framework for --enable-ZAM-profiling configuration

profiles go to zprof.log rather than stdout

cmake_templates/zeek-config.h.in

@@ -241,6 +241,9 @@
 /* Spicy analyzers built in. */
 #cmakedefine01 USE_SPICY_ANALYZERS
 
+/* Enable/disable ZAM profiling capability */
+#cmakedefine ENABLE_ZAM_PROFILE
+
 /* String with host architecture (e.g., "linux-x86_64") */
 #define HOST_ARCHITECTURE "@HOST_ARCHITECTURE@"
 

configure

@@ -65,6 +65,7 @@ Usage: $0 [OPTION]... [VAR=VALUE]...
     --enable-static-binpac build binpac statically (ignored if --with-binpac is specified)
     --enable-static-broker build Broker statically (ignored if --with-broker is specified)
     --enable-werror        build with -Werror
+    --enable-ZAM-profiling build with ZAM profiling enabled (--enable-debug implies this)
     --disable-af-packet    don't include native AF_PACKET support (Linux only)
     --disable-archiver     don't build or install zeek-archiver tool
     --disable-auxtools     don't build or install auxiliary tools
@@ -254,9 +255,11 @@ while [ $# -ne 0 ]; do
         --enable-coverage)
             append_cache_entry ENABLE_COVERAGE BOOL true
             append_cache_entry ENABLE_DEBUG BOOL true
+            append_cache_entry ENABLE_ZAM_PROFILE BOOL true
             ;;
         --enable-debug)
             append_cache_entry ENABLE_DEBUG BOOL true
+            append_cache_entry ENABLE_ZAM_PROFILE BOOL true
             ;;
         --enable-fuzzers)
             append_cache_entry ZEEK_ENABLE_FUZZERS BOOL true
@@ -280,6 +283,9 @@ while [ $# -ne 0 ]; do
         --enable-werror)
             append_cache_entry BUILD_WITH_WERROR BOOL true
             ;;
+        --enable-ZAM-profiling)
+            append_cache_entry ENABLE_ZAM_PROFILE BOOL true
+            ;;
         --disable-af-packet)
             append_cache_entry DISABLE_AF_PACKET BOOL true
             ;;

src/Options.cc

@@ -192,7 +192,7 @@ static void print_analysis_help() {
     fprintf(stderr, "    no-ZAM-opt	omit low-level ZAM optimization\n");
     fprintf(stderr, "    optimize-all	optimize all scripts, even inlined ones\n");
     fprintf(stderr, "    optimize-AST	optimize the (transformed) AST; implies xform\n");
-    fprintf(stderr, "    profile-ZAM	generate to stdout a ZAM execution profile; implies -O ZAM\n");
+    fprintf(stderr, "    profile-ZAM	generate to zprof.out a ZAM execution profile; implies -O ZAM\n");
     fprintf(stderr, "    report-recursive	report on recursive functions and exit\n");
     fprintf(stderr, "    xform	transform scripts to \"reduced\" form\n");
 
@@ -245,7 +245,7 @@ static void set_analysis_option(const char* opt, Options& opts) {
     else if ( util::streq(opt, "optimize-AST") )
         a_o.activate = a_o.optimize_AST = true;
     else if ( util::streq(opt, "profile-ZAM") )
-        a_o.activate = a_o.gen_ZAM_code = a_o.profile_ZAM = true;
+        a_o.activate = a_o.profile_ZAM = true;
     else if ( util::streq(opt, "report-C++") )
         a_o.report_CPP = true;
     else if ( util::streq(opt, "report-recursive") )

src/Var.cc

@@ -20,6 +20,7 @@
 #include "zeek/Val.h"
 #include "zeek/module_util.h"
 #include "zeek/script_opt/IDOptInfo.h"
+#include "zeek/script_opt/ScriptOpt.h"
 #include "zeek/script_opt/StmtOptInfo.h"
 #include "zeek/script_opt/UsageAnalyzer.h"
 
@@ -399,7 +400,10 @@ void add_type(ID* id, TypePtr t, std::unique_ptr<std::vector<AttrPtr>> attr) {
 
 static std::set<std::string> all_module_names;
 
