Commit Graph

3950 Commits

Author SHA1 Message Date
Dave Bartolomeo
e65a5c921e C++: Add missing QLDoc 2020-06-03 13:49:14 -04:00
Dave Bartolomeo
f93c2e4e64 C++: Remove resultType from the IPA constructors for TInstruction
Making these part of the IPA object identity changes the failure mode for cases where we assign multiple result types to an instruction. Previously, we would just have one instruction with two result types, but now we'd have two instructions, which breaks things worse. This change goes back to how things were before, to avoid any new surprises on real-world code with invalid ASTs or IR.
2020-06-03 10:11:27 -04:00
Dave Bartolomeo
53d4a8e3b2 C++: Refactor IR construction interface
Now that `TInstruction` is shared between IR stages, several of the per-stage IR construction predicates can now be moved into the `Raw` interface exposed only by the initial construction of IR from the ASTs. This also removed a couple predicates that were not used previously at all.
2020-06-01 11:15:29 -04:00
Dave Bartolomeo
1e863ac40b C++: Share TInstruction across IR stages
Each stage of the IR reuses the majority of the instructions from previous stages. Previously, we've been wrapping each reused old instruction in a branch of the `TInstruction` type for the next stage. This causes use to create roughly three times as many `TInstruction` objects as we actually need.

Now that IPA union types are supported in the compiler, we can share a single `TInstruction` IPA type across stages. We create a single `TInstruction` IPA type, with individual branches of this type for instructions created directly from the AST (`TRawInstruction`) and for instructions added by each stage of SSA construction (`T*PhiInstruction`, `T*ChiInstruction`, `T*UnreachedInstruction`). Each stage then defines a `TStageInstruction` type that is a union of all of the branches that can appear in that particular stage. The public `Instruction` class for each phase extends the `TStageInstruction` type for that stage.

The interface that each stage exposes to the pyrameterized modules in the IR is now split into three pieces:
- The `Raw` module, exposed only by the original IR construction stage. This module identifies which functions have IR, which `TRawInstruction`s exist, and which `IRVariable`s exist.
- The `SSA` module, exposed only by the two SSA construction stages. This identifiers which `Phi`, `Chi`, and `Unreached` instructions exist.
- The global module, exposed by all three stages. This module has all of the predicates whose implementation is different for each stage, like gathering definitions of `MemoryOperand`s.

Similarly, there is now a single `TIRFunction` IPA type that is shared across all three stages. There is a single `IRFunctionBase` class that exposes the stage-indepdendent predicates; the `IRFunction` class for each stage extends `IRFunctionBase`.

