i#7113 instr decode cache: move module read into raw2trace_shared

This is in preparation for further changes that include the module mapper logic in a new library (drmemtrace_instr_decode_cache) that provides an instr decode cache (instr_decode_cache_t). To avoid having to pull in the whole drmemtrace_raw2trace implementation into drmemtrace_instr_decode_cache, which will end up including it in the invariant checker and everything that depends on, we separate out the module mapper into raw2trace_shared which is intended for such cases.

Issue: #7113

clients/drcachesim/tools/opcode_mix.cpp

@@ -92,13 +92,18 @@ opcode_mix_t::initialize()
         return "";
     // Legacy trace support where binaries are needed.
     // We do not support non-module code for such traces.
-    std::string error = directory_.initialize_module_file(module_file_path_);
-    if (!error.empty())
-        return "Failed to initialize directory: " + error;
+    file_t modfile;
+    char *modfile_bytes;
+    std::string error = read_module_file_bytes(module_file_path_, modfile, modfile_bytes);
+    if (!error.empty()) {
+        return "Failed to read module file: " + error;
+    }
     module_mapper_ =
-        module_mapper_t::create(directory_.modfile_bytes_, nullptr, nullptr, nullptr,
-                                nullptr, knob_verbose_, knob_alt_module_dir_);
+        module_mapper_t::create(modfile_bytes, nullptr, nullptr, nullptr, nullptr,
+                                knob_verbose_, knob_alt_module_dir_);
     module_mapper_->get_loaded_modules();
+    delete[] modfile_bytes;
+    dr_close_file(modfile);
     error = module_mapper_->get_last_error();
     if (!error.empty())
         return "Failed to load binaries: " + error;

clients/drcachesim/tools/view.cpp

@@ -124,24 +124,22 @@ view_t::initialize_stream(memtrace_stream_t *serial_stream)
     if (module_file_path_.empty()) {
         has_modules_ = false;
     } else {
-        std::string error = directory_.initialize_module_file(module_file_path_);
+        file_t modfile;
+        char *modfile_bytes;
+        std::string error =
+            read_module_file_bytes(module_file_path_, modfile, modfile_bytes);
+        if (error.empty()) {
+            module_mapper_ =
+                module_mapper_t::create(modfile_bytes, nullptr, nullptr, nullptr, nullptr,
+                                        knob_verbose_, knob_alt_module_dir_);
+            module_mapper_->get_loaded_modules();
+            delete[] modfile_bytes;
+            dr_close_file(modfile);
+            error = module_mapper_->get_last_error();
+        }
         if (!error.empty())
             has_modules_ = false;
     }
-    if (!has_modules_) {
-        // Continue but omit disassembly to support cases where binaries are
-        // not available and OFFLINE_FILE_TYPE_ENCODINGS is not present.
-        return "";
-    }
-    // Legacy trace support where binaries are needed.
-    // We do not support non-module code for such traces.
-    module_mapper_ =
-        module_mapper_t::create(directory_.modfile_bytes_, nullptr, nullptr, nullptr,
-                                nullptr, knob_verbose_, knob_alt_module_dir_);
-    module_mapper_->get_loaded_modules();
-    std::string error = module_mapper_->get_last_error();
-    if (!error.empty())
-        return "Failed to load binaries: " + error;
     return "";
 }
 

clients/drcachesim/tracer/raw2trace_directory.cpp

@@ -1,5 +1,5 @@
 /* **********************************************************
- * Copyright (c) 2017-2023 Google, Inc.  All rights reserved.
+ * Copyright (c) 2017-2024 Google, Inc.  All rights reserved.
  * **********************************************************/
 
 /*
@@ -55,6 +55,7 @@
 #include "directory_iterator.h"
 #include "dr_api.h"              // Must be after windows.h.
 #include "raw2trace_directory.h" // Includes dr_api.h which must be after windows.h.
+#include "raw2trace_shared.h"
 #include "reader.h"
 #include "raw2trace.h"
 #include "record_file_reader.h"
@@ -435,17 +436,7 @@ raw2trace_directory_t::open_cpu_schedule_file()
 std::string
 raw2trace_directory_t::read_module_file(const std::string &modfilename)
 {
-    modfile_ = dr_open_file(modfilename.c_str(), DR_FILE_READ);
-    if (modfile_ == INVALID_FILE)
-        return "Failed to open module file " + modfilename;
-    uint64 modfile_size;
-    if (!dr_file_size(modfile_, &modfile_size))
-        return "Failed to get module file size: " + modfilename;
-    size_t modfile_size_ = (size_t)modfile_size;
-    modfile_bytes_ = new char[modfile_size_];
-    if (dr_read_file(modfile_, modfile_bytes_, modfile_size_) < (ssize_t)modfile_size_)
-        return "Didn't read whole module file " + modfilename;
-    return "";
+    return read_module_file_bytes(modfilename, modfile_, modfile_bytes_);
 }
 
 bool

clients/drcachesim/tracer/raw2trace_shared.cpp

@@ -143,5 +143,30 @@ drmemtrace_get_timestamp_from_offline_trace(const void *trace, size_t trace_size
     return DRMEMTRACE_SUCCESS;
 }
 
+std::string
+read_module_file_bytes(const std::string &modfilename, file_t &modfile,
+                       char *&modfile_bytes)
+{
+    modfile = dr_open_file(modfilename.c_str(), DR_FILE_READ);
+    if (modfile == INVALID_FILE)
+        return "Failed to open module file " + modfilename;
+    uint64 modfile_size;
+    if (!dr_file_size(modfile, &modfile_size)) {
+        dr_close_file(modfile);
+        modfile = INVALID_FILE;
+        return "Failed to get module file size: " + modfilename;
+    }
+    size_t modfile_size_ = (size_t)modfile_size;
+    modfile_bytes = new char[modfile_size_];
+    if (dr_read_file(modfile, modfile_bytes, modfile_size_) < (ssize_t)modfile_size_) {
+        dr_close_file(modfile);
+        delete[] modfile_bytes;
+        modfile = INVALID_FILE;
+        modfile_bytes = nullptr;
+        return "Didn't read whole module file " + modfilename;
+    }
+    return "";
+}
+
 } // namespace drmemtrace
 } // namespace dynamorio

clients/drcachesim/tracer/raw2trace_shared.h

@@ -154,6 +154,18 @@ class memref_counter_t : public reader_t {
     std::list<trace_entry_t> list_;
 };
 
+/**
+ * Reads the module file at the given \p modfilename path. Returns an
+ * empty string if successful, or the error string if not.
+ *
+ * If successful, the returned \p modfile must be closed using
+ * \p dr_close_file, and the returned \p modefilebytes must be freed
+ * using a delete[].
+ */
+std::string
+read_module_file_bytes(const std::string &modfilename, file_t &modfile,
+                       char *&modfile_bytes);
+
 } // namespace drmemtrace
 } // namespace dynamorio