From 3de9f39348a41ca1dc6fb72b6d41918e4bd67c4d Mon Sep 17 00:00:00 2001
From: Christian Hergert <chergert@redhat.com>
Date: Mon, 5 Jun 2023 16:32:08 -0700
Subject: [PATCH] libsysprof-profile: parse address ranges with GRegex

This uses G_REGEX_OPTIMIZE as well to get the benefit of the JIT when using
PCRE2 via GLib.

This turns out to take about half (or less) of the time that it takes to
use sscanf() to parse the template/input string. 0.03 vs 0.08 seconds on
my system.

If someone wants to take this further, and just look at the input string
rather than rely on GRegex, that'd be fine by me too.
---
 .../sysprof-linux-instrument.c                | 90 ++++++++++++-------
 1 file changed, 60 insertions(+), 30 deletions(-)

diff --git a/src/libsysprof-profile/sysprof-linux-instrument.c b/src/libsysprof-profile/sysprof-linux-instrument.c
index cc4a55fb..f43b56b7 100644
--- a/src/libsysprof-profile/sysprof-linux-instrument.c
+++ b/src/libsysprof-profile/sysprof-linux-instrument.c
@@ -40,6 +40,8 @@ enum {
 
 G_DEFINE_FINAL_TYPE (SysprofLinuxInstrument, sysprof_linux_instrument, SYSPROF_TYPE_INSTRUMENT)
 
+static GRegex *address_range_regex;
+
 static char **
 sysprof_linux_instrument_list_required_policy (SysprofInstrument *instrument)
 {
@@ -68,43 +70,64 @@ add_mmaps (SysprofRecording *recording,
   line_reader_init (&reader, (char *)mapsstr, -1);
   while ((line = line_reader_next (&reader, &line_len)))
     {
-      char file[512];
-      gulong start;
-      gulong end;
-      gulong offset;
-      gulong inode;
-      gboolean is_vdso;
-      int r;
+      g_autoptr(GMatchInfo) match_info = NULL;
 
-      r = sscanf (line,
-                  "%lx-%lx %*15s %lx %*x:%*x %lu %511[^\n]",
-                  &start, &end, &offset, &inode, file);
-      file [sizeof file - 1] = '\0';
+      if (g_regex_match_full (address_range_regex, line, line_len, 0, 0, &match_info, NULL))
+        {
+          g_autofree char *file = NULL;
+          guint64 begin_addr;
+          guint64 end_addr;
+          guint64 inode;
+          guint64 offset;
+          gboolean is_vdso;
+          int begin_addr_begin;
+          int begin_addr_end;
+          int end_addr_begin;
+          int end_addr_end;
+          int offset_begin;
+          int offset_end;
+          int inode_begin;
+          int inode_end;
+          int path_begin;
+          int path_end;
 
-      /* file has a " (deleted)" suffix if it was deleted from disk */
-      if (g_str_has_suffix (file, " (deleted)"))
-          file [strlen (file) - strlen (" (deleted)")] = '\0';
+          if (!g_match_info_fetch_pos (match_info, 1, &begin_addr_begin, &begin_addr_end) ||
+              !g_match_info_fetch_pos (match_info, 2, &end_addr_begin, &end_addr_end) ||
+              !g_match_info_fetch_pos (match_info, 3, &offset_begin, &offset_end) ||
+              !g_match_info_fetch_pos (match_info, 4, &inode_begin, &inode_end) ||
+              !g_match_info_fetch_pos (match_info, 5, &path_begin, &path_end))
+            continue;
 
-      if (r != 5)
-        continue;
+          begin_addr = g_ascii_strtoull (&line[begin_addr_begin], NULL, 16);
+          end_addr = g_ascii_strtoull (&line[end_addr_begin], NULL, 16);
+          offset = g_ascii_strtoull (&line[offset_begin], NULL, 16);
+          inode = g_ascii_strtoull (&line[inode_begin], NULL, 10);
 
-      is_vdso = strcmp ("[vdso]", file) == 0;
+          if (memcmp (" (deleted",
+                      &line[path_end] - strlen (" (deleted"),
+                      strlen (" (deleted")) == 0)
+            path_end -= strlen (" (deleted)");
 
-      if (ignore_inode || is_vdso)
-        inode = 0;
+          file = g_strndup (&line[path_begin], path_end-path_begin);
 
-      if (is_vdso)
-        offset = 0;
+          is_vdso = strcmp ("[vdso]", file) == 0;
 
-      sysprof_capture_writer_add_map (writer,
-                                      SYSPROF_CAPTURE_CURRENT_TIME,
-                                      -1,
-                                      pid,
-                                      start,
-                                      end,
-                                      offset,
-                                      inode,
-                                      file);
+          if (ignore_inode || is_vdso)
+            inode = 0;
+
+          if (is_vdso)
+            offset = 0;
+
+          sysprof_capture_writer_add_map (writer,
+                                          SYSPROF_CAPTURE_CURRENT_TIME,
+                                          -1,
+                                          pid,
+                                          begin_addr,
+                                          end_addr,
+                                          offset,
+                                          inode,
+                                          file);
+        }
     }
 }
 
@@ -379,6 +402,7 @@ sysprof_linux_instrument_class_init (SysprofLinuxInstrumentClass *klass)
 {
   GObjectClass *object_class = G_OBJECT_CLASS (klass);
   SysprofInstrumentClass *instrument_class = SYSPROF_INSTRUMENT_CLASS (klass);
+  g_autoptr(GError) error = NULL;
 
   object_class->finalize = sysprof_linux_instrument_finalize;
 
@@ -386,6 +410,12 @@ sysprof_linux_instrument_class_init (SysprofLinuxInstrumentClass *klass)
   instrument_class->prepare = sysprof_linux_instrument_prepare;
   instrument_class->record = sysprof_linux_instrument_record;
   instrument_class->process_started = sysprof_linux_instrument_process_started;
+
+  address_range_regex = g_regex_new ("^([0-9a-f]+)-([0-9a-f]+) [r\\-][w\\-][x\\-][ps\\-] [0-9a-f]+ [0-9]{2}:[0-9]{2} ([0-9]+) +(.*)$",
+                                     G_REGEX_OPTIMIZE,
+                                     G_REGEX_MATCH_DEFAULT,
+                                     &error);
+  g_assert_no_error (error);
 }
 
 static void