1

Given a server with Intel E3-1275 v6 CPU, is it possible to instruct Linux kernel to stop using the integrated graphics and restore video RAM back to OS? I'd prefer using a kernel command line flag and disable the integrated graphics during linux boot process so that I can still use graphics adapter for debugging if needed (that is, graphics adapter were still available for grub). The server is otherwise headless and does not require a display for normal use. I'm okay with a solution that requires full power cycle to restore ability to enable the graphics adapter. I'm running Ubuntu 16.04 LTS 64 bit with HWE stack in case it makes a difference (roughly equivalent to vanilla linux 4.8.17).

2
1

Ok my initial thought was no and it still kinda/sorta stands in that this has little or nothing to do with Linux itself. Certainly most BIOS's will allow you to totally disable the integrated video output but then again they usually use that for the BIOS screen/options/menu itself so - but if you can completely disable it then of course that memory then becomes available to whatever operating system is then booted - and if the OS supports headless boot then that's great.

That said I'm not entirely sure how appropriate this is anyway, it's always useful to have a proper frame-buffer/video-output on a server 'just in case' and if you tune the amount of memory down then it won't be using too much memory at all - and memory's quite cheap right now anyway. I'd be tempted to lower the memory as much as possible but leave it in place myself.

2
  • Considering that linux is taking care of all hardware after boot is complete it should be technically possible to disable the graphics and free the memory in running system. This is no different on hot swapping PCI Express graphics adapters on the fly, which is supported by the hardware in theory. Real world OS support is probably more problematic. – Mikko Rantalainen May 5 '17 at 6:48
  • I agree that the amount of RAM freed from the GPU would not be that much. However, disabling the GPU and avoiding continous screen refresh scanning (which requires reading framebuffer for each refresh, 60x the whole framebuffer per second in practice) should improve total usable memory bandwidth and may reduce memory latency a bit. – Mikko Rantalainen May 5 '17 at 6:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.