forked from onnx/onnx-mlir
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
[NNPA] zDNN extension for splitting elementwise operations (onnx#2690)
* zDNN extension for splitting elementwise operations and interface cleanup Signed-off-by: Tung D. Le <tung@jp.ibm.com> --------- Signed-off-by: Tung D. Le <tung@jp.ibm.com>
- Loading branch information
Showing
9 changed files
with
682 additions
and
133 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
340 changes: 340 additions & 0 deletions
340
src/Accelerators/NNPA/Runtime/zDNNExtension/Elementwise.c
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,340 @@ | ||
| /* | ||
| * SPDX-License-Identifier: Apache-2.0 | ||
| */ | ||
|
|
||
| //===------------------------ Elementwise.c -------------------------------===// | ||
| // | ||
| // Copyright 2024 The IBM Research Authors. | ||
| // | ||
| // ============================================================================= | ||
| // | ||
| // A wrapper of zdnn elementwise ops for ztensor partition and parallelism. | ||
| // | ||
| //===----------------------------------------------------------------------===// | ||
|
|
||
| // Include pthreads (need special treatment on z/OS). | ||
| #ifdef __MVS__ | ||
| #define _OPEN_THREADS | ||
| #endif | ||
| #include <pthread.h> | ||
|
|
||
| #include <assert.h> | ||
| #include <math.h> | ||
| #include <stdio.h> | ||
| #include <stdlib.h> | ||
| #include <sys/time.h> | ||
|
|
||
| #include "zDNNExtension.h" | ||
| #include "zdnn.h" | ||
|
|
||
| #ifdef __cplusplus | ||
| extern "C" { | ||
| #endif | ||
|
|
||
| typedef enum ElemementwiseOp { | ||
| // Binary | ||
| ZDNN_ADD_EXT, | ||
| ZDNN_DIV_EXT, | ||
| ZDNN_MAX_EXT, | ||
| ZDNN_MIN_EXT, | ||
| ZDNN_MUL_EXT, | ||
| ZDNN_SUB_EXT, | ||
| // Unary | ||
| ZDNN_EXP_EXT, | ||
| ZDNN_LOG_EXT, | ||
| ZDNN_RELU_EXT, | ||
| ZDNN_TANH_EXT, | ||
| ZDNN_SIGMOID_EXT, | ||
| } ElemementwiseOp; | ||
|
|
||
| static zdnn_status zdnn_unary_elementwise_common(const zdnn_ztensor *input, | ||
| const void *clippingValue, zdnn_ztensor *output, ElemementwiseOp opType) { | ||
| // Verify that e4, e3, e1 do not exceed the maximum dimension size. Thus, we | ||
| // will split e2 safely. | ||
| OrigShape origShapeOfX; | ||
| getOrigShape(input, &origShapeOfX); | ||
| uint32_t maxDimSize = zdnn_get_nnpa_max_dim_idx_size(); | ||
| if ((origShapeOfX.e4 > maxDimSize) || (origShapeOfX.e3 > maxDimSize) || | ||
| (origShapeOfX.e1 > maxDimSize)) { | ||
| printf("[UnaryElementwise] The input tensor dimension exceeds maximum " | ||
| "dimension index " | ||
| "size (MDIS) of %d: e4 = %d, e3 = %d, e1 = %d.\n", | ||
| maxDimSize, origShapeOfX.e4, origShapeOfX.e3, origShapeOfX.e1); | ||
| return ZDNN_EXCEEDS_MDIS; | ||
| } | ||
|
|
||
| // We split e2 in (e4, e3, e2, e1). | ||
| SplitInfo splitInfoX = { | ||
| .origZTensor = input, .axis = 2, .chunkSize = OMZTensorSplitSize}; | ||
| SplitInfo splitInfoY = { | ||
| .origZTensor = output, .axis = 2, .chunkSize = OMZTensorSplitSize}; | ||
|
|
||
| // Dim is small or ztensor split is disabled. | ||
| if (!OMZTensorSplitEnabled || !initSplitInfo(&splitInfoX) || | ||
| !initSplitInfo(&splitInfoY)) { | ||
| if (OMZTensorSplitDebug) | ||
| printf("[UnaryElementwise] Not split zTensor ...\n"); | ||
| if (opType == ZDNN_EXP_EXT) | ||
| return zdnn_exp(input, output); | ||
| else if (opType == ZDNN_LOG_EXT) | ||
| return zdnn_log(input, output); | ||
| else if (opType == ZDNN_RELU_EXT) | ||
