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 '196c0e5544'
${ noResults }
infer_clone/infer/tests/codetoanalyze/c/frontend/arithmetic/negate.c.dot

100 lines
6.3 KiB
Raw Normal View History Unescape

Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
/* @generated */
[cfg] rename iCFG to cfg in dotty files Summary: Not sure what an "iCFG" is but the dotty is only about CFGs anyway. Diff obtained by mass-`sed`. Reviewed By: sblackshear Differential Revision: D6324280 fbshipit-source-id: b7603bb
7 years ago
digraph cfg {
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_1" [label="1: Start neg_char\nFormals: a:char\nLocals: 0$?%__sil_tmpSIL_temp_conditional___n$0:int \n DECLARE_LOCALS(&return,&0$?%__sil_tmpSIL_temp_conditional___n$0); [line 12, column 1]\n " color=yellow style=filled]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_1" -> "neg_char.53ef6b31d84386046a4728d1c45b5f7a_4" ;
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_1" -> "neg_char.53ef6b31d84386046a4728d1c45b5f7a_5" ;
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_2" [label="2: Exit neg_char \n " color=yellow style=filled]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_3" [label="3: + \n " ]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_3" -> "neg_char.53ef6b31d84386046a4728d1c45b5f7a_8" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_4" [label="4: Prune (true branch) \n n$1=*&a:char [line 12, column 32]\n PRUNE(n$1, true); [line 12, column 32]\n " shape="invhouse"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_4" -> "neg_char.53ef6b31d84386046a4728d1c45b5f7a_6" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_5" [label="5: Prune (false branch) \n n$1=*&a:char [line 12, column 32]\n PRUNE(!n$1, false); [line 12, column 32]\n " shape="invhouse"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_5" -> "neg_char.53ef6b31d84386046a4728d1c45b5f7a_7" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_6" [label="6: ConditinalStmt Branch \n *&0$?%__sil_tmpSIL_temp_conditional___n$0:int=0 [line 12, column 31]\n " shape="box"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_6" -> "neg_char.53ef6b31d84386046a4728d1c45b5f7a_3" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_7" [label="7: ConditinalStmt Branch \n *&0$?%__sil_tmpSIL_temp_conditional___n$0:int=1 [line 12, column 31]\n " shape="box"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_7" -> "neg_char.53ef6b31d84386046a4728d1c45b5f7a_3" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_8" [label="8: Return Stmt \n n$2=*&0$?%__sil_tmpSIL_temp_conditional___n$0:int [line 12, column 31]\n *&return:int=n$2 [line 12, column 24]\n " shape="box"]
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_char.53ef6b31d84386046a4728d1c45b5f7a_8" -> "neg_char.53ef6b31d84386046a4728d1c45b5f7a_2" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_1" [label="1: Start neg_bool\nFormals: a:_Bool\nLocals: 0$?%__sil_tmpSIL_temp_conditional___n$0:int \n DECLARE_LOCALS(&return,&0$?%__sil_tmpSIL_temp_conditional___n$0); [line 14, column 1]\n " color=yellow style=filled]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
[debug][dotty] Fix issue in dotty output where overloaded functions were conflated Reviewed By: jvillard Differential Revision: D4117078 fbshipit-source-id: fdd8e93
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_1" -> "neg_bool.e953d6477eaaeafaa430423a26fbaac9_4" ;
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_1" -> "neg_bool.e953d6477eaaeafaa430423a26fbaac9_5" ;
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_2" [label="2: Exit neg_bool \n " color=yellow style=filled]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_3" [label="3: + \n " ]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_3" -> "neg_bool.e953d6477eaaeafaa430423a26fbaac9_8" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_4" [label="4: Prune (true branch) \n n$1=*&a:_Bool [line 14, column 33]\n PRUNE(n$1, true); [line 14, column 33]\n " shape="invhouse"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_4" -> "neg_bool.e953d6477eaaeafaa430423a26fbaac9_6" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_5" [label="5: Prune (false branch) \n n$1=*&a:_Bool [line 14, column 33]\n PRUNE(!n$1, false); [line 14, column 33]\n " shape="invhouse"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_5" -> "neg_bool.e953d6477eaaeafaa430423a26fbaac9_7" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_6" [label="6: ConditinalStmt Branch \n *&0$?