Merge pull request #6744 from rdmarsh2/rdmarsh2/dtt-subpath

MathiasVP · web-flow · commit 69541d3628c5 · 2021-09-24T11:58:31.000+01:00
C++: add subpaths to DefaultTaintTracking
diff --git a/cpp/ql/lib/semmle/code/cpp/ir/dataflow/DefaultTaintTracking.qll b/cpp/ql/lib/semmle/code/cpp/ir/dataflow/DefaultTaintTracking.qll
@@ -550,6 +550,39 @@ module TaintedWithPath {
     )
   }
 
+  /**
+   * Holds if there is flow from `arg` to `out` across a call that can by summarized by the flow
+   * from `par` to `ret` within it, in the graph of data flow path explanations.
+   */
+  query predicate subpaths(PathNode arg, PathNode par, PathNode ret, PathNode out) {
+    DataFlow3::PathGraph::subpaths(arg.(WrapPathNode).inner(), par.(WrapPathNode).inner(),
+      ret.(WrapPathNode).inner(), out.(WrapPathNode).inner())
+    or
+    // To avoid showing trivial-looking steps, we _replace_ the last node instead
+    // of adding an edge out of it.
+    exists(WrapPathNode sinkNode |
+      DataFlow3::PathGraph::subpaths(arg.(WrapPathNode).inner(), par.(WrapPathNode).inner(),
+        ret.(WrapPathNode).inner(), sinkNode.inner()) and
+      out.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
+    )
+    or
+    // Same for the first node
+    exists(WrapPathNode sourceNode |
+      DataFlow3::PathGraph::subpaths(sourceNode.inner(), par.(WrapPathNode).inner(),
+        ret.(WrapPathNode).inner(), out.(WrapPathNode).inner()) and
+      sourceNode.inner().getNode() = getNodeForExpr(arg.(InitialPathNode).inner())
+    )
+    or
+    // Finally, handle the case where the path goes directly from a source to a
+    // sink, meaning that they both need to be translated.
+    exists(WrapPathNode sinkNode, WrapPathNode sourceNode |
+      DataFlow3::PathGraph::subpaths(sourceNode.inner(), par.(WrapPathNode).inner(),
+        ret.(WrapPathNode).inner(), sinkNode.inner()) and
+      sourceNode.inner().getNode() = getNodeForExpr(arg.(InitialPathNode).inner()) and
+      out.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
+    )
+  }
+
   /** Holds if `n` is a node in the graph of data flow path explanations. */
   query predicate nodes(PathNode n, string key, string val) {
     key = "semmle.label" and val = n.toString()
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-022/SAMATE/TaintedPath/TaintedPath.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-022/SAMATE/TaintedPath/TaintedPath.expected
@@ -5,6 +5,7 @@ edges
 | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | fgets output argument | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:77:23:77:26 | (const char *)... |
 | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | fgets output argument | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:77:23:77:26 | data |
 | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | fgets output argument | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:77:23:77:26 | data indirection |
+subpaths
 nodes
 | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | ... + ... | semmle.label | ... + ... |
 | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | fgets output argument | semmle.label | fgets output argument |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-022/semmle/tests/TaintedPath.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-022/semmle/tests/TaintedPath.expected
@@ -5,6 +5,7 @@ edges
 | test.c:9:23:9:26 | argv | test.c:17:11:17:18 | fileName |
 | test.c:9:23:9:26 | argv | test.c:17:11:17:18 | fileName indirection |
 | test.c:9:23:9:26 | argv | test.c:17:11:17:18 | fileName indirection |
+subpaths
 nodes
 | test.c:9:23:9:26 | argv | semmle.label | argv |
 | test.c:9:23:9:26 | argv | semmle.label | argv |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-079/semmle/CgiXss/CgiXss.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-079/semmle/CgiXss/CgiXss.expected
