hohn/codeql
1
0
Fork 0
mirror of https://github.com/github/codeql.git synced 2025-12-19 10:23:15 +01:00
Code Issues Packages Projects Releases Wiki Activity
82,246 Commits 1,671 Branches 157 Tags
codeql-cli-2.23.0
Commit Graph

3136 Commits

Author SHA1 Message Date
Geoffrey White
2faae4dcb1 CPP: Autoformat. 2019-12-03 09:00:43 +00:00
Dave Bartolomeo
acc3d23877 Clarify comment 2019-12-02 11:53:43 -08:00
Jonas Jensen
763b18cd11 Merge remote-tracking branch 'upstream/master' into StackVariable 
Conflicts:
      change-notes/1.24/analysis-cpp.md
      cpp/ql/src/Security/CWE/CWE-131/NoSpaceForZeroTerminator.ql
 2019-11-28 17:51:20 +01:00
Dave Bartolomeo
aa6bd07971 Merge remote-tracking branch 'upstream/master' into dbartol/May-Must 2019-11-26 14:07:13 -07:00
Robert Marsh
e368d5dda0 C++: simplify getDisplayOrderInBlock 2019-11-26 16:02:30 -05:00
Jonas Jensen
c05cc77a91 Merge pull request #2421 from dbartol/dbartol/IndirectAlias 
C++/C#: Cleanup in preparation for indirect alias analysis
 2019-11-26 21:59:17 +01:00
Dave Bartolomeo
f3b4140948 C++/C#: Consistent handling of "may" vs. "must" memory accesses 
In the IR, some memory accesses are "must" accesses (the entire memory location is always read or written), and some are "may" accesses (some, all, or none of the bits in the location are written). We previously had to special case specific "may" accesses in a few places. This change regularizes our handling of "may" accesses.

The `MemoryAccessKind` enumeration now describes only the extent of the access (the set of locations potentially accessed), but does not distinguish "must" from "may". The new predicates `Operand.hasMayMemoryAccess()` and `Instruction.hasResultMayMemoryAccess()` hold when the access is a "may" access.

Unaliased SSA now correctly ignores variables that are ever accessed via a "may" access.

Aliased SSA now distinguishes `MemoryLocation`s for "may" and "must" accesses. I've refactored `getOverlap()` into the core `getExtentOverlap()`, which considers only the extent, but not the "may" vs. "must", and `getOverlap()`, which tweaks the result of `getExtentOverlap()` based on "may" vs. "must" and read-only locations.

When determining the overlap between a `Phi` operand and its definition, we now use the result of the defining `Chi` instruction, if one exists. This gives exact definitions for `Phi` operands for virtual variables.
 2019-11-26 12:13:07 -07:00
Robert Marsh
60b384a6e5 C++/C#: use line numbers for instruction IDs 
This should reduce the number of merge conflicts in the IR tests resulting
from instruction ID changes due to inserting or removing instructions
 2019-11-25 18:27:59 -05:00
Dave Bartolomeo
44c1c5a7ab C++: Update points_to.ql test to use new bit offset format 2019-11-25 11:13:02 -07:00
Dave Bartolomeo
eda47bfc51 C++: Add SSA sanity tests to IR tests 2019-11-22 16:10:51 -07:00
Dave Bartolomeo
51ff262cbc C++/C#: Add IR SSA sanity tests 2019-11-22 13:16:05 -07:00
Geoffrey White
5014432472 CPP: Tests: Add a test of NewArrayExpr.getAllocatedType() and NewArrayExpr.getExtent(). 2019-11-22 15:16:32 +00:00
Robert Marsh
05aebeff79 Merge branch 'master' into rdmarsh/cpp/ir-callee-side-effects 2019-11-21 13:45:31 -08:00
Robert Marsh
34593701b2 Merge branch 'master' into rdmarsh/cpp/ir-callee-side-effects 2019-11-20 10:03:32 -08:00
Jonas Jensen
c41114334f Merge remote-tracking branch 'upstream/master' into ir-dataflow-toString 
Solved conflicts in `*.expected` by re-running the tests.
 2019-11-19 14:27:27 +01:00
Jonas Jensen
1498499994 C++: Relax type in two tests 2019-11-19 11:31:34 +01:00
Jonas Jensen
e57f98ca64 C++: Use StackVariable in def-use libraries 
Most of the implementation was already in terms of
`SemanticStackVariable`, so not much should have changed.
 2019-11-19 11:30:59 +01:00
Jonas Jensen
95a333d28c C++: Use StackVariable in SSA libraries 
This means we'll no longer get SSA definitions for thread-local
local-scope variables.
 2019-11-19 11:30:59 +01:00
Jonas Jensen
8110039e0a C++: Bring back the StackVariable QL class 
The new `StackVariable` class actually denotes what its name suggests.
 2019-11-19 11:23:34 +01:00
Robert Marsh
180a3c9f26 C++: accept test changes 2019-11-15 11:01:18 -08:00
Ian Lynagh
4442fd8407 C++: Accept changes to syntax-zoo tests 2019-11-15 14:42:36 +00:00
Ian Lynagh
2cf714a923 C++: Follow changes in lambda locations 2019-11-15 14:42:36 +00:00
Ian Lynagh
8e00516ecf C++: Accept changes in ir test 2019-11-15 14:42:36 +00:00
Jonas Jensen
73d9cc2e7b Merge pull request #2309 from geoffw0/cpp418 
CPP: QLDoc enhancements
 2019-11-15 08:46:08 +01:00
Dave Bartolomeo
81262d5948 Merge pull request #2328 from geoffw0/routinetest 
CPP: Add a quick test of RoutineType.
 2019-11-14 15:49:13 -07:00
Robert Marsh
facbd32062 Merge branch 'master' into rdmarsh/cpp/ir-callee-side-effects 2019-11-14 11:09:13 -08:00
Robert Marsh
2fb1d4d1b1 C++: fix IR return block successors 2019-11-14 10:29:48 -08:00
Geoffrey White
7408726f41 Merge pull request #2312 from jbj/pointer-wraparound-query 
C++: New query: Pointer overflow check
 2019-11-14 16:13:04 +00:00
Geoffrey White
dc34fa366c CPP: Add a test of RoutineType. 2019-11-14 15:13:52 +00:00
Dave Bartolomeo
08620046ab Merge pull request #2068 from rdmarsh2/rdmarsh/cpp/ir-constructor-side-effects 
C++: side effect instrs for constructor qualifiers
 2019-11-13 14:56:24 -07:00
Geoffrey White
32b5501177 CPP: Add a test of ErroneousType. 2019-11-13 18:53:43 +00:00
Jonas Jensen
8d79634f8c C++: Factor out isFromMacroDefinition predicate 
This trick for excluding elements from macro bodies but not macro
arguments looks promising and should probably be used much more. With
this commit, it's now easy to use from any query.

