Summary:
The term representing an exp should not rely on more info than is
carried by the exp.
Reviewed By: jvillard
Differential Revision: D21720989
fbshipit-source-id: b65bf3678
Summary:
It is now possible to not spew dune files all over the repo, and opam
files aren't needed either.
Reviewed By: jvillard
Differential Revision: D21720978
fbshipit-source-id: 553e1d154
Summary:
```
val ( let@ ) : ('a -> 'b) -> 'a -> 'b
(** [let@ x = e in b] is equivalent to [e @@ fun x -> b], that is,
[e (fun x -> b)] *)
```
Reviewed By: jvillard
Differential Revision: D21721025
fbshipit-source-id: d8efdebbe
Summary: Cost analysis has an additional mode that checks whether the function occurs on a UI (main) thread. If so, it warns the user. This check is only supported for Java and C++ but not for ObjC, so the diff suppresses this check for ObjC, and set ```is_on_ui_thread``` to ```false``` by default.
Reviewed By: ezgicicek
Differential Revision: D21952470
fbshipit-source-id: 838dd5639
Summary:
Due to:
1. Additional dependency on kotlin-annotations,
2. nullsafe annotation being annotated as TypeQualifierDefault and
UnderMigration(status = STRICT) [which can have breaking effect on
Kotlin code],
let's bump the minor version of the artifact.
Pull Request resolved: https://github.com/facebook/infer/pull/1281
Reviewed By: jvillard
Differential Revision: D21952759
Pulled By: artempyanykh
fbshipit-source-id: 58aea14c3
Summary:
There was an issue on `make build_ant_test` that newly generated `issues.exp.test` is empty. When
infer runs with `-- ant`, if there is already built objects in `ant_out`, infer analyzes nothing,
which result in the empty `issues.exp.test`. For example,
```
~/infer/infer/tests/build_systems/ant$ make test
[11:26:49][36924] Testing ant integration...
[ 2s][36924] SUCCESS Testing ant integration
~/infer/infer/tests/build_systems/ant$ touch `which infer`
```
The makefile tries to reanalyze ant targets because `infer` binary has been changed. However, there
is nothing to build/analyze, since `ant_out` still exists as the same.
```
~/infer/infer/tests/build_systems/ant$ make test
[11:26:57][37078] Testing ant integration...
[ 1s][37078] SUCCESS Testing ant integration
diff --git a/Users/scho/infer/infer/tests/build_systems/ant/issues.exp b/Users/scho/infer/infer/tests/build_systems/ant/issues.exp.test
index 376146b5f..e69de29bb 100644
--- a/Users/scho/infer/infer/tests/build_systems/ant/issues.exp
+++ b/Users/scho/infer/infer/tests/build_systems/ant/issues.exp.test
@@ -1 +0,0 @@
[-build_systems/ant/src/Hello.java, Hello.test():int, 2, NULL_DEREFERENCE, no_bucket, ERROR, [start of procedure test()]-]
Test output (build_systems/ant/issues.exp.test) differs from expected test output build_systems/ant/issues.exp
Run the following command to replace the expected test output with the new output:
make -C build_systems/ant replace
make: *** [test] Error 1
```
To handle this issue, this cleans `ant_out` before running infer.
Reviewed By: jvillard
Differential Revision: D21950916
fbshipit-source-id: f57056f6e
Summary:
These files have started appearing in `git status` on my laptop (emacs
26.3).
Reviewed By: ezgicicek
Differential Revision: D21928308
fbshipit-source-id: f63a21493
Summary:
Before: to report an issue type, wrap it in an exception type then pass
that around and translate it back to an issue type at some point.
After: no. Except biabduction, which was built like this and uses
raising of exceptions to stop the analysis too.
Reviewed By: dulmarod
Differential Revision: D21904589
fbshipit-source-id: cb8446f38
Summary:
See previous diff: issues are always reported with the same severity so
recognise that and just use their default severity in "modern" checkers.
Reviewed By: ngorogiannis
Differential Revision: D21904591
fbshipit-source-id: fb5387e35
Summary:
By and large issues of a given type (eg NULL_DEREFERENCE) are always
reported with the same severity (eg ERROR), but that severity is hard to
tease out from the code. This makes it more explicit by favouring the
default in `IssueType.t` in many cases without changing infer's
behaviour. The checkers typically use `Reporting.log_{error,warning}`;
these are taken care of in the next diff.
Reviewed By: skcho
Differential Revision: D21904590
fbshipit-source-id: 47c76cd4c
Summary:
- "visibility" (whether an issue to report is something to show the user
or something that is only used for debugging or for detecting other
issues) is an intrinsic property of an issue type and thus belongs in
`IssueType.t`.
- "severity" (warning/error/...) is something that each issue should
have a default for ("a memory leak is by default an ERROR", a
"condition always true is by default a warning"), but can also be
overriden at run-time. Right now only nullsafe uses that capability:
when in "strict mode", some warnings become errors. We can imagine
extending this to other issue types, or even providing config flags to
turn warnings into errors, like compilers often have.
To guess the default severity (since it's dynamic it can be hard to know
for sure!), I tried to find places where it was reported as the source
of truth, but also later diffs test these defaults against our tests (by
removing most of the dynamic changes in severity).
With this diff there are 3 places where severity is set:
1. The default severity in IssueType.t: this is unused for now.
2. The severity from `Exceptions.recognize_exception`: this is
semi-statically determined and, if set, takes precedence over number 3 (which looks wrong to me!)
3. The severity passed to `Errlog.log_issue` when we actually add an
issue to the error log: this is used only when (2) is unset.
The next diffs will make 1 the default, delete 2, and make 3 optional
but override 1 when passed.
Reviewed By: skcho
Differential Revision: D21904538
fbshipit-source-id: a674b42d8
Summary:
The Java frontend translates java field accesses on an object reference as field (`.`) followed by dereference (`*`) exactly like the arrow operator in C. However, when the field is static, it generates just a field access `.`. This is a bug.
This diff mitigates its effects for printing locks in starvation.
Reviewed By: skcho
Differential Revision: D21882875
fbshipit-source-id: 79846d826
Summary:
Rather than compute the size of the llair type of the llair
initializer expression, compute the size of the llvm initializer
directly.
Reviewed By: ngorogiannis
Differential Revision: D21720982
fbshipit-source-id: 4364baf38
Summary:
Rather than translate an llvm type to a llair one and then compute the
size of the llair type, obtain the size of the llvm type directly.
Reviewed By: ngorogiannis
Differential Revision: D21720968
fbshipit-source-id: ad98112a7
Summary:
This diff implements part of the memory management for Objective-C classes in ARC, namely that `dealloc` is called when the objects become unreachable. In reality the semantics of ARC says that this happens when their reference count becomes 0, but we are not modelling this yet in Pulse. However, we could in the future.
This fixes false positives memory leaks when the memory is freed in dealloc.
`dealloc` is often implicit in Objective-C, it also calls the dealloc of instance variables and superclass. None of this is implemented yet, and will be done in a future diff. This will be added in the frontend probably, similarly to how it's done for C++ destructors.
This is an important part of modelling Objective-C semantics in Infer, I looked at whether this should be a preanalysis to be used by all analyses but this needs Pulse. So the idea is that any analysis that needs to understand Objective-C memory model well, should have Pulse as a preanalysis.
Reviewed By: jvillard
Differential Revision: D21762292
fbshipit-source-id: ced014324
Summary:
Adding a new attribute for dynamic type. It is set in the models of constructors, currently only in `alloc` in Objective-C. We use it in the following diff to figure out which `dealloc` method to call. However it could be useful for other things, such as dynamic dispatch.
#skipdeadcode
Reviewed By: jvillard
Differential Revision: D21739928
fbshipit-source-id: 9276c0a4d
Summary:
Since Java8, interfaces mays contain implementations
(default methods). We modify the resolve algorith in the Java frontend
to take care of that.
Reviewed By: jvillard
Differential Revision: D21785182
fbshipit-source-id: ffab8124c
Summary:
Closes https://github.com/facebook/infer/issues/1270
Attempt to switch from Travis to Github Actions.
Please see [CONTRIBUTING.md](./CONTRIBUTING.md) for how to set up your development environment and run tests.
Pull Request resolved: https://github.com/facebook/infer/pull/1271
Reviewed By: mityal
Differential Revision: D21881443
Pulled By: jvillard
fbshipit-source-id: f711bbf27
Summary:
We stopped relying on an arbitrary threshold. Hence,
- we don't need all the machinery for reporting at a specific node for a threshold
- we can remove
- the issue type `EXPENSIVE_EXECUTION_TIME`
- the config option `--use-cost-threshold`.
Reviewed By: skcho
Differential Revision: D21859815
fbshipit-source-id: b73a2372d
Summary: We do not use an arbitrary threshold to test cost results anymore but instead rely on `cost-issues` which do not have any trace attached. This diff adds traces to `costs-report.json` so that we can test cost issues with traces.
Reviewed By: skcho
Differential Revision: D21858846
fbshipit-source-id: e73321a92
Summary:
Now that we have a way to write cost issues, let's not rely on some arbitrary threshold (and also get rid of `EXPENSIVE_EXECUTION_TIME` issues in tests).
One consequence of this is that we will loose the cost traces in tests since `costs-report.json` doesn't have any traces. Next diff fixes that.
Reviewed By: skcho
Differential Revision: D21837574
fbshipit-source-id: 86b4d028d
Summary:
The model returns an array the length of which is the same to that of enum entries.
It takes the length of enum entries from the summary of `Enum.values` because it is not written in `tenv`. In order to do that, the model semantics should be able to request the summary of the function with `get_summary`, so I extended `model_env` to include the functionality.
Reviewed By: ezgicicek
Differential Revision: D21843319
fbshipit-source-id: d6f10eb91
Summary:
The model returns an array the length of which is the number of known
fields in `tenv`.
Reviewed By: ezgicicek
Differential Revision: D21840375
fbshipit-source-id: 891517c6e
Summary: D21816312 forgot to add the new cost testing mechanism to `fb-performance` and `performance-exlusive` directories. This diff fixes that.
Reviewed By: skcho
Differential Revision: D21837912
fbshipit-source-id: 407dafcd3
Summary: We were missing `performance-exclusive` and `fb-performance` tests in the list of cost analysis targets.
Reviewed By: jvillard
Differential Revision: D21837972
fbshipit-source-id: 550d27918
Summary: The models were too naive before since they invalidated the underlying array completely (copying C++'s push_back model), causing spurious vector invalidation issues in Java. This diff adds more reasonable models.
Reviewed By: skcho
Differential Revision: D21787543
fbshipit-source-id: a5a59ff69
Summary:
In order to test cost analysis results, currently we rely on having an arbitrary cost threshold (200) and report issues that exceed this cost. For instance, a cost of 201 is considered expensive and reported as `EXPENSIVE_EXECUTION_TIME` issue in cost tests.
This means, if we change the cost analysis in a slight way that results in some constant cost increase under 200, we wouldn't able to detect it. I find this unsatisfactory and somewhat hacky.
This diff adds the ability to write the result of `costs-report.json` into a separate `cost-issues.exp` and then compare the actual costs (not only than relying on this arbitrary threshold reporting mechanism).
Reviewed By: skcho
Differential Revision: D21816312
fbshipit-source-id: 93b531928
Josh Berdine
Qianyi Shu
Artem Pyanykh
Ezgi Çiçek
Sungkeun Cho
Jules Villard
Nikos Gorogiannis
Dulma Churchill
David Pichardie
Jesse Katsumata