On Tue, Apr 16, 2013 at 12:20 AM, Simon Jeons wrote: > Hi Jerome, > > On 04/15/2013 11:38 PM, Jerome Glisse wrote: > > On Mon, Apr 15, 2013 at 4:39 AM, Simon Jeons wrote: > >> Hi Jerome, >> On 02/10/2013 12:29 AM, Jerome Glisse wrote: >> >>> On Sat, Feb 9, 2013 at 1:05 AM, Michel Lespinasse >>> wrote: >>> >>>> On Fri, Feb 8, 2013 at 3:18 AM, Shachar Raindel >>>> wrote: >>>> >>>>> Hi, >>>>> >>>>> We would like to present a reference implementation for safely sharing >>>>> memory pages from user space with the hardware, without pinning. >>>>> >>>>> We will be happy to hear the community feedback on our prototype >>>>> implementation, and suggestions for future improvements. >>>>> >>>>> We would also like to discuss adding features to the core MM subsystem >>>>> to >>>>> assist hardware access to user memory without pinning. >>>>> >>>> This sounds kinda scary TBH; however I do understand the need for such >>>> technology. >>>> >>>> I think one issue is that many MM developers are insufficiently aware >>>> of such developments; having a technology presentation would probably >>>> help there; but traditionally LSF/MM sessions are more interactive >>>> between developers who are already quite familiar with the technology. >>>> I think it would help if you could send in advance a detailed >>>> presentation of the problem and the proposed solutions (and then what >>>> they require of the MM layer) so people can be better prepared. >>>> >>>> And first I'd like to ask, aren't IOMMUs supposed to already largely >>>> solve this problem ? (probably a dumb question, but that just tells >>>> you how much you need to explain :) >>>> >>> For GPU the motivation is three fold. With the advance of GPU compute >>> and also with newer graphic program we see a massive increase in GPU >>> memory consumption. We easily can reach buffer that are bigger than >>> 1gbytes. So the first motivation is to directly use the memory the >>> user allocated through malloc in the GPU this avoid copying 1gbytes of >>> data with the cpu to the gpu buffer. The second and mostly important >>> >> >> The pinned memory you mentioned is the memory user allocated or the >> memory of gpu buffer? >> > > Memory user allocated, we don't want to pin this memory. > > > After this idea merged, we don't need to allocate memory for integrated > GPU buffer and discrete GPU don't need to have its own memory, correct? > You need to stop considering discret and integrated GPU as different, they are not from driver point of view. Integrated GPU will keep stealing a chunk of system memory at boot because its a BIOS things and BIOS don't change like that. Both (integrated and discret) will keep allocating system memory in kernel for their own buffer because API such as OpenGL or OpenCL needs too. The transparent use of same address space on GPU as on CPU will only happen with newer API such as OpenCL 2.0 or other API that are schedule down the road. And discrete GPU will keep having its own memory, the whole point is that GDDR5 can be more than 10 times faster than the fastest DDR3. You can not beat that. GPU is all about bandwidth and that's not gonna change. Cheers, Jerome