Fedora Gets An Unofficial Kernel Based On Clear Linux

[phoronix]

Phoronix: Fedora Gets An Unofficial Kernel Based On Clear Linux

While the kernel configuration is just one part of Intel's Clear Linux optimizations for their performance-oriented distribution, a Fedora user has taken the liberty of spinning a Fedora kernel build based upon Clear Linux's kernel configuration...

http://www.phoronix.com/scan.php?page=news_item&px=Fedora-Unofficial-Clear-Kernel

[enihcam]

BTW FYI Archlinux also got a clearlinux-based kernel with extra CPU optimizations. https://aur.archlinux.org/packages/linux-clear/

[angrypie]

btw fyi archlinux also got a clearlinux-based kernel with extra cpu optimizations. https://aur.archlinux.org/packages/linux-clear/

zomgwtfbbq

[andyprough]

This has been tried multiple times for multiple distros. Last time Michael tried it in January for Ubuntu, all he managed was a decreased boot time with the Clear Linux optimized kernel. Nearly all other tests he ran were the same or showed only miniscule gains compared to the stock Ubuntu kernel.

I'm pretty sure that the reason ClearLinux wins the benchmarking tests is not because of optimizations, but because the distro, not being a real, robust distro, has very low overhead. It's built from the ground up for one thing - to win benchmarking tests.

[bitman]

This has been tried multiple times for multiple distros. Last time Michael tried it in January for Ubuntu, all he managed was a decreased boot time with the Clear Linux optimized kernel. Nearly all other tests he ran were the same or showed only miniscule gains compared to the stock Ubuntu kernel.

I'm pretty sure that the reason ClearLinux wins the benchmarking tests is not because of optimizations, but because the distro, not being a real, robust distro, has very low overhead. It's built from the ground up for one thing - to win benchmarking tests.

More like because other packages in clearlinux have similar optimizations. Kernel is just one piece of the puzzle. Not exactly vital either apparently.

[mir3x]

To get clear linux performance in most of phoronix tests its enough to export CFLAGS and CXXFLAGS

[Delgarde]

I'm pretty sure that the reason ClearLinux wins the benchmarking tests is not because of optimizations, but because the distro, not being a real, robust distro, has very low overhead.

Not really - the only "overhead" provided by any distro is basically just package management, so as long as you're not running system updates at the same time as your benchmarks, that's not going to be a factor...

[Candy]

More like because other packages in clearlinux have similar optimizations. Kernel is just one piece of the puzzle. Not exactly vital either apparently.

I remember when MorphOS received Altivec support for all sorts of memory functions (memcpy etc.) that alone this improved performance by a factor of 200x or so. If I remember correctly the Altivec support was within the gcc compiler used that time. Altivec specific code improved perofrmance everywhere.

The same applies here. Kernel optimization is just half of the entire story. You need to optimize at least glibc as well, to notice some performance gains. E.g. faster memcpy, faster memory anything etc.

Kernel, glibc and gcc improved performance will result in performant code. Specially if gcc is allowed to generate architecture specific optimizations.

[Mani]

The most interesting festure of clear linux is the function multiversioning, with this the whole system runs like it has been compiled with -march=native

This and the patched glibc would be highly appreciated in any distribution

[nanonyme]

I remember when MorphOS received Altivec support for all sorts of memory functions (memcpy etc.) that alone this improved performance by a factor of 200x or so. If I remember correctly the Altivec support was within the gcc compiler used that time. Altivec specific code improved perofrmance everywhere.

The same applies here. Kernel optimization is just half of the entire story. You need to optimize at least glibc as well, to notice some performance gains. E.g. faster memcpy, faster memory anything etc.

Kernel, glibc and gcc improved performance will result in performant code. Specially if gcc is allowed to generate architecture specific optimizations.

For some reason this reminds me of http://funroll-loops.teurasporsaat.org/