With Phoronix Test Suite 3.4-Lillesand and the new MATISK support, over one thousand revisions of the Mesa graphics library were benchmarked. Two small performance optimizations were also noted.
While S3 Texture Compression (S3TC) is widely used by many games and applications since its inclusion into OpenGL 1.3 and Microsoft DirectX 6.0, these lossy texture compression algorithms have not been implemented in the open-source Linux graphics drivers. This lack of open-source support is due to S3 Graphics holding the patent rights to this technology that they actively license to major hardware vendors. There long has been an external library that can be loaded and will work with most Mesa / Gallium3D drivers for advertising S3TC support, but it's not found by default and it's not included in leading Linux distributions due to these legal fears. There may now be a new solution for the S3TC Linux problem thanks to the advent of a new (and simpler) texture compression algorithm that can serve as a drop-in replacement.
Last month when publishing Fedora 15 vs. Ubuntu 11.04 benchmarks in some of the disk workloads the Fedora Linux release was behind that of Ubuntu Natty Narwhal. Some users speculated in our forums that SELinux was to blame, but later tests show SELinux does not cause a huge performance impact. With Security Enhanced Linux not to blame, some wondered if Fedora's use of LVM, the Logical Volume Manager, by default was the cause.
"Mobile users are urged to seriously consider these results, and possibly even avoid the Natty Narwhal...I hate to say it, especially in an Ubuntu review, but the mobile edge goes to Windows for now...There are also compelling reasons for folks to avoid [Ubuntu 11.04] at all costs. Linux gamers should see substantial improvements, while mobile users suffer a dramatic loss in battery life," were among the critical comments that Tom's Hardware had in their Ubuntu 11.04 review as they were referencing the power regressions I discovered nearly two months ago within the mainline Linux kernel. As I mentioned on Sunday, the Phoronix Test Suite stack and I have now nailed this major power regression in the Linux 2.6.38 kernel that is affecting a significant number of mobile Linux users. Here is what is happening and a way that you should be able to workaround the serious regression should it affect your computer system(s).
With the Linux 3.0 kernel carrying CleanCache support along with various improvements to the EXT4 and Btrfs file-system modules, it is time for another Phoronix file-system comparison. This time around the EXT4 vs. Btrfs performance is particularly important with Fedora 16 possibly switching to Btrfs by default. Due to this level of interest, for our Linux 3.0 kernel benchmarks of the EXT4 and Btrfs file-systems, an Intel SSD was tested as well as an old 5400RPM IDE notebook hard drive to represent two ends of the spectrum.
As mentioned last week, a plethora of Linux power tests are on the way now that we have found an AC power meter with USB interface that works under Linux and we've been able to integrate nicely into the Phoronix Test Suite and its sensor monitoring framework. In this article is one of the first tests that have been completed using this power-measuring device as we monitored the Linux kernel power consumption for an old Intel Pentium 4 and ATI Radeon 9200 system for the past several kernel releases. Even this very old desktop system looks to be affected by the kernel power problems.
Phoronix Media has announced the immediate release of Phoronix Test Suite 3.2 (codenamed "Grimstad") as the planned quarterly update to their open-source Phoronix Test Suite software. The Phoronix Test Suite provides a framework for conducting qualitative and quantitative tests in a manner that is reproducible, easy-to-use, and fully automated.
Within the free software world, GCC has long been the dominant compiler with it being backed by the Free Software Foundation, it being the most well developed free compiler suite, and is a feature rich offering that is put out under the GNU GPLv3. As of late, LLVM has also been hitting the nail on the head. The Low-Level Virtual Machine with its C/C++ Clang compiler front-end offers great performance, is successful in building code-bases like the Linux kernel, its modular design allows the compiler infrastructure to be used in areas like graphics drivers, is under a BSD-style license, and carries numerous other advantages. Other open-source compilers have advanced too, including the release of PCC 1.0. Now there is a new and extremely interesting option to shake the open-source compiler world: PathScale is freely releasing the source to the EKOPath 4 Compiler Suite. EKOPath 4 is a high-performance compiler that up until now has been proprietary and costs nearly $2000 USD per license, but now it's open-source and can sharply outperform GCC in many computationally-intense workloads.
To much dismay, the major open-source announcement we have been waiting for, did not happen this week. Yes, this is the major open-source announcement that we have codenamed Dirndl. It is really that deserving of such a fitting codename. As our early tests have shown, it can dramatically speed-up the system's performance in computationally intensive workloads. No other open-source solution comes close in many of these tests, albeit there are some other proprietary brethren. In this article are some more details and performance results for what has been called "Dirndl" in technology terms.
