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 'eaa5c32432'
${ noResults }
infer_clone/infer/tests/codetoanalyze/cpp/shared/reference/unbox.cpp.dot

91 lines
7.4 KiB
Raw Normal View History Unescape

Add @generated comment to icfg.dot files Reviewed By: jvillard Differential Revision: D3358243 fbshipit-source-id: a47cc01
9 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 {
Get rid of Declare_locals Reviewed By: jeremydubreil Differential Revision: D9234331 fbshipit-source-id: 70443eabe
6 years ago
"fun_p#7468829763884786220.ff1d58f26970dcc8ee3c2a153c5a5b85_1" [label="1: Start fun_p\nFormals: p:int*\nLocals: \n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"fun_p#7468829763884786220.ff1d58f26970dcc8ee3c2a153c5a5b85_1" -> "fun_p#7468829763884786220.ff1d58f26970dcc8ee3c2a153c5a5b85_3" ;
"fun_p#7468829763884786220.ff1d58f26970dcc8ee3c2a153c5a5b85_2" [label="2: Exit fun_p \n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[preanalysis] do not delay killing variables taken by reference Summary: Before, the liveness pre-analysis would place extra instructions in the CFG for either: 1. marking an `Ident.t` as dead, or 2. marking a `Pvar.t` as `= 0` But we have no way of marking pvars dead without setting them to 0. This is bad because setting pvars to 0 is not possible everywhere they are dead. Indeed, we only do it when we haven't seen their address being taken anyway. This prevents the following situation, recorded in our tests: ``` int address_taken() { int** x; int* y; int i = 7; y = &i; x = &y; // if we don't reason about taken addresses while adding nullify instructions, // we'll add // `nullify(y)` here and report a false NPE on the next line return **x; } ``` So we want to mark pvars as dead without nullifying them. This diff extends the `Remove_temps` SIL instruction to accept pvars as well, and so renames it to `ExitScope`. Reviewed By: da319 Differential Revision: D13102953 fbshipit-source-id: aa7f03a52
6 years ago
"fun_p#7468829763884786220.ff1d58f26970dcc8ee3c2a153c5a5b85_3" [label="3: Return Stmt \n n$0=*&p:int* [line 8, column 29]\n n$1=*n$0:int [line 8, column 28]\n *&return:int=n$1 [line 8, column 21]\n NULLIFY(&p); [line 8, column 21]\n EXIT_SCOPE(n$0,n$1,p); [line 8, column 21]\n APPLY_ABSTRACTION; [line 8, column 21]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"fun_p#7468829763884786220.ff1d58f26970dcc8ee3c2a153c5a5b85_3" -> "fun_p#7468829763884786220.ff1d58f26970dcc8ee3c2a153c5a5b85_2" ;
Get rid of Declare_locals Reviewed By: jeremydubreil Differential Revision: D9234331 fbshipit-source-id: 70443eabe
6 years ago
"fun_r#8688550998084520100.a539308a01e8443f65be5d44c29a73f6_1" [label="1: Start fun_r\nFormals: p:int&\nLocals: \n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"fun_r#8688550998084520100.a539308a01e8443f65be5d44c29a73f6_1" -> "fun_r#8688550998084520100.a539308a01e8443f65be5d44c29a73f6_3" ;
"fun_r#8688550998084520100.a539308a01e8443f65be5d44c29a73f6_2" [label="2: Exit fun_r \n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[preanalysis] do not delay killing variables taken by reference Summary: Before, the liveness pre-analysis would place extra instructions in the CFG for either: 1. marking an `Ident.t` as dead, or 2. marking a `Pvar.t` as `= 0` But we have no way of marking pvars dead without setting them to 0. This is bad because setting pvars to 0 is not possible everywhere they are dead. Indeed, we only do it when we haven't seen their address being taken anyway. This prevents the following situation, recorded in our tests: ``` int address_taken() { int** x; int* y; int i = 7; y = &i; x = &y; // if we don't reason about taken addresses while adding nullify instructions, // we'll add // `nullify(y)` here and report a false NPE on the next line return **x; } ``` So we want to mark pvars as dead without nullifying them. This diff extends the `Remove_temps` SIL instruction to accept pvars as well, and so renames it to `ExitScope`. Reviewed By: da319 Differential Revision: D13102953 fbshipit-source-id: aa7f03a52
6 years ago
"fun_r#8688550998084520100.a539308a01e8443f65be5d44c29a73f6_3" [label="3: Return Stmt \n n$0=*&p:int& [line 10, column 28]\n n$1=*n$0:int [line 10, column 28]\n *&return:int=n$1 [line 10, column 21]\n NULLIFY(&p); [line 10, column 21]\n EXIT_SCOPE(n$0,n$1,p); [line 10, column 21]\n APPLY_ABSTRACTION; [line 10, column 21]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"fun_r#8688550998084520100.a539308a01e8443f65be5d44c29a73f6_3" -> "fun_r#8688550998084520100.a539308a01e8443f65be5d44c29a73f6_2" ;
