pghs975uc
/
infer_clone
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '3eed17488a'
${ noResults }
infer_clone/infer/tests/codetoanalyze/objc/performance/cost-issues.exp

244 lines
36 KiB
Raw Normal View History Unescape

[cost] add model for iterating over collections via block Summary: Namely `enumerateObjectsUsingBlock` which takes a collection and a block as an arg and the iterates over the collection and applies the block to each element of the collection. This is a common way to iterate over Objc collections, so let's add a model for it. Reviewed By: skcho Differential Revision: D24604590 fbshipit-source-id: 1ceeb4b40
4 years ago
${XCODE_ISYSROOT}/System/Library/Frameworks/Foundation.framework/Headers/NSArray.h, NSArray.enumerateObjectsUsingBlock:[objc_blockMyBlock.call_enumerate_via_block_param_linear_FN::_3], 0, OnUIThread:false, []
${XCODE_ISYSROOT}/System/Library/Frameworks/Foundation.framework/Headers/NSArray.h, NSArray.enumerateObjectsUsingBlock:[objc_blockenumerate_via_block_linear_2], 0, OnUIThread:false, []
[cost] Report with relative paths on xcode_isysroot in tests Summary: This diff reports paths under the xcode isysroot as relative in tests. This was a problem when another machine that has a different isysroot directory is running the test. Reviewed By: ezgicicek Differential Revision: D23729222 fbshipit-source-id: 4e9681f65
4 years ago
${XCODE_ISYSROOT}/System/Library/Frameworks/Foundation.framework/Headers/NSArray.h, NSArray.indexOfObject:inSortedRange:options:usingComparator:[objc_blocknsarray_binary_search_log_FN_1], 0, OnUIThread:false, []
[preanalysis] Replace overridden method calls in ObjC Summary: This diff replaces overridden method calls in ObjC when possible, ie the first parameter of the method has a sub-class type of the method's class. For example, when `MyEnumerator` is a sub-class of `NSEnumerator` and there is overridden `nextObject`, ``` [my_enumerator nextObject] ``` in Sil, it was translated to like ``` NSEnumerator.nextObject(my_enumerator : MyEnumerator*) ``` and the analyzer missed the overridden method. This diff replaces the function call to ``` MyEnumerator.nextObject(my_enumerator : MyEnumerator*) ``` Reviewed By: ezgicicek Differential Revision: D24477290 fbshipit-source-id: 6842a76f8
4 years ago
codetoanalyze/objc/performance/MyEnumerator.m, MyEnumerator.dealloc, 1, OnUIThread:false, []
[cost] Fix on-demand generation of NSEnumerator Summary: This diff fixes on-demand symbolic value generation of a class that inherits NSEnumerator. Reviewed By: ngorogiannis Differential Revision: D24504955 fbshipit-source-id: bcb20e8aa
4 years ago
codetoanalyze/objc/performance/MyEnumerator.m, MyEnumerator.nextObject, 6 + 3 ⋅ self->n.ub + 3 ⋅ (1+max(0, self->n.ub)), OnUIThread:false, [{1+max(0, self->n.ub)},Loop,{self->n.ub},Loop]
[cost] add model for iterating over collections via block Summary: Namely `enumerateObjectsUsingBlock` which takes a collection and a block as an arg and the iterates over the collection and applies the block to each element of the collection. This is a common way to iterate over Objc collections, so let's add a model for it. Reviewed By: skcho Differential Revision: D24604590 fbshipit-source-id: 1ceeb4b40
4 years ago
codetoanalyze/objc/performance/NSArray.m, MyBlock.call_enumerate_via_block_param_linear_FN::, 6, OnUIThread:false, []
codetoanalyze/objc/performance/NSArray.m, MyBlock.dealloc, 1, OnUIThread:false, []
[preanalysis] Replace overridden method calls in ObjC Summary: This diff replaces overridden method calls in ObjC when possible, ie the first parameter of the method has a sub-class type of the method's class. For example, when `MyEnumerator` is a sub-class of `NSEnumerator` and there is overridden `nextObject`, ``` [my_enumerator nextObject] ``` in Sil, it was translated to like ``` NSEnumerator.nextObject(my_enumerator : MyEnumerator*) ``` and the analyzer missed the overridden method. This diff replaces the function call to ``` MyEnumerator.nextObject(my_enumerator : MyEnumerator*) ``` Reviewed By: ezgicicek Differential Revision: D24477290 fbshipit-source-id: 6842a76f8
4 years ago
codetoanalyze/objc/performance/NSArray.m, call_my_enumerator_next_object_linear, 9 + 3 ⋅ enumerator->n.ub + 3 ⋅ (1+max(0, enumerator->n.ub)), OnUIThread:false, [{1+max(0, enumerator->n.ub)},Call to MyEnumerator.nextObject,Loop,{enumerator->n.ub},Call to MyEnumerator.nextObject,Loop]
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/NSArray.m, call_nsarray_enumerator_param_linear, 6, OnUIThread:false, []
[cost] add model for iterating over collections via block Summary: Namely `enumerateObjectsUsingBlock` which takes a collection and a block as an arg and the iterates over the collection and applies the block to each element of the collection. This is a common way to iterate over Objc collections, so let's add a model for it. Reviewed By: skcho Differential Revision: D24604590 fbshipit-source-id: 1ceeb4b40
4 years ago
codetoanalyze/objc/performance/NSArray.m, enumerate_via_block_linear, 2 + array->elements.length.ub, OnUIThread:false, [{array->elements.length.ub},Modeled call to enumerateObjectsUsingBlock:]
[cost] Fix on-demand generation of NSEnumerator Summary: This diff fixes on-demand symbolic value generation of a class that inherits NSEnumerator. Reviewed By: ngorogiannis Differential Revision: D24504955 fbshipit-source-id: bcb20e8aa
4 years ago
codetoanalyze/objc/performance/NSArray.m, loop_with_my_enumerator_next_object_linear_FP, 2 + 3 ⋅ enumerator->n.ub × (enumerator.length + 1) + 10 ⋅ (enumerator.length + 1) + 3 ⋅ (enumerator.length + 1) × (1+max(0, enumerator->n.ub)), OnUIThread:false, [{1+max(0, enumerator->n.ub)},Call to MyEnumerator.nextObject,Loop,{enumerator.length + 1},Loop,{enumerator.length + 1},Loop,{enumerator->n.ub},Call to MyEnumerator.nextObject,Loop]
[clang] fix bad interaction between ConditionalOperator and initializers Summary: This is several inter-connected changes together to keep the tests happy. The ConditionalOperator `b?t:e` is translated by first creating a placeholder variable to temporarily store the result of the evaluation in each branch, then the real thing we want to assign to reads that variable. But, there are situations where that changes the semantics of the expression, namely when the value created is a struct on the stack (eg, a C++ temporary). This is because in SIL we cannot assign the *address* of a program variable, only its contents, so by the time we're out of the conditional operator we cannot set the struct value correctly anymore: we can only set its content, which we did, but that results in a "shifted" struct value that is one dereference away from where it should be. So a batch of changes concern `conditionalOperator_trans`: - instead of systematically creating a temporary for the conditional, use the `trans_state.var_exp_typ` provided from above if available when translating `ConditionalOperator` - don't even set anything if that variable was already initialized by merely translating the branch expression, eg when it's a constructor - fix long-standing TODO to propagate these initialization facts accurately for ConditionalOperator (used by `init_expr_trans` to also figure out if it should insert a store to the variable being initialised or not) The rest of the changes adapt some relevant other constructs to deal with conditionalOperator properly now that it can set the current variable itself, instead of storing stuff inside a temp variable. This change was a problem because some constructs, eg a variable declaration, will insert nodes that set up the variable before calling its initialization, and now the initialization happens *before* that setup, in the translation of the inner conditional operator, which naturally creates nodes above the current one. - add a generic helper to force a sequential order between two translation results, forcing node creation if necessary - use that in `init_expr_trans` and `cxxNewExpr_trans` - adjust many places where `var_exp_typ` was incorrectly not reset when translating sub-expressions The sequentiality business creates more nodes when used, and the conditionalOperator business uses fewer temporary variables, so the frontend results change quite a bit. Note that biabduction tests were invaluable in debugging this. There could be other constructs to adjust similarly to cxxNewExpr that were not covered by the tests though. Added tests in pulse that exercises the previous bug. Reviewed By: da319 Differential Revision: D24796282 fbshipit-source-id: 0790c8d17
4 years ago
codetoanalyze/objc/performance/NSArray.m, multiple_nsarray_enumerators_param_linear, 10 + 5 ⋅ (enumerator2.length + enumerator1.length + 2), OnUIThread:false, [{enumerator2.length + enumerator1.length + 2},Loop]
[frontend] Add translation of arrayWithObjects:count: Summary: This diff adds translation of `arrayWithObjects:count:`. In the previous implementation it was translated as if it was `arrayWithObjects:`, but their function parameters are different. In this diff, it translates an array literal `NSArray* a = @ [ 2, 3 ];` to ``` n$1=NSNumber.numberWithInt:(2:int) n$2=NSNumber.numberWithInt:(3:int) temp[0]:objc_object*=n$1 temp[1]:objc_object*=n$2 n$3=NSArray.arrayWithObjects:count:(temp:objc_object* const [2*8],2:int) a:NSArray*=n$3 ``` where `temp` is an additional local variable declared as array. See, https://developer.apple.com/documentation/foundation/nsarray/1460145-arraywithobjects https://developer.apple.com/documentation/foundation/nsarray/1460096-arraywithobjects?language=objc Reviewed By: jvillard Differential Revision: D22631305 fbshipit-source-id: 5be0a55d4
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_access_constant, 50, OnUIThread:false, []
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_access_linear, 3 + 7 ⋅ array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Loop]
