Change vec_scal_add examples to vec_scal_mul and cleaned up README references

programming_examples/basic/matrix_multiplication/makefile-common

@@ -100,4 +100,4 @@ clean: clean_trace
 
 .PHONY: clean_trace
 clean_trace:
-	rm -rf tmpTrace parse*.json 
+	rm -rf tmpTrace parse*.json  trace.txt

programming_guide/quick_reference.md

@@ -49,6 +49,12 @@
 | `print(ctx.module)` | Converts our ctx wrapped structural code to mlir and prints to stdout|
 | `ctx.module.operation.verify()` | Runs additional structural verficiation on the python binded source code and return result to stdout |
 
+## Common AIE API functions for Kernel Programming
+| Function Signature  | Definition | Parameters | Return Type | Example | 
+|---------------------|------------|------------|-------------|---------|
+| `aie::vector<T, vec_factor> my_vector` | Declare vector type | `T`: data type <br> `vec_factor`: vector width | n/a | aie::vector<int16_t, 32> my_vector; |
+| `aie::load_v<vec_factor>(pA1);` | Vector load | `vec_factor`: vector width | `aie::vector` | aie::vector<int16_t, 32> my_vector; |
+
 ## Helpful AI Engine Architecture References and Tables
 * [AIE2 - Table of supported data types and vector sizes (AIE API)](https://www.xilinx.com/htmldocs/xilinx2023_2/aiengine_api/aie_api/doc/group__group__basic__types.html)
 

programming_guide/section-1/Makefile

@@ -6,7 +6,7 @@
 # 
 ##===----------------------------------------------------------------------===##
 
-include ../../tutorials/makefile-common
+include ../../programming_examples/makefile-common
 
 build/aie.mlir: aie2.py
 	mkdir -p ${@D}

programming_guide/section-3/Makefile

@@ -12,11 +12,15 @@ all: build/final.xclbin build/insts.txt
 
 targetname = vectorScalar
 
+build/aie.mlir: aie2.py
+	mkdir -p ${@D}
+	python3 $< > $@
+
 build/scale.o: vector_scalar_mul.cc
 	mkdir -p ${@D}
 	cd ${@D} && xchesscc_wrapper ${CHESSCCWRAP2_FLAGS} -c $(<:%=../%) -o ${@F}
 
-build/final.xclbin: aie.mlir build/kernel1.o build/kernel2.o build/kernel3.o
+build/final.xclbin: build/aie.mlir build/scale.o
 	mkdir -p ${@D}
 	cd ${@D} && aiecc.py --aie-generate-cdo --no-compile-host --xclbin-name=${@F} \
 				--aie-generate-npu --npu-insts-name=insts.txt $(<:%=../%)

programming_guide/section-3/README.md

@@ -149,7 +149,6 @@ To compile the design and C++ testbench:
 
 ```sh
 make
-make build/vectorScalar.exe
 ```
 
 To run the design:

programming_guide/section-3/test.cpp

@@ -34,13 +34,10 @@ int main(int argc, const char *argv[]) {
 
   test_utils::parse_options(argc, argv, desc, vm);
   int verbosity = vm["verbosity"].as<int>();
-  int trace_size = vm["trace_sz"].as<int>();
 
   constexpr bool VERIFY = true;
-  constexpr bool ENABLE_TRACING = false;
-  // constexpr int TRACE_SIZE = 8192;
   constexpr int IN_SIZE = 4096;
-  constexpr int OUT_SIZE = ENABLE_TRACING ? IN_SIZE + trace_size / 4 : IN_SIZE;
+  constexpr int OUT_SIZE = IN_SIZE;
 
   // Load instruction sequence
   std::vector<uint32_t> instr_v =
@@ -64,7 +61,7 @@ int main(int argc, const char *argv[]) {
                         XRT_BO_FLAGS_HOST_ONLY, kernel.group_id(2));
   auto bo_inFactor = xrt::bo(device, 1 * sizeof(DATATYPE),
                              XRT_BO_FLAGS_HOST_ONLY, kernel.group_id(3));
-  auto bo_outC = xrt::bo(device, OUT_SIZE * sizeof(DATATYPE) + trace_size,
+  auto bo_outC = xrt::bo(device, OUT_SIZE * sizeof(DATATYPE),
                          XRT_BO_FLAGS_HOST_ONLY, kernel.group_id(4));
 
   if (verbosity >= 1)
@@ -85,7 +82,7 @@ int main(int argc, const char *argv[]) {
 
   // Zero out buffer bo_outC
   DATATYPE *bufOut = bo_outC.map<DATATYPE *>();
-  memset(bufOut, 0, OUT_SIZE * sizeof(DATATYPE) + trace_size);
+  memset(bufOut, 0, OUT_SIZE * sizeof(DATATYPE));
 
