I have a EVGA 2070 Super bought in early March last year, stock clock (I'm in Texas, I don't need more heat). That when on some occasions of gaming the screens turn black and the fans crank up and I have to reboot to get to a usable desktop again. ... Should also mention I'm running Linux mint 20.2 (Cinnamon). What's the ROOT cause? And should I RMA? Get a new PSU (that draws more power from the wall)?
 
Specs:
 
PCPartPicker Part List:

CPU: Intel Core i7-3770K 3.5 GHz Quad-Core Processor
Motherboard: Gigabyte GA-Z77X-UD5H-WB ATX LGA1155 Motherboard
Memory: G.Skill Ripjaws X Series 16 GB (2 x 8 GB) DDR3-1600 CL9 Memory  [x2, 32 GB total RAM]
Storage: SanDisk Extreme 120 GB 2.5" Solid State Drive
Storage: Seagate Barracuda 2 TB 3.5" 7200RPM Internal Hard Drive
Storage: Seagate Barracuda 3 TB 3.5" 7200RPM Internal Hard Drive
Video Card: EVGA GeForce RTX 2070 SUPER 8 GB FTW3 ULTRA+ Video Card
Case: Cooler Master HAF X ATX Full Tower Case
Power Supply: SeaSonic G 550 W 80+ Gold Certified Semi-modular ATX Power Supply

Well since this only started happening recently and didn't happen at all last year and since I can still use the computer otherwise I'd think we could rule out the PSU.  Some sort of (linux specific?) driver, possible loose connection to the GPU (one of the cables?, or the power cable to the PSU). 

This sounds absolutely like an overheating card. The solutions to this are fairly simple:
 
1. Check the GPU temperatures using a software like GPU Temp. Anything below 80C at load is fine. Anything above 80C at load can cause issues.
2. Check all connections on the card itself and make sure everything is tight and seated properly (this includes the 8 pin).
3. As odd as it sounds, run a memory test of your system (Windows Memory Diagnostic Tool).
4. Inspect the actual card itself for signs of damage, particularly bent capacitors and/or burn marks. If those are present and you are not handy with a soldering gun, you will need to send the card in for repair.
 
Either way, extreme temperatures can kill your card. The easy way to deal with this would be to improve overall case ventilation (equal push/pull fan configuration, good fan filters that you keep clean, and a temperature controlled room to no more than 75 degrees). Additionally, run the GPU fans at 100%, 100% of the time. It may be "noisy", but if you use headphones to game you won't notice it at all. My rig sounds like a jet engine and has a total of 10 fans at 100 CFM each, blowing in a 5 x 5 push/pull configuration, along with the additional 3 GPU fans, 2 CPU fans, and PSU fan. Some say overkill, but I have NEVER had a component die from thermal degradation using this method for the last 20+ years. Thermal degradation is a very real thing and over time can absolutely render your components useless.
 
Good luck, I truly hope it works out for you!