As discovered by a Phoronix reader, there is a patch in the Linux 184.108.40.206 kernel that can partially improve the system's power performance. The patch by a Nokia engineer is entitled "cpuidle: menu: fixed wrapping timers at 4.294 seconds" and initial reports have said that it will lower the power consumption compared to the stock 2.6.39 kernel.
Earlier this week there were Fedora 15 vs. Ubuntu 11.04 benchmarks looking at the overall system performance as well as the power consumption. Both of these Linux distributions had performed close to one another, as is expected considering the similarities in their kernel and other packages, but there were some discrepancies in the disk tests. Speculations in the forums were that some of the performance differences might be attributed to SELinux, so here are some tests seeing the performance impact of SELinux on Fedora 15.
On Sunday, the 5th of June, Phoronix.com will be turning seven years old. Here is to an early happy birthday for Phoronix, a look back, and what is coming up.
Those that follow my Twitter feed know that over the weekend I began running some benchmarks of the various open-source and closed-source graphics drivers. But it was not like the usual Phoronix benchmarks simply comparing the driver performance. Instead it was to see how each driver performed under the various desktops / window managers now being used by modern Linux installations. In this article are the first results of this testing of Unity with Compiz, the classic GNOME desktop with Metacity, the classic GNOME desktop with Compiz, the GNOME Shell with Mutter, and the KDE desktop with KWin. These configurations were tested with both the open and closed-source NVIDIA and ATI/AMD Linux drivers.
Last weekend I mentioned that Wayland for MeeGo Tablet UX would be discussed and showed off at the 2011 MeeGo Conference that took place this week in San Francisco. Two sessions on the topic of Wayland took place, including one by Kristian Høgsberg, Wayland's creator. In this article are the slides that Kristian presented along with a few notes. This also shows off the plans to adopt Wayland in as soon as MeeGo 1.3, which will be released this October.
While we have already delivered a number of benchmarks from the Linux 2.6.39 kernel, surprisingly we have not yet published any new file-system benchmarks from this latest stable Linux kernel release. Fortunately, that has changed today with a fresh round of Btrfs, EXT4, and XFS file-system benchmarks on the Linux 2.6.39 kernel and compared to the preceding 2.6.38 and 2.6.37 kernel releases.
As mentioned earlier on Phoronix, LinuxTag 2011 took place this past weekend in Berlin. One of the few talks I was able to make due to the Ubuntu Developer Summit in Budapest colliding with the event was the Wayland talk by SUSE's Egbert Eich. The focus of this talk was whether Wayland is on the way to becoming a new desktop standard.
With the recent look at the major Linux power regressions taking place within the Linux kernel, some initially wondered if the increase in power consumption was correlated to an increase in system performance. Unfortunately, it is clear now that is not the case. With that said though, here's some performance benchmarks of all major kernel releases going back to Linux 2.6.24 and ending with the Linux 2.6.39 kernel.
Recently there were benchmarks on Phoronix looking at the Ubuntu 11.04 boot performance relative to past Ubuntu Linux releases. This was done with five mobile systems and going back as far as Ubuntu 8.04. The tests showed around Ubuntu 10.04 LTS was where the boot performance in Ubuntu's been the best but Ubuntu 10.10 and 11.04 have slowed down a bit in how fast it's reaching the desktop. In this article we are looking at the boot performance when simply changing out the kernels. Every kernel from Linux 2.6.24 to 2.6.39-rc4 was analyzed.
Matthew Tippett and I presented at the 2011 Linux Foundation Collaboration Summit today in San Francisco about benchmarking and continuous testing of LLVM and the sub-projects that depend upon this compiler infrastructure. As the slides are somewhat generic and can be applied to many other open-source projects, the slides are now available. It's basically how to leverage the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org for driving continuous benchmarking to monitor performance regressions.
Version 4.6 of GCC was released over the weekend with a multitude of improvements and version 2.9 of the Low-Level Virtual Machine is due out in early April with its share of improvements. How though do these two leading open-source compilers compare? In this article we are providing benchmarks of GCC 4.5.2, GCC 4.6.0, DragonEgg with LLVM 2.9, and Clang with LLVM 2.9 across five distinct AMD / Intel systems to see how the compiler performance compares.
Version 2.9 of the Low-Level Virtual Machine is set to be released in a little more than a week, but what will it mean much for users in terms of performance? We will be looking at the LLVM 2.9 and Clang performance in the coming days (along with GCC 4.6, which was just released). We are beginning this weekend by providing a look at how using LLVM 2.9 affects the performance of the Mesa Gallium3D LLVMpipe driver relative to the previous LLVM 2.6, 2.7, and 2.8 releases.