| return zdnn_relu(input, clippingValue, output); | ||
| else if (opType == ZDNN_SIGMOID_EXT) | ||
| return zdnn_sigmoid(input, output); | ||
| else if (opType == ZDNN_TANH_EXT) | ||
| return zdnn_tanh(input, output); | ||
| else | ||
| return ZDNN_UNAVAILABLE_FUNCTION; | ||
| } | ||
|
|
||
| // Split input. | ||
| if (OMZTensorSplitDebug) | ||
| printf("[UnaryElementwise] Split the input ztensor along e2 into %d chunks " | ||
| "of %d elements \n", | ||
| splitInfoX.numOfChunks, splitInfoX.chunkSize); | ||
|
|
||
| double splitTime = 0.; | ||
| double mmTime = 0.; | ||
| double mergeTime = 0.; | ||
| clock_t start_time, end_time; | ||
|
|
||
| // Split input into chunks. | ||
| if (OMZTensorSplitDebug) | ||
| start_time = clock(); | ||
| splitZTensor(&splitInfoX, /*copyData=*/true); | ||
| splitZTensor(&splitInfoY, /*copyData=*/false); | ||
| if (OMZTensorSplitDebug) { | ||
| end_time = clock(); | ||
| splitTime = ((float)(end_time - start_time) / (float)CLOCKS_PER_SEC) * 1000; | ||
| } | ||
|
|
||
| // Call zdnn op on each chunk. | ||
| if (OMZTensorSplitDebug) | ||
| start_time = clock(); | ||
| for (uint32_t i = 0; i < splitInfoX.numOfChunks; ++i) { | ||
| zdnn_ztensor *zxTensor = (splitInfoX.chunks + i)->ztensor; | ||
| zdnn_ztensor *zyTensor = (splitInfoY.chunks + i)->ztensor; | ||
| zdnn_status status; | ||
| if (opType == ZDNN_EXP_EXT) | ||
| status = zdnn_exp(zxTensor, zyTensor); | ||
| else if (opType == ZDNN_LOG_EXT) | ||
| status = zdnn_log(zxTensor, zyTensor); | ||
| else if (opType == ZDNN_RELU_EXT) | ||
| status = zdnn_relu(zxTensor, clippingValue, zyTensor); | ||
| else if (opType == ZDNN_SIGMOID_EXT) | ||
| status = zdnn_sigmoid(zxTensor, zyTensor); | ||
| else if (opType == ZDNN_TANH_EXT) | ||
| status = zdnn_tanh(zxTensor, zyTensor); | ||
| else | ||
| status = ZDNN_UNAVAILABLE_FUNCTION; | ||
| assert(status == ZDNN_OK); | ||
| } | ||
| if (OMZTensorSplitDebug) { | ||
| end_time = clock(); | ||
| mmTime = ((float)(end_time - start_time) / (float)CLOCKS_PER_SEC) * 1000; | ||
| } | ||
|
|
||
| // Merging the chunks into the output. | ||
| if (OMZTensorSplitDebug) | ||
| start_time = clock(); | ||
| mergeZTensors(&splitInfoY); | ||
| if (OMZTensorSplitDebug) { | ||
| end_time = clock(); | ||
| mergeTime = ((float)(end_time - start_time) / (float)CLOCKS_PER_SEC) * 1000; | ||
| } | ||
|
|
||
| freeSplitInfoBuffer(&splitInfoX); | ||
| freeSplitInfoBuffer(&splitInfoY); | ||
|
|
||
| if (OMZTensorSplitDebug) | ||
| printf("[UnaryElementwise] split, %f, mm, %f, merge, %f (milliseconds)\n", | ||
| splitTime, mmTime, mergeTime); | ||
|
|
||
| return ZDNN_OK; | ||
| } | ||
|
|
||
| static zdnn_status zdnn_binary_elementwise_common(const zdnn_ztensor *inputA, | ||
| const zdnn_ztensor *inputB, zdnn_ztensor *output, ElemementwiseOp opType) { | ||
| // Verify that e4, e3, e1 do not exceed the maximum dimension size. Thus, we | ||
| // will split e2 safely. | ||
| OrigShape origShapeOfA, origShapeOfB; | ||
| getOrigShape(inputA, &origShapeOfA); | ||
| getOrigShape(inputB, &origShapeOfB); | ||
| uint32_t maxDimSize = zdnn_get_nnpa_max_dim_idx_size(); | ||
| if ((origShapeOfA.e4 > maxDimSize) || (origShapeOfA.e3 > maxDimSize) || | ||
| (origShapeOfA.e1 > maxDimSize)) { | ||
| printf("[BinaryElementwise] The 1st tensor dimension exceeds maximum " | ||
| "dimension index " | ||
| "size (MDIS) of %d: e4 = %d, e3 = %d, e1 = %d.\n", | ||
| maxDimSize, origShapeOfA.e4, origShapeOfA.e3, origShapeOfA.e1); | ||
