While LLVM 3.3 was released last week, there are already some performance changes for the latest LLVM 3.4 and Clang 3.4 SVN development code for this C/C++ compiler stack.
Complementing the earlier Phoronix article about optimized binaries for Intel Haswell CPUs via the -march=core-av2 Haswell compiler optimizations, in this article is a comparison of the GCC and LLVM/Clang compilers when targeting the new Core i7 4770K CPU. GCC 4.7.3, GCC 4.8.1, LLVM Clang 3.2, and LLVM Clang 3.3 were the tested compilers under Ubuntu Linux when seeing how well these different compilers optimized for Haswell.
While the Intel Haswell CPUs were just launched days ago, there's already quite a Linux story to them. The Haswell CPU is interesting and the performance is good, but there's still extra headroom to make especially when it comes to the graphics driver and performance relative to Intel's Windows driver. Even so, the Intel Haswell Linux support has already evolved a great deal.
In preparation for the upcoming release of LLVM 3.3, here is an extensive round of C/C++ benchmarks from GCC 4.8.0, LLVM Clang 3.2, and LLVM Clang 3.3-rc1 to look at the Linux compiler performance. Benchmarks happened from three different systems bearing Intel Core i7 3960X, AMD FX-8350, and Intel Core i3 3217U processors for a diverse look at the performance.
Building upon our F2FS file-system benchmarks from earlier in this week is a large comparison of four of the leading Linux file-systems at the moment: Btrfs, EXT4, XFS, and F2FS. With the four Linux kernel file-systems, each was benchmarked on the Linux 3.8, 3.9, and 3.10-rc1 kernels. The results from this large file-system comparison when backed by a solid-state drive are now published on Phoronix.
With the merge window on the feature-rich Linux 3.10 kernel having been closed, the usual roundabout of Phoronix benchmarking of the Linux kernel has commenced. In our initial testing of the F2FS file-system on Linux 3.10, however, yields negative performance changes.
One of the prominent features to be introduced with the LLVM 3.3 release this summer is the SLP Vectorizer. Introduced in the LLVM 3.2 release was the LLVM Loop Vectorizer for vectorizing loops while the new SLP Vectorizer is about optimizing straight-line code by merging multiple scalars into vectors.
Most often when delivering new compiler benchmarks on Phoronix whether it be for GCC, LLVM/Clang, or an alternative Linux code compiler, the testing is most commonly done with Intel hardware. The Intel compiler testing is done since Intel CPUs are predominantly used in the developer world and we happen to have a lot more Intel hardware samples around than AMD CPUs. However, for those curious how the LLVM/Clang 3.3 performance is stacking up, here are some GCC and LLVM/Clang benchmarks from an AMD FX-8350 "Vishera" system running Ubuntu 13.04 Linux.
Generally when delivering new Linux compiler benchmarks on Phoronix it's from x86/ARM hardware within the past two years. It's the most recent generations of hardware that excites us the most and generally where the professional Linux software developers are focusing their time and resources. However, after seeing the recent LLVM/Clang 3.3 performance improvements for this forthcoming open-source compiler release, we decided to go back a bit in CPU history.
Last month I delivered benchmarks showing LLVM/Clang 3.3 offers performance improvements and then LLVM/Clang 3.3 is very competitive to GCC 4.8. For further confirming this information, LLVM/Clang 3.3 SVN development benchmarks were carried out from an entirely different system to confirm the earlier findings. LLVM/Clang 3.3 is indeed much faster over its predecessor in a wide variety of Linux benchmarks.
The Liquorix kernel is a modified version of the Linux kernel with out-of-tree patches and a kernel configuration that is highly-optimized for desktop, multimedia, and gaming workloads. It's been one year since last benchmarking the Liquorix kernel against a vanilla Linux kernel, but now we have some benchmarks of the Liquorix 3.8 kernel compared to the latest stable Linux kernel.
Benchmarks for sharing this weekend are looking at the performance of GCC 4.7, GCC 4.8, LLVM/Clang 3.2, and the latest LLVM/Clang 3.3 development code. How does the performance of the newly released GCC 4.8.0 compare to the yet-to-be-released LLVM/Clang 3.3? It's interesting.
Last week, Wayland/Weston was forked by a long-time contributor, Scott Moreau. The fork of the Wayland/Weston display server ended up becoming known as Northfield/Norwood, following disagreements within the Wayland development camp. Scott Moreau was ultimately banned from the Wayland mailing list and IRC channel, so he's written an exclusive, independent article for Phoronix to explain his actions and why he felt a fork of the Wayland display server protocol and the reference Weston compositor were necessary.