[cost] add objc test for NSArray and NSMutableArray Summary: Add objc test for ```NSArray``` and ```NSMutableArray```. ```NSMutableArray``` is a subclass of ```NSArray```. For documentation of ```NSArray```, https://developer.apple.com/documentation/foundation/nsarray?language=objc For documentation of ```NSMutableArray```, https://developer.apple.com/documentation/foundation/nsmutablearray?language=objc The underlying mechanism for ```NSMutableArray``` is quite complicated. It changes the underlying data structure during runtime, so it is possible to have say O(log n) complexity for accessing element in array. (See here https://opensource.apple.com/source/CF/CF-855.11/CFArray.h) However, this is unlikely to happen if the engineer does not abuse the usage of the class ```NSMutableArray``` according to at least two ios engineers. So here the complexity is set to match the normal expectation of the complexity. Reviewed By: ezgicicek Differential Revision: D22041277 fbshipit-source-id: c27f43167
5 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_add_object_constant, 8, OnUIThread:false, []
[cost] add inferbo model for NSArray.arrayByAddingObjectsFromArray, NSMutableArray.initWithCapacity Summary: As title Reviewed By: skcho Differential Revision: D23396445 fbshipit-source-id: 578496bc5
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_add_objects_from_array_linear, 9 + 3 ⋅ append_array->elements.length.ub + append_array->elements.length.ub + 3 ⋅ (append_array->elements.length.ub + 1), OnUIThread:false, [{append_array->elements.length.ub + 1},Loop,{append_array->elements.length.ub},Modeled call to NSArray.arrayByAddingObjectsFromArray:,{append_array->elements.length.ub},Loop]
[cost] Fix function name in model `arrayWithObjects:` Summary: Not sure what's going on. Reviewed By: ezgicicek Differential Revision: D24446585 fbshipit-source-id: 460bb119c
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_array_with_objects_constant, 47, OnUIThread:false, []
[cost] Refactor dispatch mechanism to be uniform across cost kinds Summary: Dispatch & function call mechanism was so jumbled up together. Let's refactor it to be cleaner. Reviewed By: skcho Differential Revision: D24049889 fbshipit-source-id: 42a218016
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_binary_search_log_FN, 10, OnUIThread:false, []
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_contains_object_linear, 3 + array->elements.length.ub, OnUIThread:false, [{array->elements.length.ub},Modeled call to NSArray.containsObject:]
[inferbo] Add naive models for mutableCopy and copy Reviewed By: skcho Differential Revision: D24218517 fbshipit-source-id: 0d26e6cab
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_copy_linear, 6 + 3 ⋅ array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Loop]
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_count_bounded_linear, 3 + 3 ⋅ array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Loop]
[cost] add inferbo model for NSArray Summary: As title Reviewed By: ezgicicek Differential Revision: D22390462 fbshipit-source-id: bd9157874
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_empty_array_constant, 8, OnUIThread:false, []
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_enumerator_linear, 6 + 8 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop]
codetoanalyze/objc/performance/NSArray.m, nsarray_enumerator_param_linear, 2 + 5 ⋅ (enumerator.length + 1), OnUIThread:false, [{enumerator.length + 1},Loop]
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_find_linear, 4 + 9 ⋅ array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Loop]
[cost] add objc test for NSArray and NSMutableArray Summary: Add objc test for ```NSArray``` and ```NSMutableArray```. ```NSMutableArray``` is a subclass of ```NSArray```. For documentation of ```NSArray```, https://developer.apple.com/documentation/foundation/nsarray?language=objc For documentation of ```NSMutableArray```, https://developer.apple.com/documentation/foundation/nsmutablearray?language=objc The underlying mechanism for ```NSMutableArray``` is quite complicated. It changes the underlying data structure during runtime, so it is possible to have say O(log n) complexity for accessing element in array. (See here https://opensource.apple.com/source/CF/CF-855.11/CFArray.h) However, this is unlikely to happen if the engineer does not abuse the usage of the class ```NSMutableArray``` according to at least two ios engineers. So here the complexity is set to match the normal expectation of the complexity. Reviewed By: ezgicicek Differential Revision: D22041277 fbshipit-source-id: c27f43167
5 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_first_object_constant, 4, OnUIThread:false, []
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_get_first_constant, 27, OnUIThread:false, []
[cost] add inferbo model for NSArray Summary: As title Reviewed By: ezgicicek Differential Revision: D22390462 fbshipit-source-id: bd9157874
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_init_constant, 9, OnUIThread:false, []
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_init_with_array_constant, 14, OnUIThread:false, []
codetoanalyze/objc/performance/NSArray.m, nsarray_init_with_array_copy_linear, 7 + 3 ⋅ array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Loop]
codetoanalyze/objc/performance/NSArray.m, nsarray_init_with_array_linear, 6 + 3 ⋅ array->elements.length.ub + array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Modeled call to NSArray.initWithArray:,{array->elements.length.ub},Loop]
[cost] add cost model for most common NSString functions Summary: Add cost model for most common `NSString` functions in cost analysis Reviewed By: skcho Differential Revision: D22433005 fbshipit-source-id: 2f57bbda9
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_init_with_objects_constant, 39, OnUIThread:false, []
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_is_equal_to_array_linear, 4 + array1->elements.length.ub, OnUIThread:false, [{array1->elements.length.ub},Modeled call to NSArray.isEqualToArray:]
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_iterate_linear, 6 + 8 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop]
[cost] add objc test for NSArray and NSMutableArray Summary: Add objc test for ```NSArray``` and ```NSMutableArray```. ```NSMutableArray``` is a subclass of ```NSArray```. For documentation of ```NSArray```, https://developer.apple.com/documentation/foundation/nsarray?language=objc For documentation of ```NSMutableArray```, https://developer.apple.com/documentation/foundation/nsmutablearray?language=objc The underlying mechanism for ```NSMutableArray``` is quite complicated. It changes the underlying data structure during runtime, so it is possible to have say O(log n) complexity for accessing element in array. (See here https://opensource.apple.com/source/CF/CF-855.11/CFArray.h) However, this is unlikely to happen if the engineer does not abuse the usage of the class ```NSMutableArray``` according to at least two ios engineers. So here the complexity is set to match the normal expectation of the complexity. Reviewed By: ezgicicek Differential Revision: D22041277 fbshipit-source-id: c27f43167
5 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_last_object_constant, 4, OnUIThread:false, []
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_next_object_constant, 26, OnUIThread:false, []
codetoanalyze/objc/performance/NSArray.m, nsarray_next_object_linear, 5 + 5 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop]
[cost] Add NSArray.arrayWithObjects: and array access function Summary: As title Reviewed By: skcho Differential Revision: D22502619 fbshipit-source-id: 3af885170
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_object_at_indexed_constant, 34, OnUIThread:false, []
[cost] refactor cost model and add NSArray.sort cost model. Summary: As title Reviewed By: skcho Differential Revision: D23372082 fbshipit-source-id: 6668e3cdd
4 years ago
codetoanalyze/objc/performance/NSArray.m, nsarray_sort_using_descriptors_constant, 39, OnUIThread:false, []
codetoanalyze/objc/performance/NSArray.m, nsarray_sort_using_descriptors_nlogn, 9 + array->elements.length.ub × log(array->elements.length.ub), OnUIThread:false, [{array->elements.length.ub},Modeled call to NSArray.sortedArrayUsingDescriptors:,{array->elements.length.ub},Modeled call to NSArray.sortedArrayUsingDescriptors:]
[cost] add model for iterating over collections via block Summary: Namely `enumerateObjectsUsingBlock` which takes a collection and a block as an arg and the iterates over the collection and applies the block to each element of the collection. This is a common way to iterate over Objc collections, so let's add a model for it. Reviewed By: skcho Differential Revision: D24604590 fbshipit-source-id: 1ceeb4b40
4 years ago
codetoanalyze/objc/performance/NSArray.m, objc_blockMyBlock.call_enumerate_via_block_param_linear_FN::_3, 3 + 3 ⋅ size + 2 ⋅ (1+max(0, size)), OnUIThread:false, [{1+max(0, size)},Loop,{size},Loop]
codetoanalyze/objc/performance/NSArray.m, objc_blockenumerate_via_block_linear_2, 1, OnUIThread:false, []
[cost] Report with relative paths on xcode_isysroot in tests Summary: This diff reports paths under the xcode isysroot as relative in tests. This was a problem when another machine that has a different isysroot directory is running the test. Reviewed By: ezgicicek Differential Revision: D23729222 fbshipit-source-id: 4e9681f65
4 years ago
codetoanalyze/objc/performance/NSArray.m, objc_blocknsarray_binary_search_log_FN_1, 5, OnUIThread:false, []
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_all_keys_linear1, 3 + 3 ⋅ dict->elements.length.ub + 4 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop,{dict->elements.length.ub},Loop]
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_all_keys_linear2, 6 + 3 ⋅ dict->elements.length.ub + 3 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop,{dict->elements.length.ub},Loop]
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_all_values_linear, 3 + 3 ⋅ dict->elements.length.ub + 4 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop,{dict->elements.length.ub},Loop]
[frontend] Treat allocWithZone: same as alloc Summary: `allocWithZone` exists for historical reasons; memory zones are no longer used by Objective-C. Therefore, we treat it like `alloc`. https://developer.apple.com/documentation/objectivec/nsobject/1571945-allocwithzone Reviewed By: skcho Differential Revision: D24137019 fbshipit-source-id: aac321322
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_alloc_with_zone_init_with_dictionary_linear, 7 + 3 ⋅ dict->elements.length.ub + 4 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop,{dict->elements.length.ub},Loop]
