Message 291994 - Python tracker

Message291994

Author	mark.dickinson
Recipients	koobs, mark.dickinson, vstinner
Date	2017-04-20.18:52:16
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1492714336.27.0.224705825002.issue30104@psf.upfronthosting.co.za>
In-reply-to

Content
Hmm. Just looked at dto5.c; nice work reducing it to something so simple. This is looking a bit like either a strict-aliasing related problem, or a compiler bug, or both. My understanding of the "union trick" that dtoa.c uses is that it's well-defined under C99+TC3 (though that wasn't clear under the original C99), so it shouldn't be legitimate for an optimisation to cause different results to be produced. So this is smelling like a compiler bug to me.

Content

Hmm. Just looked at dto5.c; nice work reducing it to something so simple.

This is looking a bit like either a strict-aliasing related problem, or a compiler bug, or both. My understanding of the "union trick" that dtoa.c uses is that it's well-defined under C99+TC3 (though that wasn't clear under the original C99), so it shouldn't be legitimate for an optimisation to cause different results to be produced.

So this is smelling like a compiler bug to me.

History
Date	User	Action	Args
2017-04-20 18:52:16	mark.dickinson	set	recipients: + mark.dickinson, vstinner, koobs
2017-04-20 18:52:16	mark.dickinson	set	messageid: <1492714336.27.0.224705825002.issue30104@psf.upfronthosting.co.za>
2017-04-20 18:52:16	mark.dickinson	link	issue30104 messages
2017-04-20 18:52:16	mark.dickinson	create