gcc/analyzer/sm-sensitive.cc

   1 /* An experimental state machine, for tracking exposure of sensitive
   2    data (e.g. through logging).
   3    Copyright (C) 2019-2021 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   bool emit (rich_location *rich_loc) FINAL OVERRIDE
  95   {
  96     diagnostic_metadata m;
  97     /* CWE-532: Information Exposure Through Log Files */
  98     m.add_cwe (532);
  99     return warning_meta (rich_loc, m,
 100                          OPT_Wanalyzer_exposure_through_output_file,
 101                          "sensitive value %qE written to output file",
 102                          m_arg);
 103   }
 104
 105   label_text describe_state_change (const evdesc::state_change &change)
 106     FINAL OVERRIDE
 107   {
 108     if (change.m_new_state == m_sm.m_sensitive)
 109       {
 110         m_first_sensitive_event = change.m_event_id;
 111         return change.formatted_print ("sensitive value acquired here");
 112       }
 113     return label_text ();
 114   }
 115
 116   label_text describe_call_with_state (const evdesc::call_with_state &info)
 117     FINAL OVERRIDE
 118   {
 119     if (info.m_state == m_sm.m_sensitive)
 120       return info.formatted_print
 121         ("passing sensitive value %qE in call to %qE from %qE",
 122          info.m_expr, info.m_callee_fndecl, info.m_caller_fndecl);
 123     return label_text ();
 124   }
 125
 126   label_text describe_return_of_state (const evdesc::return_of_state &info)
 127     FINAL OVERRIDE
 128   {
 129     if (info.m_state == m_sm.m_sensitive)
 130       return info.formatted_print ("returning sensitive value to %qE from %qE",
 131                                    info.m_caller_fndecl, info.m_callee_fndecl);
 132     return label_text ();
 133   }
 134
 135   label_text describe_final_event (const evdesc::final_event &ev) FINAL OVERRIDE
 136   {
 137     if (m_first_sensitive_event.known_p ())
 138       return ev.formatted_print ("sensitive value %qE written to output file"
 139                                  "; acquired at %@",
 140                                  m_arg, &m_first_sensitive_event);
 141     else
 142       return ev.formatted_print ("sensitive value %qE written to output file",
 143                                  m_arg);
 144   }
 145
 146 private:
 147   const sensitive_state_machine &m_sm;
 148   tree m_arg;
 149   diagnostic_event_id_t m_first_sensitive_event;
 150 };
 151
 152 /* sensitive_state_machine's ctor.  */
 153
 154 sensitive_state_machine::sensitive_state_machine (logger *logger)
 155 : state_machine ("sensitive", logger)
 156 {
 157   m_sensitive = add_state ("sensitive");
 158   m_stop = add_state ("stop");
 159 }
 160
 161 /* Warn about an exposure at NODE and STMT if ARG is in the "sensitive"
 162    state.  */
 163
 164 void
 165 sensitive_state_machine::warn_for_any_exposure (sm_context *sm_ctxt,
 166                                                 const supernode *node,
 167                                                 const gimple *stmt,
 168                                                 tree arg) const
 169 {
 170   if (sm_ctxt->get_state (stmt, arg) == m_sensitive)
 171     {
 172       tree diag_arg = sm_ctxt->get_diagnostic_tree (arg);
 173       sm_ctxt->warn (node, stmt, arg,
 174                      new exposure_through_output_file (*this, diag_arg));
 175     }
 176 }
 177
 178 /* Implementation of state_machine::on_stmt vfunc for
 179    sensitive_state_machine.  */
 180
 181 bool
 182 sensitive_state_machine::on_stmt (sm_context *sm_ctxt,
 183                                   const supernode *node,
 184                                   const gimple *stmt) const
 185 {
 186   if (const gcall *call = dyn_cast <const gcall *> (stmt))
 187     if (tree callee_fndecl = sm_ctxt->get_fndecl_for_call (call))
 188       {
 189         if (is_named_call_p (callee_fndecl, "getpass", call, 1))
 190           {
 191             tree lhs = gimple_call_lhs (call);
 192             if (lhs)
 193               sm_ctxt->on_transition (node, stmt, lhs, m_start, m_sensitive);
 194             return true;
 195           }
 196         else if (is_named_call_p (callee_fndecl, "fprintf")
 197                  || is_named_call_p (callee_fndecl, "printf"))
 198           {
 199             /* Handle a match at any position in varargs.  */
 200             for (unsigned idx = 1; idx < gimple_call_num_args (call); idx++)
 201               {
 202                 tree arg = gimple_call_arg (call, idx);
 203                 warn_for_any_exposure (sm_ctxt, node, stmt, arg);
 204               }
 205             return true;
 206           }
 207         else if (is_named_call_p (callee_fndecl, "fwrite", call, 4))
 208           {
 209             tree arg = gimple_call_arg (call, 0);
 210             warn_for_any_exposure (sm_ctxt, node, stmt, arg);
 211             return true;
 212           }
 213         // TODO: ...etc.  This is just a proof-of-concept at this point.
 214       }
 215   return false;
 216 }
 217
 218 bool
 219 sensitive_state_machine::can_purge_p (state_t s ATTRIBUTE_UNUSED) const
 220 {
 221   return true;
 222 }
 223
 224 } // anonymous namespace
 225
 226 /* Internal interface to this file. */
 227
 228 state_machine *
 229 make_sensitive_state_machine (logger *logger)
 230 {
 231   return new sensitive_state_machine (logger);
 232 }
 233
 234 } // namespace ana
 235
 236 #endif /* #if ENABLE_ANALYZER */