mathieu/os-rust
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
os-rust/src/test/mir-opt/funky_arms.float_to_exponential_common.ConstProp.diff

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

147 lines
9.9 KiB
Diff
Raw Normal View History

move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
- // MIR for `float_to_exponential_common` before ConstProp
+ // MIR for `float_to_exponential_common` after ConstProp
path trimming: ignore type aliases
2021-01-28 18:01:36 +02:00
fn float_to_exponential_common(_1: &mut Formatter, _2: &T, _3: bool) -> Result<(), std::fmt::Error> {
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
debug fmt => _1; // in scope 0 at $DIR/funky_arms.rs:11:35: 11:38
debug num => _2; // in scope 0 at $DIR/funky_arms.rs:11:60: 11:63
debug upper => _3; // in scope 0 at $DIR/funky_arms.rs:11:69: 11:74
let mut _0: std::result::Result<(), std::fmt::Error>; // return place in scope 0 at $DIR/funky_arms.rs:11:85: 11:91
let _4: bool; // in scope 0 at $DIR/funky_arms.rs:15:9: 15:19
Use larger span for adjustments on method calls Currently, we use a relatively 'small' span for THIR expressions generated by an 'adjustment' (e.g. an autoderef, autoborrow, unsizing). As a result, if a borrow generated by an adustment ends up causing a borrowcheck error, for example: ```rust let mut my_var = String::new(); let my_ref = &my_var my_var.push('a'); my_ref; ``` then the span for the mutable borrow may end up referring to only the base expression (e.g. `my_var`), rather than the method call which triggered the mutable borrow (e.g. `my_var.push('a')`) Due to a quirk of the MIR borrowck implementation, this doesn't always get exposed in migration mode, but it does in many cases. This commit makes THIR building consistently use 'larger' spans for adjustment expressions The intent of this change it make it clearer to users when it's the specific way in which a variable is used (for example, in a method call) that produdes a borrowcheck error. For example, an error message claiming that a 'mutable borrow occurs here' might be confusing if it just points at a usage of a variable (e.g. `my_var`), when no `&mut` is in sight. Pointing at the entire expression should help to emphasize that the method call itself is responsible for the mutable borrow. In several cases, this makes the `#![feature(nll)]` diagnostic output match up exactly with the default (migration mode) output. As a result, several `.nll.stderr` files end up getting removed entirely.
2021-09-16 15:01:22 -05:00
let mut _5: &std::fmt::Formatter; // in scope 0 at $DIR/funky_arms.rs:15:22: 15:37
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
let mut _7: std::option::Option<usize>; // in scope 0 at $DIR/funky_arms.rs:24:30: 24:45
Use larger span for adjustments on method calls Currently, we use a relatively 'small' span for THIR expressions generated by an 'adjustment' (e.g. an autoderef, autoborrow, unsizing). As a result, if a borrow generated by an adustment ends up causing a borrowcheck error, for example: ```rust let mut my_var = String::new(); let my_ref = &my_var my_var.push('a'); my_ref; ``` then the span for the mutable borrow may end up referring to only the base expression (e.g. `my_var`), rather than the method call which triggered the mutable borrow (e.g. `my_var.push('a')`) Due to a quirk of the MIR borrowck implementation, this doesn't always get exposed in migration mode, but it does in many cases. This commit makes THIR building consistently use 'larger' spans for adjustment expressions The intent of this change it make it clearer to users when it's the specific way in which a variable is used (for example, in a method call) that produdes a borrowcheck error. For example, an error message claiming that a 'mutable borrow occurs here' might be confusing if it just points at a usage of a variable (e.g. `my_var`), when no `&mut` is in sight. Pointing at the entire expression should help to emphasize that the method call itself is responsible for the mutable borrow. In several cases, this makes the `#![feature(nll)]` diagnostic output match up exactly with the default (migration mode) output. As a result, several `.nll.stderr` files end up getting removed entirely.
