Message 256275 - Python tracker

➜

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	arigo
Recipients	arigo, mark.dickinson, rhettinger, serhiy.storchaka, skrah, tim.peters, vstinner
Date	2015-12-12.09:53:20
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<[email protected]>
In-reply-to

Content
Fwiw, I made a trivial benchmark in C that loads aligned and misaligned shorts ( http://paste.pound-python.org/show/HwnbCI3Pqsj8bx25Yfwp/ ). It shows that the memcpy() version takes only 65% of the time taken by the two-bytes-loaded version on a 2010 laptop. It takes 75% of the time on a modern server. On a recent little-endian PowerPC machine, 96%. On aarch64, only 45% faster (i.e. more than twice faster). This is all with gcc. It seems that using memcpy() is definitely a win nowadays.

Fwiw, I made a trivial benchmark in C that loads aligned and misaligned shorts ( http://paste.pound-python.org/show/HwnbCI3Pqsj8bx25Yfwp/ ).  It shows that the memcpy() version takes only 65% of the time taken by the two-bytes-loaded version on a 2010 laptop.  It takes 75% of the time on a modern server.  On a recent little-endian PowerPC machine, 96%.  On aarch64, only 45% faster (i.e. more than twice faster).  This is all with gcc.  It seems that using memcpy() is definitely a win nowadays.

History
Date	User	Action	Args
2015-12-12 09:53:21	arigo	set	recipients: + arigo, tim.peters, rhettinger, mark.dickinson, vstinner, skrah, serhiy.storchaka
2015-12-12 09:53:21	arigo	set	messageid: <[email protected]>
2015-12-12 09:53:21	arigo	link	issue25823 messages
2015-12-12 09:53:20	arigo	create