fix loading cudart.so issues

flashinfer/jit/__init__.py

@@ -76,10 +76,41 @@
 from .comm import gen_trtllm_comm_module as gen_trtllm_comm_module
 from .comm import gen_vllm_comm_module as gen_vllm_comm_module
 from .comm import gen_nvshmem_module as gen_nvshmem_module
+from typing import Optional
+
+
+def find_loaded_library(lib_name) -> Optional[str]:
+    """
+    According to according to https://man7.org/linux/man-pages/man5/proc_pid_maps.5.html,
+    the file `/proc/self/maps` contains the memory maps of the process, which includes the
+    shared libraries loaded by the process. We can use this file to find the path of the
+    a loaded library.
+    """
+    found = False
+    with open("/proc/self/maps") as f:
+        for line in f:
+            if lib_name in line:
+                found = True
+                break
+    if not found:
+        # the library is not loaded in the current process
+        return None
+    # if lib_name is libcudart, we need to match a line with:
+    # address /path/to/libcudart-hash.so.11.0
+    start = line.index("/")
+    path = line[start:].strip()
+    filename = path.split("/")[-1]
+    assert filename.rpartition(".so")[0].startswith(lib_name), (
+        f"Unexpected filename: {filename} for library {lib_name}"
+    )
+    return path
 
 
 cuda_lib_path = os.environ.get(
     "CUDA_LIB_PATH", "/usr/local/cuda/targets/x86_64-linux/lib/"
 )
-if os.path.exists(f"{cuda_lib_path}/libcudart.so.12"):
+process_cudart_path = find_loaded_library("libcudart")
+if process_cudart_path is not None:
+    ctypes.CDLL(process_cudart_path, mode=ctypes.RTLD_GLOBAL)
+elif os.path.exists(f"{cuda_lib_path}/libcudart.so.12"):
     ctypes.CDLL(f"{cuda_lib_path}/libcudart.so.12", mode=ctypes.RTLD_GLOBAL)