While the performance of the Btrfs file-system with its default mount options didn't change much with the just-released Linux 2.6.38 kernel as shown by our large HDD and SSD file-system comparison, this new kernel does bring LZO file-system compression support to Btrfs. This Oracle-sponsored file-system has supported Gzip compression for months as a means to boost performance and preserve disk space, but now there's support for using LZO compression. In this article we are looking at the Btrfs performance with its default options and then when using the transparent Zlib and LZO compression.
There have been a flurry of comments this week following my post why software defaults are important and why in the Linux benchmarks at Phoronix.com the tests are most often carried out in their default/stock configurations: it's what most everyone uses. There have been comments by Ted Ts'o on file-system default mount options and whether they are sane or not in the non-enterprise distributions and others have questioned if defaults like Compiz on in Ubuntu by default makes sense. Does using Compiz still hurt your graphics performance?
Here are the results from our largest Linux file-system comparison to date. Using the soon-to-be-released Linux 2.6.38 kernel, on a SATA hard drive and solid-state drive, we benchmarked seven file-systems on each drive with the latest kernel code as of this past weekend. The tested file-systems include EXT3, EXT4, Btrfs, XFS, JFS, ReiserFS, and NILFS2.
Last month there were benchmarks on Phoronix looking at the multi-core scaling performance of multiple operating systems, including CentOS 5.5, Fedora 14, FreeBSD 8.1, and OpenIndiana b148. CentOS 5.5 uses the long-term Linux 2.6.18 kernel while Fedora 14 has the more recent Linux 2.6.35 kernel by default, but a number of users asked how the Linux 2.6.38 kernel would fair for multi-core scaling with the removal of the Big Kernel Lock and various other low-level improvements in this forthcoming kernel. Here are some benchmarks showing just that.
VirtualBox, the Sun/Oracle virtualization platform, has supported OpenGL acceleration and Direct3D acceleration within virtual machines for more than two years. When the host system has hardware GPU acceleration, OpenGL/Direct3D calls can be passed from the guest to the host when the VirtualBox guest driver is installed. There has been the Linux 3D support since VirtualBox 2.2 and was initially limited to OpenGL 1.4 support and in the summer of 2009 it turned to OpenGL 2.0. We had not delivered any early benchmarks as the initial support was too buggy, but even with the recently released VirtualBox 4.0, while the support is usable and stable for the most part, it is still far from being very efficient and will crash under some OpenGL software.
Matthew Tippett and I talked this weekend at the Southern California Linux Expo on the matter of making more informed Linux hardware choices. While Linux hardware support has come along way, it is not perfect and there are still shortcomings. However, with Phoronix Test Suite 3.0 and OpenBenchmarking.org, which were released in Los Angeles, we believe there are now the capabilities to dramatically enhance the Linux hardware and software experience. These freely available tools are not only a game-changer for Linux, but have the capabilities to impact how projects and organizations handle their Windows, Mac OS X, BSD, and Solaris testing as well.
Earlier this month benchmarks were published on Phoronix showing the GCC 4.6 compiler performance with AVX support under Intel's new Sandy Bridge processors that are the first to provide Advanced Vector Extensions support. The Core i5 2500K CPU performance is already great under Linux, but once more Linux software supports taking advantage of this latest cross-vendor instruction set, there will be even more speed-ups. While the Low-Level Virtual Machine does not yet have full support for taking advantage of the Advanced Vector Extensions support, in this article we are looking at how the latest development code for LLVM 2.9 and the Clang compiler are performing on Intel's Sandy Bridge in relation to GCC.
February has finally arrived. Later this month Phoronix Test Suite 3.0 "Iveland" and OpenBenchmarking.org will be officially unveiled from the Southern California Linux Expo during the talk entitled "Making Better Linux Hardware Choices" by myself and Matthew Tippett, the former ATI/AMD Linux Core Engineering Manager. Before the California Linux event, there may also be a public demonstration in Munich of this major Linux testing/benchmarking breakthrough. While the Phoronix Test Suite 3.0-Iveland software can currently be downloaded as beta, OpenBenchmarking.org is not yet publicly available nor have we said much about the project. What has been said though is that it will cause Linux benchmarking to change, it will likely cause a greater impact than Phoronix.com, may result in my editorial departure from Phoronix, and will change the way that you find Linux compatible hardware. Here though is a primer of some of what you can expect out of OpenBenchmarking.org when it becomes available late in the month.
Started by one of our readers more than a week ago was a compiler deathmatch for comparing the performance of GCC, LLVM Clang, PCC (the Portable C Compiler), TCC (Tiny C Compiler), and Intel's C Compiler under Arch Linux. This user did not stop there with compiling these different x86_64 code compilers, but he also went on to look at the compiler performance with different compiler flags, among other options. The results are definitely worth looking at and here are some more.
515 software articles published on Phoronix.