[inferbo] Add model for NSDictionary.dictionary Reviewed By: roro47 Differential Revision: D23396825 fbshipit-source-id: 3d99c5e74
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_dictionary_constant, 13, OnUIThread:false, []
[cost] add inferbo model for most common function of NSDictionary Summary: As title Reviewed By: skcho Differential Revision: D22572603 fbshipit-source-id: d9871cc3f
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_dictionary_with_objects_linear, 14 + 15 ⋅ n_entries + 3 ⋅ n_entries + 2 ⋅ (1+max(0, n_entries)) + 4 ⋅ (1+max(0, n_entries)), OnUIThread:false, [{1+max(0, n_entries)},Loop,{1+max(0, n_entries)},Loop,{n_entries},Loop,{n_entries},Loop]
[frontend] Add translation of dictionaryWithObjects:forKeys:count: Summary: This diff adds translation of `dictionaryWithObjects:forKeys:count:`. In the previous implementation it was translated as if it was `dictionaryWtihObjectsAndKeys:`, but their function parameters are different. In this diff, it translates an array literal `NSDictionary* a = @ [ @"firstName": @"Foo", @"lastName":@"Bar" ];` to ``` n$1=NSString.stringWithUTF8:(@"firstName") n$2=NSNumber.stringWithUTF8:(@"Foo") n$3=NSNumber.stringWithUTF8:(@"lastName") n$4=NSNumber.stringWithUTF8:(@"Bar") temp1[0]:objc_object*=n$1 temp1[1]:objc_object*=n$3 temp2[0]:objc_object*=n$2 temp2[1]:objc_object*=n$4 n$3=NSDictionary.dictionaryWithObjects:forKeys:count:(temp2:objc_object* const [2*8], temp1:objc_object*const [2*8], 2:int) ``` where `temp` is an additional local variable declared as array. See, https://developer.apple.com/documentation/foundation/nsdictionary/1574184-dictionarywithobjects?language=objc https://developer.apple.com/documentation/foundation/nsdictionary/1574181-dictionarywithobjectsandkeys {F316854542} Reviewed By: skcho Differential Revision: D23447042 fbshipit-source-id: 14b7c3f2b
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_enumerate_call_constant, 65, OnUIThread:false, []
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_enumerate_constant, 74, OnUIThread:false, []
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_enumerator_linear, 5 + 5 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop]
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_fast_enumerate_linear, 5 + 5 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop]
[cost] add iterator support for nsdictionary Summary: As title. Reviewed By: ezgicicek Differential Revision: D23371818 fbshipit-source-id: d894c956f
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_find_key_constant, 19, OnUIThread:false, []
[cost] add objc test for NSDictionary and NSMutableDictionary Summary: Add test for ```NSDictionary``` and ```NSMutableDictionary```. ```NSMutableDictionary``` is a subclass of ```NSDictionary```. [Doc] ```NSDictionary``` https://developer.apple.com/documentation/foundation/nsdictionary?language=objc ```NSMutableDictionary``` https://developer.apple.com/documentation/foundation/nsmutabledictionary?language=objc Reviewed By: skcho Differential Revision: D22041942 fbshipit-source-id: d1a1aa781
5 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_init_dictionary_constant, 3, OnUIThread:false, []
[frontend] Add translation of dictionaryWithObjects:forKeys:count: Summary: This diff adds translation of `dictionaryWithObjects:forKeys:count:`. In the previous implementation it was translated as if it was `dictionaryWtihObjectsAndKeys:`, but their function parameters are different. In this diff, it translates an array literal `NSDictionary* a = @ [ @"firstName": @"Foo", @"lastName":@"Bar" ];` to ``` n$1=NSString.stringWithUTF8:(@"firstName") n$2=NSNumber.stringWithUTF8:(@"Foo") n$3=NSNumber.stringWithUTF8:(@"lastName") n$4=NSNumber.stringWithUTF8:(@"Bar") temp1[0]:objc_object*=n$1 temp1[1]:objc_object*=n$3 temp2[0]:objc_object*=n$2 temp2[1]:objc_object*=n$4 n$3=NSDictionary.dictionaryWithObjects:forKeys:count:(temp2:objc_object* const [2*8], temp1:objc_object*const [2*8], 2:int) ``` where `temp` is an additional local variable declared as array. See, https://developer.apple.com/documentation/foundation/nsdictionary/1574184-dictionarywithobjects?language=objc https://developer.apple.com/documentation/foundation/nsdictionary/1574181-dictionarywithobjectsandkeys {F316854542} Reviewed By: skcho Differential Revision: D23447042 fbshipit-source-id: 14b7c3f2b
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_init_literal_constant, 45, OnUIThread:false, []
[inferbo] Add model for NSDictionary.initWithDictionary Summary: Model it similar to `NSArray.initWithArray` as copying from the given dictionary elements. Removes a FP as expected. Reviewed By: ngorogiannis Differential Revision: D24136868 fbshipit-source-id: ed31c3c8f
4 years ago
codetoanalyze/objc/performance/NSDictionary.m, nsdictionary_init_with_dictionary_linear, 7 + 3 ⋅ dict->elements.length.ub + 4 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop,{dict->elements.length.ub},Loop]
[inferbo] Add model for NSFastEnumeration Reviewed By: ngorogiannis Differential Revision: D24562644 fbshipit-source-id: 71e589883
4 years ago
codetoanalyze/objc/performance/NSEnumeration.m, call_loop_enumerator_constant, 35, OnUIThread:false, []
[frontend] Track protocols in A<ObjCProtocol> Reviewed By: skcho Differential Revision: D24566435 fbshipit-source-id: 62d41c8c6
4 years ago
codetoanalyze/objc/performance/NSEnumeration.m, call_loop_id_enumerator_linear, 7 + 5 ⋅ (arr->elements.length.ub + 1), OnUIThread:false, [{arr->elements.length.ub + 1},Call to loop_id_enumerator_linear,Loop]
[inferbo] Add model for NSFastEnumeration Reviewed By: ngorogiannis Differential Revision: D24562644 fbshipit-source-id: 71e589883
4 years ago
codetoanalyze/objc/performance/NSEnumeration.m, loop_enumerator_linear, 5 + 5 ⋅ (enumeration->elements.length.ub + 1), OnUIThread:false, [{enumeration->elements.length.ub + 1},Loop]
[frontend] Track protocols in A<ObjCProtocol> Reviewed By: skcho Differential Revision: D24566435 fbshipit-source-id: 62d41c8c6
4 years ago
codetoanalyze/objc/performance/NSEnumeration.m, loop_enumerator_over_class_linear, 5 + 5 ⋅ (a->elements.length.ub + 1), OnUIThread:false, [{a->elements.length.ub + 1},Loop]
codetoanalyze/objc/performance/NSEnumeration.m, loop_id_enumerator_linear, 5 + 5 ⋅ (enumeration->elements.length.ub + 1), OnUIThread:false, [{enumeration->elements.length.ub + 1},Loop]
[cost] Add model: NSFileManager.contentsOfDirectoryAtURL Summary: This diff adds a model for NSFileManager.contentsOfDirectoryAtURL as returning a constant-length collection. The analyzer cannot know files in a directory. We have some options to handle such unknown data. 1. Use `Unknown` value, ie `top` 2. Use a symbolic value 3. Use a constant value We had been used the first option. An upside of this is that the analyzer can remain as sound. However, a downside of this is the top value can be propagated to other procedures, making their costs top, thus we may miss some cost changes of them. The second option is to introduce a symbolic value, ie. that for the number of files. A problem is that the symbolic value will never be concretized. As a result, the symbol can be propagated to other procedures, increasing the coefficient of the complexity or making top costs. Note that handling multiple symbols is somewhat limited in Inferbo's interval domain. The last option is to introduce a constant value. I think this is the best approach we can take among above. Even though we may have FNs when there are a lot of files in a directory, we cannot reason or expect about that at the analysis time anyway. Reviewed By: ezgicicek Differential Revision: D24418099 fbshipit-source-id: bf8cf3538
4 years ago
codetoanalyze/objc/performance/NSFileManager.m, contents_of_directory_at_url_constant, 11, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/NSInteger.m, nsinteger_value_linear, 3 + 3 ⋅ integer + 2 ⋅ (1+max(0, integer)), OnUIThread:false, [{1+max(0, integer)},Loop,{integer},Loop]
[cost] add inferbo model for NSNumber Summary: As title Reviewed By: ezgicicek Differential Revision: D22411709 fbshipit-source-id: 994946a85
4 years ago
codetoanalyze/objc/performance/NSInteger.m, nsnumber_number_with_int_integer_value_constant, 34, OnUIThread:false, []
[cost] Add model: NSKeyedUnarchiver.decodeObjectForKey: Reviewed By: ezgicicek Differential Revision: D24417784 fbshipit-source-id: f440ba018
4 years ago
codetoanalyze/objc/performance/NSKeyedUnarchiver.m, decode_object_for_key_linear, 9 + 3 ⋅ keyed_unarchiver->elements[*].ub + 2 ⋅ (1+max(0, keyed_unarchiver->elements[*].ub)), OnUIThread:false, [{1+max(0, keyed_unarchiver->elements[*].ub)},Loop,{keyed_unarchiver->elements[*].ub},Loop]
[inferbo] Generalize init's model to NSObject Reviewed By: roro47 Differential Revision: D23348725 fbshipit-source-id: 666472181
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsarray_new_constant, 9, OnUIThread:false, []
[cost] add NSMutableArray's add and remove function model Summary: As title Reviewed By: ezgicicek Differential Revision: D22975158 fbshipit-source-id: de4217a32
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_add_all_constant, 23, OnUIThread:false, []
[cost] make the size of collection more accurate when modifying in loop Summary: Two issues are fixed 1. For this diff, when the condition `curr_langauge_is Java` is removed in some part of the analysis, some c bufferoverrun tests are broken. This is fixed in this diff by inspecting if we are dealing with a `Size` alias referring to an `objc_internal_collection_array`. 2. Previously, when we are modifying mutable array in a loop, e.g. adding element to the array or removing element from the array, we are unable to give an estimable size of the array after exiting the loop. This is now fixed, and the corresponding FPs are resolved. Reviewed By: skcho Differential Revision: D23342350 fbshipit-source-id: 200f5261c
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_add_in_loop_constant, 152, OnUIThread:false, []
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_add_in_loop_linear, 8 + 7 ⋅ n + 3 ⋅ n + 2 ⋅ (n + 1) + 3 ⋅ (n + 1), OnUIThread:false, [{n + 1},Loop,{n + 1},Loop,{n},Loop,{n},Loop]
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_add_in_loop_quadratic, 6 + 5 ⋅ n + 8 ⋅ n × m + 2 ⋅ n × (m + 1) + 2 ⋅ (n + 1), OnUIThread:false, [{n + 1},Loop,{m + 1},Loop,{m},Loop,{n},Loop]
