-----Original Message-----
From: Wang, Haiyue <redacted>
Sent: Sunday, August 29, 2021 7:33 AM
To: Jerin Jacob <redacted>; NBU-Contact-Thomas Monjalon
[off-list ref]
Cc: Jerin Jacob <redacted>; dpdk-dev <redacted>; Stephen
Hemminger [off-list ref]; David Marchand
[off-list ref]; Andrew Rybchenko
[off-list ref]; Honnappa Nagarahalli
[off-list ref]; Yigit, Ferruh [off-list ref];
techboard@dpdk.org; Elena Agostini [off-list ref]
Subject: RE: [dpdk-dev] [RFC PATCH v2 0/7] heterogeneous computing library

quoted

-----Original Message-----
From: Jerin Jacob <redacted>
Sent: Friday, August 27, 2021 20:19
To: Thomas Monjalon <redacted>
Cc: Jerin Jacob <redacted>; dpdk-dev <redacted>; Stephen

Hemminger

quoted

[off-list ref]; David Marchand

[off-list ref]; Andrew Rybchenko

quoted

[off-list ref]; Wang, Haiyue [off-list ref];

Honnappa Nagarahalli

quoted

[off-list ref]; Yigit, Ferruh [off-list ref];

techboard@dpdk.org; Elena

quoted

Agostini [off-list ref]
Subject: Re: [dpdk-dev] [RFC PATCH v2 0/7] heterogeneous computing library

On Fri, Aug 27, 2021 at 3:14 PM Thomas Monjalon [off-list ref]

wrote:

quoted

quoted

31/07/2021 15:42, Jerin Jacob:

quoted

On Sat, Jul 31, 2021 at 1:51 PM Thomas Monjalon

[off-list ref] wrote:

quoted

quoted

quoted

quoted

31/07/2021 09:06, Jerin Jacob:

quoted

On Fri, Jul 30, 2021 at 7:25 PM Thomas Monjalon

[off-list ref] wrote:

quoted

quoted

quoted

quoted

quoted

quoted

From: Elena Agostini <redacted>

In heterogeneous computing system, processing is not only in the

CPU.

quoted

quoted

quoted

quoted

quoted

quoted

Some tasks can be delegated to devices working in parallel.

The goal of this new library is to enhance the collaboration between
DPDK, that's primarily a CPU framework, and other type of devices

like GPUs.

quoted

quoted

quoted

quoted

quoted

quoted

When mixing network activity with task processing on a non-CPU

device,

quoted

quoted

quoted

quoted

quoted

quoted

there may be the need to put in communication the CPU with the

device

quoted

quoted

quoted

quoted

quoted

quoted

in order to manage the memory, synchronize operations, exchange

info, etc..

quoted

quoted

quoted

quoted

quoted

quoted

This library provides a number of new features:
- Interoperability with device specific library with generic handlers
- Possibility to allocate and free memory on the device
- Possibility to allocate and free memory on the CPU but visible from

the device

quoted

quoted

quoted

quoted

quoted

quoted

- Communication functions to enhance the dialog between the CPU

and the device

quoted

quoted

quoted

quoted

quoted

quoted

The infrastructure is prepared to welcome drivers in drivers/hc/
as the upcoming NVIDIA one, implementing the hcdev API.

Some parts are not complete:
  - locks
  - memory allocation table
  - memory freeing
  - guide documentation
  - integration in devtools/check-doc-vs-code.sh
  - unit tests
  - integration in testpmd to enable Rx/Tx to/from GPU memory.

Since the above line is the crux of the following text, I will start
from this point.

+ Techboard

I  can give my honest feedback on this.

I can map similar  stuff  in Marvell HW, where we do machine learning
as compute offload
on a different class of CPU.

In terms of RFC patch features

1) memory API - Use cases are aligned
2) communication flag and communication list
Our structure is completely different and we are using HW ring kind of
interface to post the job to compute interface and
the job completion result happens through the event device.
Kind of similar to the DMA API that has been discussed on the mailing

list.

quoted

quoted

quoted

quoted

Interesting.

It is hard to generalize the communication mechanism.
Is other GPU vendors have a similar communication mechanism? AMD,

Intel ??

quoted

quoted

I don't know who to ask in AMD & Intel. Any ideas?

Good question.

At least in Marvell HW, the communication flag and communication list is
our structure is completely different and we are using HW ring kind of
interface to post the job to compute interface and
the job completion result happens through the event device.
kind of similar to the DMA API that has been discussed on the mailing list.

