Memory allocator on OS/X
Jason,
On May 19, 2006, at 6:37 PM, Jason Foster wrote:
A quick scan though the mailing list archives didn't reveal any reference to this article
Somehow you managed to miss it, we had a discussion about this quite recently: http://www.mail-archive.com/r-sig-mac%40stat.math.ethz.ch/msg00770.html Also based on private e-mail exchange with all involved parties the preliminary answer is no. The mentioned speed-up occurred only after patching the involved code, R was actually not even modified. It is still unclear whether a modification of R is safe and whether any speed up is to be expected. I hope to have a quiet minute during the weekend so I can test the various hypotheses... (There are two separate issues to be addressed: system malloc/free and BLAS performance). Cheers, Simon
[...] Would it make sense for the build process that generates R binaries for OS X to use the Lea allocator? Jason Foster
______________________________________________ R-devel at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel