Commit Graph

156 Commits

Author SHA1 Message Date
chrchr-github 8cd8b9c64e
Fix MSVC build and some warnings (#3334) 2021-07-16 21:55:12 +02:00
Rikard Falkeborn b013f1b3d0
Fix test case for #10186 (#3253) 2021-05-10 08:01:30 +02:00
Daniel Marjamäki 42437277dc Update Copyright year 2021-03-21 20:58:32 +01:00
Rikard Falkeborn 27316e04ed
Fix #10187 (memleak open with fd >= 0) (#3156) 2021-03-03 07:00:28 +01:00
Rikard Falkeborn 42a41e8b41
Fix #9139 (leak when variable has const qualifier) (#3159)
Also, in the tests, change socket/close to resource to get error
messages which say "resource leak" instead of "memory leak".
2021-03-03 06:58:38 +01:00
Rikard Falkeborn c3eb37972d
Fix #10182 (FN memory leak with if-statement) (#3151)
Improve leak detections in if-statements. This is done by checking
for leaks every time a scope is left. This allows cppcheck to catch
more memory leaks, as well as improve some error messages which now
contain the line where the variable goes out of scope, instead of the
end of the function.
2021-02-26 12:58:52 +01:00
Rikard Falkeborn 951ca2c8b9
Fix #10186 (FP memleak with cast and assignment) (#3148) 2021-02-25 10:55:34 +01:00
PKEuS 141d2ac215 Refactorization: Improved internal implementation of severity and certainty levels
Backported from LCppC.
2021-02-24 22:00:06 +01:00
PKEuS 4b01d5bdc6 Re-enabled a bunch of disabled unit tests, some as TODO tests
Merged from LCppC.
2021-02-20 13:02:14 +01:00
PKEuS cf1937294a Refactorization: Removed unnecessary \n and spaces in strings
Merged from LCppC.
2021-02-20 12:58:42 +01:00
Rikard Falkeborn f018163551
Improve handling of realloc in memory leak checker (#3036)
Mark realloced variables as realloced instead of freed. This allows
improved checking for code with error realloc handling.

If cppcheck finds an if-statement which checks the validity of the
allocated memory or resource, check if the memory/resource is
reallocated from another variable. If so, we can add checking of that
variable in the if-statement instead. This allows to check that variable
for memleaks and double frees.

This fixes #9292 and #9990 which both concern FPs with double frees
after correct error handling.
2021-01-11 07:55:05 +01:00
Rikard Falkeborn fb35756126
TestLeakAutoVar: Split recursiveCountLimit test to separate class (#2995)
This test is by far the slowest one to run. Split it to a separate class
to make it easy to select if it should be run during development.
2020-12-31 09:34:53 +01:00
Ken-Patrick Lehrmann 79bdd64689
Fix false positive memoryleak (#2882)
```
int *f() {
  int *p = static_cast<int *>(realloc(nullptr, 10));
  if (!!(!p)) {
    return nullptr;
  }
  return p;
}

```
would give
```
memleak2.cpp:4:5: error: Memory leak: p [memleak]
    return nullptr;
    ^
```
Because of the additional `!!̀ .
2020-11-10 15:59:51 +01:00
Daniel Marjamäki 6d0b78433b astyle formatting 2020-10-29 21:43:56 +01:00
Rikard Falkeborn de19dc9e3e
Fix #8327 (Memleak with mmap return value check) (#2864)
Also fix a broken test case related to checking the return value -1
where socket was not defined in the library used in the tests.

This was tested running test-my-pr with 500 packages. The difference was
six fewer FPs.
2020-10-29 13:17:33 +01:00
Rikard Falkeborn 1e679cc5d1 Fix #9635 (FP: Memory leak with comma operator in if-statement)
When checking for comparisons in if-statements, if there are comma
operators in the if-statement, skip until after the last comma.
2020-07-09 21:31:43 +02:00
Rikard Falkeborn 4996ec190e Fix #9652 (fp memleak with function call with cast)
When the first argument was (void *)(1), at the start of the second
iteration, arg was pointing to the "1", which caused problems for
nextArgument(), which saw the ")" as the next token and returned
nullptr, signalling that there are no more arguments.

