forked from vortexgpgpu/vortex
-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Ivan khromov #1
Open
vano105
wants to merge
8
commits into
master
Choose a base branch
from
ivan_khromov
base: master
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Ivan khromov #1
Changes from all commits
Commits
Show all changes
8 commits
Select commit
Hold shift + click to select a range
9beecfd
Add tests from sgemm tutor
vano105 9c2a057
Add new kernel
vano105 72f2738
Fix input args function
vano105 0777e5d
clean code
vano105 03630ad
Add jupiter notebook
vano105 e004fb8
Remove graph from j_stat
vano105 7dfaec3
Delete output file in j_stat
vano105 826da5c
Fix issues for pull request
vano105 File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1 @@ | ||
| -DNUM_CLUSTERS=1 -DNUM_CORES=16 -DNUM_WARPS=16 -DNUM_THREADS=16 +0+0 |
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,268 @@ | ||
| { | ||
| "cells": [ | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 37, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "import subprocess\n", | ||
| "import matplotlib.pyplot as plt\n", | ||
| "from dataclasses import dataclass, field\n", | ||
| "import pandas as pd\n", | ||
| "from tqdm import tqdm\n", | ||
| "from pathlib import Path" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 38, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "# architecture parameters\n", | ||
| "@dataclass\n", | ||
| "class arch:\n", | ||
| " warps: int = 1\n", | ||
| " cores: int = 1\n", | ||
| " threads: int = 1\n", | ||
| "# running parameters \n", | ||
| "@dataclass\n", | ||
| "class run:\n", | ||
| " arch: arch\n", | ||
| " kernel: str\n", | ||
| " driver: str = \"simx\"\n", | ||
| " args: dict = field(default_factory=dict)" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 39, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "path_to_vortex = Path.cwd().parent.parent.parent\n", | ||
| "tile_size = 'TS'\n", | ||
| "work_per_thread = 'WPT'\n", | ||
| "width = 'WIDTH'" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 40, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "def error_running (run_params: run, error_text: str) -> str:\n", | ||
| " return f\"error in running in {run_params.kernel} : warps={run_params.arch.warps} cores={run_params.arch.cores} threads={run_params.arch.threads}\" \\\n", | ||
| " f\" driver={run_params.driver} args=-N{run_params.args['N']} -M{run_params.args['M']} -K{run_params.args['K']} error message - {error_text}/n\"" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 41, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "def error_verification (run_params: run, number_of_errors: str) -> str:\n", | ||
| " return f\"error in verifing results {run_params.kernel} : warps={run_params.arch.warps} cores={run_params.arch.cores} threads={run_params.arch.threads}\" \\\n", | ||
| " f\" driver={run_params.driver} args=-N{run_params.args['N']} -M{run_params.args['M']} -K{run_params.args['K']} Number of errors : {number_of_errors}'\\n'\" " | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 42, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "def create_common_h (params: dict, kernel_name: str):\n", | ||
| " file_name = f\"{path_to_vortex}/tests/opencl/{kernel_name}/common.h\"\n", | ||
| " with open(file_name, 'w') as file:\n", | ||
| " file.write(\"#ifndef COMMON_H\\n\" + \"#define COMMON_H\\n\" + \"\\n\")\n", | ||
| " if tile_size in params:\n", | ||
| " file.write(f\"#define TS {params[tile_size]}\\n\")\n", | ||
| " if work_per_thread in params:\n", | ||
| " file.write(f\"#define WPT {params[work_per_thread]}\\n\")\n", | ||
| " file.write(\"#define RTS (TS/WPT)\\n\")\n", | ||
| " if width in params:\n", | ||
| " file.write(f\"#define WIDTH {params[width]}\\n\")\n", | ||
| " file.write('\\n' + \"#endif // COMMON_H\")\n", | ||
| " # open main.cc file to recompile before run with new common.h\n", | ||
| " Path(f\"{path_to_vortex}/tests/opencl/{kernel_name}/main.cc\").touch(exist_ok=True)" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 43, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "def perf (run_params: run, path_to_output_file: str) -> pd.DataFrame:\n", | ||
