Make various updates and fixes:
- Add support for legacy CUDA versions; now compatible with CUDA 12.3 and newer - Add support for NVRTC compilation - Other fixes and code refactoring
This commit is contained in:
+1
-1
@@ -9,7 +9,7 @@
|
||||
namespace deep_gemm {
|
||||
|
||||
class KernelRuntimeCache {
|
||||
std::unordered_map<std::filesystem::path, std::shared_ptr<KernelRuntime>> cache;
|
||||
std::unordered_map<std::string, std::shared_ptr<KernelRuntime>> cache;
|
||||
|
||||
public:
|
||||
// TODO: consider cache capacity
|
||||
|
||||
Reference in New Issue
Block a user