Instead, save the first token in the argument, which makes
nextArgument() do the right thing.
2020-07-08 00:02:39 +02:00
orbitcowboy 9861a5291e Formatted the code, there are no functional changes [ci skip] 2020-05-20 23:45:00 +02:00
PKEuS fb1afe2345 Fixed test suite: Do no longer apply simplifyTokenList2 to token lists, except for those tests that test those simplifications, because checks are no longer run on that simplified token list
Changed failing unit test to TODO tests, as they indicate patterns we do no longer understand properly.
2020-05-20 18:54:16 +02:00
PKEuS c9d8f607df Optimization: Reduced peak memory usage (30% in my test case) by immediately deleting simplecpp::TokenList while creating the cppcheck TokenList. 2020-05-19 12:08:17 +02:00
Daniel Marjamäki 08ddd84780 Update copyright year 2020-05-10 11:16:32 +02:00
Daniel Marjamäki 3e0218299b Revert "Update copyright year"
This reverts commit 6eec6c4bd5.
2020-05-10 11:13:05 +02:00
Daniel Marjamäki 6eec6c4bd5 Update copyright year 2020-05-10 11:11:34 +02:00
Rikard Falkeborn 1f668c8dc4
Add regression test for #9187 (#2548)
This was fixed in ae0a73a53.
2020-02-22 21:26:26 +01:00
Ken-Patrick Lehrmann 7044c17599
9575: false positive in fclose (#2540)
Fix false positive introduced by
0b7649ca9b

Only return the function token from checkTokenInsideExpression when it
might be one the argument (hence keeping a pointer to one of them).
Otherwise, we can directly skip to the token after the function call.
2020-02-17 10:25:30 +01:00
Ken-Patrick Lehrmann 0b7649ca9b Fix 9298 (#2476)
* Fix 9298

Tell cppcheck that strcpy returns its first argument, and use that
knowledge in checkTokenInsideExpression.

* Add missing unit tests in cmake
2020-01-09 08:47:36 +01:00
Rikard Falkeborn 2d326c011d Add regression test for #9028 (#2343)
Ticket #9028 was fixed in 5800692fa1, add a test case where the
class definition is seen.
2019-11-10 08:28:46 +01:00
Rikard Falkeborn bb5ac32872 Fix #7031 (improve error message for memory related warnings) (#2204)
Printout both the locations on double free errors, mismatching
alloc/dealloc and dealloc return error.
2019-09-22 21:50:02 +02:00
Rikard Falkeborn df800e35d4 Fix memleak FP with return with parenthesis (#2202)
* Fix memleak FP with return  with parenthesis

Fix FPs pointed out by daca@home on the following form:

    void* f(void) {
        void* x = malloc(1);
        return(x);
    }

Fix it by only skipping tokens if there is an actual match with a
variable. This allows to remove the special casing of "return;".

* Add testcase with cast
2019-09-22 19:18:31 +02:00
Rikard Falkeborn 46ac0d79c1 Checkmemleakautovar: fix crash and FP (#2196)
This fixes crashes found by daca where valueType() is NULL. Also,
somewhat related, it removes warnings when casting to a type that is
unknown to cppcheck, for example, there is no longer a warning for the
following code:

	void* f() {
		void *x = malloc(1);
		return (mytype)x;
	}
2019-09-21 14:59:54 +02:00
Oliver Stöneberg 9028b4a81d do not access static methods through instance (#2189) 2019-09-20 21:54:30 +02:00
Rikard Falkeborn 007b5d3e8d Fix #9343 (memleak FP when return with cast) (#2162)
This was most likely introduced when the checks were changed to run on
the full tokenlist instead of the simplified one.

Take care to warn about cases where casts destroy the pointer, such as

	uint8_t f() {
		void* x = malloc(1);
		return (uint8_t)x;
	}
2019-09-20 15:09:27 +02:00
Daniel Marjamäki 9fa1a5a540 astyle formatting
[ci skip]
2019-08-07 08:05:46 +02:00
Ken-Patrick cb0b057595 8706 (#2066)
* 8706

* Add test case for #8706
2019-08-07 08:04:10 +02:00
Rikard Falkeborn f95d193e22 Fix #8282 (memleak FP when returning std::pair) (#2039) 2019-07-26 07:03:58 +02:00
Rikard Falkeborn abea580b78 Fix FP memory leak with unknown function call in condition (#2012)
* Fix FP memory leak with unknown function call in condition

This was introduced in 8513fb81d2 when
fixing memory leaks for global variables allocated in condition. The
refactored code had an inconsistency where c and c++ code behaved
slightly differently when `var` is NULL. This seemed to not have an
impact as the code was written prior to 8513fb81d2,
but when the same code was used for conditions, FPs were introduced.

The introduced FPs were memleak warnings when there should have been an
information message about missing configurations for code like

	void f() {
		char *p = malloc(10);
		if (set_data(p)) {}
	}

Fix this by always returning true if varTok->Variable() is NULL for
both c and c++ code.

* Improve function name
2019-07-18 15:23:19 +02:00
Daniel Marjamäki 5800692fa1 Move and refactor the CheckMemoryLeak::isclass 2019-07-17 08:59:09 +02:00
Rikard Falkeborn 8513fb81d2 Fix #9206 (FP with global variable allocated in condition) (#2007) 2019-07-17 07:43:07 +02:00
IOBYTE c2ccfd5f8b Increase Windows stack size to 8M (#1998)
* Increase Windows stack size to 8M

* try to add stack size cmake
2019-07-15 12:39:58 +02:00
IOBYTE 526a86dc60 Fix recursiveCount in CheckLeakAutoVar to really be recursive count. (#1988) 2019-07-13 07:40:24 +02:00
Daniel Marjamäki ae933c20c7 Fixed #1765 (Improve check: delete not handled well when there are extra parentheses) 2019-07-12 18:25:20 +02:00
IOBYTE 74e3114a64 Fix #9097 (Crash on thousands of "else ifs"s in gcc-avr package) (#1982)
* Fix #9097 (Crash on thousands of "else ifs"s in gcc-avr package)

* increase recursion count maximum to 512 because cppcheck was hitting the 256 limit

* 512 was too much for windows
2019-07-12 07:56:05 +02:00
Rikard Falkeborn a1a14b8465 Fix FP with cast pointer to free() (#1961)
This fixes false positives when the pointer passed to free() (or similar
deallocation functions) is cast using a c-style cast.
2019-07-10 09:13:59 +02:00
Rikard Falkeborn 839fcddd8a Fix #6115 (Add support to realloc to cfg files) (#1953)
* Allow to configure realloc like functions

* memleakonrealloc: Bring back tests.

The old memleak checker was removed, and the tests for it was removed in
commit 9765a2dfab. This also removed the
tests for memleakOnRealloc. Bring back those tests, somewhat modified
since the checker no longer checks for memory leaks.

* Add realloc to mem leak check

* Add tests of realloc buffer size

* Configure realloc functions

* Add test of freopen

* Allow to configure which element is realloc argument

* Fix wrong close in test

cppcheck now warns for this

* Update manual

* Update docs

* Rename alloc/dalloc/realloc functions

Naming the member function realloc caused problems on appveyor. Rename
the alloc and dealloc functions as well for consistency.

* Change comparisson order

* Remove variable and use function call directly

* Create temporary variable to simplify

* Throw mismatchError on mismatching allocation/reallocation

* Refactor to separate function

* Fix potential nullptr dereference

As pointed out by cppcheck.
2019-07-05 12:44:52 +02:00
Rikard Falkeborn 60a213e6a5 Fix #9047 (c-style casts before malloc) (#1930)
* Fix #9047 (c-style casts before malloc)

Note that there are still no warnings for c++-style casts

* Fix memleak check with casts of assignments in if-statements

* Fix possible null pointer dereference

As pointed out by cppcheck.

* Add check of astOperand2 when removing casts

This is similar to how it is done in other checks.
2019-07-03 08:39:44 +02:00
Daniel Marjamäki 7a3302b3e3 CheckLeakAutoVar: Fixed FP seen in daca@home 2019-06-02 15:25:54 +02:00
Daniel Marjamäki 4da50942b0 Fixed #8120 (False positive: Memory pointed to by 'p' is freed twice) 2019-05-30 16:22:41 +02:00
Rikard Falkeborn ada881ccdf Fix #9130 (FP memory leak with NULL pointer cast) (#1839) 2019-05-18 06:22:25 +02:00
Daniel Marjamäki da46bff1b3 CheckLeakAutoVar: Use Library::isSmartPointer() 2019-04-24 15:35:47 +02:00
Daniel Marjamäki 3dc34f1515 Disable all simplified checks 2019-03-16 09:17:50 +01:00