| " # run kernel\n", | ||
| " vortex = f\"--warps={run_params.arch.warps} --cores={run_params.arch.cores} --threads={run_params.arch.threads}\"\n", | ||
| " run_args = f\"-N{run_params.args['N']} -M{run_params.args['M']} -K{run_params.args['K']}\"\n", | ||
| " command = f\"cd {path_to_vortex}/build && ./ci/blackbox.sh {vortex} --driver={run_params.driver} --app={run_params.kernel} --args=\\\"{run_args}\\\"\"\n", | ||
| " print(command)\n", | ||
| " result = subprocess.run(f\"{command} > {path_to_output_file}\", shell=True)\n", | ||
| "\n", | ||
| " # collect statistic \n", | ||
| " with open(path_to_output_file, 'r') as file:\n", | ||
| " lines = file.readlines()\n", | ||
| " error_message = \"\"\n", | ||
| " general_perf_stat = \"\"\n", | ||
| " for line in lines:\n", | ||
| " if \"PERF:\" in line:\n", | ||
| " general_perf_stat = line\n", | ||
| " # check for errors\n", | ||
| " if result != 0:\n", | ||
| " if \"FAILED\" in line: \n", | ||
| " error_message = error_verification(run_params, line[line.find(\"FAILED! - \"):])\n", | ||
| " if \"Error\" in line:\n", | ||
| " error_message = error_running(run_params, line[line.find(\"Error:\"):])\n", | ||
| " # pars string with general perf statistic of running kernel\n", | ||
| " pairs = general_perf_stat.replace(\"PERF: \", \"\").split(\", \")\n", | ||
| " perf_dict = {key_value.split(\"=\")[0]: float(key_value.split(\"=\")[1]) for key_value in pairs}\n", | ||
| " if perf_dict[\"cycles\"] <= 0:\n", | ||
| " error_message = error_running(run_params, \"Invalid number of cycles\")\n", | ||
| " # write result to data frame\n", | ||
| " run_result = pd.DataFrame([{\"kernel\": run_params.kernel[-1], \"driver\": run_params.driver, \"cores\": run_params.arch.cores, \n", | ||
| " \"warps\": run_params.arch.warps, \"threads\": run_params.arch.threads, \"M\": run_params.args[\"M\"], \n", | ||
| " \"N\": run_params.args[\"N\"], \"K\": run_params.args[\"K\"], \"instrs\": perf_dict[\"instrs\"], \"cycles\": perf_dict[\"cycles\"],\n", | ||
| " \"IPC\": perf_dict[\"IPC\"], \"error\": error_message}])\n", | ||
| " return run_result" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 44, | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "def draw (data_frame: pd.DataFrame, x_label: str, y_label: str, title: str, path: str):\n", | ||
| " data_frame.plot(kind = \"bar\", x = x_label, y = y_label)\n", | ||
| " plt.title(title)\n", | ||
| " plt.xlabel(x_label)\n", | ||
| " plt.ylabel(y_label)\n", | ||
| " plt.savefig(path)" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 45, | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stderr", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| " 0%| | 0/4 [00:00<?, ?it/s]" | ||
| ] | ||
| }, | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "cd /home/jblab/ivan_khromov/release/vortex/build && ./ci/blackbox.sh --warps=2 --cores=1 --threads=8 --driver=simx --app=kernel1 --args=\"-N16 -M16 -K16\"\n" | ||
| ] | ||
| }, | ||
| { | ||
| "name": "stderr", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "In file included from /home/jblab/tools/pocl/include/CL/cl.h:20,\n", | ||
| " from /home/jblab/tools/pocl/include/CL/opencl.h:24,\n", | ||
| " from /home/jblab/ivan_khromov/release/vortex/tests/opencl/kernel1/main.cc:1:\n", | ||
| "/home/jblab/tools/pocl/include/CL/cl_version.h:22:104: note: #pragma message: cl_version.h: CL_TARGET_OPENCL_VERSION is not defined. Defaulting to 300 (OpenCL 3.0)\n", | ||
| " 22 | #pragma message(\"cl_version.h: CL_TARGET_OPENCL_VERSION is not defined. Defaulting to 300 (OpenCL 3.0)\")\n", | ||
| " | ^\n", | ||
| "/home/jblab/ivan_khromov/release/vortex/tests/opencl/kernel1/main.cc: In function ‘int main(int, char**)’:\n", | ||
| "/home/jblab/ivan_khromov/release/vortex/tests/opencl/kernel1/main.cc:161:25: warning: unused variable ‘context_properties’ [-Wunused-variable]\n", | ||
| " 161 | cl_context_properties context_properties[]{\n", | ||
| " | ^~~~~~~~~~~~~~~~~~\n", | ||
| "/home/jblab/ivan_khromov/release/vortex/tests/opencl/kernel1/main.cc:163:16: warning: unused variable ‘devices’ [-Wunused-variable]\n", | ||
| " 163 | cl_device_id devices[]{device_id};\n", | ||
| " | ^~~~~~~\n", | ||