Get rid of Declare_locals Reviewed By: jeremydubreil Differential Revision: D9234331 fbshipit-source-id: 70443eabe
6 years ago
"fun_v#125358748374922080.2b082c989a86eb6a918b15eb596c685a_1" [label="1: Start fun_v\nFormals: p:int\nLocals: \n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"fun_v#125358748374922080.2b082c989a86eb6a918b15eb596c685a_1" -> "fun_v#125358748374922080.2b082c989a86eb6a918b15eb596c685a_3" ;
"fun_v#125358748374922080.2b082c989a86eb6a918b15eb596c685a_2" [label="2: Exit fun_v \n " color=yellow style=filled]
[preanalysis] do not delay killing variables taken by reference Summary: Before, the liveness pre-analysis would place extra instructions in the CFG for either: 1. marking an `Ident.t` as dead, or 2. marking a `Pvar.t` as `= 0` But we have no way of marking pvars dead without setting them to 0. This is bad because setting pvars to 0 is not possible everywhere they are dead. Indeed, we only do it when we haven't seen their address being taken anyway. This prevents the following situation, recorded in our tests: ``` int address_taken() { int** x; int* y; int i = 7; y = &i; x = &y; // if we don't reason about taken addresses while adding nullify instructions, // we'll add // `nullify(y)` here and report a false NPE on the next line return **x; } ``` So we want to mark pvars as dead without nullifying them. This diff extends the `Remove_temps` SIL instruction to accept pvars as well, and so renames it to `ExitScope`. Reviewed By: da319 Differential Revision: D13102953 fbshipit-source-id: aa7f03a52
6 years ago
"fun_v#125358748374922080.2b082c989a86eb6a918b15eb596c685a_3" [label="3: Return Stmt \n n$0=*&p:int [line 9, column 27]\n *&return:int=n$0 [line 9, column 20]\n NULLIFY(&p); [line 9, column 20]\n EXIT_SCOPE(n$0,p); [line 9, column 20]\n APPLY_ABSTRACTION; [line 9, column 20]\n " shape="box"]
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"fun_v#125358748374922080.2b082c989a86eb6a918b15eb596c685a_3" -> "fun_v#125358748374922080.2b082c989a86eb6a918b15eb596c685a_2" ;
Get rid of Declare_locals Reviewed By: jeremydubreil Differential Revision: D9234331 fbshipit-source-id: 70443eabe
6 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_1" [label="1: Start unbox_ptr\nFormals: \nLocals: p:int* a:int \n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[clang] update clang plugin to hash mangled names Summary: With this change and the previous facebook-clang-plugins change, infer no longer exhausts the biniou buffer when reading the serialized C++ AST. update-submodule: facebook-clang-plugins Reviewed By: mbouaziz Differential Revision: D5891081 fbshipit-source-id: cf48eac
7 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_1" -> "unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_7" ;
[SIL] change `variable_initialization()` builtin to a new auxiliary instruction Summary: Instead of emitting an ad-hoc builtin on variable declaration emit a new metadata instruction. This allows us to remove the code matching on that ad-hoc builtin that had to be inserted in several checkers. Inferbo & pulse used that information meaningfully and had to undergo some minor changes to cope with the new metada instruction. Reviewed By: ezgicicek Differential Revision: D14833100 fbshipit-source-id: 9b3009d22
6 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_2" [label="2: Exit unbox_ptr \n NULLIFY(&a); [line 32, column 1]\n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[SIL] change `variable_initialization()` builtin to a new auxiliary instruction Summary: Instead of emitting an ad-hoc builtin on variable declaration emit a new metadata instruction. This allows us to remove the code matching on that ad-hoc builtin that had to be inserted in several checkers. Inferbo & pulse used that information meaningfully and had to undergo some minor changes to cope with the new metada instruction. Reviewed By: ezgicicek Differential Revision: D14833100 fbshipit-source-id: 9b3009d22
6 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_3" [label="3: Call _fun_fun_r \n n$1=*&p:int* [line 31, column 10]\n n$2=_fun_fun_r(n$1:int&) [line 31, column 3]\n NULLIFY(&p); [line 31, column 3]\n EXIT_SCOPE(n$1,n$2,p); [line 31, column 3]\n APPLY_ABSTRACTION; [line 31, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[clang] update clang plugin to hash mangled names Summary: With this change and the previous facebook-clang-plugins change, infer no longer exhausts the biniou buffer when reading the serialized C++ AST. update-submodule: facebook-clang-plugins Reviewed By: mbouaziz Differential Revision: D5891081 fbshipit-source-id: cf48eac