Performance is still good because the new predicate gets appropriately
magiced.
 2019-11-13 14:07:33 +01:00
Robert Marsh
47f87c214c Merge branch 'master' into rdmarsh/cpp/ir-constructor-side-effects 2019-11-12 10:31:04 -08:00
Ziemowit Laski
faf4342d8e [CPP-418] Update references to BuiltInOperationBuiltInOffsetOf and __builtin_offsetof. 2019-11-12 17:35:50 +00:00
Jonas Jensen
eb55d964a8 C++: Fix semantic merge conflict 
This test output must have been wrong because I produced it with an
extractor that didn't have #2153 applied.
 2019-11-11 15:39:53 +01:00
Jonas Jensen
279fc16b60 C++: ConvertToBase -> ConvertToNonVirtualBase 
This rename was done with

    perl -p -i -e's/ConvertToBase/ConvertToNonVirtualBase/g' **/*.ql* **/*.expected

followed by re-running the affected tests.
 2019-11-10 10:35:53 +01:00
Robert Marsh
64b34ad975 Merge branch 'master' of github.com:Semmle/ql into rdmarsh/cpp/ir-constructor-side-effects 2019-11-08 14:06:36 -08:00
Robert Marsh
1dc0cb89d0 Merge branch 'master' of github.com:Semmle/ql into rdmarsh/cpp/ir-constructor-side-effects 2019-11-08 12:47:27 -08:00
Dave Bartolomeo
c365b2f2f0 Merge from master 
Resolve conflicts in test output
 2019-11-08 10:42:29 -07:00
Dave Bartolomeo
17f76c2516 C++: Fix merge conflicts 2019-11-07 22:02:15 -07:00
Robert Marsh
2582b69e17 Merge branch 'master' of github.com:Semmle/ql into rdmarsh/cpp/ir-constructor-side-effects 2019-11-07 15:46:08 -08:00
Robert Marsh
e93dcdb16c Merge branch 'master' into rdmarsh/cpp/ir-constructor-side-effects 2019-11-07 15:19:46 -08:00
Robert Marsh
f483ec152b Merge branch 'master' of github.com:Semmle/ql into rdmarsh/cpp/uninit-string-initializers 2019-11-07 14:36:58 -08:00
Robert Marsh
ae1377447e C++: only generate uninits when needed 2019-11-07 13:55:49 -08:00
Dave Bartolomeo
6c1d219c86 Merge from master 2019-11-07 14:50:04 -07:00
Robert Marsh
c5396d9980 Merge pull request #2262 from jbj/ir-virtual-dispatch-local 
C++: Rudimentary support for IR data flow virtual dispatch
 2019-11-07 13:09:24 -08:00
Dave Bartolomeo
f808dcefab Merge pull request #2277 from ian-semmle/cfg_diffs 
C++: Remove tests for CFG differences
 2019-11-07 12:41:40 -07:00
Dave Bartolomeo
64480c2ace Merge pull request #1999 from jbj/ir-copy-unloaded-result 
C++: Make sure there's a Instruction for each Expr
 2019-11-07 12:31:54 -07:00
Ian Lynagh
b5af4e5acd C++: Remove tests for CFG differences 
Now that we have switched over, they are no longer interesting.
 2019-11-07 16:32:18 +00:00
Matthew Gretton-Dann
ddf1ef8a7d C++: Add new test case for template member change 
We now output literals for accesses to members of template parameters:

So for `foo` in the following example:

```
template<typename T> void bar(T& t) {
  T.foo(1)
}
```
 2019-11-07 14:08:25 +00:00