| " 0%| | 0/4 [00:23<?, ?it/s]\n" | ||
| ] | ||
| }, | ||
| { | ||
| "ename": "IndexError", | ||
| "evalue": "list index out of range", | ||
| "output_type": "error", | ||
| "traceback": [ | ||
| "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", | ||
| "\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)", | ||
| "Cell \u001b[0;32mIn[45], line 37\u001b[0m\n\u001b[1;32m 35\u001b[0m output_file \u001b[38;5;241m=\u001b[39m \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mpath_to_vortex\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m/tests/opencl/j_stat/output.txt\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m 36\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m params \u001b[38;5;129;01min\u001b[39;00m tqdm(run_p):\n\u001b[0;32m---> 37\u001b[0m data_frames\u001b[38;5;241m.\u001b[39mappend(\u001b[43mperf\u001b[49m\u001b[43m(\u001b[49m\u001b[43mparams\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43moutput_file\u001b[49m\u001b[43m)\u001b[49m)\n\u001b[1;32m 38\u001b[0m data_frame \u001b[38;5;241m=\u001b[39m pd\u001b[38;5;241m.\u001b[39mconcat(data_frames, ignore_index\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n\u001b[1;32m 40\u001b[0m \u001b[38;5;66;03m# draw graph based on the recived statistic\u001b[39;00m\n", | ||
| "Cell \u001b[0;32mIn[43], line 26\u001b[0m, in \u001b[0;36mperf\u001b[0;34m(run_params, path_to_output_file)\u001b[0m\n\u001b[1;32m 24\u001b[0m \u001b[38;5;66;03m# pars string with general perf statistic of running kernel\u001b[39;00m\n\u001b[1;32m 25\u001b[0m pairs \u001b[38;5;241m=\u001b[39m general_perf_stat\u001b[38;5;241m.\u001b[39mreplace(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mPERF: \u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m)\u001b[38;5;241m.\u001b[39msplit(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m, \u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m---> 26\u001b[0m perf_dict \u001b[38;5;241m=\u001b[39m \u001b[43m{\u001b[49m\u001b[43mkey_value\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msplit\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m:\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mfloat\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43mkey_value\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msplit\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m)\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43;01mfor\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mkey_value\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;129;43;01min\u001b[39;49;00m\u001b[43m \u001b[49m\u001b[43mpairs\u001b[49m\u001b[43m}\u001b[49m\n\u001b[1;32m 27\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m perf_dict[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcycles\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m<\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m:\n\u001b[1;32m 28\u001b[0m error_message \u001b[38;5;241m=\u001b[39m error_running(run_params, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mInvalid number of cycles\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n", | ||
| "Cell \u001b[0;32mIn[43], line 26\u001b[0m, in \u001b[0;36m<dictcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 24\u001b[0m \u001b[38;5;66;03m# pars string with general perf statistic of running kernel\u001b[39;00m\n\u001b[1;32m 25\u001b[0m pairs \u001b[38;5;241m=\u001b[39m general_perf_stat\u001b[38;5;241m.\u001b[39mreplace(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mPERF: \u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m)\u001b[38;5;241m.\u001b[39msplit(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m, \u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m---> 26\u001b[0m perf_dict \u001b[38;5;241m=\u001b[39m {key_value\u001b[38;5;241m.\u001b[39msplit(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m=\u001b[39m\u001b[38;5;124m\"\u001b[39m)[\u001b[38;5;241m0\u001b[39m]: \u001b[38;5;28mfloat\u001b[39m(\u001b[43mkey_value\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msplit\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m]\u001b[49m) \u001b[38;5;28;01mfor\u001b[39;00m key_value \u001b[38;5;129;01min\u001b[39;00m pairs}\n\u001b[1;32m 27\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m perf_dict[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcycles\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m<\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m:\n\u001b[1;32m 28\u001b[0m error_message \u001b[38;5;241m=\u001b[39m error_running(run_params, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mInvalid number of cycles\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n", | ||