Fluendo, the well-known company that backs the development of GStreamer and has also sponsored projects like PiTiVi and other open-source multimedia projects, has released Codec Pack 18. Special about Codec Pack 18 is that it's intended for use with GStreamer 1.0.
In the benchmarking that has happened since the release of the Linux 3.8 kernel, there's been many tests that occurred of Samsung's Flash-Friendly File-System (F2FS). With that testing has also come many requests to compare the performance of this file-system designed for flash storage devices to Microsoft's exFAT file-system as well as NTFS. In this article are those benchmark results.
The latest in our series of ARM Linux benchmarking is looking at the impact of GCC compiler optimizations on the ARM Cortex A15-based Samsung Exynos 5 Dual.
Last week when benchmarking the new F2FS file-system from Samsung that was introduced in the Linux 3.8 kernel its performance was compared to Btrfs, EXT3, EXT4, XFS, JFS, and ReiserFS. For those hoping to see file-system performance results of NILFS2, those results are available today.
Earlier this week I performed some F2FS file-system benchmarks on the Linux 3.8 kernel with an Intel X25 Solid-State Drive (SSD) compared to the EXT4, Btrfs, and other file-systems. Out today are benchmarks of the Flash-Friendly File-System from an SDHC card on Linux.
Being released soon is the Linux 3.8 kernel and one of its many new features is the introduction of the F2FS file-system. The "Flash-Friendly File-System" was developed by Samsung and is showing promise as a new Linux file-system designed around the characteristics of flash-based storage devices. In this article are the first benchmarks of F2FS compared to Btrfs, EXT3, EXT4, XFS, JFS, and ReiserFS file-systems.
With the final release of the Linux 3.8 kernel coming in the very near future, here are file-system benchmarks of EXT4 and Btrfs on the Linux 3.8 development code compared to recent Linux kernel releases.
GCC 4.8 is set to introduce a new optimization level that provides fast compilation performance and decent run-time performance of the resulting binary while still providing a superior debugging experience. Here are benchmarks of this new GCC general optimization level (-Og) compared to the other long-standing compiler optimization levels.
Going on two years ago PathScale open-sourced their EKOPath 4 Compiler Suite. This Fortran/C/C++ compiler suite hasn't seen widespread adoption since then outside of some scientific circles and other select high-performance areas, but PathScale hasn't stalled in advancing their compiler software that is also still available commercially. PathScale has been preparing to release EKOPath 5.0, which is the subject of today's benchmarks.
The first development release of Phoronix Test Suite 4.4-Forsand is now available for those interested in open-source benchmarking and automated testing. This major quarterly update is poised to introduce new features that will benefit all users from hardware enthusiasts to enterprise customers.
Courtesy of Intel's Open-Source Technology Center are 13 new Linux micro-benchmarks that have been created based upon the Phoronix Test Suite and OpenBenchmarking.org. These brand new test profiles provide test coverage of systemd boot performance, timing of various common system tasks, GPU residency times, PowerTop wake-up monitoring, and much more.
While there's already been articles looking at the Nouveau NVIDIA driver and AMD Radeon driver on the forthcoming Linux 3.8 kernel, up today are some early computational benchmarks of this new kernel. From an Intel Core i7 "Ivy Bridge" system, some general-purpose Linux benchmarks were conducted from the Linux 3.5 kernel release through the latest Linux 3.8 Git development kernel.
With last week's release of LLVM 3.2, here are new benchmarks of LLVM 3.2 with the Clang C/C++ compiler front-end. The LLVM/Clang 3.2 performance using last week's source code releases were compared to the earlier LLVM/Clang 3.1 release and then for competition was the GCC 4.7.2 stable release and the latest GCC 4.8.0 development snapshot.
After already sharing the free response survey results, here are the results from the structured part of this year's annual GNOME User Survey.
After yesterday sharing the general feedback submitted by over a thousand GNOME users (Part 1, Part 2) from the 2012 GNOME User Survey about their views on the popular Linux desktop environment, here's all of the responses to another one of the questions. The question came down to what features of GNOME are most important from your personal use and would not like them to go away.
The first batch of user feedback was published this morning from the 2012 GNOME User Survey. Here's now the second and final batch of GNOME user feedback collected through the annual desktop survey. The results from the actual survey questions will be published later this week.
With the 2012 GNOME User Survey now officially over, here's the start of the results. In this posting are the first (of two) batches of feedback that users supplied while filling out the survey. This year there were 4,494 people participating in the annual yet independent GNOME survey. Of the nearly 4.5k respondents, 1,950 of them also provided feedback with this first batch consisting of the first one thousand responses. The results from the survey in full will also be published this week.
492 software articles published on Phoronix.