-void add_module(const char* module_name) { all_module_names.emplace(module_name); }
+void add_module(const char* module_name) {
+    all_module_names.emplace(module_name);
+    switch_to_module(module_name);
+}
 
 const std::set<std::string>& module_names() { return all_module_names; }
 

src/script_opt/ScriptOpt.cc

@@ -17,6 +17,7 @@
 #include "zeek/script_opt/UsageAnalyzer.h"
 #include "zeek/script_opt/UseDefs.h"
 #include "zeek/script_opt/ZAM/Compile.h"
+#include "zeek/script_opt/ZAM/Profile.h"
 
 namespace zeek::detail {
 
@@ -310,6 +311,22 @@ static void init_options() {
             add_file_analysis_pattern(analysis_options, zo);
     }
 
+    if ( analysis_options.profile_ZAM ) {
+        auto zsamp = getenv("ZEEK_ZAM_PROF_SAMPLING_RATE");
+        if ( zsamp ) {
+            analysis_options.profile_sampling_rate = atoi(zsamp);
+            if ( analysis_options.profile_sampling_rate == 0 ) {
+                fprintf(stderr, "bad ZAM sampling profile rate from $ZEEK_ZAM_PROF_SAMPLING_RATE: %s\n", zsamp);
+                analysis_options.profile_ZAM = false;
+            }
+        }
+
+        // If no ZAM generation options have been specified, default to
+        // the usual "-O ZAM" profile. But if they have, honor those.
+        if ( ! analysis_options.gen_ZAM_code )
+            analysis_options.gen_ZAM = true;
+    }
+
     if ( analysis_options.gen_ZAM ) {
         analysis_options.gen_ZAM_code = true;
         analysis_options.inliner = true;
@@ -437,6 +454,19 @@ static void analyze_scripts_for_ZAM() {
 
     auto pfs = std::make_shared<ProfileFuncs>(funcs, nullptr, true);
 
+    if ( analysis_options.profile_ZAM ) {
+#ifdef ENABLE_ZAM_PROFILE
+        blocks = std::make_unique<BlockAnalyzer>(funcs);
+        const auto prof_filename = "zprof.out";
+        analysis_options.profile_file = fopen(prof_filename, "w");
+        if ( ! analysis_options.profile_file )
+            reporter->FatalError("cannot create ZAM profiling log %s", prof_filename);
+#else
+        fprintf(stderr, "warning: zeek was not built with --enable-ZAM-profiling\n");
+        analysis_options.profile_ZAM = false;
+#endif
+    }
+
     bool report_recursive = analysis_options.report_recursive;
     std::unique_ptr<Inliner> inl;
     if ( analysis_options.inliner )
@@ -599,11 +629,20 @@ void profile_script_execution() {
     if ( analysis_options.profile_ZAM ) {
         report_ZOP_profile();
 
+        ProfMap module_prof;
+
         for ( auto& f : funcs ) {
-            if ( f.Body()->Tag() == STMT_ZAM )
-                cast_intrusive<ZBody>(f.Body())->ProfileExecution();
+            if ( f.Body()->Tag() == STMT_ZAM ) {
+                auto zb = cast_intrusive<ZBody>(f.Body());
+                zb->ProfileExecution(module_prof);
             }
         }
+
+        for ( auto& mp : module_prof )
+            if ( mp.second.first > 0 )
+                fprintf(analysis_options.profile_file, "module %s sampled CPU time %.06f, %d sampled instructions\n",
+                        mp.first.c_str(), mp.second.second, static_cast<int>(mp.second.first));
+    }
 }
 
 void finish_script_execution() { profile_script_execution(); }

src/script_opt/ScriptOpt.h

@@ -75,6 +75,12 @@ struct AnalyOpt {
     // Produce a profile of ZAM execution.
     bool profile_ZAM = false;
 
+    // ZAM profiling sampling rate. Set via ZEEK_ZAM_PROF_SAMPLING_RATE.
+    int profile_sampling_rate = 100;
+
+    // An associated file to which to write the profile.
+    FILE* profile_file = nullptr;
+
     // If true, dump out transformed code: the results of reducing
     // interpreted scripts, and, if optimize is set, of then optimizing
     // them.
@@ -217,6 +223,10 @@ extern void analyze_global_stmts(Stmt* stmts);
 // Returns the body and scope for the previously analyzed global statements.
 extern std::pair<StmtPtr, ScopePtr> get_global_stmts();
 
+// Informs script optimization that parsing is switching to the given module.
+// Used to associate module names with profiling information.
+extern void switch_to_module(const char* module);
+
 // Add a pattern to the "only_funcs" list.
 extern void add_func_analysis_pattern(AnalyOpt& opts, const char* pat);