Add build-time publish step to cpu build script

build.sh

@@ -37,6 +37,8 @@ HELP="$0 [clean] [libcudf] [cudf] [dask_cudf] [benchmarks] [tests] [libcudf_kafk
    --disable_nvtx                - disable inserting NVTX profiling ranges
    --show_depr_warn              - show cmake deprecation warnings
    --ptds                        - enable per-thread default stream
+   --build_metrics               - generate build metrics report for libcudf
+   --incl_cache_stats            - include cache statistics in build metrics report
    --cmake-args=\\\"<args>\\\"   - pass arbitrary list of CMake configuration options (escape all quotes in argument)
    -h | --h[elp]                 - print this text
 
@@ -61,6 +63,8 @@ BUILD_NVTX=ON
 BUILD_TESTS=OFF
 BUILD_DISABLE_DEPRECATION_WARNING=ON
 BUILD_PER_THREAD_DEFAULT_STREAM=OFF
+BUILD_REPORT_METRICS=ON
+BUILD_REPORT_INCL_CACHE_STATS=ON
 
 # Set defaults for vars that may not have been defined externally
 #  FIXME: if INSTALL_PREFIX is not set, check PREFIX, then check
@@ -144,6 +148,14 @@ fi
 if hasArg --ptds; then
     BUILD_PER_THREAD_DEFAULT_STREAM=ON
 fi
+if hasArg --build_metrics; then
+    BUILD_REPORT_METRICS=ON
+fi
+
+if hasArg --incl_cache_stats; then
+    BUILD_REPORT_INCL_CACHE_STATS=ON
+fi
+
 
 # If clean given, run it prior to any other steps
 if hasArg clean; then
@@ -174,8 +186,11 @@ if buildAll || hasArg libcudf; then
 
     # get the current count before the compile starts
     FILES_IN_CCACHE=""
-    if [ -x "$(command -v ccache)" ]; then
+    if [[ "$BUILD_REPORT_INCL_CACHE_STATS"=="ON" && -x "$(command -v ccache)" ]]; then
         FILES_IN_CCACHE=$(ccache -s | grep "files in cache")
+        echo "$FILES_IN_CCACHE"
+        # zero the ccache statistics
+        ccache -z
     fi
 
     cmake -S $REPODIR/cpp -B ${LIB_BUILD_DIR} \
@@ -189,6 +204,7 @@ if buildAll || hasArg libcudf; then
           -DCMAKE_BUILD_TYPE=${BUILD_TYPE} \
           ${CMAKE_ARGS}
 
+    echo "LIB_BUILD_DIR=$LIB_BUILD_DIR"
     cd ${LIB_BUILD_DIR}
 
     compile_start=$(date +%s)
@@ -197,12 +213,24 @@ if buildAll || hasArg libcudf; then
     compile_total=$(( compile_end - compile_start ))
 
     # Record build times
-    if [[ -f "${LIB_BUILD_DIR}/.ninja_log" ]]; then
-        echo "Formatting build times"
+    if [[ "$BUILD_REPORT_METRICS"=="ON" && -f "${LIB_BUILD_DIR}/.ninja_log" ]]; then
+        echo "Formatting build metrics"
         python ${REPODIR}/cpp/scripts/sort_ninja_log.py ${LIB_BUILD_DIR}/.ninja_log --fmt xml > ${LIB_BUILD_DIR}/ninja_log.xml
-        message="$FILES_IN_CCACHE <p>$PARALLEL_LEVEL parallel build time is $compile_total seconds"
-        echo "$message"
-        python ${REPODIR}/cpp/scripts/sort_ninja_log.py ${LIB_BUILD_DIR}/.ninja_log --fmt html --msg "$message" > ${LIB_BUILD_DIR}/ninja_log.html
+        MSG="<p>"
+        # get some ccache stats after the compile
+        if [[ "$BUILD_REPORT_INCL_CACHE_STATS"=="ON" && -x "$(command -v ccache)" ]]; then
+           MSG="${MSG}<br/>$FILES_IN_CCACHE"
+           HIT_RATE=$(ccache -s | grep "cache hit rate")
+           MSG="${MSG}<br/>${HIT_RATE}"
+        fi
+        MSG="${MSG}<br/>parallel setting: $PARALLEL_LEVEL"
+        MSG="${MSG}<br/>parallel build time: $compile_total seconds"
+        if [[ -f "${LIB_BUILD_DIR}/libcudf.so" ]]; then
+           LIBCUDF_FS=$(ls -lh ${LIB_BUILD_DIR}/libcudf.so | awk '{print $5}')
+           MSG="${MSG}<br/>libcudf.so size: $LIBCUDF_FS"
+        fi
+        echo "$MSG"
+        python ${REPODIR}/cpp/scripts/sort_ninja_log.py ${LIB_BUILD_DIR}/.ninja_log --fmt html --msg "$MSG" > ${LIB_BUILD_DIR}/ninja_log.html
     fi
 
     if [[ ${INSTALL_TARGET} != "" ]]; then

ci/cpu/build.sh

@@ -78,6 +78,14 @@ if [ "$BUILD_LIBCUDF" == '1' ]; then
   mkdir -p ${CONDA_BLD_DIR}/libcudf/work
   cp -r ${CONDA_BLD_DIR}/work/* ${CONDA_BLD_DIR}/libcudf/work
 