[cost] add NSMutableArray's add and remove function model Summary: As title Reviewed By: ezgicicek Differential Revision: D22975158 fbshipit-source-id: de4217a32
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_add_then_loop_constant, 108, OnUIThread:false, []
[cost] add objc test for NSArray and NSMutableArray Summary: Add objc test for ```NSArray``` and ```NSMutableArray```. ```NSMutableArray``` is a subclass of ```NSArray```. For documentation of ```NSArray```, https://developer.apple.com/documentation/foundation/nsarray?language=objc For documentation of ```NSMutableArray```, https://developer.apple.com/documentation/foundation/nsmutablearray?language=objc The underlying mechanism for ```NSMutableArray``` is quite complicated. It changes the underlying data structure during runtime, so it is possible to have say O(log n) complexity for accessing element in array. (See here https://opensource.apple.com/source/CF/CF-855.11/CFArray.h) However, this is unlikely to happen if the engineer does not abuse the usage of the class ```NSMutableArray``` according to at least two ios engineers. So here the complexity is set to match the normal expectation of the complexity. Reviewed By: ezgicicek Differential Revision: D22041277 fbshipit-source-id: c27f43167
5 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_empty_ok_costant, 7, OnUIThread:false, []
[cost] add inferbo model for NSArray.arrayByAddingObjectsFromArray, NSMutableArray.initWithCapacity Summary: As title Reviewed By: skcho Differential Revision: D23396445 fbshipit-source-id: 578496bc5
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_init_with_capacity_constant, 4617, OnUIThread:false, []
[cost] add NSMutableArray's add and remove function model Summary: As title Reviewed By: ezgicicek Differential Revision: D22975158 fbshipit-source-id: de4217a32
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_remove_all_linear, 4 + 3 ⋅ array->elements.length.ub + array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Modeled call to NSArray.removeAllObjects,{array->elements.length.ub},Loop]
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_remove_constant, 17, OnUIThread:false, []
[cost] make the size of collection more accurate when modifying in loop Summary: Two issues are fixed 1. For this diff, when the condition `curr_langauge_is Java` is removed in some part of the analysis, some c bufferoverrun tests are broken. This is fixed in this diff by inspecting if we are dealing with a `Size` alias referring to an `objc_internal_collection_array`. 2. Previously, when we are modifying mutable array in a loop, e.g. adding element to the array or removing element from the array, we are unable to give an estimable size of the array after exiting the loop. This is now fixed, and the corresponding FPs are resolved. Reviewed By: skcho Differential Revision: D23342350 fbshipit-source-id: 200f5261c
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_remove_in_loop_constant, 182, OnUIThread:false, []
[cost] add NSMutableArray's add and remove function model Summary: As title Reviewed By: ezgicicek Differential Revision: D22975158 fbshipit-source-id: de4217a32
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_set_in_loop_constant, 51, OnUIThread:false, []
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/NSMutableArray.m, nsmarray_set_linear, 3 + 11 ⋅ array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Loop]
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/NSMutableDictionary.m, nsmutabledictionary_set_element_in_loop_linear, 5 + 9 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop]
[inferbo] Add naive models for mutableCopy and copy Reviewed By: skcho Differential Revision: D24218517 fbshipit-source-id: 0d26e6cab
4 years ago
codetoanalyze/objc/performance/NSMutableString.m, copy_string_constant_FP, 11 + 3 ⋅ str3.length.ub + str3.length.ub + 3 ⋅ (str3.length.ub + 1), OnUIThread:false, [{str3.length.ub + 1},Loop,{str3.length.ub},Modeled call to NSMutableString.appendString:,{str3.length.ub},Loop]
[inferbo][cost] Check supertype in ObjC models Summary: We check for supertypes in Java. Why not ObjC? Would be good to get dulmarod's input here. Reviewed By: roro47 Differential Revision: D22817126 fbshipit-source-id: 52c1c3f3c
4 years ago
codetoanalyze/objc/performance/NSMutableString.m, nsmstring_append_string_constant, 14 + 3 ⋅ (str.length.ub + 5) + 3 ⋅ (str.length.ub + 6), OnUIThread:false, [{str.length.ub + 6},Loop,{str.length.ub + 5},Loop]
codetoanalyze/objc/performance/NSMutableString.m, nsmstring_append_string_linear, 5 + str2.length.ub + 3 ⋅ (str2.length.ub + str1.length.ub) + 3 ⋅ (str2.length.ub + str1.length.ub + 1), OnUIThread:false, [{str2.length.ub + str1.length.ub + 1},Loop,{str2.length.ub + str1.length.ub},Loop,{str2.length.ub},Modeled call to NSMutableString.appendString:]
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/NSOrderedSet.m, nsordered_set_empty_constant, 14, OnUIThread:false, []
codetoanalyze/objc/performance/NSOrderedSet.m, nsordered_set_iterate_linear, 5 + 5 ⋅ (ordered_set->elements.length.ub + 1), OnUIThread:false, [{ordered_set->elements.length.ub + 1},Loop]
codetoanalyze/objc/performance/NSOrderedSet.m, nsordered_set_with_array_linear, 10 + 5 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Call to nsordered_set_iterate_linear,Loop]
codetoanalyze/objc/performance/NSSet.m, nsset_enumerator_linear, 6 + 8 ⋅ (set->elements.length.ub + 1), OnUIThread:false, [{set->elements.length.ub + 1},Loop]
[inferbo] Model NSSet/NSOrderedSet as collections Reviewed By: skcho Differential Revision: D24138823 fbshipit-source-id: 7e8eddf34
4 years ago
codetoanalyze/objc/performance/NSSet.m, nsset_init_constant, 9, OnUIThread:false, []
codetoanalyze/objc/performance/NSSet.m, nsset_init_with_array_linear, 7 + 3 ⋅ array->elements.length.ub + 3 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop,{array->elements.length.ub},Loop]
codetoanalyze/objc/performance/NSSet.m, nsset_init_with_set_constant, 6, OnUIThread:false, []
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/NSSet.m, nsset_iterate_linear, 6 + 8 ⋅ (set->elements.length.ub + 1), OnUIThread:false, [{set->elements.length.ub + 1},Loop]
codetoanalyze/objc/performance/NSSet.m, nsset_next_object_linear, 5 + 5 ⋅ (set->elements.length.ub + 1), OnUIThread:false, [{set->elements.length.ub + 1},Loop]
[frontend] Revise translation of message expression in ObjC frontend Summary: This diff revises the translation of message expression's arguments in ObjC frontend. In the frontend, it massages the arguments when calling a static method, so the class or object value is not given to the static method as the first parameter. The problem is that it used a raise-exception-and-catch way to detect where we remove the first parameter. This way of using an exception is not only hard to understand, but also incorrectly removed the first parameter, with breaking abstract semantics sometimes. (See the added test.) This diff avoids using the exception. Reviewed By: jvillard Differential Revision: D24565513 fbshipit-source-id: 0a84ca394
4 years ago
codetoanalyze/objc/performance/NSString.m, DummyClass.call_string_by_appending_string_constant_FP, ⊤, OnUIThread:false, [Unbounded loop,Modeled call to NSString.stringByAppendingString:]
codetoanalyze/objc/performance/NSString.m, DummyClass.dealloc, 1, OnUIThread:false, []
[cost] add more inferbo and cost analysis models for NSString Summary: As title Reviewed By: skcho Differential Revision: D22663505 fbshipit-source-id: 61ee2d346
4 years ago
codetoanalyze/objc/performance/NSString.m, call_component_separated_by_char_constant, 46, OnUIThread:false, []
[cost] add inferbo model for init function of NSString and NSArray Summary: As title Reviewed By: ezgicicek Differential Revision: D22844475 fbshipit-source-id: 0eb757bb0
4 years ago
codetoanalyze/objc/performance/NSString.m, call_init_with_string_constant, 15, OnUIThread:false, []
[cost] add more inferbo and cost analysis models for NSString Summary: As title Reviewed By: skcho Differential Revision: D22663505 fbshipit-source-id: 61ee2d346
4 years ago
codetoanalyze/objc/performance/NSString.m, component_seperated_by_char_linear, 6 + m.length.ub + 3 ⋅ (-1+max(2, m.length.ub)) + 3 ⋅ (max(2, m.length.ub)), OnUIThread:false, [{max(2, m.length.ub)},Loop,{-1+max(2, m.length.ub)},Loop,{m.length.ub},Modeled call to NSString.componentsSeparatedByString:]
codetoanalyze/objc/performance/NSString.m, component_seperated_by_string_linear, 6 + sep.length.ub × m.length.ub + 3 ⋅ (-1+max(2, m.length.ub)) + 3 ⋅ (max(2, m.length.ub)), OnUIThread:false, [{max(2, m.length.ub)},Loop,{-1+max(2, m.length.ub)},Loop,{m.length.ub},Modeled call to NSString.componentsSeparatedByString:,{sep.length.ub},Modeled call to NSString.componentsSeparatedByString:]
[clang] fix bad interaction between ConditionalOperator and initializers Summary: This is several inter-connected changes together to keep the tests happy. The ConditionalOperator `b?t:e` is translated by first creating a placeholder variable to temporarily store the result of the evaluation in each branch, then the real thing we want to assign to reads that variable. But, there are situations where that changes the semantics of the expression, namely when the value created is a struct on the stack (eg, a C++ temporary). This is because in SIL we cannot assign the *address* of a program variable, only its contents, so by the time we're out of the conditional operator we cannot set the struct value correctly anymore: we can only set its content, which we did, but that results in a "shifted" struct value that is one dereference away from where it should be. So a batch of changes concern `conditionalOperator_trans`: - instead of systematically creating a temporary for the conditional, use the `trans_state.var_exp_typ` provided from above if available when translating `ConditionalOperator` - don't even set anything if that variable was already initialized by merely translating the branch expression, eg when it's a constructor - fix long-standing TODO to propagate these initialization facts accurately for ConditionalOperator (used by `init_expr_trans` to also figure out if it should insert a store to the variable being initialised or not) The rest of the changes adapt some relevant other constructs to deal with conditionalOperator properly now that it can set the current variable itself, instead of storing stuff inside a temp variable. This change was a problem because some constructs, eg a variable declaration, will insert nodes that set up the variable before calling its initialization, and now the initialization happens *before* that setup, in the translation of the inner conditional operator, which naturally creates nodes above the current one. - add a generic helper to force a sequential order between two translation results, forcing node creation if necessary - use that in `init_expr_trans` and `cxxNewExpr_trans` - adjust many places where `var_exp_typ` was incorrectly not reset when translating sub-expressions The sequentiality business creates more nodes when used, and the conditionalOperator business uses fewer temporary variables, so the frontend results change quite a bit. Note that biabduction tests were invaluable in debugging this. There could be other constructs to adjust similarly to cxxNewExpr that were not covered by the tests though. Added tests in pulse that exercises the previous bug. Reviewed By: da319 Differential Revision: D24796282 fbshipit-source-id: 0790c8d17
