Summary: Reports on reads that have one or more conflicting writes. When you report, say which other methods race with it.
Reviewed By: sblackshear
Differential Revision: D4538793
fbshipit-source-id: 47ce700
Summary:
It seems that what `Checkers.ST.store_summary` was called witin on-demand was actually redundant with what `Ondemand` is doing before storing the summaries.
This also makes the `Ondemand` module no longer depend on `Checkers` as the dependency is expected to be the other way around.
Reviewed By: sblackshear
Differential Revision: D4595006
fbshipit-source-id: d62187e
Summary: Thread-local variables can't be shared between threads, so it's safe to mutate them outside of synchronization
Reviewed By: jeremydubreil
Differential Revision: D4568316
fbshipit-source-id: 0634cad
Summary:
Enrich the domain of SIOF to contain, as well as the globals needed by a
procedure, the globals that the procedure initializes. Also add the possibility
to model some procedures as initializing some variables. Use that mechanism to
teach the checker about `std::ios_base::Init`.
Reviewed By: mbouaziz
Differential Revision: D4588284
fbshipit-source-id: d72fc87
Summary:
Reimplement whitelists as a match against a single regexp. This allows one to
precompile the whitelist regexp to make fast check against a whitelist of fuzzy
qualifiers, instead of checks linear in the number of items in the whitelist.
Reviewed By: akotulski
Differential Revision: D4588278
fbshipit-source-id: 3bac614
Summary:
This is useful to make sure `./build-infer.sh clang` and `./build-infer.sh
java` work correctly.
I had to move one unit test inside a new unit/clang/ directory used only if clang is enabled, and create a stub for it in unit/clang_stubs/ for when clang is disabled.
Renamed `OCAML_SOURCES` to the more consistent and descriptive `OCAML_CONFIG_SOURCES`.
Reviewed By: jeremydubreil
Differential Revision: D4571997
fbshipit-source-id: 9502114
Summary:
It's annoying that changing the Makefile or running `./configure` with
different options does not always retrigger compilation of some targets, eg
infer.
Reviewed By: akotulski
Differential Revision: D4571988
fbshipit-source-id: dda080f
Summary:
This gives a way for users to flag safe methods regarding SIOF, for instance if
the problematic paths in the method cannot happen before `main()` has started.
Reviewed By: akotulski
Differential Revision: D4578700
fbshipit-source-id: 6542dcf
Summary: Shorter is better. `--compilation-database` was taken, renaming it to `--buck-compilation-database`.
Reviewed By: dulmarod
Differential Revision: D4567028
fbshipit-source-id: 011cd6f
Summary:
Two options to do the same thing, one for Java and and one for clang, become
one option to do the same thing.
Reviewed By: akotulski
Differential Revision: D4578472
fbshipit-source-id: fb0f21b
Summary:
Before: `make clean` followed by running `infer -- make`. If infer fails, it is
rerun automatically (by the `silent_on_success` Makefile function) to show the
output to the user, but by then there is nothing to build and `make` does
nothing.
Now: run directly `infer -- make clean all`. If infer fails, the command is
rerun and rebuilds all the source files, so there is a higher chance that the
same error will be displayed to the user than the one that originally caused
the command to fail.
Reviewed By: dulmarod
Differential Revision: D4578477
fbshipit-source-id: 774f45c
Summary:
A good first step in order to run multiple checkers together is to prevent the analysis the analysis to side effect on the summaries of the method being analyzed from disk, or the shared specs summary. The idea is that `Ondemand` creates a summary for the procedure being analyzed and only saves the summary once all the checkers have been run. The summary for the caller (i.e. the procedure being analyzed) should never be looked up from disk during the analysis. In other words, the analysis should only ever lookup the summaries of the callees and the proposed solution to enforce this is to have `Ondemand.analyze_proc_name` be the only way to lookup the summary of a procedure.
Another objective is to make sure that the summaries are never saved to disk more than once.
Reviewed By: sblackshear
Differential Revision: D4549764
fbshipit-source-id: f0a6e21
Summary:
We waste a lot of space storing the types of field accesses and comparing them sets/maps with access paths.
Yet almost none of the code ever looks at these types (only a tiny piece of code in thread-safety).
If we know the base type, we have enough information to recover the type of the field.
Let's do that instead.
Reviewed By: jeremydubreil
Differential Revision: D4567996
fbshipit-source-id: e7fd2da
Summary:
This simplifies a bit the code to run the analysis on all the prcedures in the cluster. Before, the functions procedure_should_be_analyzed, which loads the attributes, and get_proc_desc were called twice for the analysis of every procedure.
The objective is to remove the calls to procedure_should_be_analyzed and hide it from the ondemand API since it is already called before the analysis of every procedure.
Reviewed By: sblackshear
Differential Revision: D4553397
fbshipit-source-id: 02cffaf
Summary: In C++ there are types that contain `<>` in their names (templates). When printing type to `html` those should be escaped
Reviewed By: jeremydubreil
Differential Revision: D4572506
fbshipit-source-id: a180537
Summary:
One gets very obscure errors when trying to run infer for clang when it was
compiled for Java, or vice-versa. This diff makes sure we crash early with the
appropriate error message. For instance:
```
$ ./build-infer java
$ infer -- clang -c hello.c
Uncaught exception:
(Failure
"Unsupported build mode: make/cc\
\nInfer was built with clang analyzers disabled.\
\nPlease rebuild infer with clang enabled.\
\n")
Raised at file "pervasives.ml", line 30, characters 22-33
Called from file "backend/infer.ml", line 398, characters 6-48
Called from file "backend/infer.ml", line 449, characters 20-38
$ infer --clang-compilation-db-files foo.json
Uncaught exception:
(Failure
"Unsupported build mode: clang compilation database\
\nInfer was built with clang analyzers disabled.\
\nPlease rebuild infer with clang enabled.\
\n")
Raised at file "pervasives.ml", line 30, characters 22-33
Called from file "backend/infer.ml", line 392, characters 8-65
Called from file "backend/infer.ml", line 449, characters 20-38
```
Reviewed By: sblackshear
Differential Revision: D4566641
fbshipit-source-id: d9a118f
Summary:
- inferbo introduced a dependency to extlib. When building Java analyzers, this
is implicitly pulled in by javalib, but it's missing when building only the
clang analyzers. Add `extlib` to the packages we build against.
- infer.ml and Javac.ml depend on Javalib, but it's easy to push down the code
that needs it to `jMain.ml` so that we can build without javalib for the
clang-only case.
- jMain.mli had 2 copies: one in java/ and one in java_stubs/. Make one a symlink to the other.
Reviewed By: jeremydubreil
Differential Revision: D4566581
fbshipit-source-id: 214a4eb
Summary:
Sevel auxiliary files made it to the output directory of the analysis of individual targets when analyzing Java projects build with Buck. However, these files are then taken into account= to compute the target rule key and then to decide whether to analyze the dependent targets. Since these auxiliary files were containing time sentive information, every cach miss on a given target would then invalitate the cache entries for all the dependent targets.
This diff cleans up the output directory to only keep the specs files, the `global.tenv` and the `report.json` files which are the only artifacts needed to analyze the dependent targets
This diff makes a minimal number of changes to see how it behaves in prod, but I intend to refoctor this more when continuing to add support for running Infer with genrules
Reviewed By: sblackshear
Differential Revision: D4562615
fbshipit-source-id: 4628420
Summary:
Hi!
It's quite common to have collections of delegates. The collection itself is usually named like "delegatesHash", "delegatesStorage" or simply "delegates". Obviously, there is common part in all these cases, but currently you're excluding property only if it contains "queue".
I've added a simple exclusion by common part. It solved false-positive warnings for me and I think for others it'll be quite helpful too.
Closes https://github.com/facebook/infer/pull/582
Reviewed By: ddino
Differential Revision: D4565221
Pulled By: dulmarod
fbshipit-source-id: c48242e
Summary:
Xcode's compilation databases follows a different convention than cmake's and
escape the `"file"` and `"dir"` fields of each unit to make them shell-ready.
We need to treat them differently when reading them.
This adds a new `--clang-compilation-db-files-escaped` option and makes the
code related to reading compilation databases deal correctly with both
conventions.
Reviewed By: akotulski
Differential Revision: D4559239
fbshipit-source-id: 51120ae
Summary: Some compilation databases give relatives paths for the `"file"` field. This is not ambiguous as there is also a `"dir"` field, so use that to make the path absolute when needed.
Reviewed By: dulmarod
Differential Revision: D4559145
fbshipit-source-id: be36a16
Summary:
This avoids spurious warnings on projects using gcc with optimization flags
that are ignored by clang.
Reviewed By: akotulski
Differential Revision: D4559326
fbshipit-source-id: 14a2431
Summary:
I couldn't figure out why, but from within an infer release the traces we get
for this test are different than the expected ones. This is even consistent
across osx and linux.
In order to restore sanity, let's just hide this incomprehensible fact. Let's
come back to it if more tests exhibit this, maybe traces are not guaranteed to
be exactly the same across runs.
Reviewed By: sblackshear
Differential Revision: D4559405
fbshipit-source-id: dd88c59
Summary: Should stop us from reporting on benign races of fields that are caching resources.
Reviewed By: peterogithub
Differential Revision: D4538037
fbshipit-source-id: 15236b4
Summary: This annotation can then be used to suppress the warnings on non-android Java projects.
Reviewed By: sblackshear
Differential Revision: D4544858
fbshipit-source-id: 8a0b8fa
Summary:
It seems that the `close()` method that should normally be called on an object `obj` of type `java.io.Closeable` is sometimes called on `obj` of type `java.lang.Object`. It did not fully understand in which case this happens but it could be coming from a bug in Sawja since the type of `obj` in the bytecode is correct, but the Sawja reciever expression given to the Java frontend has the type `java.lang.Object`.
In any case, it does not hurt to always consider that `obj.close()` will replace the `FILE` attribute on `obj` by a `MEM` attribute.
Reviewed By: sblackshear
Differential Revision: D4540627
fbshipit-source-id: 71f9c95
Summary:
This reorganises the contents of `infer --help`:
- Headings are more prominent (start with `**`)
- New "Java" section
- Delete "Analysis" section, distribute contents over other sections
- New "Quandary" section
- Under the hood, new "Buffer Overruns" and "Crashcontext" sections, but do not show them as we don't expect external use yet, although that may be a bit arbitrary
- typo: `--bufferoverrn` -> `--bufferoverrun`
- move some options from one section to another
Reviewed By: jberdine
Differential Revision: D4537500
fbshipit-source-id: a789375
Summary:
Native compilation seems to take a couple of seconds to build the unit test
binary, but we don't need it to be native (and neither do users building infer
to install it on their machines). In fact, the test mode rebuilds it in
bytecode mode. For the sake of being able to run the unit tests when developing
on infer (and thus possibly while having some fatal warnings showing up in the
code, preventing us from building the test target), also build InferUnit in
bytecode mode.
Reviewed By: sblackshear, martinoluca
Differential Revision: D4537455
fbshipit-source-id: 374c84c
Summary:
The Java models for resources are way to complex. The main issue I am facing with these models is that small changes in the analysis can affect the generation of the models in some weird ways. For instance, I get different specs for some of the models between my devserver and my devvm, which seems to be mostly related with the backend treatment of `instanceof`.
The objective here is to simplify the models as much as possible in order to:
1) make debugging regressions easier
2) get simpler specs and less modeled methods shipped in `models.jar`
Reviewed By: sblackshear
Differential Revision: D4536115
fbshipit-source-id: 577183a
Summary: Better documentation, and could perhaps be checked instead of trusted later if the analysis understands threads better.
Reviewed By: jaegs
Differential Revision: D4537463
fbshipit-source-id: 4323c78
Summary: In some cases where a function is called directly on a formal (e.g, `def foo(o) { callSomething(o) }`, we were failing to propagate the footprint trace to the caller.
Reviewed By: jeremydubreil
Differential Revision: D4502404
fbshipit-source-id: d4d632f
Summary: This case was already working but there was no tests for it
Reviewed By: sblackshear
Differential Revision: D4529473
fbshipit-source-id: ca3ff02
Summary: This will be important for maintaining ownership of `View`'s, which involve a lot of casting.
Reviewed By: peterogithub
Differential Revision: D4520441
fbshipit-source-id: fdef226
Summary:
Previously, we would lose track of ownership in code like
```
Obj owned = new Obj();
Obj stillOwned = id(owned); // would lose ownership here
stillOwned.f = ... // would report false alarm here
```
This diff partially addresses the problem by adding a notion of "unconditional" (always owned) or "conditional" (owned if some formal at index i is owned) ownership.
Now we can handle simple examples like the one above.
I say "partially" because we still can't handle cases where there are different reasons for conditional ownership, such as
```
oneOrTwo(Obj o1, Obj o2) { if (*) return o1; else return o2; } // we won't understand that this maintains ownership if both formals are owned
Obj stillOwned = oneOrTwo(owned1, owned2);
stillOwned.f = ... // we'll report a false alarm here
```
This can be addressed in the future, but will require slightly more work
Reviewed By: peterogithub
Differential Revision: D4520069
fbshipit-source-id: 99c7418
Summary: This will make it a cinch to track new "attributes" of memory locations, and to propagate more complex attributes such as conditional ownership (coming in a future diff).
Reviewed By: peterogithub
Differential Revision: D4523143
fbshipit-source-id: 57aa133
Summary: The diff remove the no-op model for `Cursor.close()` by the frontend-based `Closeable` as resources mechanism where every call of the form `object.close()` removes the file attribute on `object` when `object` is of type `Closeable`.
Reviewed By: sblackshear
Differential Revision: D4519386
fbshipit-source-id: 83633d4
Summary: This fixes a wrong level of indirection when performing the type substitution.
Reviewed By: sblackshear
Differential Revision: D4521008
fbshipit-source-id: 7324ea6
Summary: Those should be treated angelically during the analysis with the same end results
Reviewed By: sblackshear
Differential Revision: D4518930
fbshipit-source-id: ee5bae8
Summary: Being forced to separately define `pp_element`/`pp_key` is uneccessary and makes it more cumbersome to create a set/map from an existing module that already defines `pp`.
Reviewed By: jeremydubreil
Differential Revision: D4517308
fbshipit-source-id: 9b17c9c
Summary: Not clear why we need to disable this case and in which case is Infer creating too many disjunctions.
Reviewed By: sblackshear
Differential Revision: D4509394
fbshipit-source-id: fbc106d
Summary: This method can return `null` if the parameter is not a supported system service. However, since this method tends to be called with a constant value as parameter, it does seem to be returning null often in practice.
Reviewed By: sblackshear
Differential Revision: D4509185
fbshipit-source-id: 4cb80ce
Summary:
At one point I thought we'd want to have lots of different schedulers for things like exploring loops in different orders, but that hasn't materialized.
Let's make the common use-case simpler by hiding the `Scheduler` parameter inside the `AbstractInterpreter` module.
We can always expose `MakeWithScheduler` later if we want to.
Reviewed By: jberdine
Differential Revision: D4508095
fbshipit-source-id: 726e051
Summary:
This fixes false positives we had in fields written by callees of a constructor (see new E2E test).
This is also a bit cleaner than what we did before; instead of special-casing constructors, we just use the existing ownership concept.
Reviewed By: peterogithub
Differential Revision: D4505161
fbshipit-source-id: a739ebc
Summary:
Constants are always "owned" in the sense that no one can mutate them.
In code like
```
Obj getX(boolean b) {
if (b) {
return null;
}
return new Obj();
}
```
, we need to understand this in order to infer that the returned value is owned.
This should fix a few FP's that I've seen.
Reviewed By: peterogithub
Differential Revision: D4485452
fbshipit-source-id: beae15b
Summary:
Propagating the arguments read in .inferconfig shouldn't be necessary as they
are parsed by each executable. There are corner cases where I think this diff
could change the behaviour of infer (eg, .inferconfig redefines project-root
and subsequent exes read a different .inferconfig thanks to the project-root in
INFER_ARGS), but I don't think there are good use cases like that.
Reviewed By: jberdine
Differential Revision: D4475092
fbshipit-source-id: 5c020d0
Summary:
Clients should use `Config.parse_action` instead to figure in what mode they
are operating.
In particular, the biggest change is in logging. Take the `parse_action` into
account instead of the exe, and change the log/ subdirectories to be "capture",
"driver", "analyze", and "print", corresponding to the various phases of an
infer run.
Reviewed By: jberdine
Differential Revision: D4474943
fbshipit-source-id: 6d33ad3
Summary:
Support several parsing modes: Infer, Javac, NoParse
The "Infer" mode specifies a list of sections, ie the parts of infer that are affected by an option (corresponds to the old notion of "exes"):
analysis, clang frontend, print, ...
- .inferconfig and INFER_ARGS always parsed
- outside .inferconfig and INFER_ARGS, do not parse subcommand arguments before the subcommand has been activated
- command-line is parsed or not based on the subcommand/executable selected
- executable dictates subcommand, so almost nothing depends on the executable outside of Config. Another diff will restrict the API around exes to reflect this.
Reviewed By: jberdine
Differential Revision: D4474886
fbshipit-source-id: 442dfef
Summary:
Some classes may have deleted new operator for them. To fix it, run global `new` operator instead
```
struct X {
void* operator new(size_t) = delete;
};
X *p = new X; // compilation error
X *p = ::new X; // no compilation error
```
This change is following same strategy standard headers follow.
Reviewed By: jvillard
Differential Revision: D4500977
fbshipit-source-id: 20babfa
Summary:
Infer used to report null dereference when field was accessed later:
```
vector<int> v;
int& a = v[0]; // should be EMPTY_VECTOR_ACCESS here, but it wasn't reported
int b = a; // was NULL_DEREFERENCE here
```
To avoid this problem, model all accesses to vector as dereference of its internal `beginPtr` field.
Reviewed By: jberdine
Differential Revision: D4481942
fbshipit-source-id: 2142894
Summary: This should fix the issue with broken invariants when the method specialization on pointer ends up doing a substitution on non pointer types
Reviewed By: sblackshear
Differential Revision: D4487232
fbshipit-source-id: f3fce84
Summary:
When the receiver type and return type of an unknown call are the same, propagate taint to both the receiver and the return type.
This does the right thing for common "builder-style" methods that both update and return the receiver.
We already had custom models for a few such methods (e.g., `StringBuilder.append`), but we can remove them now.
Reviewed By: jeremydubreil
Differential Revision: D4490071
fbshipit-source-id: 325ea88
Summary: The Buck integration assumes that the output jar file of all the dependency targets are available locally in order to retrieve the analysis from these targets. However, this is not guaranteed to be true when there is a cache hit on some targets. Running `buck build` with the option `--deep` forces this property.
Reviewed By: jvillard
Differential Revision: D4474036
fbshipit-source-id: accabfa
Summary: The method `junit.framework.TestCase.setUp()` is always run before the other methods by the JUnit testing framework. So the method act as a class initializer.
Reviewed By: sblackshear
Differential Revision: D4487371
fbshipit-source-id: 1998801
Peter O'Hearn
Andrzej Kotulski
Jeremy Dubreil
Sam Blackshear
Cristiano Calcagno
Jules Villard
Dulma Churchill
Nikita Anisimov
Dino Distefano
Kihong Heo