Re: [PATCH net-next 00/12] Add support for PSE port priority

On Thu, Oct 17, 2024 at 12:35:57PM +0200, Kory Maincent wrote:

On Tue, 15 Oct 2024 11:43:52 +0200
Kory Maincent [off-list ref] wrote:

quoted

quoted

Policy Variants and Implementation

In cases where we are discussing prioritization, we are fundamentally
talking about over-provisioning. This typically means that while a device
advertises a certain maximum per-port power capacity (e.g., 95W), the total
system power budget (e.g., 300W) is insufficient to supply maximum power to
all ports simultaneously. This is often due to various system limitations,
and if there were no power limits, prioritization wouldn't be necessary.

The challenge then becomes how to squeeze more Powered Devices (PDs) onto
one PSE system. Here are two methods for over-provisioning:

1. Static Method:
 
   This method involves distributing power based on PD classification. It’s
   straightforward and stable, with the software (probably within the PSE
   framework) keeping track of the budget and subtracting the power
requested by each PD’s class. 
 
   Advantages: Every PD gets its promised power at any time, which
guarantees reliability. 

   Disadvantages: PD classification steps are large, meaning devices request
   much more power than they actually need. As a result, the power supply
may only operate at, say, 50% capacity, which is inefficient and wastes
money.

2. Dynamic Method:  

   To address the inefficiencies of the static method, vendors like
Microchip have introduced dynamic power budgeting, as seen in the PD692x0
firmware. This method monitors the current consumption per port and
subtracts it from the available power budget. When the budget is exceeded,
lower-priority ports are shut down.  

   Advantages: This method optimizes resource utilization, saving costs.

   Disadvantages: Low-priority devices may experience instability. A
possible improvement could involve using LLDP protocols to dynamically
configure power limits per port, thus allowing us to reduce power on
over-consuming ports rather than shutting them down entirely.  

Indeed we will have only static method for PSE controllers not supporting
system power budget management like the TPS2388x or LTC426.
Both method could be supported for "smart" PSE controller like PD692x0.

Let's begin with the static method implementation in the PSE framework for
now. It will need the power domain notion you have talked about.

While developing the software support for port priority in static method, I
faced an issue.

Supposing we are exceeding the power budget when we plug a new PD.
The port power should not be enabled directly or magic smoke will appear.
So we have to separate the detection part to know the needs of the PD from the
power enable part.

Currently the port power is enabled on the hardware automatically after the
detection process. There is no way to separate power port process and detection
process with the PD692x0 controller and it could be done on the TPS23881 by
configuring it to manual mode but: "The use of this mode is intended for system
diagnostic purposes only in the event that ports cannot be powered in
accordance with the IEEE 802.3bt standard from semiauto or auto modes."
Not sure we want that.

So in fact the workaround you talked about above will be needed for the two PSE
controllers.

For the TPS23881, "9.1.1.2 Semiauto", seems to be exactly what we wont:
"The port performs detection and classification (if valid detection
occurs) continuously. Registers are updated each time a detection or
classification occurs. The port power is not automatically turned on. A
Power Enable command is required to turn on the port"

For PD692x0 controller, i'm not 100% sure. There is "4.3.5 Set Enable/Disable
Channels" command, "Sets individual port Enable (Delivering power
enable) or Disable (Delivering power disable)." 

For my understanding, "Delivering power" is the state after
classification. So, it is what we wont too.

If, it works in both cases, it would be a more elegant way to go. THe
controller do auto- detection and classification, what we should do in
the software is do decide if the PD can be enabled based on
classification results, priority and available budget.

quoted

Both methods have their pros and cons. Since the dynamic method is not always
desirable, and if there's no way to disable it in the PD692x0's firmware, one
potential workaround could be handling the budget in software and dynamically
setting per-port limits. For instance, with a total budget of 300W and unused
ports, we could initially set 95W limits per port. As high-priority PDs (e.g.,
three 95W devices) are powered, we could dynamically reduce the power limit on
the remaining ports to 15W, ensuring that no device exceeds that
classification threshold.

We would set port overcurrent limit for all unpowered ports when the power
budget available is less than max PI power 100W as you described.
If a new PD plugged exceed the overcurrent limit then it will raise an interrupt
and we could deal with the power budget to turn off low priority ports at that
time. 

Mmh in fact I could not know if the overcurrent event interrupt comes from a
newly plugged PD or not.

Hm..  in case of PD692x0, may be using event counters?

An option: When we get new PD device plug interrupt event, we wait the end of
classification time (Tpon 400ms) and read the interrupt states again to know if
there is an overcurrent or not on the port.

Let's try Semiauto mode for TPS23881 first, I assume it is designed
exactly for this use case.

And then, test if PD692x0 supports a way to disable auto power delivery
in the 4.3.5 command.

Regards,
Oleksij
-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |