diff --git a/Makefile b/Makefile
old mode 100644
new mode 100755
index f7a4d65..a568995
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,8 @@
 MKFILE_PATH := $(abspath $(lastword $(MAKEFILE_LIST)))
 ROOT_DIR := $(patsubst %/,%,$(dir $(MKFILE_PATH)))
-
-GPP:= /usr/bin/g++
+ROCM_DIR := /opt/rocm/
+HIPCC := $(ROCM_DIR)/bin/hipcc
+GPP:= g++
 ifeq ($(CUDA_HOME),)
 	CUDA_HOME:= $(shell which nvcc | rev | cut -d'/' -f3- | rev)
 endif
@@ -113,12 +114,12 @@ cpuonly: $(BUILD_DIR) env
 
 HIP_INCLUDE := -I $(ROOT_DIR)/csrc -I $(ROOT_DIR)/include 
 # -I /opt/rocm-5.3.0/hipcub/include
-HIP_LIB := -L/opt/rocm-5.3.0/lib -L/opt/rocm-5.3.0/llvm/bin/../lib/clang/15.0.0/lib/linux -L/usr/lib/gcc/x86_64-linux-gnu/11 -L/usr/lib/gcc/x86_64-linux-gnu/11/../../../../lib64 -L/lib/x86_64-linux-gnu -L/lib/../lib64 -L/usr/lib/x86_64-linux-gnu -L/usr/lib/../lib64 -L/lib -L/usr/lib -lgcc_s -lgcc -lpthread -lm -lrt -lamdhip64 -lhipblas -lhipsparse -lclang_rt.builtins-x86_64 -lstdc++ -lm -lgcc_s -lgcc -lc -lgcc_s -lgcc
+HIP_LIB := -L$(ROCM_DIR)/lib -L$(ROCM_DIR)/llvm/bin/../lib/clang/15.0.0/lib/linux -L/usr/lib/gcc/x86_64-linux-gnu/11 -L/usr/lib/gcc/x86_64-linux-gnu/11/../../../../lib64 -L/lib/x86_64-linux-gnu -L/lib/../lib64 -L/usr/lib/x86_64-linux-gnu -L/usr/lib/../lib64 -L/lib -L/usr/lib -lgcc_s -lgcc -lpthread -lm -lrt -lamdhip64 -lhipblas -lhipsparse -lclang_rt.builtins-x86_64 -lstdc++ -lm -lgcc_s -lgcc -lc -lgcc_s -lgcc
 
 hip: $(BUILD_DIR)
-	/usr/bin/hipcc -std=c++14 -c -fPIC --amdgpu-target=gfx1030 $(HIP_INCLUDE) -o $(BUILD_DIR)/ops.o -D NO_CUBLASLT $(CSRC)/ops.cu
-	/usr/bin/hipcc -std=c++14 -c -fPIC --amdgpu-target=gfx1030 $(HIP_INCLUDE) -o $(BUILD_DIR)/kernels.o -D NO_CUBLASLT $(CSRC)/kernels.cu
-	# /usr/bin/hipcc -fPIC -static $(BUILD_DIR)/ops.o $(BUILD_DIR)/kernels.o -o $(BUILD_DIR)/link.so 
+	$(HIPCC) -std=c++14 -c -fPIC --amdgpu-target=gfx1030 $(HIP_INCLUDE) -o $(BUILD_DIR)/ops.o -D NO_CUBLASLT $(CSRC)/ops.cu
+	$(HIPCC) -std=c++14 -c -fPIC --amdgpu-target=gfx1030 $(HIP_INCLUDE) -o $(BUILD_DIR)/kernels.o -D NO_CUBLASLT $(CSRC)/kernels.cu
+	# $(HIPCC) -fPIC -static $(BUILD_DIR)/ops.o $(BUILD_DIR)/kernels.o -o $(BUILD_DIR)/link.so 
 	$(GPP) -std=c++14 -D__HIP_PLATFORM_AMD__ -DBUILD_CUDA -shared -fPIC -I /opt/rocm/include $(HIP_INCLUDE) $(BUILD_DIR)/ops.o $(BUILD_DIR)/kernels.o $(FILES_CPP) $(HIP_LIB) -o ./bitsandbytes/libbitsandbytes_hip_nocublaslt.so 
 
 env:
diff --git a/README.md b/README.md
old mode 100644
new mode 100755
index d420e6c..f535ba4
--- a/README.md
+++ b/README.md
@@ -1,3 +1,29 @@
+# bitsandbytes-rocm
+
+Credits to [broncotc/bitsandbytes-rocm](https://github.com/broncotc/bitsandbytes-rocm) for the original fork, and [0cc4m/bitsandbytes-rocm](https://github.com/0cc4m/bitsandbytes-rocm) for updating it. I'm only hosting a local copy as this has some slight tweaks in the Makefile, and instructions
+
+## Pre-Requisite
+
+This assumes you're an Arch Linux user like me (or derivation). If you're using a different distro, you can easily adapt it.
+
+In `Makefile`:
+* edit line 3 to point to your ROCm install folder (default: `/opt/rocm/`), if your distro installs it in an exotic place
+* that should be it!
+
+## Compiling
+
+```
+# activate your VENV, if using this within a VENV
+git clone https://git.ecker.tech/mrq/bitsandbytes-rocm
+make hip
+CUDA_VERSION=gfx1030 python setup.py install # assumes you're using a 6XXX series card
+python3 -m bitsandbytes # to validate it works
+```
+
+**!**NOTE**!**: this assumes you have a AMD 6XXX series card. Adapt this to your proper GFX version if different.
+
+---
+
 # bitsandbytes
 
 The bitsandbytes is a lightweight wrapper around CUDA custom functions, in particular 8-bit optimizers, matrix multiplication (LLM.int8()), and quantization functions.