7 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_3" -> "unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_2" ;
[preanalysis] do not delay killing variables taken by reference Summary: Before, the liveness pre-analysis would place extra instructions in the CFG for either: 1. marking an `Ident.t` as dead, or 2. marking a `Pvar.t` as `= 0` But we have no way of marking pvars dead without setting them to 0. This is bad because setting pvars to 0 is not possible everywhere they are dead. Indeed, we only do it when we haven't seen their address being taken anyway. This prevents the following situation, recorded in our tests: ``` int address_taken() { int** x; int* y; int i = 7; y = &i; x = &y; // if we don't reason about taken addresses while adding nullify instructions, // we'll add // `nullify(y)` here and report a false NPE on the next line return **x; } ``` So we want to mark pvars as dead without nullifying them. This diff extends the `Remove_temps` SIL instruction to accept pvars as well, and so renames it to `ExitScope`. Reviewed By: da319 Differential Revision: D13102953 fbshipit-source-id: aa7f03a52
6 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_4" [label="4: Call _fun_fun_v \n n$3=*&p:int* [line 30, column 10]\n n$4=*n$3:int [line 30, column 9]\n n$5=_fun_fun_v(n$4:int) [line 30, column 3]\n EXIT_SCOPE(n$3,n$4,n$5); [line 30, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[clang] update clang plugin to hash mangled names Summary: With this change and the previous facebook-clang-plugins change, infer no longer exhausts the biniou buffer when reading the serialized C++ AST. update-submodule: facebook-clang-plugins Reviewed By: mbouaziz Differential Revision: D5891081 fbshipit-source-id: cf48eac
7 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_4" -> "unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_3" ;
[preanalysis] do not delay killing variables taken by reference Summary: Before, the liveness pre-analysis would place extra instructions in the CFG for either: 1. marking an `Ident.t` as dead, or 2. marking a `Pvar.t` as `= 0` But we have no way of marking pvars dead without setting them to 0. This is bad because setting pvars to 0 is not possible everywhere they are dead. Indeed, we only do it when we haven't seen their address being taken anyway. This prevents the following situation, recorded in our tests: ``` int address_taken() { int** x; int* y; int i = 7; y = &i; x = &y; // if we don't reason about taken addresses while adding nullify instructions, // we'll add // `nullify(y)` here and report a false NPE on the next line return **x; } ``` So we want to mark pvars as dead without nullifying them. This diff extends the `Remove_temps` SIL instruction to accept pvars as well, and so renames it to `ExitScope`. Reviewed By: da319 Differential Revision: D13102953 fbshipit-source-id: aa7f03a52
6 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_5" [label="5: Call _fun_fun_p \n n$6=*&p:int* [line 29, column 9]\n n$7=_fun_fun_p(n$6:int*) [line 29, column 3]\n EXIT_SCOPE(n$6,n$7); [line 29, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[clang] update clang plugin to hash mangled names Summary: With this change and the previous facebook-clang-plugins change, infer no longer exhausts the biniou buffer when reading the serialized C++ AST. update-submodule: facebook-clang-plugins Reviewed By: mbouaziz Differential Revision: D5891081 fbshipit-source-id: cf48eac
7 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_5" -> "unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_4" ;
[SIL] change `variable_initialization()` builtin to a new auxiliary instruction Summary: Instead of emitting an ad-hoc builtin on variable declaration emit a new metadata instruction. This allows us to remove the code matching on that ad-hoc builtin that had to be inserted in several checkers. Inferbo & pulse used that information meaningfully and had to undergo some minor changes to cope with the new metada instruction. Reviewed By: ezgicicek Differential Revision: D14833100 fbshipit-source-id: 9b3009d22
6 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_6" [label="6: DeclStmt \n VARIABLE_DECLARED(p:int*); [line 27, column 3]\n *&p:int*=&a [line 27, column 3]\n EXIT_SCOPE(a); [line 27, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[clang] update clang plugin to hash mangled names Summary: With this change and the previous facebook-clang-plugins change, infer no longer exhausts the biniou buffer when reading the serialized C++ AST. update-submodule: facebook-clang-plugins Reviewed By: mbouaziz Differential Revision: D5891081 fbshipit-source-id: cf48eac