4 years ago
codetoanalyze/objc/performance/NSString.m, has_prefix_constant, 11, OnUIThread:false, []
[cost] add alloc for NSObject and NSString.init as an example function using alloc Summary: Implement `alloc` and implement initialisation method that uses the return value of `alloc`, i.e. `NSString.init`. Reviewed By: ezgicicek Differential Revision: D22840080 fbshipit-source-id: 47a7523e3
4 years ago
codetoanalyze/objc/performance/NSString.m, init_string_constant, 9, OnUIThread:false, []
[cost] add inferbo model for init function of NSString and NSArray Summary: As title Reviewed By: ezgicicek Differential Revision: D22844475 fbshipit-source-id: 0eb757bb0
4 years ago
codetoanalyze/objc/performance/NSString.m, init_with_bytes_linear, 9 + 3 ⋅ length + 3 ⋅ (length + 1), OnUIThread:false, [{length + 1},Loop,{length},Loop]
codetoanalyze/objc/performance/NSString.m, init_with_string_constant, 39, OnUIThread:false, []
codetoanalyze/objc/performance/NSString.m, init_with_string_linear, 7 + 3 ⋅ s.length.ub + 3 ⋅ (s.length.ub + 1), OnUIThread:false, [{s.length.ub + 1},Loop,{s.length.ub},Loop]
[cost] add more inferbo and cost analysis models for NSString Summary: As title Reviewed By: skcho Differential Revision: D22663505 fbshipit-source-id: 61ee2d346
4 years ago
codetoanalyze/objc/performance/NSString.m, rangeof_character_from_set_linear, 5 + m.length.ub, OnUIThread:false, [{m.length.ub},Modeled call to NSString.rangeOfString:]
codetoanalyze/objc/performance/NSString.m, rangeof_string_quadratic, 5 + n.length.ub × m.length.ub, OnUIThread:false, [{m.length.ub},Modeled call to NSString.rangeOfString:,{n.length.ub},Modeled call to NSString.rangeOfString:]
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/NSString.m, replace_linear_FP, ⊤, OnUIThread:false, [Unbounded loop,Loop]
[cost] add cost model for most common NSString functions Summary: Add cost model for most common `NSString` functions in cost analysis Reviewed By: skcho Differential Revision: D22433005 fbshipit-source-id: 2f57bbda9
4 years ago
codetoanalyze/objc/performance/NSString.m, string_by_appending_path_component_linear, 4 + path.length.ub + file.length.ub, OnUIThread:false, [{file.length.ub},Modeled call to NSString.stringByAppendingPathComponent:,{path.length.ub},Modeled call to NSString.stringByAppendingPathComponent:]
[cost] add more inferbo and cost analysis models for NSString Summary: As title Reviewed By: skcho Differential Revision: D22663505 fbshipit-source-id: 61ee2d346
4 years ago
codetoanalyze/objc/performance/NSString.m, string_by_appending_same_string_linear, 11 + s.length.ub + 3 ⋅ (s.length.ub + 2) + 3 ⋅ (s.length.ub + 3), OnUIThread:false, [{s.length.ub + 3},Loop,{s.length.ub + 2},Loop,{s.length.ub},Modeled call to NSString.stringByAppendingString:]
codetoanalyze/objc/performance/NSString.m, string_by_appending_string_linear, 8 + m.length.ub + 3 ⋅ (m.length.ub + s.length.ub) + s.length.ub + 3 ⋅ (m.length.ub + s.length.ub + 1), OnUIThread:false, [{m.length.ub + s.length.ub + 1},Loop,{s.length.ub},Modeled call to NSString.stringByAppendingString:,{m.length.ub + s.length.ub},Loop,{m.length.ub},Modeled call to NSString.stringByAppendingString:]
[cost] add cost model for most common NSString functions Summary: Add cost model for most common `NSString` functions in cost analysis Reviewed By: skcho Differential Revision: D22433005 fbshipit-source-id: 2f57bbda9
4 years ago
codetoanalyze/objc/performance/NSString.m, string_has_prefix_linear, 4 + str.length.ub, OnUIThread:false, [{str.length.ub},Modeled call to NSString.hasPrefix:]
codetoanalyze/objc/performance/NSString.m, string_is_equal_to_string_linear, 4 + str1.length.ub, OnUIThread:false, [{str1.length.ub},Modeled call to NSString.isEqualToString:]
[cost] add model for NSString.string_with_utf8_string_linear and NSString.length Summary: As title Model `NSString` as `JavaString`. Since `NSArray` does not contain information about its type of element, we do not use associate string with collection as in Java and C++. In Java, String model is implemented using java collection, and for C++, string model is implemented using vector. So instead, we use existing `JavaString` model. Reviewed By: skcho Differential Revision: D22431949 fbshipit-source-id: 7cdde1ad7
4 years ago
codetoanalyze/objc/performance/NSString.m, string_length_linear, 3 + 3 ⋅ s.length.ub + 4 ⋅ (s.length.ub + 1), OnUIThread:false, [{s.length.ub + 1},Loop,{s.length.ub},Loop]
[cost] add cost model for most common NSString functions Summary: Add cost model for most common `NSString` functions in cost analysis Reviewed By: skcho Differential Revision: D22433005 fbshipit-source-id: 2f57bbda9
4 years ago
codetoanalyze/objc/performance/NSString.m, string_with_utf8_string_linear, 7 + 3 ⋅ p->strlen.ub + p->strlen.ub + 4 ⋅ (p->strlen.ub + 1), OnUIThread:false, [{p->strlen.ub + 1},Loop,{p->strlen.ub},Modeled call to NSString.stringWithUTF8String:,{p->strlen.ub},Loop]
[cost] add more inferbo and cost analysis models for NSString Summary: As title Reviewed By: skcho Differential Revision: D22663505 fbshipit-source-id: 61ee2d346
4 years ago
codetoanalyze/objc/performance/NSString.m, substring_no_end_linear, 10 + 6 ⋅ (-x + s.length.ub), OnUIThread:false, [{-x + s.length.ub},Loop]
[cost] Add models: NSURL.getResourceValue, path Reviewed By: ezgicicek Differential Revision: D24417544 fbshipit-source-id: 96249dd14
4 years ago
codetoanalyze/objc/performance/NSURL.m, get_resource_value_constant, 16, OnUIThread:false, []
codetoanalyze/objc/performance/NSURL.m, path_linear, 6 + 3 ⋅ url.length.ub + 3 ⋅ (url.length.ub + 1), OnUIThread:false, [{url.length.ub + 1},Loop,{url.length.ub},Loop]
[clang] Add implicit implementations for ObjC getters and setters as a preanalysis Summary: Move the implementation of implicit getters and setters from the biabduction to the clang frontend so these methods are accessible to all the checkers. *Background*: In Objective-C when properties are created in the interface of a class, the compiler creates automatically the instance variable for it and also the getter and setter in the implementation of the class. In the frontend we collect the information about which method is the implicit getter and setter of which instance variable (we get the method declaration but not the implementation), and here we add the implicit implementation. Reviewed By: skcho Differential Revision: D22187238 fbshipit-source-id: 76e0508ed
4 years ago
codetoanalyze/objc/performance/araii.m, Araii.buffer, 4, OnUIThread:false, []
[cost] Record cost traces in cost-report.json 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
5 years ago
codetoanalyze/objc/performance/araii.m, Araii.dealloc, 4, OnUIThread:false, []
codetoanalyze/objc/performance/araii.m, Araii.initWithBuffer, 15, OnUIThread:false, []
[clang] Add implicit implementations for ObjC getters and setters as a preanalysis Summary: Move the implementation of implicit getters and setters from the biabduction to the clang frontend so these methods are accessible to all the checkers. *Background*: In Objective-C when properties are created in the interface of a class, the compiler creates automatically the instance variable for it and also the getter and setter in the implementation of the class. In the frontend we collect the information about which method is the implicit getter and setter of which instance variable (we get the method declaration but not the implementation), and here we add the implicit implementation. Reviewed By: skcho Differential Revision: D22187238 fbshipit-source-id: 76e0508ed
4 years ago
codetoanalyze/objc/performance/araii.m, Araii.setBuffer:, 4, OnUIThread:false, []
[cost] Record cost traces in cost-report.json 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
5 years ago
codetoanalyze/objc/performance/araii.m, memory_leak_raii_main, 18, OnUIThread:false, []
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/block.m, block_multiply_array_linear, 13 + 9 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Call to objc_blockblock_multiply_array_linear_1,Loop]
[cost] Support closure symbols in operation/allocation cost kinds Summary: This diff adds closure symbols to operation/allocation costs, when function pointer is called. Reviewed By: ezgicicek Differential Revision: D24308550 fbshipit-source-id: 6c5889d41
4 years ago
codetoanalyze/objc/performance/block.m, call_wrapper_runBlockA_linear, 11 + 3 ⋅ a + 2 ⋅ (1+max(0, a)), OnUIThread:false, [{1+max(0, a)},Call to loop_linear,Loop,{a},Call to loop_linear,Loop]
[closures preanalysis] [3/n] Adding models to dispatch functions and remove code for previous specialization and dispatch models Summary: This diff finishes the migration from the specialization of methods that take blocks as arguments. Here we delete all the old code and change the way we model dispatch functions so that the tests pass. - Remove the code for specializing the methods in biabduction. - Remove the call flags `cf_with_block_parameters` that was only used in this algorithm. - Removes models for dispatch functions. - Adds models for dispatch functions as program transformation only in biabduction. To be added in other checkers in the future. Reviewed By: ngorogiannis Differential Revision: D23345342 fbshipit-source-id: b5e8542ce
