On 01/22/2018 06:23 PM, Andrew Morton wrote:
On Thu, 18 Jan 2018 11:47:48 -0500 Andrey Grodzovsky <andrey.grodzovsky@amd.com> wrote:


Hi, this series is a revised version of an RFC sent by Christian König
a few years ago. The original RFC can be found at 
https://lists.freedesktop.org/archives/dri-devel/2015-September/089778.html

This is the same idea and I've just adressed his concern from the original RFC 
and switched to a callback into file_ops instead of a new member in struct file.

Should be in address_space_operations, I suspect.  If an application
opens a file twice, we only want to count it once?

Makes sense 


But we're putting the cart ahead of the horse here.  Please provide us
with a detailed description of the problem which you are addressing so
that the MM developers can better consider how to address your
requirements.

I will just reiterate the problem statement from the original RFC, should have
put it in the body of the RFC and not just a link, as already commented by Michal.
Bellow is the quoted RFC.

Thanks,
Andrey

P.S You can also check the follow up discussion after this first email.

"
I'm currently working on the issue that when device drivers allocate memory on
behalf of an application the OOM killer usually doesn't knew about that unless
the application also get this memory mapped into their address space.

This is especially annoying for graphics drivers where a lot of the VRAM
usually isn't CPU accessible and so doesn't make sense to map into the
address space of the process using it.

The problem now is that when an application starts to use a lot of VRAM those
buffers objects sooner or later get swapped out to system memory, but when we
now run into an out of memory situation the OOM killer obviously doesn't knew
anything about that memory and so usually kills the wrong process

"