Intel never said that the processor won't run Linux, but that they aren't going to support it.
Huge difference.
The reason they aren't going to support it is because Windows uses different system calls relating to shutting down cores on the processor in order to save battery life, and they designed the chip specifically for Windows.
If you used Linux with the processor, you'd get horrible battery life, and Intel understandably does not want the blame for that.
I had an inkling the headline might be sensational after RTFA. So it looks like this is just one more poorly-documented hoop for kernel devs to jump through?
Sorry but this sounds like marketing speak. I highly doubt that Intel designs anything specifically for Windows at the CPU level.
There is no direct mapping between Windows System Calls and functionality of a specific chip. There is a ton of code between that system call and what actually happens on the chip or in its firmware.
The top layer sure is Windows specific because it runs as part of Windows. However, the bottom layer is just bits to fiddle with or instructions to send to the CPU and any OS can implement that. Including Linux.
If you click through to the source article, they even describe what I'm referring to:
>"Intel went to great lengths to highlight the new P-states and C-states in which it can completely shut down the clock of a core. The firm said the operating system needs to provide "hints" to the processor in order to make use of power states and it seems likely that such hints are presently not provided by the Linux kernel in order to properly make use of Clover Trail."
Atom contains the superawful PowerVR chip that is a nightmare in Linux and Intel never got around to release open source drivers (because it's only licensed form PowerVR).
i'm not really surprised. I remember intel guys talking about hybrid tablets in the near future, with a iX CPU's running on the docks and atoms in the tablet part.
don't forget that we're talking about atom CPU's here. for all I care atoms are shit anyway, the AMD fusion alternatives are much better.
on the one hand, why would the linux kernel support some stuff that isn't there yet and only relevant to new windows 8 tablets.
on the other hand I don't see the intel linux driver teams disappearing anytime soon.
12 comments
[ 3.4 ms ] story [ 38.6 ms ] threadHuge difference.
The reason they aren't going to support it is because Windows uses different system calls relating to shutting down cores on the processor in order to save battery life, and they designed the chip specifically for Windows.
If you used Linux with the processor, you'd get horrible battery life, and Intel understandably does not want the blame for that.
Seriously though, this is why I don't have /. on my bookmarks bar anymore. It's almost as bad as Fox News.
There is no direct mapping between Windows System Calls and functionality of a specific chip. There is a ton of code between that system call and what actually happens on the chip or in its firmware.
The top layer sure is Windows specific because it runs as part of Windows. However, the bottom layer is just bits to fiddle with or instructions to send to the CPU and any OS can implement that. Including Linux.
Again, marketing blah.
>"Intel went to great lengths to highlight the new P-states and C-states in which it can completely shut down the clock of a core. The firm said the operating system needs to provide "hints" to the processor in order to make use of power states and it seems likely that such hints are presently not provided by the Linux kernel in order to properly make use of Clover Trail."
Also: http://msdn.microsoft.com/en-us/library/windows/desktop/ff19...
How long until the commits start rolling in for this?
Atom contains the superawful PowerVR chip that is a nightmare in Linux and Intel never got around to release open source drivers (because it's only licensed form PowerVR).
http://www.unix-heaven.org/intel-powervr-gma3600-for-linux
don't forget that we're talking about atom CPU's here. for all I care atoms are shit anyway, the AMD fusion alternatives are much better.
on the one hand, why would the linux kernel support some stuff that isn't there yet and only relevant to new windows 8 tablets.
on the other hand I don't see the intel linux driver teams disappearing anytime soon.