2021-09-16 15:01:22 -05:00
let mut _8: &std::fmt::Formatter; // in scope 0 at $DIR/funky_arms.rs:24:30: 24:45
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
let mut _9: isize; // in scope 0 at $DIR/funky_arms.rs:24:12: 24:27
let mut _11: &mut std::fmt::Formatter; // in scope 0 at $DIR/funky_arms.rs:26:43: 26:46
let mut _12: &T; // in scope 0 at $DIR/funky_arms.rs:26:48: 26:51
let mut _13: core::num::flt2dec::Sign; // in scope 0 at $DIR/funky_arms.rs:26:53: 26:57
let mut _14: u32; // in scope 0 at $DIR/funky_arms.rs:26:59: 26:79
let mut _15: u32; // in scope 0 at $DIR/funky_arms.rs:26:59: 26:75
let mut _16: usize; // in scope 0 at $DIR/funky_arms.rs:26:59: 26:68
let mut _17: bool; // in scope 0 at $DIR/funky_arms.rs:26:81: 26:86
let mut _18: &mut std::fmt::Formatter; // in scope 0 at $DIR/funky_arms.rs:28:46: 28:49
let mut _19: &T; // in scope 0 at $DIR/funky_arms.rs:28:51: 28:54
let mut _20: core::num::flt2dec::Sign; // in scope 0 at $DIR/funky_arms.rs:28:56: 28:60
let mut _21: bool; // in scope 0 at $DIR/funky_arms.rs:28:62: 28:67
scope 1 {
debug force_sign => _4; // in scope 1 at $DIR/funky_arms.rs:15:9: 15:19
let _6: core::num::flt2dec::Sign; // in scope 1 at $DIR/funky_arms.rs:19:9: 19:13
scope 2 {
debug sign => _6; // in scope 2 at $DIR/funky_arms.rs:19:9: 19:13
scope 3 {
debug precision => _10; // in scope 3 at $DIR/funky_arms.rs:24:17: 24:26
Revert "More minimal changes" This reverts commit fe0dedcb06947317d41a8570b7fff7f8690dcbff.
2022-06-09 23:09:04 -07:00
let _10: usize; // in scope 3 at $DIR/funky_arms.rs:24:17: 24:26
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
}
}
}
bb0: {
StorageLive(_4); // scope 0 at $DIR/funky_arms.rs:15:9: 15:19
Use larger span for adjustments on method calls Currently, we use a relatively 'small' span for THIR expressions generated by an 'adjustment' (e.g. an autoderef, autoborrow, unsizing). As a result, if a borrow generated by an adustment ends up causing a borrowcheck error, for example: ```rust let mut my_var = String::new(); let my_ref = &my_var my_var.push('a'); my_ref; ``` then the span for the mutable borrow may end up referring to only the base expression (e.g. `my_var`), rather than the method call which triggered the mutable borrow (e.g. `my_var.push('a')`) Due to a quirk of the MIR borrowck implementation, this doesn't always get exposed in migration mode, but it does in many cases. This commit makes THIR building consistently use 'larger' spans for adjustment expressions The intent of this change it make it clearer to users when it's the specific way in which a variable is used (for example, in a method call) that produdes a borrowcheck error. For example, an error message claiming that a 'mutable borrow occurs here' might be confusing if it just points at a usage of a variable (e.g. `my_var`), when no `&mut` is in sight. Pointing at the entire expression should help to emphasize that the method call itself is responsible for the mutable borrow. In several cases, this makes the `#![feature(nll)]` diagnostic output match up exactly with the default (migration mode) output. As a result, several `.nll.stderr` files end up getting removed entirely.
2021-09-16 15:01:22 -05:00
StorageLive(_5); // scope 0 at $DIR/funky_arms.rs:15:22: 15:37
_5 = &(*_1); // scope 0 at $DIR/funky_arms.rs:15:22: 15:37
pretty: trim paths of unique symbols If a symbol name can only be imported from one place for a type, and as long as it was not glob-imported anywhere in the current crate, we can trim its printed path and print only the name. This has wide implications on error messages with types, for example, shortening `std::vec::Vec` to just `Vec`, as long as there is no other `Vec` importable anywhere. This adds a new '-Z trim-diagnostic-paths=false' option to control this feature. On the good path, with no diagnosis printed, we should try to avoid issuing this query, so we need to prevent trimmed_def_paths query on several cases. This change also relies on a previous commit that differentiates between `Debug` and `Display` on various rustc types, where the latter is trimmed and presented to the user and the former is not.