@@ -24,6 +24,7 @@ edges
 | search.c:55:17:55:25 | raw_query indirection | search.c:14:24:14:28 | *query |
 | search.c:57:5:57:15 | raw_query | search.c:22:24:22:28 | query |
 | search.c:57:17:57:25 | raw_query indirection | search.c:22:24:22:28 | *query |
+subpaths
 nodes
 | search.c:14:24:14:28 | *query | semmle.label | *query |
 | search.c:14:24:14:28 | query | semmle.label | query |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-089/SqlTainted/SqlTainted.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-089/SqlTainted/SqlTainted.expected
@@ -13,6 +13,7 @@ edges
 | test.cpp:43:27:43:30 | argv | test.cpp:43:27:43:33 | access to array |
 | test.cpp:43:27:43:30 | argv | test.cpp:43:27:43:33 | access to array indirection |
 | test.cpp:43:27:43:30 | argv | test.cpp:43:27:43:33 | access to array indirection |
+subpaths
 nodes
 | test.c:15:20:15:23 | argv | semmle.label | argv |
 | test.c:15:20:15:23 | argv | semmle.label | argv |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-114/SAMATE/UncontrolledProcessOperation/UncontrolledProcessOperation.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-114/SAMATE/UncontrolledProcessOperation/UncontrolledProcessOperation.expected
@@ -12,6 +12,7 @@ edges
 | test.cpp:64:30:64:35 | call to getenv | test.cpp:73:24:73:27 | data indirection |
 | test.cpp:73:17:73:22 | data | test.cpp:37:73:37:76 | data |
 | test.cpp:73:24:73:27 | data indirection | test.cpp:37:73:37:76 | *data |
+subpaths
 nodes
 | test.cpp:37:73:37:76 | *data | semmle.label | *data |
 | test.cpp:37:73:37:76 | data | semmle.label | data |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-114/semmle/UncontrolledProcessOperation/UncontrolledProcessOperation.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-114/semmle/UncontrolledProcessOperation/UncontrolledProcessOperation.expected
@@ -47,6 +47,7 @@ edges
 | test.cpp:106:17:106:22 | recv output argument | test.cpp:107:15:107:20 | (const char *)... |
 | test.cpp:106:17:106:22 | recv output argument | test.cpp:107:15:107:20 | buffer |
 | test.cpp:106:17:106:22 | recv output argument | test.cpp:107:15:107:20 | buffer indirection |
+subpaths
 nodes
 | test.cpp:24:30:24:36 | *command | semmle.label | *command |
 | test.cpp:24:30:24:36 | command | semmle.label | command |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-119/SAMATE/UnboundedWrite.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-119/SAMATE/UnboundedWrite.expected
@@ -1,3 +1,4 @@
 edges
+subpaths
 nodes
 #select
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-119/semmle/tests/UnboundedWrite.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-119/semmle/tests/UnboundedWrite.expected
@@ -1,3 +1,4 @@
 edges
+subpaths
 nodes
 #select
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-120/semmle/tests/UnboundedWrite.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-120/semmle/tests/UnboundedWrite.expected
@@ -49,6 +49,7 @@ edges
 | tests.c:34:10:34:13 | argv | tests.c:34:10:34:16 | access to array |
 | tests.c:34:10:34:13 | argv | tests.c:34:10:34:16 | access to array indirection |
 | tests.c:34:10:34:13 | argv | tests.c:34:10:34:16 | access to array indirection |
+subpaths
 nodes
 | tests.c:28:22:28:25 | argv | semmle.label | argv |
 | tests.c:28:22:28:25 | argv | semmle.label | argv |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-134/SAMATE/UncontrolledFormatString.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-134/SAMATE/UncontrolledFormatString.expected
@@ -17,6 +17,7 @@ edges
 | char_environment_fprintf_01_bad.c:27:30:27:35 | call to getenv | char_environment_fprintf_01_bad.c:36:21:36:24 | data |
 | char_environment_fprintf_01_bad.c:27:30:27:35 | call to getenv | char_environment_fprintf_01_bad.c:36:21:36:24 | data indirection |