%__sil_tmpSIL_temp_conditional___n$0:int=0 [line 14, column 32]\n " shape="box"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_6" -> "neg_bool.e953d6477eaaeafaa430423a26fbaac9_3" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_7" [label="7: ConditinalStmt Branch \n *&0$?%__sil_tmpSIL_temp_conditional___n$0:int=1 [line 14, column 32]\n " shape="box"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_7" -> "neg_bool.e953d6477eaaeafaa430423a26fbaac9_3" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_8" [label="8: Return Stmt \n n$2=*&0$?%__sil_tmpSIL_temp_conditional___n$0:int [line 14, column 32]\n *&return:int=n$2 [line 14, column 25]\n " shape="box"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_bool.e953d6477eaaeafaa430423a26fbaac9_8" -> "neg_bool.e953d6477eaaeafaa430423a26fbaac9_2" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_1" [label="1: Start neg_int\nFormals: a:int\nLocals: 0$?%__sil_tmpSIL_temp_conditional___n$0:int \n DECLARE_LOCALS(&return,&0$?%__sil_tmpSIL_temp_conditional___n$0); [line 10, column 1]\n " color=yellow style=filled]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_1" -> "neg_int.2aa25aca565c41dd997912d11504462c_4" ;
"neg_int.2aa25aca565c41dd997912d11504462c_1" -> "neg_int.2aa25aca565c41dd997912d11504462c_5" ;
"neg_int.2aa25aca565c41dd997912d11504462c_2" [label="2: Exit neg_int \n " color=yellow style=filled]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_3" [label="3: + \n " ]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_3" -> "neg_int.2aa25aca565c41dd997912d11504462c_8" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_4" [label="4: Prune (true branch) \n n$1=*&a:int [line 10, column 30]\n PRUNE(n$1, true); [line 10, column 30]\n " shape="invhouse"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_4" -> "neg_int.2aa25aca565c41dd997912d11504462c_6" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_5" [label="5: Prune (false branch) \n n$1=*&a:int [line 10, column 30]\n PRUNE(!n$1, false); [line 10, column 30]\n " shape="invhouse"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
[debug][dotty] Fix issue in dotty output where overloaded functions were conflated Reviewed By: jvillard Differential Revision: D4117078 fbshipit-source-id: fdd8e93
8 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_5" -> "neg_int.2aa25aca565c41dd997912d11504462c_7" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_6" [label="6: ConditinalStmt Branch \n *&0$?%__sil_tmpSIL_temp_conditional___n$0:int=0 [line 10, column 29]\n " shape="box"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_6" -> "neg_int.2aa25aca565c41dd997912d11504462c_3" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_7" [label="7: ConditinalStmt Branch \n *&0$?%__sil_tmpSIL_temp_conditional___n$0:int=1 [line 10, column 29]\n " shape="box"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_7" -> "neg_int.2aa25aca565c41dd997912d11504462c_3" ;
[siof] one access per sink, better report deduplication Summary: The previous domain for SIOF was duplicating some work with the generic Trace domain, and basically was a bit confused and confusing. A sink was a set of global accesses, and a state contains a set of sinks. Then the checker has to needlessly jump through hoops to normalize this set of sets of accesses into a set of accesses. The new domain has one sink = one access, as suggested by sblackshear. This simplifies a few things, and makes the dedup logic much easier: just grab the first report of the list of reports for a function. We only report on the fake procedures generated to initialise a global, and the filtering means that we keep only one report per global. Reviewed By: sblackshear Differential Revision: D5932138 fbshipit-source-id: acb7285
7 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_8" [label="8: Return Stmt \n n$2=*&0$?%__sil_tmpSIL_temp_conditional___n$0:int [line 10, column 29]\n *&return:int=n$2 [line 10, column 22]\n " shape="box"]
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
Sort nodes when printing cfg to dot file Summary: Currently cfg nodes are written into dot files in whatever order they appear in a hash table. This seems unnecessarily sensitive, so this diff sorts the nodes. Reviewed By: dulmarod Differential Revision: D4232377 fbshipit-source-id: a907cc6
8 years ago
"neg_int.2aa25aca565c41dd997912d11504462c_8" -> "neg_int.2aa25aca565c41dd997912d11504462c_2" ;
Fix translation of unary operator! for all integral types Summary: Backend has the same treatment for all Typ.Int types and so UnaryOperator should be translated the same for all of them Reviewed By: jvillard Differential Revision: D3534277 fbshipit-source-id: 8569b65
8 years ago
}