A significant patch-set was published on Saturday night that implements the driver-independent bits of OpenGL 4's ARB_tessellation_shader extension inside Mesa.
The tessellation support has been one of the big pieces missing from Mesa's OpenGL 4 implementation and fortunately it's getting close to mainline. Chris Forbes of Intel published fifty-six patches this weekend that implement the driver-independent portions of the extension inside Mesa. Of course, the driver portions still need to follow for it to be useful.
Marek Olšák has published another big set of fixes/improvements to the open-source RadeonSI Gallium3D driver.
On Friday, Marek posted a set of 15 patches against the RadeonSI Gallium3D driver, which for those out of the loop is the driver that handles AMD's Radeon HD 7000 series GPUs and newer. These 15 patches vary in nature but come down to "random improvements" ranging from improvements/clean-ups to DB registers and MSAA, code clean-ups, and OpenGL Geometry Shader fixes.
Mesa 10.3 has been released! Mesa 10.3 is a feature release that
includes many updates and enhancements. The full list is available in
the release notes file in docs/relnotes/10.3.html.
The tag in the GIT repository for Mesa 10.3 is 'mesa-10.3'. I have
verified that the tag is in the correct place in the tree.
Mesa 10.3 is available for download at
For those able to spend $1000+ (USD) on a processor, the Intel Core i7 5960X is a fantastic offering that is still leaving me pleased with the performance after extensive Linux testing.
After the initial X99 motherboard failure, everything has been working fine with a different motherboard. Aside from Legit Reviews' motherboard also failing, I haven't heard of any other X99 issues from others so it looks like it could be an isolated incident. However, the root cause still appears to have been undetermined. I'm still waiting for MSI on a direct update to the matter but over in Taiwan, MSI reportedly told one of my SilverStone contacts that the Raven RV05 fan is faulty and caused the problem (though the case/fan works fine with a different motherboard and SilverStone has received no reports of other Raven RV05 issues in the few months that the chassis has been on the market). I also heard that Kingston ended up lowering their DDR4 voltages on the memory kit that was supplied to Legit Reviews; their DDR4 kit is now rated for 1.35V instead of 1.5V. In regards to the ASUS X99 motherboard failure, I haven't heard anything new but just a rumor of a bad soldering job. Anyhow, it fortunately looks like no widespread problems but just some isolated incidents.
Testing was done with the Intel Core i7 5960X at stock speeds. The Linux 3.17 Git kernel atop Ubuntu 14.10 was used for this round of testing. The Core i7 5960X has eight physical cores plus Hyper Threading, 3.0GHz base frequency, and 3.5GHz turbo frequency (that's the reported difference in the system information table due to P-State advertising the turbo frequency where as CPUFreq just reports the base frequency).
While the Linux 3.17 kernel isn't being released for a few weeks, we already have a good idea for the DRM graphics driver improvements coming for the Linux 3.18 cycle.
Linux 3.17 has many new features, including many DRM graphics improvements, with Linux 3.18 there's of course more changes to get excited about; it's a never-ending cycle in improving Linux graphics drivers and the kernel stack as a whole. With Linux 3.18 though, it's going to be the first release where the drm-next merge window is closing early. Usually David Airlie, the DRM subsystem maintainer, allows new DRM graphics driver code to be introduced up until the start of the next kernel merge window, with that drm-next code-base then being sent in for mainline inclusion. Beginning with Linux 3.18, Airlie is planning to close the merge window of drm-next around the -rc5 state of the previous release. As a result, this week is likely the last that major new DRM graphics driver code has a chance to land for making the 3.18 window.
With X.Org Server 1.16 having landed in Ubuntu 14.10, it's time for some benchmarks comparing the 1.15 and 1.16 releases on Ubuntu while using the GLAMOR 2D acceleration library.
For some basic X.Org 2D benchmarks I tested a Radeon HD 7950 and R7 260X while running various Linux 2D desktop benchmarks on Ubuntu 14.10 with the Linux 3.16 kernel and Mesa 10.4-devel. In testing the two graphics cards, I was using X.Org Server 1.15.1 that was previously found in the Ubuntu Utopic archive and then switched to X.Org Server 1.16.0 with the rebuilt DDX driver packages too.