 | char_environment_fprintf_01_bad.c:27:30:27:35 | call to getenv | char_environment_fprintf_01_bad.c:36:21:36:24 | data indirection |
+subpaths
 nodes
 | char_connect_socket_w32_vsnprintf_01_bad.c:94:46:94:69 | recv output argument | semmle.label | recv output argument |
 | char_connect_socket_w32_vsnprintf_01_bad.c:94:55:94:68 | ... + ... | semmle.label | ... + ... |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/argv/argvLocal.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/argv/argvLocal.expected
@@ -247,6 +247,13 @@ edges
 | argvLocal.c:168:18:168:21 | argv | argvLocal.c:170:24:170:26 | i10 |
 | argvLocal.c:168:18:168:21 | argv | argvLocal.c:170:24:170:26 | i10 |
 | argvLocal.c:170:15:170:26 | i10 indirection | argvLocal.c:9:25:9:31 | *correct |
+subpaths
+| argvLocal.c:117:2:117:13 | i3 | argvLocal.c:9:25:9:31 | correct | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:117:15:117:16 | printWrapper output argument |
+| argvLocal.c:117:15:117:16 | i3 indirection | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:117:15:117:16 | printWrapper output argument |
+| argvLocal.c:122:2:122:13 | i4 | argvLocal.c:9:25:9:31 | correct | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:122:15:122:16 | printWrapper output argument |
+| argvLocal.c:122:15:122:16 | i4 indirection | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:122:15:122:16 | printWrapper output argument |
+| argvLocal.c:128:2:128:13 | i5 | argvLocal.c:9:25:9:31 | correct | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:128:15:128:16 | printWrapper output argument |
+| argvLocal.c:128:15:128:16 | i5 indirection | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:9:25:9:31 | *correct | argvLocal.c:128:15:128:16 | printWrapper output argument |
 nodes
 | argvLocal.c:9:25:9:31 | *correct | semmle.label | *correct |
 | argvLocal.c:9:25:9:31 | *correct | semmle.label | *correct |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/funcs/funcsLocal.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/funcs/funcsLocal.expected
@@ -51,6 +51,7 @@ edges
 | funcsLocal.c:41:18:41:20 | i61 | funcsLocal.c:42:9:42:10 | (const char *)... |
 | funcsLocal.c:41:18:41:20 | i61 | funcsLocal.c:42:9:42:10 | i6 |
 | funcsLocal.c:41:18:41:20 | i61 | funcsLocal.c:42:9:42:10 | i6 indirection |
+subpaths
 nodes
 | funcsLocal.c:16:8:16:9 | fread output argument | semmle.label | fread output argument |
 | funcsLocal.c:16:8:16:9 | i1 | semmle.label | i1 |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/globalVars/UncontrolledFormatString.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/globalVars/UncontrolledFormatString.expected
@@ -1,3 +1,4 @@
 edges
+subpaths
 nodes
 #select
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/globalVars/UncontrolledFormatStringThroughGlobalVar.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/globalVars/UncontrolledFormatStringThroughGlobalVar.expected
@@ -46,6 +46,7 @@ edges
 | globalVars.c:50:9:50:13 | copy2 | globalVars.c:50:9:50:13 | (const char *)... |
 | globalVars.c:50:9:50:13 | copy2 | globalVars.c:50:9:50:13 | copy2 |
 | globalVars.c:50:9:50:13 | copy2 | globalVars.c:50:9:50:13 | copy2 indirection |
+subpaths
 nodes
 | globalVars.c:8:7:8:10 | copy | semmle.label | copy |
 | globalVars.c:9:7:9:11 | copy2 | semmle.label | copy2 |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/ifs/ifs.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-134/semmle/ifs/ifs.expected
@@ -71,6 +71,7 @@ edges
 | ifs.c:123:8:123:11 | argv | ifs.c:124:9:124:10 | i9 |
 | ifs.c:123:8:123:11 | argv | ifs.c:124:9:124:10 | i9 indirection |
 | ifs.c:123:8:123:11 | argv | ifs.c:124:9:124:10 | i9 indirection |
+subpaths
 nodes
 | ifs.c:61:8:61:11 | argv | semmle.label | argv |
