Author skrah
Recipients brian.curtin, loewis, nadeem.vawda, skrah, steve.dower, vstinner
Date 2014-06-10.17:34:56
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1402421697.25.0.360890572514.issue15993@psf.upfronthosting.co.za>
In-reply-to
Content
"I ran a quick test with profile-guided optimization (PGO, pronounced "pogo"), which has supposedly been improved since VC9, and
+saw a very unscientific 20% speed improvement on pybench.py and 10% size reduction in python35.dll. I'm not sure what we used
+to get from VC9, but it certainly seems worth enabling provided it doesn't break anything. (Interestingly, PGO decided that
+only 1% of functions needed to be compiled for speed. Not sure if I can find out which ones those are but if anyone's
+interested I can give it a shot?)"


Steve, could you try if this is still a problem?  The (presumed) Visual Studio bug only surfaced when training the instrumented build with "profiletests.txt" and then doing the PGupdate build.


I'm opening this issue again and leave it at release blocker for 3.5.
History
Date User Action Args
2014-06-10 17:34:57skrahsetrecipients: + skrah, loewis, vstinner, nadeem.vawda, brian.curtin, steve.dower
2014-06-10 17:34:57skrahsetmessageid: <1402421697.25.0.360890572514.issue15993@psf.upfronthosting.co.za>
2014-06-10 17:34:57skrahlinkissue15993 messages
2014-06-10 17:34:56skrahcreate