| return ZDNN_EXCEEDS_MDIS; | ||
| } | ||
| if ((origShapeOfB.e4 > maxDimSize) || (origShapeOfB.e3 > maxDimSize) || | ||
| (origShapeOfB.e1 > maxDimSize)) { | ||
| printf("[BinaryElementwise] The 2nd tensor dimension exceeds maximum " | ||
| "dimension index " | ||
| "size (MDIS) of %d: e4 = %d, e3 = %d, e1 = %d.\n", | ||
| maxDimSize, origShapeOfB.e4, origShapeOfB.e3, origShapeOfB.e1); | ||
| return ZDNN_EXCEEDS_MDIS; | ||
| } | ||
|
|
||
| // We split e2 in (e4, e3, e2, e1). | ||
| SplitInfo splitInfoA = { | ||
| .origZTensor = inputA, .axis = 2, .chunkSize = OMZTensorSplitSize}; | ||
| SplitInfo splitInfoB = { | ||
| .origZTensor = inputB, .axis = 2, .chunkSize = OMZTensorSplitSize}; | ||
| SplitInfo splitInfoY = { | ||
| .origZTensor = output, .axis = 2, .chunkSize = OMZTensorSplitSize}; | ||
|
|
||
| // Dim is small or ztensor split is disabled. | ||
| if (!OMZTensorSplitEnabled || !initSplitInfo(&splitInfoA) || | ||
| !initSplitInfo(&splitInfoB) || !initSplitInfo(&splitInfoY)) { | ||
| if (OMZTensorSplitDebug) | ||
| printf("[BinaryElementwise] Not split zTensor ...\n"); | ||
| if (opType == ZDNN_ADD_EXT) | ||
| return zdnn_add(inputA, inputB, output); | ||
| else if (opType == ZDNN_SUB_EXT) | ||
| return zdnn_sub(inputA, inputB, output); | ||
| else if (opType == ZDNN_MUL_EXT) | ||
| return zdnn_mul(inputA, inputB, output); | ||
| else if (opType == ZDNN_MAX_EXT) | ||
| return zdnn_max(inputA, inputB, output); | ||
| else if (opType == ZDNN_MIN_EXT) | ||
| return zdnn_min(inputA, inputB, output); | ||
| else | ||
| return ZDNN_UNAVAILABLE_FUNCTION; | ||
| } | ||
|
|
||
| // Split input. | ||
| if (OMZTensorSplitDebug) | ||
| printf( | ||
| "[BinaryElementwise] Split the input ztensors along e2 into %d chunks " | ||
| "of %d elements \n", | ||
| splitInfoA.numOfChunks, splitInfoA.chunkSize); | ||
|
|
||
| double splitTime = 0.; | ||
| double mmTime = 0.; | ||
| double mergeTime = 0.; | ||
| clock_t start_time, end_time; | ||
|
|
||
| // Split input into chunks. | ||
| if (OMZTensorSplitDebug) | ||
| start_time = clock(); | ||
| splitZTensor(&splitInfoA, /*copyData=*/true); | ||
| splitZTensor(&splitInfoB, /*copyData=*/true); | ||
| splitZTensor(&splitInfoY, /*copyData=*/false); | ||
| if (OMZTensorSplitDebug) { | ||
| end_time = clock(); | ||
| splitTime = ((float)(end_time - start_time) / (float)CLOCKS_PER_SEC) * 1000; | ||
| } | ||
|
|
||
| // Call zdnn op on each chunk. | ||
| if (OMZTensorSplitDebug) | ||
| start_time = clock(); | ||
| for (uint32_t i = 0; i < splitInfoA.numOfChunks; ++i) { | ||
| zdnn_ztensor *zaTensor = (splitInfoA.chunks + i)->ztensor; | ||
| zdnn_ztensor *zbTensor = (splitInfoB.chunks + i)->ztensor; | ||
| zdnn_ztensor *zyTensor = (splitInfoY.chunks + i)->ztensor; | ||
| zdnn_status status; | ||
| if (opType == ZDNN_ADD_EXT) | ||
| status = zdnn_add(zaTensor, zbTensor, zyTensor); | ||
| else if (opType == ZDNN_SUB_EXT) | ||
| status = zdnn_sub(zaTensor, zbTensor, zyTensor); | ||
| else if (opType == ZDNN_MUL_EXT) | ||
| status = zdnn_mul(zaTensor, zbTensor, zyTensor); | ||
| else if (opType == ZDNN_DIV_EXT) | ||
| status = zdnn_div(zaTensor, zbTensor, zyTensor); | ||
| else if (opType == ZDNN_MAX_EXT) | ||
| status = zdnn_max(zaTensor, zbTensor, zyTensor); | ||
| else if (opType == ZDNN_MIN_EXT) | ||
| status = zdnn_min(zaTensor, zbTensor, zyTensor); | ||
| else | ||
| status = ZDNN_UNAVAILABLE_FUNCTION; | ||
| assert(status == ZDNN_OK); | ||
| } | ||
| if (OMZTensorSplitDebug) { | ||
