[Refactor] Phaseout LLVM Dependency by Making it Optional (#247)

* remove llvm build

* [Refactor] Update kernel compilation and profiling in examples

- Replaced `tilelang.lower` with `tilelang.compile` in multiple example scripts to streamline kernel compilation.
- Updated profiling calls to utilize the new `get_profiler` method, enhancing performance measurement consistency.
- Adjusted assertions and benchmarking methods to align with the new profiling structure across various examples, ensuring correctness and clarity in performance evaluations.

* lint fix

* License Update

* [Refactor] Improve code formatting and documentation in CUDA header and HIP runtime files

- Adjusted formatting in `cuda.h` for better readability, including alignment of comments and struct fields.
- Cleaned up whitespace and improved comment clarity in `rt_mod_hip.cc` to enhance code maintainability.

* [Refactor] Enhance formatting and clarity in CUDA header and HIP runtime files

- Improved comment alignment and readability in `cuda.h`.
- Cleaned up whitespace and formatting in `rt_mod_hip.cc` to enhance maintainability.

* lint fix

* lint fix

* lint fix

* lint fix

* fix

* License update

* [Enhancement] Update JITKernel to use artifact for kernel source

- Assigned the generated artifact to `self.artifact` for better management.
- Updated kernel source references to use `artifact.kernel_source` for consistency in execution backend handling.

* lint fix

* Add @tilelang.testing.requires_llvm decorator to vectorization tests

* Enhance setup.py and env.py for library management

- Added functionality to remove original files after copying in CMakeBuild.
- Updated TVM_LIBRARY_PATH in env.py to include the PyPI build library path for better integration.

* Refactor TVM_LIBRARY_PATH assignment for improved readability in env.py

* Refactor CMakeBuild file handling in setup.py

- Added a check to ensure the target library directory exists before copying .so files.
- Improved the logic for creating the target directory and copying files to enhance robustness.

* bugfix

* Rename BuildTLDebug to BuildTileLangCUDAWithoutCompile and update registration. Add @tilelang.testing.requires_llvm decorator to multiple tests for LLVM requirement.

* lint fix

* Enhance TileLang code generation by adding support for device code generation without compilation. Updated `host_codegen` and `device_codegen` functions to include new transformations and registration for `tilelang_hip_without_compile`. Refactored JIT kernel adapters to accommodate host and device modules, improving overall integration and flexibility.

* lint fix

* Add support for C target in device code generation

- Updated `device_codegen_without_compile` to include handling for the C target by registering the `tilelang_cpp` function.

* [Enhancement] Implement auto-clear cache feature based on environment variable

* Added TILELANG_CLEAR_CACHE environment variable to control cache clearing.
* Updated CI workflow to set TILELANG_CLEAR_CACHE during testing.
* Modified cache initialization to clear cache if TILELANG_CLEAR_CACHE is set to true.

* [Refactor] Update kernel invocation and import paths in tests and cache

* Changed kernel invocation in `test_tilelang_kernel_dequantize_gemm.py` to return the result.
* Updated import statements in `test_tilelang_kernel_int4_gemm_mma.py` to use `bitblas` instead of `tilelang`.
* Refactored paths for artifact and parameters in `kernel_cache.py` for better maintainability.

* [Refactor] Clean up whitespace and improve code formatting in kernel_cache.py

* Removed unnecessary blank lines and adjusted spacing for better readability in the KernelCache class.
* Enhanced overall code formatting to align with project standards.

* [Enhancement] Add bfloat16 test case and improve kernel caching logic

* Introduced a new test case for bfloat16 matrix multiplication in `test_tilelang_kernel_gemm_mma_intrinsic.py`.
* Updated `KernelCache` to handle multiple kernel source files and improve error handling during saving and loading.
* Refactored `JITKernel` to support instantiation from a database, enhancing flexibility in kernel management.
* Adjusted `CtypesKernelAdapter` and `CythonKernelAdapter` to utilize the new kernel loading mechanism from the database.
* Improved code formatting and readability across several files.