+  # Copy libcudf build metrics results
+  LIBCUDF_BUILD_DIR=$CONDA_BLD_DIR/libcudf/work/cpp/build
+  echo "Checking for build metrics log $LIBCUDF_BUILD_DIR/ninja_log.html"
+  if [[ -f "$LIBCUDF_BUILD_DIR/ninja_log.html" ]]; then
+      gpuci_logger "Copying build metrics results"
+      mkdir -p "$WORKSPACE/build-metrics"
+      cp "$LIBCUDF_BUILD_DIR/ninja_log.html" "$WORKSPACE/build-metrics/BuildMetrics.html"
+  fi
 
   gpuci_logger "Build conda pkg for libcudf_kafka"
   gpuci_conda_retry build --no-build-id --croot ${CONDA_BLD_DIR} conda/recipes/libcudf_kafka $CONDA_BUILD_ARGS

ci/gpu/build.sh

@@ -181,12 +181,11 @@ else
     done
 
     # Copy libcudf build time results
-    echo "Checking for build time log $LIB_BUILD_DIR/ninja_log.html"
-    if [[ -f "$LIB_BUILD_DIR/ninja_log.html" ]]; then
+    echo "Checking for build time log $LIB_BUILD_DIR/ninja_log.xml"
+    if [[ -f "$LIB_BUILD_DIR/ninja_log.xml" ]]; then
         gpuci_logger "Copying build time results"
+        echo "LIB_BUILD_DIR=$LIB_BUILD_DIR"
         cp "$LIB_BUILD_DIR/ninja_log.xml" "$WORKSPACE/test-results/buildtimes-junit.xml"
-        mkdir -p "$WORKSPACE/build-metrics"
-        cp "$LIB_BUILD_DIR/ninja_log.html" "$WORKSPACE/build-metrics/BuildMetrics.html"
     fi
 
     ################################################################################

cpp/scripts/sort_ninja_log.py

@@ -88,21 +88,38 @@
 elif output_fmt == "html":
     # output results in HTML format
     print("<html><head><title>Sorted Ninja Build Times</title>")
-    print("<style>", "table, th, td { border:1px solid black; }", "</style>")
+    # Note: Jenkins does not support style defined in the html
+    # https://www.jenkins.io/doc/book/security/configuring-content-security-policy/
     print("</head><body>")
     if args.msg is not None:
         print("<p>", args.msg, "</p>")
     print("<table>")
     print(
         "<tr><th>File</th>",
-        "<th align='right'>Compile time (ms)</th>",
-        "<th align='right'>Size (bytes)</th><tr>",
+        "<th>Compile time<br/>(ms)</th>",
+        "<th>Size<br/>(bytes)</th><tr>",
         sep="",
     )
+    summary = {"red": 0, "yellow": 0, "green": 0}
+    red = "bgcolor='#FFBBD0'"
+    yellow = "bgcolor='#FFFF80'"
+    green = "bgcolor='#AAFFBD'"
     for key in sl:
         result = entries[key]
+        elapsed = result[0]
+        color = green
+        if elapsed > 300000:  # 5 minutes
+            color = red
+            summary["red"] += 1
+        elif elapsed > 120000:  # 2 minutes
+            color = yellow
+            summary["yellow"] += 1
+        else:
+            summary["green"] += 1
         print(
-            "<tr><td>",
+            "<tr ",
+            color,
+            "><td>",
             key,
             "</td><td align='right'>",
             result[0],
@@ -111,6 +128,14 @@
             "</td></tr>",
             sep="",
         )
+    print("</table><br/><table border='2'>")
+    # include summary table with color legend
+    print("<tr><td", red, ">time &gt; 5 minutes</td>")
+    print("<td align='right'>", summary["red"], "</td></tr>")
+    print("<tr><td", yellow, ">2 minutes &lt; time &lt; 5 minutes</td>")
+    print("<td align='right'>", summary["yellow"], "</td></tr>")
+    print("<tr><td", green, ">time &lt; 2 minutes</td>")
+    print("<td align='right'>", summary["green"], "</td></tr>")
     print("</table></body></html>")
 
 else: