gcc/analyzer/sm-sensitive.cc

   1 /* An experimental state machine, for tracking exposure of sensitive
   2    data (e.g. through logging).
   3    Copyright (C) 2019-2022 Free Software Foundation, Inc.
   4    Contributed by David Malcolm <dmalcolm@redhat.com>.
   5
   6 This file is part of GCC.
   7
   8 GCC is free software; you can redistribute it and/or modify it
   9 under the terms of the GNU General Public License as published by
  10 the Free Software Foundation; either version 3, or (at your option)
  11 any later version.
  12
  13 GCC is distributed in the hope that it will be useful, but
  14 WITHOUT ANY WARRANTY; without even the implied warranty of
  15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  16 General Public License for more details.
  17
  18 You should have received a copy of the GNU General Public License
  19 along with GCC; see the file COPYING3.  If not see
  20 <http://www.gnu.org/licenses/>.  */
  21
  22 #include "config.h"
  23 #include "system.h"
  24 #include "coretypes.h"
  25 #include "tree.h"
  26 #include "function.h"
  27 #include "function.h"
  28 #include "basic-block.h"
  29 #include "gimple.h"
  30 #include "options.h"
  31 #include "diagnostic-path.h"
  32 #include "diagnostic-metadata.h"
  33 #include "function.h"
  34 #include "json.h"
  35 #include "analyzer/analyzer.h"
  36 #include "diagnostic-event-id.h"
  37 #include "analyzer/analyzer-logging.h"
  38 #include "analyzer/sm.h"
  39 #include "analyzer/pending-diagnostic.h"
  40
  41 #if ENABLE_ANALYZER
  42
  43 namespace ana {
  44
  45 namespace {
  46
  47 /* An experimental state machine, for tracking exposure of sensitive
  48    data (e.g. through logging).  */
  49
  50 class sensitive_state_machine : public state_machine
  51 {
  52 public:
  53   sensitive_state_machine (logger *logger);
  54
  55   bool inherited_state_p () const FINAL OVERRIDE { return true; }
  56
  57   bool on_stmt (sm_context *sm_ctxt,
  58                 const supernode *node,
  59                 const gimple *stmt) const FINAL OVERRIDE;
  60
  61   bool can_purge_p (state_t s) const FINAL OVERRIDE;
  62
  63   /* State for "sensitive" data, such as a password.  */
  64   state_t m_sensitive;
  65
  66   /* Stop state, for a value we don't want to track any more.  */
  67   state_t m_stop;
  68
  69 private:
  70   void warn_for_any_exposure (sm_context *sm_ctxt,
  71                               const supernode *node,
  72                               const gimple *stmt,
  73                               tree arg) const;
  74 };
  75
  76 class exposure_through_output_file
  77   : public pending_diagnostic_subclass<exposure_through_output_file>
  78 {
  79 public:
  80   exposure_through_output_file (const sensitive_state_machine &sm, tree arg)
  81   : m_sm (sm), m_arg (arg)
  82   {}
  83
  84   const char *get_kind () const FINAL OVERRIDE
  85   {
  86     return "exposure_through_output_file";
  87   }
  88
  89   bool operator== (const exposure_through_output_file &other) const
  90   {
  91     return same_tree_p (m_arg, other.m_arg);
  92   }
  93
  94   int get_controlling_option () const FINAL OVERRIDE
  95   {
  96     return OPT_Wanalyzer_exposure_through_output_file;
  97   }
  98
  99   bool emit (rich_location *rich_loc) FINAL OVERRIDE
 100   {
 101     diagnostic_metadata m;
 102     /* CWE-532: Information Exposure Through Log Files */
 103     m.add_cwe (532);
 104     return warning_meta (rich_loc, m, get_controlling_option (),
 105                          "sensitive value %qE written to output file",
 106                          m_arg);
 107   }
 108
 109   label_text describe_state_change (const evdesc::state_change &change)
 110     FINAL OVERRIDE
 111   {
 112     if (change.m_new_state == m_sm.m_sensitive)
 113       {
 114         m_first_sensitive_event = change.m_event_id;
 115         return change.formatted_print ("sensitive value acquired here");
 116       }
 117     return label_text ();
 118   }
 119
 120   label_text describe_call_with_state (const evdesc::call_with_state &info)
 121     FINAL OVERRIDE
 122   {
 123     if (info.m_state == m_sm.m_sensitive)
 124       return info.formatted_print
 125         ("passing sensitive value %qE in call to %qE from %qE",
 126          info.m_expr, info.m_callee_fndecl, info.m_caller_fndecl);
 127     return label_text ();
 128   }
 129
 130   label_text describe_return_of_state (const evdesc::return_of_state &info)
 131     FINAL OVERRIDE
 132   {
 133     if (info.m_state == m_sm.m_sensitive)
 134       return info.formatted_print ("returning sensitive value to %qE from %qE",
 135                                    info.m_caller_fndecl, info.m_callee_fndecl);
 136     return label_text ();
 137   }
 138
 139   label_text describe_final_event (const evdesc::final_event &ev) FINAL OVERRIDE
 140   {
 141     if (m_first_sensitive_event.known_p ())
 142       return ev.formatted_print ("sensitive value %qE written to output file"
 143                                  "; acquired at %@",
 144                                  m_arg, &m_first_sensitive_event);
 145     else
 146       return ev.formatted_print ("sensitive value %qE written to output file",
 147                                  m_arg);
 148   }
 149
 150 private:
 151   const sensitive_state_machine &m_sm;
 152   tree m_arg;
 153   diagnostic_event_id_t m_first_sensitive_event;
 154 };
 155
 156 /* sensitive_state_machine's ctor.  */
 157
 158 sensitive_state_machine::sensitive_state_machine (logger *logger)
 159 : state_machine ("sensitive", logger)
 160 {
 161   m_sensitive = add_state ("sensitive");
 162   m_stop = add_state ("stop");
 163 }
 164
 165 /* Warn about an exposure at NODE and STMT if ARG is in the "sensitive"
 166    state.  */
 167
 168 void
 169 sensitive_state_machine::warn_for_any_exposure (sm_context *sm_ctxt,
 170                                                 const supernode *node,
 171                                                 const gimple *stmt,
 172                                                 tree arg) const
 173 {
 174   if (sm_ctxt->get_state (stmt, arg) == m_sensitive)
 175     {
 176       tree diag_arg = sm_ctxt->get_diagnostic_tree (arg);
 177       sm_ctxt->warn (node, stmt, arg,
 178                      new exposure_through_output_file (*this, diag_arg));
 179     }
 180 }
 181
 182 /* Implementation of state_machine::on_stmt vfunc for
 183    sensitive_state_machine.  */
 184
 185 bool
 186 sensitive_state_machine::on_stmt (sm_context *sm_ctxt,
 187                                   const supernode *node,
 188                                   const gimple *stmt) const
 189 {
 190   if (const gcall *call = dyn_cast <const gcall *> (stmt))
 191     if (tree callee_fndecl = sm_ctxt->get_fndecl_for_call (call))
 192       {
 193         if (is_named_call_p (callee_fndecl, "getpass", call, 1))
 194           {
 195             tree lhs = gimple_call_lhs (call);
 196             if (lhs)
 197               sm_ctxt->on_transition (node, stmt, lhs, m_start, m_sensitive);
 198             return true;
 199           }
 200         else if (is_named_call_p (callee_fndecl, "fprintf")
 201                  || is_named_call_p (callee_fndecl, "printf"))
 202           {
 203             /* Handle a match at any position in varargs.  */
 204             for (unsigned idx = 1; idx < gimple_call_num_args (call); idx++)
 205               {
 206                 tree arg = gimple_call_arg (call, idx);
 207                 warn_for_any_exposure (sm_ctxt, node, stmt, arg);
 208               }
 209             return true;
 210           }
 211         else if (is_named_call_p (callee_fndecl, "fwrite", call, 4))
 212           {
 213             tree arg = gimple_call_arg (call, 0);
 214             warn_for_any_exposure (sm_ctxt, node, stmt, arg);
 215             return true;
 216           }
 217         // TODO: ...etc.  This is just a proof-of-concept at this point.
 218       }
 219   return false;
 220 }
 221
 222 bool
 223 sensitive_state_machine::can_purge_p (state_t s ATTRIBUTE_UNUSED) const
 224 {
 225   return true;
 226 }
 227
 228 } // anonymous namespace
 229
 230 /* Internal interface to this file. */
 231
 232 state_machine *
 233 make_sensitive_state_machine (logger *logger)
 234 {
 235   return new sensitive_state_machine (logger);
 236 }
 237
 238 } // namespace ana
 239
 240 #endif /* #if ENABLE_ANALYZER */