7 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_6" -> "unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_5" ;
[SIL] change `variable_initialization()` builtin to a new auxiliary instruction Summary: Instead of emitting an ad-hoc builtin on variable declaration emit a new metadata instruction. This allows us to remove the code matching on that ad-hoc builtin that had to be inserted in several checkers. Inferbo & pulse used that information meaningfully and had to undergo some minor changes to cope with the new metada instruction. Reviewed By: ezgicicek Differential Revision: D14833100 fbshipit-source-id: 9b3009d22
6 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_7" [label="7: DeclStmt \n VARIABLE_DECLARED(a:int); [line 26, column 3]\n *&a:int=3 [line 26, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[clang] update clang plugin to hash mangled names Summary: With this change and the previous facebook-clang-plugins change, infer no longer exhausts the biniou buffer when reading the serialized C++ AST. update-submodule: facebook-clang-plugins Reviewed By: mbouaziz Differential Revision: D5891081 fbshipit-source-id: cf48eac
7 years ago
"unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_7" -> "unbox_ptr#3550280956167916174.75d50cc2e2dfffd1cc23613b01fc878b_6" ;
Get rid of Declare_locals Reviewed By: jeremydubreil Differential Revision: D9234331 fbshipit-source-id: 70443eabe
6 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_1" [label="1: Start unbox_ref\nFormals: \nLocals: r:int& a:int \n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_1" -> "unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_7" ;
[SIL] change `variable_initialization()` builtin to a new auxiliary instruction Summary: Instead of emitting an ad-hoc builtin on variable declaration emit a new metadata instruction. This allows us to remove the code matching on that ad-hoc builtin that had to be inserted in several checkers. Inferbo & pulse used that information meaningfully and had to undergo some minor changes to cope with the new metada instruction. Reviewed By: ezgicicek Differential Revision: D14833100 fbshipit-source-id: 9b3009d22
6 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_2" [label="2: Exit unbox_ref \n NULLIFY(&a); [line 22, column 1]\n " color=yellow style=filled]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[SIL] change `variable_initialization()` builtin to a new auxiliary instruction Summary: Instead of emitting an ad-hoc builtin on variable declaration emit a new metadata instruction. This allows us to remove the code matching on that ad-hoc builtin that had to be inserted in several checkers. Inferbo & pulse used that information meaningfully and had to undergo some minor changes to cope with the new metada instruction. Reviewed By: ezgicicek Differential Revision: D14833100 fbshipit-source-id: 9b3009d22
6 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_3" [label="3: Call _fun_fun_r \n n$1=*&r:int& [line 21, column 9]\n n$2=_fun_fun_r(n$1:int&) [line 21, column 3]\n NULLIFY(&r); [line 21, column 3]\n EXIT_SCOPE(n$1,n$2,r); [line 21, column 3]\n APPLY_ABSTRACTION; [line 21, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_3" -> "unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_2" ;
[preanalysis] do not delay killing variables taken by reference Summary: Before, the liveness pre-analysis would place extra instructions in the CFG for either: 1. marking an `Ident.t` as dead, or 2. marking a `Pvar.t` as `= 0` But we have no way of marking pvars dead without setting them to 0. This is bad because setting pvars to 0 is not possible everywhere they are dead. Indeed, we only do it when we haven't seen their address being taken anyway. This prevents the following situation, recorded in our tests: ``` int address_taken() { int** x; int* y; int i = 7; y = &i; x = &y; // if we don't reason about taken addresses while adding nullify instructions, // we'll add // `nullify(y)` here and report a false NPE on the next line return **x; } ``` So we want to mark pvars as dead without nullifying them. This diff extends the `Remove_temps` SIL instruction to accept pvars as well, and so renames it to `ExitScope`. Reviewed By: da319 Differential Revision: D13102953 fbshipit-source-id: aa7f03a52
