Post binary sizes to dashboard database (#1517)

Python script and necessary changes in the azure-pipelines yaml file to post the binary size data from NuGet package build. Currently only posted from CPU pipeline. GPU and other pipelines may be added as necessary.

tools/ci_build/github/azure-pipelines/nuget/templates/cpu.yml

@@ -100,6 +100,7 @@ jobs:
   - MacOS_CI_Dev
   condition: succeeded()
   steps:  
+
   - task: DownloadPipelineArtifact@0
     displayName: 'Download Pipeline Artifact - NuGet'
     inputs:
@@ -107,6 +108,7 @@ jobs:
       targetPath: '$(Build.BinariesDirectory)/nuget-artifact'
     continueOnError: true
 
+
   - task: DownloadPipelineArtifact@0
     displayName: 'Download Pipeline Artifact - Win-x86'
     inputs:
@@ -150,4 +152,25 @@ jobs:
       artifactName: 'drop-signed-nuget'
       targetPath: '$(Build.ArtifactStagingDirectory)'
 
+
 - template: test_all_os.yml
+
+- job: Publish_NuGet_Package_And_Report
+  variables:
+  - group: Dashboard_MySQL_Secret
+  condition: ${{ parameters.DoEsrp }}
+  pool: 
+    name: Hosted Windows 2019 with VS2019
+    demands: azureps
+  dependsOn:
+  - NuGet_Test_Win
+  - NuGet_Test_Linux
+  - NuGet_Test_MacOS
+  steps:
+
+  - template: ../../templates/set-version-number-variables-step.yml
+  - template: upload-binary-sizes-from-nuget-package.yml
+    parameters:
+      gitCommitHash: $(OnnxRuntimeGitCommitHashShort)  
+
+

tools/ci_build/github/azure-pipelines/nuget/templates/upload-binary-sizes-from-nuget-package.yml

@@ -0,0 +1,48 @@
+parameters:
+  gitCommitHash: ''
+
+steps:
+- task: DownloadPipelineArtifact@0
+  displayName: 'Download Pipeline Artifact - Signed NuGet Package'
+  inputs:
+    artifactName: 'drop-signed-nuget'
+    targetPath: '$(Build.BinariesDirectory)/nuget-artifact/final-package'
+
+- task: UsePythonVersion@0
+  inputs:
+    versionSpec: '3.7' 
+    addToPath: true 
+    architecture: 'x64'
+
+- task: CmdLine@1
+  displayName: 'Install conda modules mysql-connector-python'
+  inputs:
+    filename: '%CONDA%\condabin\conda.bat'
+    arguments: 'install -q --insecure -y mysql-connector-python'
+  timeoutInMinutes: 10
+
+- task: CmdLine@2
+  displayName: 'Post binary sizes to the dashboard database using command line'
+  inputs:
+    script: |
+      echo changing directory to artifact download path
+      pushd $(Build.BinariesDirectory)\nuget-artifact\final-package
+      echo processing nupkg
+      FOR /R %%i IN (*.nupkg) do (
+        echo processing %%~ni.nupkg
+        copy %%~ni.nupkg %%~ni.zip
+        echo copied to zip
+        echo listing lib files in the zip
+        REM use a single .csv file to put the data 
+        echo os,arch,build_config,size > binary_size_data.txt
+        7z.exe l -slt %%~ni.zip runtimes\linux-x64\native\libonnxruntime.so | findstr /R /C:"^Size = [0-9]*" | for /F "tokens=3" %%a  in ('more') do if not "%%a" == "" echo linux,x64,openmp,%%a >> binary_size_data.txt
+        7z.exe l -slt %%~ni.zip runtimes\linux-x86\native\libonnxruntime.so | findstr /R /C:"^Size = [0-9]*" | for /F "tokens=3" %%a  in ('more') do if not "%%a" == "" echo linux,x86,openmp,%%a >> binary_size_data.txt
+        7z.exe l -slt %%~ni.zip runtimes\osx-x64\native\libonnxruntime.dylib | findstr /R /C:"^Size = [0-9]*" | for /F "tokens=3" %%a  in ('more') do if not "%%a" == "" echo osx,x64,openmp,%%a >> binary_size_data.txt
+        7z.exe l -slt %%~ni.zip runtimes\win-x64\native\onnxruntime.dll | findstr /R /C:"^Size = [0-9]*" | for /F "tokens=3" %%a  in ('more') do if not "%%a" == "" echo win,x64,openmp,%%a >> binary_size_data.txt
+        7z.exe l -slt %%~ni.zip runtimes\win-x86\native\onnxruntime.dll | findstr /R /C:"^Size = [0-9]*" | for /F "tokens=3" %%a  in ('more') do if not "%%a" == "" echo win,x86,openmp,%%a >> binary_size_data.txt
+        echo calling python script to post to database
+        %CONDA%\python.exe $(Build.SourcesDirectory)\tools\ci_build\github\windows\post_binary_sizes_to_dashboard.py --commit_hash=${{ parameters.gitCommitHash }} --size_data_file=binary_size_data.txt --build_project=Lotus --build_id=$(Build.BuildId) 
+      )
+
+  env: 
+    DASHBOARD_MYSQL_ORT_PASSWORD: $(dashboard-mysql-ort-password)

tools/ci_build/github/azure-pipelines/templates/set-version-number-variables-step.yml

