This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author serhiy.storchaka
Recipients Demur Rumed, mark.dickinson, serhiy.storchaka
Date 2016-05-24.19:05:10
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1464116710.35.0.535977915553.issue27097@psf.upfronthosting.co.za>
In-reply-to
Content
What if change the type of next_instr to unsigned short*? This would guarantee that the read is aligned since the start of code data is aligned.

I have small doubt about PREDICT(). I doubt that the compiler is smart enough to infer that *next_instr and *(unsigned short*)next_instr & 255 is the same. It can do unneeded work. I don't know whether this have negative effect.
History
Date User Action Args
2016-05-24 19:05:10serhiy.storchakasetrecipients: + serhiy.storchaka, mark.dickinson, Demur Rumed
2016-05-24 19:05:10serhiy.storchakasetmessageid: <1464116710.35.0.535977915553.issue27097@psf.upfronthosting.co.za>
2016-05-24 19:05:10serhiy.storchakalinkissue27097 messages
2016-05-24 19:05:10serhiy.storchakacreate