Remove DPSII patches

airootfs/home/tori/.local/share/tori/patches/README.md

@@ -1,6 +1,6 @@
 > [!IMPORTANT]
 > The DPSII patches have been superseded by the `nvidia-pstated` daemon, which can be found [here](https://github.com/sasha0552/nvidia-pstated). The daemon does not require application patching and switches performance states automatically.  
-> **Patches in this directory will be removed at the end of this year.**
+> **DPSII patches in this directory will be removed at the end of this year.**
 
 # ToriLinux patches
 
@@ -23,22 +23,22 @@ Don't forget to recompile the project if necessary!
 
 ## Patch types
 
-### Drop PState In Idle (DPSII)
+### ~~Drop PState In Idle (DPSII)~~
 
-These patches can be used to automatically manage the performance states of NVIDIA GPUs using the [nvidia-pstate](https://github.com/sasha0552/nvidia-pstate).
+~~These patches can be used to automatically manage the performance states of NVIDIA GPUs using the [nvidia-pstate](https://github.com/sasha0552/nvidia-pstate).~~
 
-Remember that you need to install `nvidia-pstate` in the virtual environment (if applicable), and/or add the executable to the system PATH.  
-*Currently only `llama.cpp` uses `nvidia-pstate` executables from PATH. Other projects use `nvidia-pstate` as a library.*
+~~Remember that you need to install `nvidia-pstate` in the virtual environment (if applicable), and/or add the executable to the system PATH.  
+*Currently only `llama.cpp` uses `nvidia-pstate` executables from PATH. Other projects use `nvidia-pstate` as a library.*~~
 
-This can reduce power consumption/temperatures, especially on server GPUs (note that SXM versions of server GPUs may not work).
+~~This can reduce power consumption/temperatures, especially on server GPUs (note that SXM versions of server GPUs may not work).~~
 
-For example, on my NVIDIA Tesla P40, these patches reduce power consumption from 50W to 10W (and temperature from 50C to ~25C (room temperature)) by setting the performance state to 8 when the application is idle. When a request comes in, it restores the driver behavior, so inference can run as usual.
+~~For example, on my NVIDIA Tesla P40, these patches reduce power consumption from 50W to 10W (and temperature from 50C to ~25C (room temperature)) by setting the performance state to 8 when the application is idle. When a request comes in, it restores the driver behavior, so inference can run as usual.~~
 
-This type of patch is currently available for these projects:
+~~This type of patch is currently available for these projects:~~
 
-* [automatic](https://github.com/vladmandic/automatic)
-* [llama.cpp](https://github.com/ggerganov/llama.cpp)
-* [vllm](https://github.com/vllm-project/vllm)
+* ~~[automatic](https://github.com/vladmandic/automatic)~~
+* ~~[llama.cpp](https://github.com/ggerganov/llama.cpp)~~
+* ~~[vllm](https://github.com/vllm-project/vllm)~~
 
 ### There were a few other patches earlier ...