1. Computers
  2. Display Drivers
  3. Graphics Cards
  4. Memory
  5. Motherboards
  6. Processors
  7. Software
  8. Storage
  9. Operating Systems


Facebook RSS Twitter Twitter Google Plus


Phoronix Test Suite

OpenBenchmarking.org

LLVM 3.3 Improves Its Loop Vectorizer

Compiler

Published on 15 February 2013 01:07 AM EST
Written by Michael Larabel in Compiler
2 Comments

The release of LLVM 3.3 is still months away, but one of its features already are notable improvements to its loop vectorizer.

Introduced in LLVM 3.2 was a loop vectorizer. From the earlier Phoronix article on the subject, "The LLVM Loop Vectorizer is the second vectorizer for the compiler infrastructure after its Basic Block Vectorizer for dealing with straight-line code. The Loop Vectorizer is meant to widen instructions in an original loop for operating on multiple consecutive loop iterations. As mentioned in the earlier Phoronix articles on the subject, this feature isn't enabled by default with the forthcoming LLVM 3.2 but can be flipped on via the -mllvm -vectorize-loops compiler switches." The functionality has also been described by LLVM developers.

I did run benchmarks of the LLVM Loop Vectorizer back in December and it could enhance the performance for some workloads. There's previously been talk of enabling the vectorizer by default in LLVM 3.3, but that hasn't happened yet.

What has happened in the code-base of LLVM 3.3 are a number of improvements to this loop vectorizer. With the latest development code, the LLVM 3.3 Loop Vectorizer now supports loops with an unknown trip count, run-time checks of pointers, reductions/inductions, if conversions, pointer induction variables, reverse iterators, vectorization of mixed types, vectorization of function calls, and partial unrolling during vectorization.

These LLVM 3.3 improvements should allow the loop vectorizer to be used more widely and applicable to more code-bases. Benchmarks of the LLVM Loop Vectorizer improvements are forthcoming as we begin to look at the LLVM/Clang 3.3 performance changes at Phoronix.

About The Author
Michael Larabel is the principal author of Phoronix.com and founded the web-site in 2004 with a focus on enriching the Linux hardware experience and being the largest web-site devoted to Linux hardware reviews, particularly for products relevant to Linux gamers and enthusiasts but also commonly reviewing servers/workstations and embedded Linux devices. Michael has written more than 10,000 articles covering the state of Linux hardware support, Linux performance, graphics hardware drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated testing software. He can be followed via and or contacted via .
Latest Linux Hardware Reviews
  1. The Power Consumption & Efficiency Of Open-Source GPU Drivers
  2. AMD R600g/RadeonSI Performance On Linux 3.16 With Mesa 10.3-devel
  3. Intel Pentium G3258 On Linux
  4. SilverStone Precision PS10
Latest Linux Articles
  1. Nouveau vs. Radeon vs. Intel Tests On Linux 3.16, Mesa 10.3-devel
  2. KVM Benchmarks On Ubuntu 14.10
  3. X.Org Server 1.16 Officially Released With Terrific Features
  4. Ubuntu With Linux 3.16 Smashes OS X 10.9.4 On The MacBook Air
Latest Linux News
  1. Linus Torvalds On GCC 4.9: Pure & Utter Crap
  2. Cauldron 2014: GCC & LLVM Will Look To Collaborate More
  3. GCC Receives ACM Programming Languages Software Award
  4. KDE 4.14 Beta 3 Released
  5. A New Video Has Us Real Excited About The New UT For Linux
  6. CoreOS Experiences Its First Stable Release
  7. GNOME 3.13.4 Continues Working On Wayland & More
  8. Wine 1.7.23 Has Initial Support For 64-bit Android Builds
  9. FreeBSD Developing A New Automounter, Other Features
  10. Guix 0.7 Can Now Install The GNU Operating System
Latest Forum Discussions
  1. Updated and Optimized Ubuntu Free Graphics Drivers
  2. Debian + radeonsi
  3. How To Setup Radeon DPM On Ubuntu Linux
  4. Open-source drivers on ATI R7 260X
  5. New build, first Linux PC, what could go wrong? ;)
  6. AMD "Hawaii" Open-Source GPU Acceleration Still Not Working Right
  7. Radeon related kernel bug??
  8. how the US intellegentia operates: