Summary:
Unknown functions may create false positives as well as false negatives
for Pulse. Let's consider that unknown functions behave "functionally",
or at least that a functional behaviour is a possible behaviour for
them: when called with the same parameter values, they should return the
same value.
This is implemented purely in the arithmetic domain by recording
`v_return = f_unknown(v1, v2, ..., vN)` for each call to unknown
functions `f_unknown` with values `v1`, `v2`, ..., `vN` (and return
`v_return`). The hope is that this will create more false negatives than
false positives, as several FPs have been observed on real code that
would be suppressed with this heuristic.
The other effect this has on reports is to record hypotheses made on the
return values of unknown functions into the "pruned" part of formulas,
which inhibits reporting on paths whose feasibility depends on the
return value of unknown functions (by making these issues latent
instead). This should allow us to control the amount of FPs until we
model more functions.
Reviewed By: skcho
Differential Revision: D27798275
fbshipit-source-id: d31cfb8b6
master
Jules Villard4 years agocommitted byFacebook GitHub Bot
codetoanalyze/c/pulse/traces.c, access_use_after_free_bad, 2, USE_AFTER_FREE, no_bucket, ERROR, [invalidation part of the trace starts here,parameter `l` of access_use_after_free_bad,was invalidated by call to `free()`,use-after-lifetime part of the trace starts here,parameter `l` of access_use_after_free_bad,invalid access occurs here]
codetoanalyze/c/pulse/traces.c, call_makes_null_deref_manifest_bad, 4, NULLPTR_DEREFERENCE, no_bucket, ERROR, [calling context starts here,in call to `something_about_strings_latent`,null pointer dereference part of the trace starts here,in call to `malloc` (modelled),is the null pointer,assigned,invalid access occurs here]
codetoanalyze/c/pulse/traces.c, access_use_after_free_bad, 2, USE_AFTER_FREE, no_bucket, ERROR, [invalidation part of the trace starts here,was invalidated by call to `free()`,use-after-lifetime part of the trace starts here,invalid access occurs here]
codetoanalyze/c/pulse/traces.c, call_makes_null_deref_manifest_bad, 4, NULLPTR_DEREFERENCE, no_bucket, ERROR, [calling context starts here,in call to `something_about_strings_latent`,null pointer dereference part of the trace starts here,in call to `malloc` (modelled),is the null pointer,assigned,invalid access occurs here]
codetoanalyze/cpp/pulse/nullptr.cpp, test_after_dereference2_latent, 1, NULLPTR_DEREFERENCE, no_bucket, ERROR, [*** LATENT ***,source of the null value part of the trace starts here,is the null pointer,null pointer dereference part of the trace starts here,parameter `x` of test_after_dereference2_latent,invalid access occurs here]
codetoanalyze/cpp/pulse/optional.cpp, FP_smart_pointer, 2, OPTIONAL_EMPTY_ACCESS, no_bucket, ERROR, [in call to `Node::getShared`,in call to `folly::Optional::Optional(=None)` (modelled),is optional empty,return from call to `Node::getShared`,invalid access occurs here]
codetoanalyze/cpp/pulse/path.cpp, faulty_call_bad, 0, NULLPTR_DEREFERENCE, no_bucket, ERROR, [calling context starts here,in call to `only_bad_on_42_latent`,null pointer dereference part of the trace starts here,in call to `may_return_null`,is the null pointer,returned,return from call to `may_return_null`,assigned,invalid access occurs here]
codetoanalyze/cpp/pulse/path.cpp, only_bad_on_42_latent, 2, NULLPTR_DEREFERENCE, no_bucket, ERROR, [*** LATENT ***,in call to `may_return_null`,is the null pointer,returned,return from call to `may_return_null`,assigned,invalid access occurs here]
codetoanalyze/cpp/pulse/reference_wrapper.cpp, reference_wrapper_heap_bad, 2, USE_AFTER_DELETE, no_bucket, ERROR, [invalidation part of the trace starts here,when calling `getwrapperHeap` here,in call to `WrapsB::WrapsB`,allocated by call to `new` (modelled),assigned,return from call to `WrapsB::WrapsB`,when calling `WrapsB::~WrapsB` here,parameter `this` of WrapsB::~WrapsB,when calling `WrapsB::__infer_inner_destructor_~WrapsB` here,parameter `this` of WrapsB::__infer_inner_destructor_~WrapsB,was invalidated by `delete`,use-after-lifetime part of the trace starts here,in call to `getwrapperHeap`,in call to `WrapsB::WrapsB`,allocated by call to `new` (modelled),assigned,return from call to `WrapsB::WrapsB`,in call to `ReferenceWrapperHeap::ReferenceWrapperHeap`,parameter `a` of ReferenceWrapperHeap::ReferenceWrapperHeap,in call to `WrapsB::getb`,parameter `this` of WrapsB::getb,returned,return from call to `WrapsB::getb`,assigned,return from call to `ReferenceWrapperHeap::ReferenceWrapperHeap`,return from call to `getwrapperHeap`,invalid access occurs here]
codetoanalyze/cpp/pulse/reference_wrapper.cpp, reference_wrapper_stack_bad, 2, USE_AFTER_LIFETIME, no_bucket, ERROR, [invalidation part of the trace starts here,variable `rw` declared here,when calling `getwrapperStack` here,variable `b` declared here,is the address of a stack variable `b` whose lifetime has ended,use-after-lifetime part of the trace starts here,variable `rw` declared here,in call to `getwrapperStack`,variable `b` declared here,in call to `ReferenceWrapperStack::ReferenceWrapperStack`,parameter `bref` of ReferenceWrapperStack::ReferenceWrapperStack,assigned,return from call to `ReferenceWrapperStack::ReferenceWrapperStack`,return from call to `getwrapperStack`,invalid access occurs here]
codetoanalyze/cpp/pulse/reference_wrapper.cpp, reference_wrapper_stack_bad, 2, USE_AFTER_LIFETIME, no_bucket, ERROR, [invalidation part of the trace starts here,when calling `getwrapperStack` here,variable `b` declared here,is the address of a stack variable `b` whose lifetime has ended,use-after-lifetime part of the trace starts here,in call to `getwrapperStack`,variable `b` declared here,in call to `ReferenceWrapperStack::ReferenceWrapperStack`,parameter `bref` of ReferenceWrapperStack::ReferenceWrapperStack,assigned,return from call to `ReferenceWrapperStack::ReferenceWrapperStack`,return from call to `getwrapperStack`,invalid access occurs here]
Jules Villard
4 years ago
committed by
Facebook GitHub Bot