   // sync host to device memories
   bo_instr.sync(XCL_BO_SYNC_BO_TO_DEVICE);
@@ -120,11 +117,6 @@ int main(int argc, const char *argv[]) {
     }
   }
 
-  if (trace_size > 0) {
-    test_utils::write_out_trace(((char *)bufOut) + (IN_SIZE * sizeof(DATATYPE)),
-                                trace_size, vm["trace_file"].as<std::string>());
-  }
-
   // Print Pass/Fail result of our test
   if (!errors) {
     std::cout << std::endl << "PASS!" << std::endl << std::endl;

programming_guide/section-3/test.py

@@ -15,7 +15,6 @@
 from aie.extras.dialects.ext import memref, arith
 
 import aie.utils.test as test_utils
-import aie.utils.trace as trace_utils
 
 
 def main(opts):
@@ -41,7 +40,7 @@ def main(opts):
     INOUT1_SIZE = INOUT1_VOLUME * INOUT1_DATATYPE().itemsize
     INOUT2_SIZE = INOUT2_VOLUME * INOUT2_DATATYPE().itemsize
 
-    OUT_SIZE = INOUT2_SIZE + int(opts.trace_size)
+    OUT_SIZE = INOUT2_SIZE
 
     # ------------------------------------------------------
     # Get device, load the xclbin & kernel and register them
@@ -99,11 +98,6 @@ def main(opts):
         e = np.equal(output_buffer, ref)
         errors = errors + np.size(e) - np.count_nonzero(e)
 
-    # Write trace values if trace_size > 0
-    if opts.trace_size > 0:
-        trace_buffer = entire_buffer[INOUT2_VOLUME:]
-        trace_utils.write_out_trace(trace_buffer, str(opts.trace_file))
-
     # ------------------------------------------------------
     # Print verification and timing results
     # ------------------------------------------------------

programming_guide/section-4/section-4a/Makefile

@@ -16,10 +16,14 @@ build/aie.mlir: aie2.py
 	mkdir -p ${@D}
 	python3 $< > $@
 
-build/final.xclbin: build/aie.mlir
+build/scale.o: vector_scalar_mul.cc
 	mkdir -p ${@D}
-	cd ${@D} && aiecc.py --aie-generate-cdo --aie-generate-npu --no-compile-host \
-		--xclbin-name=${@F} --npu-insts-name=insts.txt ${<F}
+	cd ${@D} && xchesscc_wrapper ${CHESSCCWRAP2_FLAGS} -c $(<:%=../%) -o ${@F}
+
+build/final.xclbin: build/aie.mlir build/scale.o
+	mkdir -p ${@D}
+	cd ${@D} && aiecc.py --aie-generate-cdo --no-compile-host --xclbin-name=${@F} \
+				--aie-generate-npu --npu-insts-name=insts.txt $(<:%=../%)
 
 ${targetname}.exe: test.cpp
 	rm -rf _build

programming_guide/section-4/section-4a/README.md

@@ -24,7 +24,7 @@ Adding the application timer is as simple as noting a start and stop time surrou
 
 ```c++
     auto start = std::chrono::high_resolution_clock::now();
-    auto run = kernel(bo_instr, instr_v.size(), bo_inout0, bo_inout1, bo_inout2);
+    auto run = kernel(bo_instr, instr_v.size(), bo_inA, bo_inFactor, bo_outC);
     run.wait();
     auto stop = std::chrono::high_resolution_clock::now();
 
@@ -78,9 +78,6 @@ We can then compute and print the actual average, minimum and maximum run times.
 
 1. Let's set our iterations to 10 and run again with `make run` which recompiles our host code for `test.cpp`. What reported Avg NPU time do you see this time? <img src="../../../mlir_tutorials/images/answer1.jpg" title="Answer can be anywhere from 430-480us but is likely different than before" height=25>
 
-1. Let's change our design and increase the loop size of our kernel by a factor of 10. This involves changing the outer loop from 8 to 80. What reported times do you see now? <img src="../../../mlir_tutorials/images/answer1.jpg" title="? us" height=25>
-
-
 -----
 [[Up]](../../section-4) [[Next]](../section-4b)