4 years ago
codetoanalyze/objc/performance/block.m, doBlockA_direct_block_linear, 10 + 3 ⋅ a + 2 ⋅ (1+max(0, a)), OnUIThread:false, [{1+max(0, a)},Call to runBlockA[objc_blockdoBlockA_direct_block_linear_3],Call to objc_blockdoBlockA_direct_block_linear_3,Call to loop_linear,Loop,{a},Call to runBlockA[objc_blockdoBlockA_direct_block_linear_3],Call to objc_blockdoBlockA_direct_block_linear_3,Call to loop_linear,Loop]
[closure] Substitute closure parameter when given via variable Summary: This diff substitutes closure parameter when it is given via variable. For example, ``` x = ^{ ... } foo(x) ``` this diff substitutes the closure variable `x`, ``` x = ^{ ... } foo(^{ ... }) ``` so that the specialization of `foo` can be done by `CCallSpecializaedWithClosures.process`. Reviewed By: jvillard Differential Revision: D23814595 fbshipit-source-id: a89f1530f
4 years ago
codetoanalyze/objc/performance/block.m, doBlockA_linear, 12 + 3 ⋅ a + 2 ⋅ (1+max(0, a)), OnUIThread:false, [{1+max(0, a)},Call to runBlockA[objc_blockdoBlockA_linear_2],Call to objc_blockdoBlockA_linear_2,Call to loop_linear,Loop,{a},Call to runBlockA[objc_blockdoBlockA_linear_2],Call to objc_blockdoBlockA_linear_2,Call to loop_linear,Loop]
[cost] Add FN test for functions that take blocks as arguments Reviewed By: skcho Differential Revision: D23345830 fbshipit-source-id: 9c5b1e0c8
4 years ago
codetoanalyze/objc/performance/block.m, loop_linear, 3 + 3 ⋅ x + 2 ⋅ (1+max(0, x)), OnUIThread:false, [{1+max(0, x)},Loop,{x},Loop]
[cost] Enable passing enumerator as paramter in ObjC Summary: This diff revises memory model of enumerator in ObjC to enable passing it as a parameter. The cost checker was not able to analyze a function precisely when it gets an enumerator as a parameter because the offset of an enumerator was available only when the analyzer knew the correct relation between the enumerator and an array. This diff simplifies the enumerator to have a similar value with `array->elements`, so its offset can be taken without the relation between enumerator and array to get them. Reviewed By: ezgicicek Differential Revision: D24446574 fbshipit-source-id: 27cdc051e
4 years ago
codetoanalyze/objc/performance/block.m, objc_blockblock_multiply_array_linear_1, 8 + 9 ⋅ (array->elements.length.ub + 1), OnUIThread:false, [{array->elements.length.ub + 1},Loop]
[cost] Support closure symbols in operation/allocation cost kinds Summary: This diff adds closure symbols to operation/allocation costs, when function pointer is called. Reviewed By: ezgicicek Differential Revision: D24308550 fbshipit-source-id: 6c5889d41
4 years ago
codetoanalyze/objc/performance/block.m, objc_blockcall_wrapper_runBlockA_linear_4, 5 + 3 ⋅ a + 2 ⋅ (1+max(0, a)), OnUIThread:false, [{1+max(0, a)},Call to loop_linear,Loop,{a},Call to loop_linear,Loop]
[closures preanalysis] [3/n] Adding models to dispatch functions and remove code for previous specialization and dispatch models Summary: This diff finishes the migration from the specialization of methods that take blocks as arguments. Here we delete all the old code and change the way we model dispatch functions so that the tests pass. - Remove the code for specializing the methods in biabduction. - Remove the call flags `cf_with_block_parameters` that was only used in this algorithm. - Removes models for dispatch functions. - Adds models for dispatch functions as program transformation only in biabduction. To be added in other checkers in the future. Reviewed By: ngorogiannis Differential Revision: D23345342 fbshipit-source-id: b5e8542ce
4 years ago
codetoanalyze/objc/performance/block.m, objc_blockdoBlockA_direct_block_linear_3, 5 + 3 ⋅ a + 2 ⋅ (1+max(0, a)), OnUIThread:false, [{1+max(0, a)},Call to loop_linear,Loop,{a},Call to loop_linear,Loop]
[closure] Substitute closure parameter when given via variable Summary: This diff substitutes closure parameter when it is given via variable. For example, ``` x = ^{ ... } foo(x) ``` this diff substitutes the closure variable `x`, ``` x = ^{ ... } foo(^{ ... }) ``` so that the specialization of `foo` can be done by `CCallSpecializaedWithClosures.process`. Reviewed By: jvillard Differential Revision: D23814595 fbshipit-source-id: a89f1530f
4 years ago
codetoanalyze/objc/performance/block.m, objc_blockdoBlockA_linear_2, 5 + 3 ⋅ a + 2 ⋅ (1+max(0, a)), OnUIThread:false, [{1+max(0, a)},Call to loop_linear,Loop,{a},Call to loop_linear,Loop]
[cost] Support closure symbols in operation/allocation cost kinds Summary: This diff adds closure symbols to operation/allocation costs, when function pointer is called. Reviewed By: ezgicicek Differential Revision: D24308550 fbshipit-source-id: 6c5889d41
4 years ago
codetoanalyze/objc/performance/block.m, runBlockA, 2 + |block|, OnUIThread:false, []
[closures preanalysis] [3/n] Adding models to dispatch functions and remove code for previous specialization and dispatch models Summary: This diff finishes the migration from the specialization of methods that take blocks as arguments. Here we delete all the old code and change the way we model dispatch functions so that the tests pass. - Remove the code for specializing the methods in biabduction. - Remove the call flags `cf_with_block_parameters` that was only used in this algorithm. - Removes models for dispatch functions. - Adds models for dispatch functions as program transformation only in biabduction. To be added in other checkers in the future. Reviewed By: ngorogiannis Differential Revision: D23345342 fbshipit-source-id: b5e8542ce
4 years ago
codetoanalyze/objc/performance/block.m, runBlockA[objc_blockdoBlockA_direct_block_linear_3], 8 + 3 ⋅ a[doBlockA_direct_block_linear()] + 2 ⋅ (1+max(0, a[doBlockA_direct_block_linear()])), OnUIThread:false, [{1+max(0, a[doBlockA_direct_block_linear()])},Call to objc_blockdoBlockA_direct_block_linear_3,Call to loop_linear,Loop,{a[doBlockA_direct_block_linear()]},Call to objc_blockdoBlockA_direct_block_linear_3,Call to loop_linear,Loop]
[closure] Substitute closure parameter when given via variable Summary: This diff substitutes closure parameter when it is given via variable. For example, ``` x = ^{ ... } foo(x) ``` this diff substitutes the closure variable `x`, ``` x = ^{ ... } foo(^{ ... }) ``` so that the specialization of `foo` can be done by `CCallSpecializaedWithClosures.process`. Reviewed By: jvillard Differential Revision: D23814595 fbshipit-source-id: a89f1530f
4 years ago
codetoanalyze/objc/performance/block.m, runBlockA[objc_blockdoBlockA_linear_2], 8 + 3 ⋅ a[doBlockA_linear()] + 2 ⋅ (1+max(0, a[doBlockA_linear()])), OnUIThread:false, [{1+max(0, a[doBlockA_linear()])},Call to objc_blockdoBlockA_linear_2,Call to loop_linear,Loop,{a[doBlockA_linear()]},Call to objc_blockdoBlockA_linear_2,Call to loop_linear,Loop]
[cost] Support closure symbols in operation/allocation cost kinds Summary: This diff adds closure symbols to operation/allocation costs, when function pointer is called. Reviewed By: ezgicicek Differential Revision: D24308550 fbshipit-source-id: 6c5889d41
4 years ago
codetoanalyze/objc/performance/block.m, wrapper_runBlockA, 4 + |block|, OnUIThread:false, []
codetoanalyze/objc/performance/block.m, wrapper_runBlockA[objc_blockcall_wrapper_runBlockA_linear_4], 4 + |block|, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/break.m, break_constant_FP, 8 + 5 ⋅ p + 2 ⋅ (1+max(0, p)), OnUIThread:false, [{1+max(0, p)},Call to break_loop,Loop,{p},Call to break_loop,Loop]
codetoanalyze/objc/performance/break.m, break_loop, 5 + 5 ⋅ p + 2 ⋅ (1+max(0, p)), OnUIThread:false, [{1+max(0, p)},Loop,{p},Loop]
codetoanalyze/objc/performance/break.m, break_loop_with_t, 7 + 5 ⋅ p + 2 ⋅ (1+max(0, p)), OnUIThread:false, [{1+max(0, p)},Loop,{p},Loop]
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/cf.m, array_count_linear, 6 + 3 ⋅ arr->elements.length.ub + 2 ⋅ (arr->elements.length.ub + 1), OnUIThread:false, [{arr->elements.length.ub + 1},Loop,{arr->elements.length.ub},Loop]
codetoanalyze/objc/performance/cf.m, cf_array_create_copy_linear, 8 + 3 ⋅ arrayValues.elements.length + 2 ⋅ (arrayValues.elements.length + 1), OnUIThread:false, [{arrayValues.elements.length + 1},Loop,{arrayValues.elements.length},Loop]
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/cf.m, cf_array_create_linear, 11 + 3 ⋅ x + 2 ⋅ (1+max(0, x)), OnUIThread:false, [{1+max(0, x)},Loop,{x},Loop]
[cost] write ObjC's NSCollection model separately and refactor Summary: When implementing iterator, we find out that because some semantics of inferbo is Java-specific, we cannot simply use Java's `Collection` model for `NSCollection`. So this diff writing `NSCollection` model separately. This diff also extracts the common parts of `NSCollection` and `Collection` into `AbstractCollection` to combine the duplicate the parts. Reviewed By: ezgicicek Differential Revision: D22975159 fbshipit-source-id: daed3f99f
4 years ago
codetoanalyze/objc/performance/cf.m, dict_count_linear, 6 + 3 ⋅ dict->elements.length.ub + 2 ⋅ (dict->elements.length.ub + 1), OnUIThread:false, [{dict->elements.length.ub + 1},Loop,{dict->elements.length.ub},Loop]
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/compound_loop_guard.m, compound_while, 7 + 3 ⋅ m + 4 ⋅ (1+max(0, m)), OnUIThread:false, [{1+max(0, m)},Loop,{m},Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/compound_loop_guard.m, nested_while_and_or_constant, 20, OnUIThread:false, []
codetoanalyze/objc/performance/compound_loop_guard.m, simplified_simulated_while_with_and_constant, 605, OnUIThread:false, []
codetoanalyze/objc/performance/compound_loop_guard.m, simulated_nested_loop_with_and_constant, 3529, OnUIThread:false, []
codetoanalyze/objc/performance/compound_loop_guard.m, simulated_while_shortcut_constant, 9, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/compound_loop_guard.m, simulated_while_with_and_linear, 6 + 3 ⋅ p + 4 ⋅ (1+max(0, p)), OnUIThread:false, [{1+max(0, p)},Loop,{p},Loop]
codetoanalyze/objc/performance/compound_loop_guard.m, while_and_or, ⊤, OnUIThread:false, [Unbounded loop,Loop]