* lint fix

* Update bfloat16 matrix multiplication test case to use larger dimensions for improved coverage

src/layout/utils.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file layout/utils.cc
  * \brief Some arith tools for layout & fragment inference

src/layout/utils.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file layout/utils.h
  * \brief Some arith tools for layout & fragment inference

src/op/bulk_copy.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/bulk_copy.h
  * \brief Bulk copy operator.

src/op/elem.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/elem.cc
  *

src/op/elem.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/elem.h
  * \brief Define elment-wise operators.

src/op/gemm.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/gemm.cc
  *

src/op/gemm.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/gemm.h
  * \brief Define gemm operator.

src/op/op.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/op.cc
  *

src/op/parallel.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/parallel.h
  * \brief Infer layout from ops and parallel for

src/op/reduce.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/reduce.cc
  *

src/op/reduce.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/op/reduce.h
  * \brief Define reduce operator.

src/runtime/runtime.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/runtime/runtime.h
  * \brief Runtime functions.

src/runtime/runtime.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file tl/runtime/runtime.h
  * \brief Runtime functions.

src/target/codegen_cuda.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file target/codegen.cc
  */

src/target/codegen_cuda.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file target/codegen.h
  * \brief Utility to generate code

src/target/codegen_hip.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file target/codegen.cc
  */

src/target/codegen_hip.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 /*!
  * \file target/codegen.h
  * \brief Utility to generate code

src/target/cuda.h

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
 /*
  * Copyright 1993-2023 NVIDIA Corporation.  All rights reserved.
  *
@@ -1150,7 +1148,7 @@ typedef enum CUpointer_attribute_enum {
              from a mempool. Otherwise returns NULL. **/
   CU_POINTER_ATTRIBUTE_MAPPING_SIZE =
       18, /**< Size of the actual underlying mapping that the pointer belongs to
-             **/
+           **/
   CU_POINTER_ATTRIBUTE_MAPPING_BASE_ADDR =
       19, /**< The start address of the mapping that the pointer belongs to **/
   CU_POINTER_ATTRIBUTE_MEMORY_BLOCK_ID =
@@ -2230,12 +2228,14 @@ typedef struct CUgraphEdgeData_st {
                     ::CU_GRAPH_KERNEL_NODE_PORT_LAUNCH_ORDER. */
   unsigned char
       to_port;        /**< This indicates what portion of the downstream node is
-                         dependent on        the upstream node or portion thereof (indicated
-                         by \c from_port). The        meaning is specific to the node type. A
-                         value of 0 in all cases means        the entirety of the downstream
-                         node is dependent on the upstream work.        <br>        Currently no node
-                         types define non-zero ports. Accordingly, this field        must be
-                         set to zero. */
+                         dependent on        the upstream node or portion thereof
+                         (indicated        by \c from_port). The        meaning is
+                         specific to        the node type. A        value of 0 in all
+                         cases        means        the        entirety of the
+                         downstream        node        is        dependent on the
+                         upstream        work.        <br>        Currently        no
+                         node        types define        non-zero ports.        Accordingly,
+                         this        field        must be        set to        zero. */
   unsigned char type; /**< This should be populated with a value from
                          ::CUgraphDependencyType. (It is typed as char due to
                          compiler-specific layout of bitfields.) See
@@ -2495,15 +2495,17 @@ typedef enum CUlaunchAttributeID_enum {
 typedef union CUlaunchAttributeValue_union {
   char pad[64]; /* Pad to 64 bytes */
   CUaccessPolicyWindow
-      accessPolicyWindow;             /**< Value of launch attribute
-                                         ::CU_LAUNCH_ATTRIBUTE_ACCESS_POLICY_WINDOW. */
-  int cooperative;                    /**< Value of launch attribute
-                                         ::CU_LAUNCH_ATTRIBUTE_COOPERATIVE. Nonzero indicates a
-                                         cooperative                    kernel (see ::cuLaunchCooperativeKernel). */
-  CUsynchronizationPolicy syncPolicy; /**< Value of launch attribute
-                                         ::CU_LAUNCH_ATTRIBUTE_SYNCHRONIZATION_POLICY.
-                                         ::CUsynchronizationPolicy for work
-                                         queued up in this stream */
+      accessPolicyWindow; /**< Value of launch attribute
+                             ::CU_LAUNCH_ATTRIBUTE_ACCESS_POLICY_WINDOW. */
+  int cooperative;        /**< Value of launch attribute
+                             ::CU_LAUNCH_ATTRIBUTE_COOPERATIVE. Nonzero indicates a
+                             cooperative                    kernel (see
+                             ::cuLaunchCooperativeKernel). */
+  CUsynchronizationPolicy
+      syncPolicy; /**< Value of launch attribute
+                     ::CU_LAUNCH_ATTRIBUTE_SYNCHRONIZATION_POLICY.
+                     ::CUsynchronizationPolicy for work
+                     queued up in this stream */
   /**
    *  Value of launch attribute ::CU_LAUNCH_ATTRIBUTE_CLUSTER_DIMENSION that
@@ -2524,8 +2526,8 @@ typedef union CUlaunchAttributeValue_union {
   CUclusterSchedulingPolicy
       clusterSchedulingPolicyPreference;      /**< Value of launch attribute
                                                  ::CU_LAUNCH_ATTRIBUTE_CLUSTER_SCHEDULING_POLICY_PREFERENCE.
-                                                 Cluster      scheduling policy preference
-                                                 for the kernel. */
+                                                 Cluster      scheduling policy
+                                                 preference      for the kernel. */
   int programmaticStreamSerializationAllowed; /**< Value of launch attribute
                                                 ::CU_LAUNCH_ATTRIBUTE_PROGRAMMATIC_STREAM_SERIALIZATION.
                                               */
@@ -4844,7 +4846,7 @@ typedef struct CUgraphNodeParams_st {
     CUDA_CHILD_GRAPH_NODE_PARAMS graph;    /**< Child graph node parameters. */
     CUDA_EVENT_WAIT_NODE_PARAMS eventWait; /**< Event wait node parameters. */
     CUDA_EVENT_RECORD_NODE_PARAMS
-        eventRecord; /**< Event record node parameters. */
+    eventRecord; /**< Event record node parameters. */
     CUDA_EXT_SEM_SIGNAL_NODE_PARAMS_v2
         extSemSignal; /**< External semaphore signal node parameters. */
     CUDA_EXT_SEM_WAIT_NODE_PARAMS_v2
@@ -4854,7 +4856,7 @@ typedef struct CUgraphNodeParams_st {
     CUDA_MEM_FREE_NODE_PARAMS free; /**< Memory free node parameters. */
     CUDA_BATCH_MEM_OP_NODE_PARAMS_v2 memOp; /**< MemOp node parameters. */
     CUDA_CONDITIONAL_NODE_PARAMS
-        conditional; /**< Conditional node parameters. */
+    conditional; /**< Conditional node parameters. */
   };
   long long reserved2; /**< Reserved bytes. Must be zero. */

src/target/rt_mod_cpp.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 #include "codegen_cpp.h"
 
 namespace tvm {

src/target/rt_mod_cuda.cc

-// Copyright (c) Microsoft Corporation.
-// Licensed under the MIT License.
-
 #include "codegen_cuda.h"
 #include "runtime/cuda/cuda_module.h"
 
@@ -70,7 +67,7 @@ runtime::Module BuildTileLangCUDA(IRModule mod, Target target) {
   return runtime::CUDAModuleCreate(ptx, fmt, ExtractFuncInfo(mod), code);
 }
 
-String BuildTLDebug(IRModule mod, Target target) {
+runtime::Module BuildTileLangCUDAWithoutCompile(IRModule mod, Target target) {
   using tvm::runtime::Registry;
   bool output_ssa = false;
   CodeGenTileLangCUDA cg;
@@ -90,13 +87,13 @@ String BuildTLDebug(IRModule mod, Target target) {
   if (const auto *f = Registry::Get("tilelang_callback_cuda_postproc")) {
     code = (*f)(code, target).operator std::string();
   }
-  return String(code);
+  return runtime::CUDAModuleCreate("ptx", "ptx", ExtractFuncInfo(mod), code);
 }
 
 TVM_REGISTER_GLOBAL("target.build.tilelang_cuda")
     .set_body_typed(BuildTileLangCUDA);
-TVM_REGISTER_GLOBAL("target.build.tl_debug_codegen")
-    .set_body_typed(BuildTLDebug);
+TVM_REGISTER_GLOBAL("target.build.tilelang_cuda_without_compile")
+    .set_body_typed(BuildTileLangCUDAWithoutCompile);
 
 } // namespace codegen
 } // namespace tvm