use pypy 3.12 for the home page graphs

Author: mattip

codespeed/views.py

@@ -114,7 +114,9 @@ def embed_comparison(request):
 
 @require_GET
 def gethistoricaldata(request):
-    data = {'results': {}, 'benchmarks': []}
+    # 'labels' maps each key in tagged_revs (plus baseline and 'latest') to
+    # a two-line label [tag, project] for the x axis of the history chart.
+    data = {'results': {}, 'benchmarks': [], 'labels': {}}
     env = Environment.objects.all()
     if settings.DEF_ENVIRONMENT:
         env = env.get(name=settings.DEF_ENVIRONMENT)
@@ -134,6 +136,7 @@ def gethistoricaldata(request):
             )
         rev0 = rev[0]
         resname = '{} {}'.format(b['executable'], rev0.tag)
+        data['labels'][resname] = [rev0.tag, baseline_exe.project.name]
         baseline_results.append((resname, Result.objects.filter(
             executable=baseline_exe, revision=rev0, environment=env,
             benchmark__source='legacy')))
@@ -165,12 +168,14 @@ def gethistoricaldata(request):
                 logger.info("no results for '%s' '%s' '%s'" % (str(_default_exe), str(rev), str(env)))
                 continue
             default_results[rev.tag] = res
+            data['labels'][rev.tag] = [rev.tag, _default_exe.project.name]
     data['tagged_revs'] = [rev.tag for rev in all_taggedrevs if rev.tag in default_results]
     # Fetch data for latest results
     executable = settings.DEF_EXECUTABLES[0]
     def_name = executable['name']
     def_project = Project.objects.get(name=executable['project'])
     default_exe = Executable.objects.get(name=def_name, project=def_project)
+    data['labels']['latest'] = ['latest', default_exe.project.name]
     default_branch = Branch.objects.get(
             name=default_exe.project.default_branch,
             project=default_exe.project)

speed_pypy/settings.py

@@ -93,6 +93,7 @@
                  {'executable': 'cpython', 'revision': '3.11.15'},
                 ]
 DEF_EXECUTABLES = [
+                   {'name': 'pypy3.12-jit-64', 'project': 'PyPy3.12'},
                    {'name': 'pypy3.11-jit-64', 'project': 'PyPy3.11'},
                   ]
 DEF_ENVIRONMENT = 'benchmarker2'

speed_pypy/templates/embed_comparison.html

@@ -6,11 +6,13 @@
   <meta name="viewport" content="width=device-width, initial-scale=1">
   <title>PyPy vs CPython benchmark comparison</title>
   <style>
-    html, body { margin: 0; padding: 0; background: transparent; }
+    html, body { margin: 0; padding: 0; background: transparent; height: 100%; overflow: hidden; }
+    /* Fill the embedding iframe so the rotated benchmark labels are never
+       clipped: the host page controls the height via the iframe. */
     #baseline-comparison-plot {
       position: relative;
       width: 100%;
-      height: 480px;
+      height: 100vh;
       font-family: sans-serif;
     }
   </style>

speed_pypy/templates/home.html

@@ -31,7 +31,7 @@ <h2>Comparison</h2>
     <h3>How fast is {{ default_exe.project }}?</h3>
     <iframe src="{% url 'embed-comparison' %}"
             title="{{ default_exe.project }} ({{ default_exe }}) vs {{ baseline }} benchmark comparison"
-            style="border:0;width:100%;height:400px"
+            style="border:0;width:100%;height:480px"
             loading="lazy"></iframe>
     <p class="plot-caption">Plot 1: The above plot represents {{ default_exe.project }} ({{ default_exe }}) benchmark times normalized to {{ baseline }}. Smaller is better.</p>
     <p>It depends greatly on the type of task being performed. The geometric average of all benchmarks is <span id="geomean"></span> or <strong id="geofaster"></strong> times <em>faster</em> than {{ baseline }}</p>
@@ -120,9 +120,11 @@ <h3>How has PyPy performance evolved over time?</h3>
         geomeans.push(1 / trunk_geomean);
         $('#num_of_benchs').html(num_of_benchs);
 
-        var ticks2 = [data['baseline']];
-        for (var i in data['tagged_revs']) { ticks2.push(data['tagged_revs'][i]); }
-        ticks2.push('latest {{ default_exe.project }}');
+        // Two-line labels: [tag, project]; fall back to the key itself.
+        var labelFor = function(key) { return (data['labels'] || {})[key] || key; };
+        var ticks2 = [labelFor(data['baseline'])];
+        for (var i in data['tagged_revs']) { ticks2.push(labelFor(data['tagged_revs'][i])); }
+        ticks2.push(labelFor('latest'));
 
         var wrap2 = document.getElementById('historical-plot');
         var canvas2 = document.createElement('canvas');