2020-09-02 10:40:56 +03:00
_4 = Formatter::sign_plus(move _5) -> bb1; // scope 0 at $DIR/funky_arms.rs:15:22: 15:37
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
// mir::Constant
// + span: $DIR/funky_arms.rs:15:26: 15:35
tweak names and output and bless
2022-07-06 10:14:46 -04:00
// + literal: Const { ty: for<'r> fn(&'r Formatter) -> bool {Formatter::sign_plus}, val: Value(<ZST>) }
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
}
bb1: {
StorageDead(_5); // scope 0 at $DIR/funky_arms.rs:15:36: 15:37
StorageLive(_6); // scope 1 at $DIR/funky_arms.rs:19:9: 19:13
Fix debugger stepping behavior around `match` expressions Previously, we would set up the source lines for `match` expressions so that the code generated to perform the test of the scrutinee was matched to the line of the arm that required the test and then jump from the arm block to the "next" block was matched to all of the lines in the `match` expression. While that makes sense, it has the side effect of causing strange stepping behavior in debuggers. I've changed the source information so that all of the generated tests are sourced to `match {scrutinee}` and the jumps are sourced to the last line of the block they are inside. This resolves the weird stepping behavior in all debuggers and resolves some instances of "ambiguous symbol" errors in WinDbg preventing the user from setting breakpoints at `match` expressions.
2021-07-23 18:55:36 -04:00
switchInt(_4) -> [false: bb3, otherwise: bb2]; // scope 1 at $DIR/funky_arms.rs:19:16: 19:32
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
}
bb2: {
Fix tests broken by deaggregation change
2022-04-05 19:02:57 -04:00
Deinit(_6); // scope 1 at $DIR/funky_arms.rs:21:17: 21:41
Preserve signed zero on roundtrip This commit removes the previous mechanism of differentiating between "Debug" and "Display" formattings for the sign of -0 so as to comply with the IEEE 754 standard's requirements on external character sequences preserving various attributes of a floating point representation. In addition, numerous tests are fixed.
2020-10-31 13:16:14 -07:00
discriminant(_6) = 1; // scope 1 at $DIR/funky_arms.rs:21:17: 21:41
Fix debugger stepping behavior around `match` expressions Previously, we would set up the source lines for `match` expressions so that the code generated to perform the test of the scrutinee was matched to the line of the arm that required the test and then jump from the arm block to the "next" block was matched to all of the lines in the `match` expression. While that makes sense, it has the side effect of causing strange stepping behavior in debuggers. I've changed the source information so that all of the generated tests are sourced to `match {scrutinee}` and the jumps are sourced to the last line of the block they are inside. This resolves the weird stepping behavior in all debuggers and resolves some instances of "ambiguous symbol" errors in WinDbg preventing the user from setting breakpoints at `match` expressions.
2021-07-23 18:55:36 -04:00
goto -> bb4; // scope 1 at $DIR/funky_arms.rs:21:17: 21:41
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
}
bb3: {
Fix tests broken by deaggregation change
2022-04-05 19:02:57 -04:00
Deinit(_6); // scope 1 at $DIR/funky_arms.rs:20:18: 20:38
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
discriminant(_6) = 0; // scope 1 at $DIR/funky_arms.rs:20:18: 20:38
Fix debugger stepping behavior around `match` expressions Previously, we would set up the source lines for `match` expressions so that the code generated to perform the test of the scrutinee was matched to the line of the arm that required the test and then jump from the arm block to the "next" block was matched to all of the lines in the `match` expression. While that makes sense, it has the side effect of causing strange stepping behavior in debuggers. I've changed the source information so that all of the generated tests are sourced to `match {scrutinee}` and the jumps are sourced to the last line of the block they are inside. This resolves the weird stepping behavior in all debuggers and resolves some instances of "ambiguous symbol" errors in WinDbg preventing the user from setting breakpoints at `match` expressions.
2021-07-23 18:55:36 -04:00
goto -> bb4; // scope 1 at $DIR/funky_arms.rs:20:18: 20:38
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
}
bb4: {
Revert "More minimal changes" This reverts commit fe0dedcb06947317d41a8570b7fff7f8690dcbff.
