On Tue, Nov 3, 2020 at 4:04 PM Alexei Starovoitov
[off-list ref] wrote:

On Tue, Nov 03, 2020 at 02:19:22PM -0500, Kenny Ho wrote:

quoted

On Tue, Nov 3, 2020 at 12:43 AM Alexei Starovoitov
[off-list ref] wrote:

quoted

On Mon, Nov 2, 2020 at 9:39 PM Kenny Ho [off-list ref] wrote:

Sounds like either bpf_lsm needs to be made aware of cgv2 (which would
be a great thing to have regardless) or cgroup-bpf needs a drm/gpu specific hook.
I think generic ioctl hook is too broad for this use case.
I suspect drm/gpu internal state would be easier to access inside
bpf program if the hook is next to gpu/drm. At ioctl level there is 'file'.
It's probably too abstract for the things you want to do.
Like how VRAM/shader/etc can be accessed through file?
Probably possible through a bunch of lookups and dereferences, but
if the hook is custom to GPU that info is likely readily available.
Then such cgroup-bpf check would be suitable in execution paths where
ioctl-based hook would be too slow.

Just to clarify, when you say drm specific hook, did you mean just a
unique attach_type or a unique prog_type+attach_type combination?  (I
am still a bit fuzzy on when a new prog type is needed vs a new attach
type.  I think prog type is associated with a unique type of context
that the bpf prog will get but I could be missing some nuances.)

When I was thinking of doing an ioctl wide hook, the file would be the
device file and the thinking was to have a helper function provided by
device drivers to further disambiguate.  For our (AMD's) driver, we
have a bunch of ioctls for set/get/create/destroy
(https://elixir.bootlin.com/linux/latest/source/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c#L1763)
so the bpf prog can make the decision after the disambiguation.  For
example, we have an ioctl called "kfd_ioctl_set_cu_mask."  You can
think of cu_mask like cpumask but for the cores/compute-unit inside a
GPU.  The ioctl hook will get the file, the bpf prog will call a
helper function from the amdgpu driver to return some data structure
specific to the driver and then the bpf prog can make a decision on
gating the ioctl or not.  From what you are saying, sounds like this
kind of back and forth lookup and dereferencing should be avoided for
performance considerations?

Having a DRM specific hook is certainly an alternative.  I just wasn't
sure which level of trade off on abstraction/generic is acceptable.  I
am guessing a new BPF_PROG_TYPE_CGROUP_AMDGPU is probably too
specific?  But sounds like BPF_PROG_TYPE_CGROUP_DRM may be ok?

Regards,
Kenny