[PATCH] D25722: Improved cost model for FDIV and FSQRT

Michael Kuperstein via llvm-commits llvm-commits at lists.llvm.org
Wed Oct 26 00:46:33 PDT 2016


I apologize, didn't realize that.

Never mind IACA then, let's just go with Agner's numbers for now.


On Oct 26, 2016 00:43, "Andrew V. Tischenko" <andrew.v.tischenko at gmail.com>
wrote:

avt77 added a comment.

If I understood correctly I should replace all IACA numbers with Agner's
numbers, right? OK, I'll do it.
JFYI, I'm not working in Intel since July but of course I know a lot of
guys from Intel and I'll try to ask them about IACA future.


https://reviews.llvm.org/D25722
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20161026/717bf9e0/attachment.html>


More information about the llvm-commits mailing list