2022-06-09 23:09:04 -07:00
StorageLive(_7); // scope 3 at $DIR/funky_arms.rs:24:30: 24:45
StorageLive(_8); // scope 3 at $DIR/funky_arms.rs:24:30: 24:45
_8 = &(*_1); // scope 3 at $DIR/funky_arms.rs:24:30: 24:45
_7 = Formatter::precision(move _8) -> bb5; // scope 3 at $DIR/funky_arms.rs:24:30: 24:45
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
// mir::Constant
// + span: $DIR/funky_arms.rs:24:34: 24:43
tweak names and output and bless
2022-07-06 10:14:46 -04:00
// + literal: Const { ty: for<'r> fn(&'r Formatter) -> Option<usize> {Formatter::precision}, val: Value(<ZST>) }
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
}
bb5: {
Revert "More minimal changes" This reverts commit fe0dedcb06947317d41a8570b7fff7f8690dcbff.
2022-06-09 23:09:04 -07:00
StorageDead(_8); // scope 3 at $DIR/funky_arms.rs:24:44: 24:45
_9 = discriminant(_7); // scope 3 at $DIR/funky_arms.rs:24:12: 24:27
switchInt(move _9) -> [1_isize: bb6, otherwise: bb8]; // scope 3 at $DIR/funky_arms.rs:24:12: 24:27
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
}
bb6: {
Fix `SourceScope` for `if let` bindings.
2022-06-09 11:06:44 -07:00
StorageLive(_10); // scope 3 at $DIR/funky_arms.rs:24:17: 24:26
_10 = ((_7 as Some).0: usize); // scope 3 at $DIR/funky_arms.rs:24:17: 24:26
Actually fix
2022-06-09 23:16:56 -07:00
StorageLive(_11); // scope 3 at $DIR/funky_arms.rs:26:43: 26:46
_11 = &mut (*_1); // scope 3 at $DIR/funky_arms.rs:26:43: 26:46
StorageLive(_12); // scope 3 at $DIR/funky_arms.rs:26:48: 26:51
_12 = _2; // scope 3 at $DIR/funky_arms.rs:26:48: 26:51
StorageLive(_13); // scope 3 at $DIR/funky_arms.rs:26:53: 26:57
_13 = _6; // scope 3 at $DIR/funky_arms.rs:26:53: 26:57
StorageLive(_14); // scope 3 at $DIR/funky_arms.rs:26:59: 26:79
StorageLive(_15); // scope 3 at $DIR/funky_arms.rs:26:59: 26:75
StorageLive(_16); // scope 3 at $DIR/funky_arms.rs:26:59: 26:68
_16 = _10; // scope 3 at $DIR/funky_arms.rs:26:59: 26:68
_15 = move _16 as u32 (Misc); // scope 3 at $DIR/funky_arms.rs:26:59: 26:75
StorageDead(_16); // scope 3 at $DIR/funky_arms.rs:26:74: 26:75
_14 = Add(move _15, const 1_u32); // scope 3 at $DIR/funky_arms.rs:26:59: 26:79
StorageDead(_15); // scope 3 at $DIR/funky_arms.rs:26:78: 26:79
StorageLive(_17); // scope 3 at $DIR/funky_arms.rs:26:81: 26:86
_17 = _3; // scope 3 at $DIR/funky_arms.rs:26:81: 26:86
_0 = float_to_exponential_common_exact::<T>(move _11, move _12, move _13, move _14, move _17) -> bb7; // scope 3 at $DIR/funky_arms.rs:26:9: 26:87
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
// mir::Constant
// + span: $DIR/funky_arms.rs:26:9: 26:42
tweak names and output and bless
2022-07-06 10:14:46 -04:00
// + literal: Const { ty: for<'r, 's, 't0> fn(&'r mut Formatter<'s>, &'t0 T, Sign, u32, bool) -> Result<(), std::fmt::Error> {float_to_exponential_common_exact::<T>}, val: Value(<ZST>) }
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
}
Fix drop handling for `if let` expressions MIR lowering for `if let` expressions is now more complicated now that `if let` exists in HIR. This PR adds a scope for the variables bound in an `if let` expression and then uses an approach similar to how we handle loops to ensure that we reliably drop the correct variables.