Most of the other changes are largely mechanical.
2020-06-01 11:15:29 -04:00
Mathias Vorreiter Pedersen
cd574e8569 Merge pull request #3589 from rdmarsh2/ir-placement-new-consistency
C++: fix IR control flow for cast in placement new
2020-05-30 13:27:34 +02:00
Robert Marsh
e17adf14dc C++: autoformat 2020-05-29 16:13:40 -07:00
Jonas Jensen
91da0d5567 Merge pull request #3592 from geoffw0/strlen
CPP: Don't taint the return value of strlen
2020-05-29 19:23:47 +02:00
Robert Marsh
6c9051ae6f C++: accept consistency fixes 2020-05-29 09:49:28 -07:00
Geoffrey White
f534f09784 C++: Autoformat. 2020-05-29 14:05:08 +01:00
Geoffrey White
19c33ab41c C++: Refine StrLenFunction, including removal of taint flow. 2020-05-29 14:04:27 +01:00
Geoffrey White
705529cdf7 C++: Split StrLenFunction from PureStrFunction (without changes). 2020-05-29 14:04:27 +01:00
Geoffrey White
59cb5f9b1e C++: Remove a special case for strlen in DefaultTaintTracking. 2020-05-29 14:04:26 +01:00
Geoffrey White
408e38a4d4 C++: Clarify which taint tracking libraries should be used somewhat. 2020-05-29 14:04:26 +01:00
Geoffrey White
d77092c931 C++: Add taint tests for strlen. 2020-05-29 13:39:40 +01:00
Jonas Jensen
453de6bf4e Merge pull request #3583 from MathiasVP/qldoc-for-unix-constants
C++: QLDoc for Constants
2020-05-29 12:27:59 +02:00
Jonas Jensen
7d4d435f25 Merge remote-tracking branch 'upstream/master' into Expr-location-workaround
Conflicts:
	cpp/ql/test/library-tests/dataflow/fields/dataflow-ir-consistency.expected
2020-05-29 10:04:12 +02:00
Mathias Vorreiter Pedersen
a305d39111 Merge pull request #3577 from dbartol/github/codeql-c-analysis-team/69
C++: Fix `MemoryLocation` with multiple `VirtualVariables`
2020-05-29 09:40:58 +02:00
Mathias Vorreiter Pedersen
0467995f4f C++: Make explicit that O_CREAT and O_EXCL are Linux-specific 2020-05-29 09:36:08 +02:00
Jonas Jensen
9813258a3e Merge remote-tracking branch 'upstream/master' into Expr-location-workaround
Conflicts and semantic conflicts in `library-tests/dataflow/fields` and
`library-tests/ir/ir`.
2020-05-29 08:44:37 +02:00
Jonas Jensen
87ad519541 Merge pull request #3569 from geoffw0/strftime
C++: Taint flow consistency change for strftime
2020-05-29 08:05:25 +02:00
Robert Marsh
f82c97b84a C++: fix IR control flow for cast in placement new 2020-05-28 16:53:21 -07:00
Robert Marsh
56d0762380 C++: add test for placement new with cast 2020-05-28 16:36:26 -07:00
Robert Marsh
732da9cc4c Merge pull request #3586 from MathiasVP/qldoc-for-remaining-controlflow
C++: QLDoc for the remaining elements in the controlflow directory
2020-05-28 15:59:19 -07:00
Dave Bartolomeo
476f27e427 Merge from master 2020-05-28 17:27:08 -04:00
Dave Bartolomeo
01ef8795bf C++: Updated fixed test expectation 2020-05-28 17:24:38 -04:00
Robert Marsh
d8b5d3bce8 C++: accept test fixes 2020-05-28 08:45:01 -07:00
Robert Marsh
693789c2cc Merge branch 'master' into ir-this-parameter
Bring in new tests so their output can be fixed
2020-05-28 08:32:10 -07:00
Mathias Vorreiter Pedersen
7b2c9c5aed C++: Add quotes to improve readability. 2020-05-28 16:48:48 +02:00
Mathias Vorreiter Pedersen
5fb79cde9a C++: Sync identical files 2020-05-28 16:45:52 +02:00
Mathias Vorreiter Pedersen
0671586aac C++: QLDoc for the remaining elements in the controlflow directory 2020-05-28 16:35:46 +02:00
Geoffrey White
6fcfd0310f C++: Autoformat. 2020-05-28 15:23:48 +01:00
Geoffrey White
c9e1ccf320 Merge branch 'master' into strftime 2020-05-28 15:22:16 +01:00
Jonas Jensen
688f540843 Merge pull request #3582 from MathiasVP/qldoc-for-controlflow
C++: QLDoc for BasicBlock, ControlFlowGraph and Dataflow
2020-05-28 13:52:43 +02:00
Mathias Vorreiter Pedersen
1ef0643b60 C++: QLDoc for Constants 2020-05-28 12:24:23 +02:00
Mathias Vorreiter Pedersen
52da5755b3 C++: Respond to review comments. 2020-05-28 11:20:13 +02:00
Jonas Jensen
1b23f3ec90 C++: Accept two more changed tests 2020-05-28 11:18:14 +02:00
Mathias Vorreiter Pedersen
3d27b6bbde C++: QLDoc for BasicBlock, ControlFlowGraph and Dataflow 2020-05-28 10:10:26 +02:00
Jonas Jensen
9153f568be C++: Accept test results with location fixes 2020-05-28 09:42:49 +02:00
Jonas Jensen
6eaf64c896 Merge remote-tracking branch 'upstream/master' into Expr-location-workaround 2020-05-28 09:37:15 +02:00
Robert Marsh
54ed5d647a C++:autoformat 2020-05-27 19:30:02 -07:00
Robert Marsh
58673c449a C++: switch to TranslatedThisParameter 2020-05-27 19:29:29 -07:00
Robert Marsh
593d4c0f32 Merge pull request #3567 from MathiasVP/ir-partial-definition
Implement `asPartialDefinition` for IR dataflow nodes
2020-05-27 13:51:41 -07:00
Robert Marsh
be74616b2b C++: accept consistency test fixes 2020-05-27 12:39:54 -07:00
Dave Bartolomeo
533eeff7e8 C++: Fix MemoryLocation with multiple VirtualVariables
While investigating a bug with `TInstruction` sharing, I discovered that we had a case where alias analysis could create two `VirtualVariable`s for the same `Allocation`. For an indirect parameter allocation, we were using the type of the pointer variable as the type of the indirect allocation, instead of just `Unknown`. If the `IRType` of the pointer variable was the same type as the type of at least one access to the indirect allocation, we'd create both an `EntireAllocationVirtualVariable` and a `VariableVirtualVariable` for the allocation.

I added a new consistency test to guard against this in the future. This also turned out to be the root cause of the one existing known consistency failure in the IR tests.
2020-05-27 14:06:59 -04:00
Mathias Vorreiter Pedersen
bd97fe627c Merge branch 'master' into remove-field-conflation-from-ir-fieldflow 2020-05-27 17:08:19 +02:00
Mathias Vorreiter Pedersen
db557a45e7 Merge pull request #3570 from geoffw0/mysprintftest
C++: Fix mysprintf in taint test
2020-05-27 09:19:54 +02:00
Mathias Vorreiter Pedersen
97edd97778 C++: Add getLocation to TNode IPA type in testcase 2020-05-27 08:28:18 +02:00
Robert Marsh
b45473ec4c C++: more specific type in IndirectParameterAlloc 2020-05-26 14:07:26 -07:00
Robert Marsh
70f62538af C++: autoformat 2020-05-26 14:06:22 -07:00
Robert Marsh
fb46002332 C++: Fix ThisParameterNode after IR changes 2020-05-26 13:35:08 -07:00