LLVM Clang 3.4 Compiler Performance Is Doing Good

Written by Michael Larabel in Software on 21 December 2013 at 04:22 PM EST. Page 3 of 4. 18 Comments.

Most of the SciMark2 tests exhibited stable performance between the LLVM Clang 3.3 and 3.4 compiler versions, but the Sparse Matrix Multiply and Dense LU Matrix Factorization tests were much faster on the new liberally licensed compiler code.

While LLVM Clang keeps adding in more compiler features, they still are doing a phenomenal job at fast compile times... In fact, building PHP is now even faster with LLVM Clang 3.4! LLVM Clang 3.3 was already easily beating the GNU Compiler Collection when it came to compile times.


Related Articles