Please correct me if I'm wrong but what you are describing is a specific way 
to submit work on the device. Communication flag/list here is a direct data 
communication between the CPU and some kind of workload (e.g. GPU kernel)
that's already running on the device.

The rationale here is that:
- some work has been already submitted on the device and it's running
- CPU needs a real-time direct interaction through memory with the device
- the workload on the device needs some info from the CPU it can't get at submission time

This is good enough for NVIDIA and AMD GPU.
Need to double check for Intel GPU.

quoted

quoted

quoted

quoted

quoted

Now the bigger question is why need to Tx and then Rx something to
compute the device
Isn't  ot offload something? If so, why not add the those offload in
respective subsystem
to improve the subsystem(ethdev, cryptiodev etc) features set to adapt
new features or
introduce new subsystem (like ML, Inline Baseband processing) so that
it will be an opportunity to
implement the same in  HW or compute device. For example, if we take
this path, ML offloading will
be application code like testpmd, which deals with "specific" device
commands(aka glorified rawdev)
to deal with specific computing device offload "COMMANDS"
(The commands will be specific to  offload device, the same code wont
run on  other compute device)

Having specific features API is convenient for compatibility
between devices, yes, for the set of defined features.
Our approach is to start with a flexible API that the application
can use to implement any processing because with GPU programming,
there is no restriction on what can be achieved.
This approach does not contradict what you propose,
it does not prevent extending existing classes.

It does prevent extending the existing classes as no one is going to
extent it there is the path of not doing do.

I disagree. Specific API is more convenient for some tasks,
so there is an incentive to define or extend specific device class APIs.
But it should not forbid doing custom processing.

This is the same as the raw device is in DPDK where the device
personality is not defined.

Even if define another API and if the personality is not defined,
it comes similar to the raw device as similar
to rawdev enqueue and dequeue.

To summarize,

1)  My _personal_ preference is to have specific subsystems
to improve the DPDK instead of the raw device kind of path.

Something like rte_memdev to focus on device (GPU) memory management ?

The new DPDK auxiliary bus maybe make life easier to solve the complex
heterogeneous computing library. ;-)

To get a concrete idea about what's the best and most comprehensive
approach we should start with something that's flexible and simple enough.

A dedicated library it's a good starting point: easy to implement and embed in DPDK applications,
isolated from other components and users can play with it learning from the code.
As a second step we can think to embed the functionality in some other way 
within DPDK (e.g. split memory management and communication features).

quoted

2) If the device personality is not defined, use rawdev
3) All computing devices do not use  "communication flag" and
"communication list"
kind of structure. If are targeting a generic computing device then
that is not a portable scheme.
For GPU abstraction if "communication flag" and "communication list"
is the right kind of mechanism
then we can have a separate library for GPU communication specific to GPU <-

DPDK communication needs and explicit for GPU.

I think generic DPDK applications like testpmd should not
pollute with device-specific functions. Like, call device-specific
messages from the application
which makes the application runs only one device. I don't have a
strong opinion(expect
standardizing  "communication flag" and "communication list" as
generic computing device
communication mechanism) of others think it is OK to do that way in DPDK.

I'd like to introduce (with a dedicated option) the memory API in testpmd to 
provide an example of how to TX/RX packets using device memory.

I agree to not embed communication flag/list features.

quoted

quoted

quoted

If an application can run only on a specific device, it is similar to
a raw device,
where the device definition is not defined. (i.e JOB metadata is not defined

and

quoted

quoted

quoted

it is specific to the device).

quoted

quoted

Just my _personal_ preference is to have specific subsystems to
improve the DPDK instead of raw device kind of
path. If we decide another path as a community it is _fine_ too(as a
_project manager_ point of view it will be an easy path to dump SDK
stuff to DPDK without introducing the pain of the subsystem nor
improving the DPDK).

Adding a new class API is also improving DPDK.

But the class is similar as raw dev class. The reason I say,
Job submission and response is can be abstracted as queue/dequeue APIs.
Taks/Job metadata is specific to compute devices (and it can not be
generalized).
If we generalize it makes sense to have a new class that does
"specific function".

Computing device programming is already generalized with languages like

OpenCL.

quoted

quoted

We should not try to reinvent the same.
We are just trying to properly integrate the concept in DPDK
and allow building on top of it.

Agree.

quoted

See above.

quoted