6 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_4" [label="4: Call _fun_fun_v \n n$3=*&r:int& [line 20, column 9]\n n$4=*n$3:int [line 20, column 9]\n n$5=_fun_fun_v(n$4:int) [line 20, column 3]\n EXIT_SCOPE(n$3,n$4,n$5); [line 20, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_4" -> "unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_3" ;
[preanalysis] do not delay killing variables taken by reference Summary: Before, the liveness pre-analysis would place extra instructions in the CFG for either: 1. marking an `Ident.t` as dead, or 2. marking a `Pvar.t` as `= 0` But we have no way of marking pvars dead without setting them to 0. This is bad because setting pvars to 0 is not possible everywhere they are dead. Indeed, we only do it when we haven't seen their address being taken anyway. This prevents the following situation, recorded in our tests: ``` int address_taken() { int** x; int* y; int i = 7; y = &i; x = &y; // if we don't reason about taken addresses while adding nullify instructions, // we'll add // `nullify(y)` here and report a false NPE on the next line return **x; } ``` So we want to mark pvars as dead without nullifying them. This diff extends the `Remove_temps` SIL instruction to accept pvars as well, and so renames it to `ExitScope`. Reviewed By: da319 Differential Revision: D13102953 fbshipit-source-id: aa7f03a52
6 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_5" [label="5: Call _fun_fun_p \n n$6=*&r:int& [line 19, column 10]\n n$7=_fun_fun_p(n$6:int*) [line 19, column 3]\n EXIT_SCOPE(n$6,n$7); [line 19, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_5" -> "unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_4" ;
[SIL] change `variable_initialization()` builtin to a new auxiliary instruction Summary: Instead of emitting an ad-hoc builtin on variable declaration emit a new metadata instruction. This allows us to remove the code matching on that ad-hoc builtin that had to be inserted in several checkers. Inferbo & pulse used that information meaningfully and had to undergo some minor changes to cope with the new metada instruction. Reviewed By: ezgicicek Differential Revision: D14833100 fbshipit-source-id: 9b3009d22
6 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_6" [label="6: DeclStmt \n VARIABLE_DECLARED(r:int&); [line 17, column 3]\n *&r:int&=&a [line 17, column 3]\n EXIT_SCOPE(a); [line 17, column 3]\n " shape="box"]
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_6" -> "unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_5" ;
[SIL] change `variable_initialization()` builtin to a new auxiliary instruction Summary: Instead of emitting an ad-hoc builtin on variable declaration emit a new metadata instruction. This allows us to remove the code matching on that ad-hoc builtin that had to be inserted in several checkers. Inferbo & pulse used that information meaningfully and had to undergo some minor changes to cope with the new metada instruction. Reviewed By: ezgicicek Differential Revision: D14833100 fbshipit-source-id: 9b3009d22
6 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_7" [label="7: DeclStmt \n VARIABLE_DECLARED(a:int); [line 16, column 3]\n *&a:int=3 [line 16, column 3]\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
[debug] print procedures in alphabetical order in cfgs Summary: When looking at large CFGs, at least in `xdot`, it's often difficult to find the procedure you're looking for. Sorting the proc names puts them in alphabetical order, which makes searching one procedure easier. Reviewed By: mbouaziz Differential Revision: D7758521 fbshipit-source-id: 8e9997f
7 years ago
"unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_7" -> "unbox_ref#9977470601320200599.91094dce9e5b43dc4c89abcbc69b2c70_6" ;
Fix handling of C++ references Summary: @​public This diff changes following things: 1. expression_info.type_ptr has type than decl_ref_info.type_ptr for reference types. Use type from decl_ref_info as a source of truth 2. reference types need to have one extra dereference that is not in AST. Add handling for this. 3. [small refactor] create function that creates temporary variable from res_trans expression and returns new res_trans. Some caveats: 1. types are not quite right yet (see .dot files). 2. decl_ref_info might not be set for DeclRefExpr, make frontend crash in that case to catch when this happens This is high risk change since it changes behavior of every translation on very widely used expr. Reviewed By: @dulmarod Differential Revision: D2540632 fb-gh-sync-id: aa28936
9 years ago
}