[cost] Record cost traces in cost-report.json 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
5 years ago
codetoanalyze/objc/performance/control.m, __infer_globals_initializer_gvar, 2, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/control.m, wrong_cvar_FP, ⊤, OnUIThread:false, [Unbounded loop,Loop]
[preanalysis] Add a new preanalysis for ObjC's copy and mutableCopy Summary: In ObjC, `NSObject.copy` returns the object returned by `copyWithZone:` on the given class. This method must be implemented if the class complies with `NSCopying` protocol. Since we don't have access to `NSObject`'s code, to follow calls into `copyWithZone:`, we replace such `copy` calls with calls to `copyWithZone:` when a) such a method exists in the class and b) the class conforms to `NSCopying` protocol. This is done in the preanalysis because - we need to know if there is a `copyWithZone:` method in the class. - so that other analyses also benefit (as opposed to doing this in cost and inferbo models). Note that `NSObject` doesn't itself conform to `NSCopying` but all its subclasses must confrom to the protocol and support the same behavior as above. https://developer.apple.com/documentation/objectivec/nsobject/1418807-copy Similarly for `mutableCopy` -> `mutableCopyWithZone:` for classes implementing `NSMutableCopying` protocol. Reviewed By: skcho Differential Revision: D24218102 fbshipit-source-id: 42900760e
4 years ago
codetoanalyze/objc/performance/copy_test.m, MyCopyObj.copyWithZone:, 22, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test.m, MyCopyObj.dealloc, 4, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test.m, MyCopyObj.mObjects, 4, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test.m, MyCopyObj.objects, 7, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test.m, MyCopyObj.setMObjects:, 4, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test.m, loop_over_copied_objects_linear, 36 + 3 ⋅ b->_mObjects->elements.length.ub + 3 ⋅ (b->_mObjects->elements.length.ub + 1), OnUIThread:false, [{b->_mObjects->elements.length.ub + 1},Loop,{b->_mObjects->elements.length.ub},Loop]
codetoanalyze/objc/performance/copy_test_object.m, MyObj.copyWithZone:, 22, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test_object.m, MyObj.dealloc, 4, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test_object.m, MyObj.mObjects, 4, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test_object.m, MyObj.objects, 7, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test_object.m, MyObj.setMObjects:, 4, OnUIThread:false, []
codetoanalyze/objc/performance/copy_test_object.m, loop_over_copied_myobj_linear, 36 + 3 ⋅ b->_mObjects->elements.length.ub + 3 ⋅ (b->_mObjects->elements.length.ub + 1), OnUIThread:false, [{b->_mObjects->elements.length.ub + 1},Loop,{b->_mObjects->elements.length.ub},Loop]
[clang] fix bad interaction between ConditionalOperator and initializers Summary: This is several inter-connected changes together to keep the tests happy. The ConditionalOperator `b?t:e` is translated by first creating a placeholder variable to temporarily store the result of the evaluation in each branch, then the real thing we want to assign to reads that variable. But, there are situations where that changes the semantics of the expression, namely when the value created is a struct on the stack (eg, a C++ temporary). This is because in SIL we cannot assign the *address* of a program variable, only its contents, so by the time we're out of the conditional operator we cannot set the struct value correctly anymore: we can only set its content, which we did, but that results in a "shifted" struct value that is one dereference away from where it should be. So a batch of changes concern `conditionalOperator_trans`: - instead of systematically creating a temporary for the conditional, use the `trans_state.var_exp_typ` provided from above if available when translating `ConditionalOperator` - don't even set anything if that variable was already initialized by merely translating the branch expression, eg when it's a constructor - fix long-standing TODO to propagate these initialization facts accurately for ConditionalOperator (used by `init_expr_trans` to also figure out if it should insert a store to the variable being initialised or not) The rest of the changes adapt some relevant other constructs to deal with conditionalOperator properly now that it can set the current variable itself, instead of storing stuff inside a temp variable. This change was a problem because some constructs, eg a variable declaration, will insert nodes that set up the variable before calling its initialization, and now the initialization happens *before* that setup, in the translation of the inner conditional operator, which naturally creates nodes above the current one. - add a generic helper to force a sequential order between two translation results, forcing node creation if necessary - use that in `init_expr_trans` and `cxxNewExpr_trans` - adjust many places where `var_exp_typ` was incorrectly not reset when translating sub-expressions The sequentiality business creates more nodes when used, and the conditionalOperator business uses fewer temporary variables, so the frontend results change quite a bit. Note that biabduction tests were invaluable in debugging this. There could be other constructs to adjust similarly to cxxNewExpr that were not covered by the tests though. Added tests in pulse that exercises the previous bug. Reviewed By: da319 Differential Revision: D24796282 fbshipit-source-id: 0790c8d17
4 years ago
codetoanalyze/objc/performance/cost_test.m, always, 7, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/cost_test.m, call_infinite, ⊤, OnUIThread:false, [Call to infinite,Unbounded loop,Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/cost_test.m, call_while_upto20_10_constant, 56, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test.m, call_while_upto20_minus100_constant, 606, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/cost_test.m, call_while_upto20_unsigned, 6 + 4 ⋅ (-x + 20) + (21-min(20, x)), OnUIThread:false, [{21-min(20, x)},Call to while_upto20,Loop,{-x + 20},Call to while_upto20,Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/cost_test.m, cond_constant, 14, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test.m, div_const, 3, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test.m, foo_constant, 6, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/cost_test.m, infinite, ⊤, OnUIThread:false, [Unbounded loop,Loop]
[clang] fix bad interaction between ConditionalOperator and initializers Summary: This is several inter-connected changes together to keep the tests happy. The ConditionalOperator `b?t:e` is translated by first creating a placeholder variable to temporarily store the result of the evaluation in each branch, then the real thing we want to assign to reads that variable. But, there are situations where that changes the semantics of the expression, namely when the value created is a struct on the stack (eg, a C++ temporary). This is because in SIL we cannot assign the *address* of a program variable, only its contents, so by the time we're out of the conditional operator we cannot set the struct value correctly anymore: we can only set its content, which we did, but that results in a "shifted" struct value that is one dereference away from where it should be. So a batch of changes concern `conditionalOperator_trans`: - instead of systematically creating a temporary for the conditional, use the `trans_state.var_exp_typ` provided from above if available when translating `ConditionalOperator` - don't even set anything if that variable was already initialized by merely translating the branch expression, eg when it's a constructor - fix long-standing TODO to propagate these initialization facts accurately for ConditionalOperator (used by `init_expr_trans` to also figure out if it should insert a store to the variable being initialised or not) The rest of the changes adapt some relevant other constructs to deal with conditionalOperator properly now that it can set the current variable itself, instead of storing stuff inside a temp variable. This change was a problem because some constructs, eg a variable declaration, will insert nodes that set up the variable before calling its initialization, and now the initialization happens *before* that setup, in the translation of the inner conditional operator, which naturally creates nodes above the current one. - add a generic helper to force a sequential order between two translation results, forcing node creation if necessary - use that in `init_expr_trans` and `cxxNewExpr_trans` - adjust many places where `var_exp_typ` was incorrectly not reset when translating sub-expressions The sequentiality business creates more nodes when used, and the conditionalOperator business uses fewer temporary variables, so the frontend results change quite a bit. Note that biabduction tests were invaluable in debugging this. There could be other constructs to adjust similarly to cxxNewExpr that were not covered by the tests though. Added tests in pulse that exercises the previous bug. Reviewed By: da319 Differential Revision: D24796282 fbshipit-source-id: 0790c8d17
4 years ago
codetoanalyze/objc/performance/cost_test.m, infinite_FN, 17, OnUIThread:false, []
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/cost_test.m, iter_div_const_constant, 109, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test.m, loop0_constant, 1005, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test.m, loop3_constant, 97, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/cost_test.m, loop_character_symbols_linear, 5 + 4 ⋅ (122-min(97, c)), OnUIThread:false, [{122-min(97, c)},Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/cost_test.m, unit_cost_function, 1, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/cost_test.m, while_upto20, 4 + 4 ⋅ (-m + 20) + (21-min(20, m)), OnUIThread:false, [{21-min(20, m)},Loop,{-m + 20},Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/cost_test_deps.m, if_bad_constant, 75, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, if_bad_loop_constant, 203, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, loop_despite_inferbo_constant, 1208, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, nested_loop_constant, 2547, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, real_while_constant, 218, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, simulated_nested_loop_cond_in_goto_constant, 3534, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, simulated_nested_loop_constant, 2529, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, simulated_nested_loop_more_expensive_constant, 2534, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, simulated_while_constant, 218, OnUIThread:false, []