2021-09-01 22:52:17 +01:00
bb7: {
Actually fix
2022-06-09 23:16:56 -07:00
StorageDead(_17); // scope 3 at $DIR/funky_arms.rs:26:86: 26:87
StorageDead(_14); // scope 3 at $DIR/funky_arms.rs:26:86: 26:87
StorageDead(_13); // scope 3 at $DIR/funky_arms.rs:26:86: 26:87
StorageDead(_12); // scope 3 at $DIR/funky_arms.rs:26:86: 26:87
StorageDead(_11); // scope 3 at $DIR/funky_arms.rs:26:86: 26:87
Fix drop handling for `if let` expressions MIR lowering for `if let` expressions is now more complicated now that `if let` exists in HIR. This PR adds a scope for the variables bound in an `if let` expression and then uses an approach similar to how we handle loops to ensure that we reliably drop the correct variables.
2021-09-01 22:52:17 +01:00
StorageDead(_10); // scope 2 at $DIR/funky_arms.rs:27:5: 27:6
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
goto -> bb10; // scope 2 at $DIR/funky_arms.rs:24:5: 29:6
}
Fix drop handling for `if let` expressions MIR lowering for `if let` expressions is now more complicated now that `if let` exists in HIR. This PR adds a scope for the variables bound in an `if let` expression and then uses an approach similar to how we handle loops to ensure that we reliably drop the correct variables.
2021-09-01 22:52:17 +01:00
bb8: {
StorageLive(_18); // scope 2 at $DIR/funky_arms.rs:28:46: 28:49
_18 = &mut (*_1); // scope 2 at $DIR/funky_arms.rs:28:46: 28:49
StorageLive(_19); // scope 2 at $DIR/funky_arms.rs:28:51: 28:54
_19 = _2; // scope 2 at $DIR/funky_arms.rs:28:51: 28:54
StorageLive(_20); // scope 2 at $DIR/funky_arms.rs:28:56: 28:60
_20 = _6; // scope 2 at $DIR/funky_arms.rs:28:56: 28:60
StorageLive(_21); // scope 2 at $DIR/funky_arms.rs:28:62: 28:67
_21 = _3; // scope 2 at $DIR/funky_arms.rs:28:62: 28:67
_0 = float_to_exponential_common_shortest::<T>(move _18, move _19, move _20, move _21) -> bb9; // scope 2 at $DIR/funky_arms.rs:28:9: 28:68
// mir::Constant
// + span: $DIR/funky_arms.rs:28:9: 28:45
tweak names and output and bless
2022-07-06 10:14:46 -04:00
// + literal: Const { ty: for<'r, 's, 't0> fn(&'r mut Formatter<'s>, &'t0 T, Sign, bool) -> Result<(), std::fmt::Error> {float_to_exponential_common_shortest::<T>}, val: Value(<ZST>) }
Fix drop handling for `if let` expressions MIR lowering for `if let` expressions is now more complicated now that `if let` exists in HIR. This PR adds a scope for the variables bound in an `if let` expression and then uses an approach similar to how we handle loops to ensure that we reliably drop the correct variables.
2021-09-01 22:52:17 +01:00
}
move Deaggregate pass to post_borrowck_cleanup
2020-05-17 15:39:35 +02:00
bb9: {
StorageDead(_21); // scope 2 at $DIR/funky_arms.rs:28:67: 28:68
StorageDead(_20); // scope 2 at $DIR/funky_arms.rs:28:67: 28:68
StorageDead(_19); // scope 2 at $DIR/funky_arms.rs:28:67: 28:68
StorageDead(_18); // scope 2 at $DIR/funky_arms.rs:28:67: 28:68
goto -> bb10; // scope 2 at $DIR/funky_arms.rs:24:5: 29:6
}
bb10: {
StorageDead(_6); // scope 1 at $DIR/funky_arms.rs:30:1: 30:2
StorageDead(_4); // scope 0 at $DIR/funky_arms.rs:30:1: 30:2
StorageDead(_7); // scope 0 at $DIR/funky_arms.rs:30:1: 30:2
return; // scope 0 at $DIR/funky_arms.rs:30:2: 30:2
}
}
Reference in a new issue Copy permalink