I'll svnmerge it over to 3.0.1 once Barry unfreezes the branches. Martin, let me know if you analyze this any more and find anything interesting. -- stage: commit rev

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Raymond Hettinger <[EMAIL PROTECTED]> added the comment: Jeffrey, I've run my own timings and verified the improvement. Please go ahead and apply the trunk (2.7 head). It's too late for 3.0 but once that is out the door, Barry has approved adding the patch for 3.0.1. -- assignee: rhe

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Changes by Raymond Hettinger <[EMAIL PROTECTED]>: -- assignee: -> rhettinger versions: +Python 3.0 ___ Python tracker <[EMAIL PROTECTED]> ___ _

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Raymond Hettinger <[EMAIL PROTECTED]> added the comment: +1 on the patch. It makes good sense to me. -- nosy: +rhettinger ___ Python tracker <[EMAIL PROTECTED]> ___ __

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Martin v. Löwis <[EMAIL PROTECTED]> added the comment: Before it is accepted, I'd like to see some independent analysis (which I might have to carry out myself). It would also be good if some deeper understanding of the effect could be provided - I'm fairly skeptical that such a minor change shou

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Benjamin Peterson <[EMAIL PROTECTED]> added the comment: I think it should stay out of the trunk until 3.0 is released only because that simplifies merging and 3.0 is only 3 days away. -- nosy: +benjamin.peterson ___ Python tracker <[EMAIL PROTECTED]>

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Jeffrey Yasskin <[EMAIL PROTECTED]> added the comment: Martin, does your comment mean you think this should stay out of the trunk(s) until 3.0 is released? That's fine. How about after that: should it go in the 2.6.x and 3.0.x branches too? ___ Python tracker

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Changes by Skip Montanaro <[EMAIL PROTECTED]>: -- nosy: +skip.montanaro ___ Python tracker <[EMAIL PROTECTED]> ___ ___ Python-bugs-list

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Martin v. Löwis <[EMAIL PROTECTED]> added the comment: I don't think anything performance-related should be checked in before 3.0. -- nosy: +loewis ___ Python tracker <[EMAIL PROTECTED]> _

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

Changes by Brett Cannon <[EMAIL PROTECTED]>: -- nosy: +brett.cannon ___ Python tracker <[EMAIL PROTECTED]> ___ ___ Python-bugs-list mail

[issue4477] Speed up PyEval_EvalFrameEx when tracing is off.

New submission from Jeffrey Yasskin <[EMAIL PROTECTED]>: Tracing support shows up fairly heavily an a Python profile, even though it's nearly always turned off. The attached patch against the trunk speeds up PyBench by 2% for me. All tests pass. I have 2 questions: 1) Can other people corroborat