| "\u001b[0;31mIndexError\u001b[0m: list index out of range" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "# create common.h files for each kernel\n", | ||
| "params1 = {\n", | ||
| " tile_size: 4\n", | ||
| "}\n", | ||
| "create_common_h(params1, \"kernel1\")\n", | ||
| "create_common_h(params1, \"kernel2\")\n", | ||
| "\n", | ||
| "params3 = {\n", | ||
| " tile_size: 4,\n", | ||
| " work_per_thread: 4\n", | ||
| "}\n", | ||
| "create_common_h(params3, \"kernel3\")\n", | ||
| "\n", | ||
| "params4 = {\n", | ||
| " tile_size: 8,\n", | ||
| " width: 4\n", | ||
| "}\n", | ||
| "create_common_h(params4, \"kernel4\")\n", | ||
| "\n", | ||
| "# fill running params data class for each kernel\n", | ||
| "run_p = []\n", | ||
| "arg = {\n", | ||
| " \"M\": 16,\n", | ||
| " \"N\": 16,\n", | ||
| " \"K\": 16\n", | ||
| "}\n", | ||
| "arch_p = arch(threads=8, cores=1, warps=2)\n", | ||
| "run_p.append(run(arch_p, kernel=\"kernel1\", driver=\"simx\", args=arg))\n", | ||
| "run_p.append(run(arch_p, kernel=\"kernel2\", driver=\"simx\", args=arg))\n", | ||
| "run_p.append(run(arch_p, kernel=\"kernel3\", driver=\"simx\", args=arg))\n", | ||
| "run_p.append(run(arch_p, kernel=\"kernel4\", driver=\"simx\", args=arg))\n", | ||
| "\n", | ||
| "# run all kernels and collect statistic in data frame\n", | ||
| "data_frames = []\n", | ||
| "output_file = f\"{path_to_vortex}/tests/opencl/j_stat/output.txt\"\n", | ||
| "for params in tqdm(run_p):\n", | ||
| " data_frames.append(perf(params, output_file))\n", | ||
| "data_frame = pd.concat(data_frames, ignore_index=True)\n", | ||
| "\n", | ||
| "# draw graph based on the recived statistic\n", | ||
| "draw(data_frame, \"kernel\", \"cycles\", \"number of cycles per kernel\", f\"graphics/graph.png\")" | ||
| ] | ||
| } | ||
| ], | ||
| "metadata": { | ||
| "kernelspec": { | ||
| "display_name": "Python 3", | ||
| "language": "python", | ||
| "name": "python3" | ||
| }, | ||
| "language_info": { | ||
| "codemirror_mode": { | ||
| "name": "ipython", | ||
| "version": 3 | ||
| }, | ||
| "file_extension": ".py", | ||
| "mimetype": "text/x-python", | ||
| "name": "python", | ||
| "nbconvert_exporter": "python", | ||
| "pygments_lexer": "ipython3", | ||
| "version": "3.11.9" | ||
| } | ||
| }, | ||
| "nbformat": 4, | ||
| "nbformat_minor": 2 | ||
| } | ||
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 +1,17 @@ | ||
| ROOT_DIR := $(realpath ../../..) | ||
| include $(ROOT_DIR)/config.mk | ||
|
|
||
| PROJECT := kernel1 | ||
|
|
||
| SRC_DIR := $(VORTEX_HOME)/tests/opencl/$(PROJECT) | ||
|
|
||
| SRCS := $(SRC_DIR)/main.cc | ||
|
|
||
| kernel.cl: $(SRC_DIR)/kernel.cl | ||
| cp $< $@ | ||
|
|
||
| KERNEL_SRCS := kernel.cl | ||
|
|
||
| OPTS ?= | ||
|
|
||
| include ../common.mk |
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 +1,6 @@ | ||
| #ifndef COMMON_H | ||
| #define COMMON_H | ||
|
|
||
| #define TS 4 | ||
|
|
||
| #endif // COMMON_H |
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 +1,18 @@ | ||
| __kernel void myGEMM1(const int M, const int N, const int K, | ||
| const __global float* A, | ||
| const __global float* B, | ||
| __global float* C) { | ||
| // Thread identifiers | ||
| const int globalRow = get_global_id(0); // Row ID of C (0..M) | ||
| const int globalCol = get_global_id(1); // Col ID of C (0..N) | ||
|
|
||
| // Compute a single element (loop over K) | ||
| float acc = 0.0f; | ||
| for (int k=0; k<K; k++) { | ||
| acc += A[k*M + globalRow] * B[globalCol*K + k]; | ||
| } | ||
|
|
||
| // Store the result | ||
| C[globalCol*M + globalRow] = acc; | ||
| } | ||
|
|
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.