 | ifs.c:61:8:61:11 | argv | semmle.label | argv |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-190/SAMATE/ArithmeticTainted.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-190/SAMATE/ArithmeticTainted.expected
@@ -3,6 +3,7 @@ edges
 | examples.cpp:63:26:63:30 | & ... | examples.cpp:66:11:66:14 | data |
 | examples.cpp:63:26:63:30 | fscanf output argument | examples.cpp:66:11:66:14 | data |
 | examples.cpp:63:26:63:30 | fscanf output argument | examples.cpp:66:11:66:14 | data |
+subpaths
 nodes
 | examples.cpp:63:26:63:30 | & ... | semmle.label | & ... |
 | examples.cpp:63:26:63:30 | fscanf output argument | semmle.label | fscanf output argument |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-190/semmle/TaintedAllocationSize/TaintedAllocationSize.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-190/semmle/TaintedAllocationSize/TaintedAllocationSize.expected
@@ -73,6 +73,7 @@ edges
 | test.cpp:305:18:305:21 | Chi | test.cpp:308:10:308:27 | ... * ... |
 | test.cpp:305:18:305:21 | Chi | test.cpp:308:10:308:27 | ... * ... |
 | test.cpp:305:18:305:21 | get_size output argument [[]] | test.cpp:305:18:305:21 | Chi |
+subpaths
 nodes
 | test.cpp:40:21:40:24 | argv | semmle.label | argv |
 | test.cpp:40:21:40:24 | argv | semmle.label | argv |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-190/semmle/tainted/ArithmeticTainted.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-190/semmle/tainted/ArithmeticTainted.expected
@@ -23,6 +23,7 @@ edges
 | test.c:51:17:51:20 | argv | test.c:54:7:54:10 | len3 |
 | test.c:51:17:51:20 | argv | test.c:54:7:54:10 | len3 |
 | test.c:51:17:51:20 | argv | test.c:54:7:54:10 | len3 |
+subpaths
 nodes
 | test2.cpp:12:21:12:21 | v | semmle.label | v |
 | test2.cpp:14:11:14:11 | v | semmle.label | v |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-290/semmle/AuthenticationBypass/AuthenticationBypass.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-290/semmle/AuthenticationBypass/AuthenticationBypass.expected
@@ -17,6 +17,7 @@ edges
 | test.cpp:38:25:38:42 | (const char *)... | test.cpp:42:14:42:20 | address |
 | test.cpp:38:25:38:42 | (const char *)... | test.cpp:42:14:42:20 | address |
 | test.cpp:38:25:38:42 | (const char *)... | test.cpp:42:14:42:20 | address indirection |
+subpaths
 nodes
 | test.cpp:16:25:16:30 | call to getenv | semmle.label | call to getenv |
 | test.cpp:16:25:16:42 | (const char *)... | semmle.label | (const char *)... |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-311/semmle/tests/CleartextBufferWrite.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-311/semmle/tests/CleartextBufferWrite.expected
@@ -5,6 +5,7 @@ edges
 | test.cpp:54:17:54:20 | argv | test.cpp:58:25:58:29 | input |
 | test.cpp:54:17:54:20 | argv | test.cpp:58:25:58:29 | input indirection |
 | test.cpp:54:17:54:20 | argv | test.cpp:58:25:58:29 | input indirection |
+subpaths
 nodes
 | test.cpp:54:17:54:20 | argv | semmle.label | argv |
 | test.cpp:54:17:54:20 | argv | semmle.label | argv |
diff --git a/cpp/ql/test/query-tests/Security/CWE/CWE-807/semmle/TaintedCondition/TaintedCondition.expected b/cpp/ql/test/query-tests/Security/CWE/CWE-807/semmle/TaintedCondition/TaintedCondition.expected
@@ -7,6 +7,7 @@ edges
 | test.cpp:20:29:20:47 | (const char *)... | test.cpp:24:11:24:16 | call to strcmp |
 | test.cpp:20:29:20:47 | (const char *)... | test.cpp:41:10:41:38 | ! ... |
 | test.cpp:20:29:20:47 | (const char *)... | test.cpp:41:11:41:16 | call to strcmp |
+subpaths
 nodes
 | test.cpp:20:29:20:34 | call to getenv | semmle.label | call to getenv |
 | test.cpp:20:29:20:47 | (const char *)... | semmle.label | (const char *)... |

-Original file line number
+Diff line change
@@ @@ -1,3 +1,4 @@ @@
 edges
 +subpaths
 nodes
 #select