| end_time = clock(); | ||
| mmTime = ((float)(end_time - start_time) / (float)CLOCKS_PER_SEC) * 1000; | ||
| } | ||
|
|
||
| // Merging the chunks into the output. | ||
| if (OMZTensorSplitDebug) | ||
| start_time = clock(); | ||
| mergeZTensors(&splitInfoY); | ||
| if (OMZTensorSplitDebug) { | ||
| end_time = clock(); | ||
| mergeTime = ((float)(end_time - start_time) / (float)CLOCKS_PER_SEC) * 1000; | ||
| } | ||
|
|
||
| freeSplitInfoBuffer(&splitInfoA); | ||
| freeSplitInfoBuffer(&splitInfoB); | ||
| freeSplitInfoBuffer(&splitInfoY); | ||
|
|
||
| if (OMZTensorSplitDebug) | ||
| printf("[BinaryElementwise] split, %f, mm, %f, merge, %f (milliseconds)\n", | ||
| splitTime, mmTime, mergeTime); | ||
|
|
||
| return ZDNN_OK; | ||
| } | ||
|
|
||
| // ----------------------------------------------------------------------------- | ||
| // Extension Functions | ||
| // Same name as zdnn functions but with the `_ext` postfix. | ||
| // ----------------------------------------------------------------------------- | ||
|
|
||
| zdnn_status zdnn_add_ext(const zdnn_ztensor *inputA, const zdnn_ztensor *inputB, | ||
| zdnn_ztensor *output) { | ||
| return zdnn_binary_elementwise_common(inputA, inputB, output, ZDNN_ADD_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_sub_ext(const zdnn_ztensor *inputA, const zdnn_ztensor *inputB, | ||
| zdnn_ztensor *output) { | ||
| return zdnn_binary_elementwise_common(inputA, inputB, output, ZDNN_SUB_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_mul_ext(const zdnn_ztensor *inputA, const zdnn_ztensor *inputB, | ||
| zdnn_ztensor *output) { | ||
| return zdnn_binary_elementwise_common(inputA, inputB, output, ZDNN_MUL_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_div_ext(const zdnn_ztensor *inputA, const zdnn_ztensor *inputB, | ||
| zdnn_ztensor *output) { | ||
| return zdnn_binary_elementwise_common(inputA, inputB, output, ZDNN_DIV_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_min_ext(const zdnn_ztensor *inputA, const zdnn_ztensor *inputB, | ||
| zdnn_ztensor *output) { | ||
| return zdnn_binary_elementwise_common(inputA, inputB, output, ZDNN_MIN_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_max_ext(const zdnn_ztensor *inputA, const zdnn_ztensor *inputB, | ||
| zdnn_ztensor *output) { | ||
| return zdnn_binary_elementwise_common(inputA, inputB, output, ZDNN_MAX_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_exp_ext(const zdnn_ztensor *input, zdnn_ztensor *output) { | ||
| return zdnn_unary_elementwise_common(input, NULL, output, ZDNN_EXP_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_log_ext(const zdnn_ztensor *input, zdnn_ztensor *output) { | ||
| return zdnn_unary_elementwise_common(input, NULL, output, ZDNN_LOG_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_relu_ext(const zdnn_ztensor *input, const void *clippingValue, | ||
| zdnn_ztensor *output) { | ||
| return zdnn_unary_elementwise_common( | ||
| input, clippingValue, output, ZDNN_RELU_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_sigmoid_ext(const zdnn_ztensor *input, zdnn_ztensor *output) { | ||
| return zdnn_unary_elementwise_common(input, NULL, output, ZDNN_SIGMOID_EXT); | ||
| } | ||
|
|
||
| zdnn_status zdnn_tanh_ext(const zdnn_ztensor *input, zdnn_ztensor *output) { | ||
| return zdnn_unary_elementwise_common(input, NULL, output, ZDNN_TANH_EXT); | ||
| } | ||
|
|
||
| #ifdef __cplusplus | ||
| } | ||
| #endif |
Oops, something went wrong.