Here is the circuit I am using.
I try to tune the input voltage on the gate of the lower device, so that the current pass through the resistor(0.25 Ohm) to certain level (let's say 7A for example). If the voltage of the power supply (the 50V one in the diagram) is low, like 30V, there is no oscillation detected. But to increase the voltage to like 50V, then the oscillation can be detected.
The oscillation is just like the following diagrams show.

I don't know where this oscillation comes from, and why is so "stable".

 

Schematic please!
(Two of your images have vanished)

 

 

Schematic please!
(Two of your images have vanished)

Attached.
BTW, can I delete the thread? I want to repost it. I am a new one here

 

Is the gate of the upper MOSFET being held to its source between pulses?

 

Is the gate of the upper MOSFET being held to its source between pulses?

I think yeah. But I'm not quite sure what do u mean "held", you mean the voltage is stable?

 

MOSFETs tend to be adversely affected by gate drive that doesn't remain strictly within the absolute ±gate drive voltage specifications. One way to do that conveniently is to reference gate drive from the MOSFET source. That is a reason why floating gate driver ICs are common.

 

What are you using as the source of the pulse?
Is it isolated from, and low capacitance to, the ground of your circuit?

 

What are you using as the source of the pulse?
Is it isolated from, and low capacitance to, the ground of your circuit?

I use a SI8264 for generating the pulse. It's powered by an isolated DC/DC converter.