build: update fallback gencodes

add hopper/blackwell sm90/sm100 fallback gencodes, guarded by >=12.0 and
>=12.6 respectively.

src/Makefile

@@ -16,10 +16,28 @@ CUDARTLIB ?= cudart
 
 CUDA_VERSION = $(strip $(shell which $(NVCC) >/dev/null && $(NVCC) --version | grep release | sed 's/.*release //' | sed 's/\,.*//'))
 CUDA_MAJOR = $(shell echo $(CUDA_VERSION) | cut -d "." -f 1)
+CUDA_MINOR = $(shell echo $(CUDA_VERSION) | cut -d "." -f 2)
 
 # Better define NVCC_GENCODE in your environment to the minimal set
 # of archs to reduce compile time.
-ifeq ($(shell test "0$(CUDA_MAJOR)" -ge 11; echo $$?),0)
+ifeq ($(shell test "0$(CUDA_MAJOR)" -ge 12 -a "0$(CUDA_MINOR)" -ge 6; echo $$?),0)
+NVCC_GENCODE ?= -gencode=arch=compute_60,code=sm_60 \
+                -gencode=arch=compute_61,code=sm_61 \
+                -gencode=arch=compute_70,code=sm_70 \
+                -gencode=arch=compute_80,code=sm_80 \
+                -gencode=arch=compute_80,code=sm_80 \
+                -gencode=arch=compute_90,code=sm_90 \
+                -gencode=arch=compute_100,code=sm_100 \
+                -gencode=arch=compute_100,code=compute_100
+else ifeq ($(shell test "0$(CUDA_MAJOR)" -ge 12; echo $$?),0)
+NVCC_GENCODE ?= -gencode=arch=compute_60,code=sm_60 \
+                -gencode=arch=compute_61,code=sm_61 \
+                -gencode=arch=compute_70,code=sm_70 \
+                -gencode=arch=compute_80,code=sm_80 \
+                -gencode=arch=compute_80,code=sm_80 \
+                -gencode=arch=compute_90,code=sm_90 \
+                -gencode=arch=compute_90,code=compute_90
+else ifeq ($(shell test "0$(CUDA_MAJOR)" -ge 11; echo $$?),0)
 NVCC_GENCODE ?= -gencode=arch=compute_60,code=sm_60 \
                 -gencode=arch=compute_61,code=sm_61 \
                 -gencode=arch=compute_70,code=sm_70 \