@@ -12,6 +12,10 @@ steps:
         FOR /F "tokens=* USEBACKQ" %%F IN (`git rev-parse HEAD`) DO (
             @echo ##vso[task.setvariable variable=OnnxRuntimeGitCommitHash;]%%F
         )
+
+        FOR /F "tokens=* USEBACKQ" %%F IN (`git rev-parse --short HEAD`) DO (
+            @echo ##vso[task.setvariable variable=OnnxRuntimeGitCommitHashShort;]%%F
+        )
         
     workingDirectory: '$(Build.SourcesDirectory)'
   condition: eq(variables['Agent.OS'], 'Windows_NT') 
@@ -26,5 +30,8 @@ steps:
         _OnnxRuntimeGitCommitHash=$(git rev-parse HEAD)
         echo "##vso[task.setvariable variable=OnnxRuntimeGitCommitHash;]$_OnnxRuntimeGitCommitHash"
 
+        _OnnxRuntimeGitCommitHash=$(git rev-parse --short=8 HEAD)
+        echo "##vso[task.setvariable variable=OnnxRuntimeGitCommitHashShort;]$_OnnxRuntimeGitCommitHash"
+
     workingDirectory: '$(Build.SourcesDirectory)'
   condition: not(eq(variables['Agent.OS'], 'Windows_NT'))

tools/ci_build/github/azure-pipelines/templates/win-ci.yml

@@ -13,10 +13,12 @@ parameters:
   MsbuildArguments: '/m'
   EnvSetupScript: 'setup_env.bat'
   CudaVersion: ''
+  AgentPool: 'Win-CPU'
 
 jobs:
 - job: ${{ parameters.JobName }}
   timeoutInMinutes: 120
+  pool: ${{ parameters.AgentPool }}
   variables:
     buildDirectory: '$(Build.BinariesDirectory)'
     BuildCommand: ${{ parameters.BuildCommand }}

tools/ci_build/github/windows/post_binary_sizes_to_dashboard.py

@@ -0,0 +1,110 @@
+#!/usr/bin/env python3
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# Licensed under the MIT License.
+
+
+import argparse
+import mysql.connector
+import xml.etree.ElementTree as ET
+import sys
+import os
+
+def parse_arguments():
+    parser = argparse.ArgumentParser(description="ONNXRuntime binary size uploader for dashboard")
+    parser.add_argument("--commit_hash", help="Full Git commit hash")
+    parser.add_argument("--build_project", default='Lotus', choices=['Lotus','onnxruntime'], help="Lotus or onnxruntime build project, to construct the build URL")
+    parser.add_argument("--build_id", help="Build Id")
+    parser.add_argument("--size_data_file", help="Path to file that contains the binary size data")
+
+    return parser.parse_args()
+
+# Assumes size_data_file is a csv file with a header line, containing binary sizes and other attributes
+# CSV fields are:
+#    os,arch,build_config,size
+# No empty line or space between fields expected
+def get_binary_sizes(size_data_file):
+    binary_size = []
+    with open(size_data_file, 'r') as f:
+        line = f.readline()
+        headers = line.strip().split(',')
+        while line:
+            line = f.readline()
+            if not line:
+                break;    
+            linedata = line.strip().split(',') 
+            tablerow = {}
+            for i in range(0,len(headers)):
+                if headers[i] == 'size':
+                    tablerow[headers[i]] = int(linedata[i])
+                else:
+                    tablerow[headers[i]] = linedata[i]
+            binary_size.append(tablerow)
+    return binary_size
+
+
+def write_to_db(binary_size_data, args):
+    # connect to database
+
+    cnx = mysql.connector.connect(
+        user='ort@onnxruntimedashboard', 
+        password=os.environ.get('DASHBOARD_MYSQL_ORT_PASSWORD'), 
+        host='onnxruntimedashboard.mysql.database.azure.com', 
+        database='onnxruntime')
+
+    try:
+        cursor = cnx.cursor()
+
+        #delete old records
+        delete_query = ('DELETE FROM onnxruntime.binary_size '
+            'WHERE build_time < DATE_SUB(Now(), INTERVAL 30 DAY);'
+        )
+        
+        cursor.execute(delete_query)
+
+        #insert current records
+        for row in binary_size_data:
+            insert_query = ('INSERT INTO onnxruntime.binary_size '
+                '(build_time, build_project, build_id, commit_id, os, arch, build_config, size) '
+                'VALUES (Now(), "%s", "%s", "%s", "%s", "%s", "%s", %d) '
+                'ON DUPLICATE KEY UPDATE '
+                'build_time=Now(), build_project="%s", build_id="%s", size=%d;'
+            ) % (
+                args.build_project,
+                args.build_id,
+                args.commit_hash, 
+                row['os'],
+                row['arch'],
+                row['build_config'],
+                row['size'],
+
+                args.build_project,
+                args.build_id,
+                row['size']
+            )
+            cursor.execute(insert_query) 
+
+        cnx.commit()
+
+        # # Use below for debugging:
+        # cursor.execute('select * from onnxruntime.binary_size')
+        # for r in cursor:
+        #     print(r)
+            
+        cursor.close()
+        cnx.close()
+    except BaseException as e:
+        cnx.close()
+        raise e
+
+
+if __name__ == "__main__":
+    try:
+        args = parse_arguments()
+        binary_size_data = get_binary_sizes(args.size_data_file)
+        write_to_db(binary_size_data, args)
+    except BaseException as e:
+        print(str(e))
+        sys.exit(1)
+
+
+