codetoanalyze/objc/performance/cost_test_deps.m, two_loops, 549, OnUIThread:false, []
codetoanalyze/objc/performance/exit.m, call_exit_unreachable, ⊥, OnUIThread:false, [Unreachable node]
codetoanalyze/objc/performance/exit.m, call_unreachable_constant, 2, OnUIThread:false, []
codetoanalyze/objc/performance/exit.m, compute_exit_unreachable, ⊥, OnUIThread:false, [Unreachable node]
codetoanalyze/objc/performance/exit.m, exit_unreachable, ⊥, OnUIThread:false, [Unreachable node]
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/exit.m, inline_exit_unreachable_FP, 3 ⋅ p + 2 ⋅ (1+max(0, p)), OnUIThread:false, [{1+max(0, p)},Loop,{p},Loop]
codetoanalyze/objc/performance/exit.m, linear, 3 + 3 ⋅ p + 2 ⋅ (1+max(0, p)), OnUIThread:false, [{1+max(0, p)},Loop,{p},Loop]
[cost] add objc test for customized_class and blocks Summary: Add objc test for customized class and blocks. Mostly sanity test. Reviewed By: ezgicicek Differential Revision: D22043918 fbshipit-source-id: 917deeea7
4 years ago
codetoanalyze/objc/performance/field_access.m, Person.add_income_constant:, 12, OnUIThread:false, []
codetoanalyze/objc/performance/field_access.m, Person.bank_account, 4, OnUIThread:false, []
codetoanalyze/objc/performance/field_access.m, Person.dealloc, 4, OnUIThread:false, []
codetoanalyze/objc/performance/field_access.m, Person.init_with_name_constant:, 22, OnUIThread:false, []
codetoanalyze/objc/performance/field_access.m, Person.name, 4, OnUIThread:false, []
codetoanalyze/objc/performance/field_access.m, Person.setBank_account:, 4, OnUIThread:false, []
codetoanalyze/objc/performance/field_access.m, Person.setName:, 4, OnUIThread:false, []
[cost] add cost model for most common NSString functions Summary: Add cost model for most common `NSString` functions in cost analysis Reviewed By: skcho Differential Revision: D22433005 fbshipit-source-id: 2f57bbda9
4 years ago
codetoanalyze/objc/performance/field_access.m, Person.species_name_constant, 7, OnUIThread:false, []
[cost] add objc test for customized_class and blocks Summary: Add objc test for customized class and blocks. Mostly sanity test. Reviewed By: ezgicicek Differential Revision: D22043918 fbshipit-source-id: 917deeea7
4 years ago
codetoanalyze/objc/performance/field_access.m, Test1.dealloc, 1, OnUIThread:false, []
codetoanalyze/objc/performance/field_access.m, Test1.setX:, 4, OnUIThread:false, []
codetoanalyze/objc/performance/field_access.m, Test1.x, 4, OnUIThread:false, []
[cost] add cost model for most common NSString functions Summary: Add cost model for most common `NSString` functions in cost analysis Reviewed By: skcho Differential Revision: D22433005 fbshipit-source-id: 2f57bbda9
4 years ago
codetoanalyze/objc/performance/field_access.m, create_common_person_constant, 57, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/field_access.m, iterate_upto_field_size_linear, 3 + 3 ⋅ test->_x.ub + 6 ⋅ (1+max(0, test->_x.ub)), OnUIThread:false, [{1+max(0, test->_x.ub)},Loop,{test->_x.ub},Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/instantiate.m, do_2_times_constant, 20, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/instantiate.m, do_half_m2_times_quadratic, 3 + 5 ⋅ (m - 1) × m + 7 ⋅ m + 2 ⋅ m × (max(1, m)) + 2 ⋅ (1+max(0, m)), OnUIThread:false, [{1+max(0, m)},Loop,{max(1, m)},Call to do_n_times,Loop,{m},Loop,{m},Loop,{m - 1},Call to do_n_times,Loop]
codetoanalyze/objc/performance/instantiate.m, do_m2_times_quadratic, 3 + 7 ⋅ m + 5 ⋅ m × m + 2 ⋅ m × (1+max(0, m)) + 2 ⋅ (1+max(0, m)), OnUIThread:false, [{1+max(0, m)},Loop,{1+max(0, m)},Call to do_n_times,Loop,{m},Call to do_n_times,Loop,{m},Loop]
codetoanalyze/objc/performance/instantiate.m, do_n_times, 3 + 5 ⋅ n + 2 ⋅ (1+max(0, n)), OnUIThread:false, [{1+max(0, n)},Loop,{n},Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/instantiate.m, no_op, 2, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/invariant.m, do_n_m_times_nested, 7 + 5 ⋅ n + 3 ⋅ n × m + 2 ⋅ n × (1+max(0, m)) + 2 ⋅ (1+max(0, n)), OnUIThread:false, [{1+max(0, n)},Loop,{1+max(0, m)},Loop,{m},Loop,{n},Loop]
codetoanalyze/objc/performance/invariant.m, two_loops_nested_invariant, 6 + 23 ⋅ p + 2 ⋅ (1+max(0, p)), OnUIThread:false, [{1+max(0, p)},Loop,{p},Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/invariant.m, while_infinite_FN, 2, OnUIThread:false, []
codetoanalyze/objc/performance/invariant.m, while_unique_def_FN, 15, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/jump_inside_loop.m, jump_inside_loop_constant_linear, 9 + (k - 1) + 4 ⋅ (max(1, k)), OnUIThread:false, [{max(1, k)},Loop,{k - 1},Loop]
codetoanalyze/objc/performance/jump_inside_loop.m, loop_always_linear, 7 + k + 2 ⋅ (max(1, k)) + 2 ⋅ (1+max(1, k)), OnUIThread:false, [{1+max(1, k)},Loop,{max(1, k)},Loop,{k},Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/loops.m, __infer_globals_initializer_array1, 4, OnUIThread:false, []
codetoanalyze/objc/performance/loops.m, __infer_globals_initializer_array2, 2, OnUIThread:false, []
codetoanalyze/objc/performance/loops.m, do_while_independent_of_p, 228, OnUIThread:false, []
codetoanalyze/objc/performance/loops.m, if_in_loop, 324, OnUIThread:false, []
codetoanalyze/objc/performance/loops.m, if_out_loop, 515, OnUIThread:false, []
codetoanalyze/objc/performance/loops.m, larger_state_FN, 1005, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/loops.m, loop_use_global_vars, 4 + 4 ⋅ x + 2 ⋅ (1+max(0, x)), OnUIThread:false, [{1+max(0, x)},Loop,{x},Loop]
codetoanalyze/objc/performance/loops.m, ptr_cmp, 5 + 5 ⋅ size + 2 ⋅ (2+max(-1, size)), OnUIThread:false, [{2+max(-1, size)},Loop,{size},Loop]
[preanalysis] Add a new preanalysis for ObjC's copy and mutableCopy Summary: In ObjC, `NSObject.copy` returns the object returned by `copyWithZone:` on the given class. This method must be implemented if the class complies with `NSCopying` protocol. Since we don't have access to `NSObject`'s code, to follow calls into `copyWithZone:`, we replace such `copy` calls with calls to `copyWithZone:` when a) such a method exists in the class and b) the class conforms to `NSCopying` protocol. This is done in the preanalysis because - we need to know if there is a `copyWithZone:` method in the class. - so that other analyses also benefit (as opposed to doing this in cost and inferbo models). Note that `NSObject` doesn't itself conform to `NSCopying` but all its subclasses must confrom to the protocol and support the same behavior as above. https://developer.apple.com/documentation/objectivec/nsobject/1418807-copy Similarly for `mutableCopy` -> `mutableCopyWithZone:` for classes implementing `NSMutableCopying` protocol. Reviewed By: skcho Differential Revision: D24218102 fbshipit-source-id: 42900760e
4 years ago
codetoanalyze/objc/performance/mutable_copy_test.m, MyMutableObj.dealloc, 4, OnUIThread:false, []
codetoanalyze/objc/performance/mutable_copy_test.m, MyMutableObj.mObjects, 4, OnUIThread:false, []
codetoanalyze/objc/performance/mutable_copy_test.m, MyMutableObj.mutableCopyWithZone:, 22, OnUIThread:false, []
codetoanalyze/objc/performance/mutable_copy_test.m, MyMutableObj.objects, 7, OnUIThread:false, []
codetoanalyze/objc/performance/mutable_copy_test.m, MyMutableObj.setMObjects:, 4, OnUIThread:false, []
codetoanalyze/objc/performance/mutable_copy_test.m, loop_over_mutable_copy_linear, 36 + 3 ⋅ b->_mObjects->elements.length.ub + 3 ⋅ (b->_mObjects->elements.length.ub + 1), OnUIThread:false, [{b->_mObjects->elements.length.ub + 1},Loop,{b->_mObjects->elements.length.ub},Loop]
[cost] Support closure symbols in operation/allocation cost kinds Summary: This diff adds closure symbols to operation/allocation costs, when function pointer is called. Reviewed By: ezgicicek Differential Revision: D24308550 fbshipit-source-id: 6c5889d41
4 years ago
codetoanalyze/objc/performance/purity.m, loop, 6007 + 1000 ⋅ |fun_ptr|, OnUIThread:false, []
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/switch_continue.m, test_switch_FN, 601, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/switch_continue.m, unroll_loop, 16 + (n - 1) + 11 ⋅ (max(1, n)), OnUIThread:false, [{max(1, n)},Loop,{n - 1},Loop]
[cost] import c performance test to objc Summary: Basically copying test file from c/performance to objc/performance. Reviewed By: skcho Differential Revision: D21972945 fbshipit-source-id: 6e4412598
5 years ago
codetoanalyze/objc/performance/two_loops_symbolic.m, nop, 2, OnUIThread:false, []
[cost] Do not print loop line number in trace message Summary: This diff prevents printing line numbers of loop in the trace description, which helps to keep the same descriptions even when the line number of a function is changed in tests. Reviewed By: ezgicicek Differential Revision: D22375584 fbshipit-source-id: 676d1a7cc
4 years ago
codetoanalyze/objc/performance/two_loops_symbolic.m, two_loops_symb_diff, 8 + 5 ⋅ k + 5 ⋅ m + 2 ⋅ (1+max(0, k)) + 2 ⋅ (1+max(0, m)), OnUIThread:false, [{1+max(0, m)},Loop,{1+max(0, k)},Loop,{m},Loop,{k},Loop]
[cost] Fix unreachable node Top poisoning Summary: If a node is unreachable and the cost of the node is Top, we were giving Top cost :( Let's fix it. Reviewed By: skcho Differential Revision: D22548269 fbshipit-source-id: d79743669
4 years ago
codetoanalyze/objc/performance/unreachable.m, always_ret_false, 2, OnUIThread:false, []
codetoanalyze/objc/performance/unreachable.m, linear_loop, 3 + 3 ⋅ x + 2 ⋅ (1+max(0, x)), OnUIThread:false, [{1+max(0, x)},Loop,{x},Loop]
codetoanalyze/objc/performance/unreachable.m, unreachable_branch_constant, 4, OnUIThread:false, []
codetoanalyze/objc/performance/unreachable.m, unreachable_branch_constant2, 4, OnUIThread:false, []
codetoanalyze/objc/performance/unreachable.m, unreachable_branch_linear_loop, 8 + 3 ⋅ x + 2 ⋅ (1+max(0, x)), OnUIThread:false, [{1+max(0, x)},Call to linear_loop,Loop,{x},Call to linear_loop,Loop]