maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/DebugInfo/X86/debug-info-block-captured-s...

113 lines
6.8 KiB
LLVM
Raw Normal View History

Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
; RUN: llc -mtriple x86_64-apple-darwin -filetype=obj -o %t.o < %s
; RUN: llvm-dwarfdump %t.o | FileCheck %s
;
; Test that DW_AT_location is generated for a captured "self" inside a
; block.
;
; This test is split into two parts, the frontend part can be found at
; llvm/tools/clang/test/CodeGenObjC/debug-info-block-captured-self.m
;
DebugInfo: Following up to r209677, refactor local variable emission to delay the choice between emitting the definition attributes or using DW_AT_abstract_definition This doesn't fix the abstract variable handling yet, but it introduces a similar delay mechanism as was added for subprograms, causing DW_AT_location to be reordered to the beginning of the attribute list for local variables, and fixes all the test fallout for that. A subsequent commit will remove the abstract variable handling in DbgVariable and just do the abstract variable lookup at module end to ensure that abstract variables introduced after their concrete counterparts are appropriately referenced by the concrete variable. llvm-svn: 210943
2014-06-14 06:18:23 +08:00
; CHECK: {{.*}}DW_AT_name{{.*}}_block_invoke{{.*}}
; CHECK: DW_TAG_variable
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
; CHECK-NOT: DW_TAG
DebugInfo: Following up to r209677, refactor local variable emission to delay the choice between emitting the definition attributes or using DW_AT_abstract_definition This doesn't fix the abstract variable handling yet, but it introduces a similar delay mechanism as was added for subprograms, causing DW_AT_location to be reordered to the beginning of the attribute list for local variables, and fixes all the test fallout for that. A subsequent commit will remove the abstract variable handling in DbgVariable and just do the abstract variable lookup at module end to ensure that abstract variables introduced after their concrete counterparts are appropriately referenced by the concrete variable. llvm-svn: 210943
2014-06-14 06:18:23 +08:00
; CHECK: DW_AT_location
; CHECK-NOT: DW_TAG
; CHECK: DW_AT_name{{.*}}"self"{{.*}}
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
;
DebugInfo: Following up to r209677, refactor local variable emission to delay the choice between emitting the definition attributes or using DW_AT_abstract_definition This doesn't fix the abstract variable handling yet, but it introduces a similar delay mechanism as was added for subprograms, causing DW_AT_location to be reordered to the beginning of the attribute list for local variables, and fixes all the test fallout for that. A subsequent commit will remove the abstract variable handling in DbgVariable and just do the abstract variable lookup at module end to ensure that abstract variables introduced after their concrete counterparts are appropriately referenced by the concrete variable. llvm-svn: 210943
2014-06-14 06:18:23 +08:00
; CHECK: {{.*}}DW_AT_name{{.*}}_block_invoke{{.*}}
; CHECK: DW_TAG_variable
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
; CHECK-NOT: DW_TAG
DebugInfo: Following up to r209677, refactor local variable emission to delay the choice between emitting the definition attributes or using DW_AT_abstract_definition This doesn't fix the abstract variable handling yet, but it introduces a similar delay mechanism as was added for subprograms, causing DW_AT_location to be reordered to the beginning of the attribute list for local variables, and fixes all the test fallout for that. A subsequent commit will remove the abstract variable handling in DbgVariable and just do the abstract variable lookup at module end to ensure that abstract variables introduced after their concrete counterparts are appropriately referenced by the concrete variable. llvm-svn: 210943
2014-06-14 06:18:23 +08:00
; CHECK: DW_AT_location
; CHECK-NOT: DW_TAG
; CHECK: DW_AT_name{{.*}}"self"{{.*}}
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
;
; Generated (and then reduced) from
; ----------------------------------------------------------------------
;
; @class T;
; @interface S
; @end
; @interface Mode
; -(int) count;
; @end
; @interface Context
; @end
; @interface ViewController
; @property (nonatomic, readwrite, strong) Context *context;
; @end
; typedef enum {
; Unknown = 0,
; } State;
; @interface Main : ViewController
; {
; T * t1;
; T * t2;
; }
; @property(readwrite, nonatomic) State state;
; @end
; @implementation Main
; - (id) initWithContext:(Context *) context
; {
; t1 = [self.context withBlock:^(id obj){
; id *mode1;
; t2 = [mode1 withBlock:^(id object){
; Mode *mode2 = object;
; if ([mode2 count] != 0) {
; self.state = 0;
; }
; }];
; }];
; }
; @end
; ----------------------------------------------------------------------
; ModuleID = 'llvm/tools/clang/test/CodeGenObjC/debug-info-block-captured-self.m'
%0 = type opaque
%struct.__block_descriptor = type { i64, i64 }
Move the complex address expression out of DIVariable and into an extra argument of the llvm.dbg.declare/llvm.dbg.value intrinsics. Previously, DIVariable was a variable-length field that has an optional reference to a Metadata array consisting of a variable number of complex address expressions. In the case of OpPiece expressions this is wasting a lot of storage in IR, because when an aggregate type is, e.g., SROA'd into all of its n individual members, the IR will contain n copies of the DIVariable, all alike, only differing in the complex address reference at the end. By making the complex address into an extra argument of the dbg.value/dbg.declare intrinsics, all of the pieces can reference the same variable and the complex address expressions can be uniqued across the CU, too. Down the road, this will allow us to move other flags, such as "indirection" out of the DIVariable, too. The new intrinsics look like this: declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr) declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr) This patch adds a new LLVM-local tag to DIExpressions, so we can detect and pretty-print DIExpression metadata nodes. What this patch doesn't do: This patch does not touch the "Indirect" field in DIVariable; but moving that into the expression would be a natural next step. http://reviews.llvm.org/D4919 rdar://problem/17994491 Thanks to dblaikie and dexonsmith for reviewing this patch! Note: I accidentally committed a bogus older version of this patch previously. llvm-svn: 218787
2014-10-02 02:55:02 +08:00
declare void @llvm.dbg.declare(metadata, metadata, metadata) #1
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
define internal void @"__24-[Main initWithContext:]_block_invoke"(i8* %.block_descriptor, i8* %obj) #0 {
%block = bitcast i8* %.block_descriptor to <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0* }>*, !dbg !84
%block.captured-self = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0* }>* %block, i32 0, i32 5, !dbg !84
Move the complex address expression out of DIVariable and into an extra argument of the llvm.dbg.declare/llvm.dbg.value intrinsics. Previously, DIVariable was a variable-length field that has an optional reference to a Metadata array consisting of a variable number of complex address expressions. In the case of OpPiece expressions this is wasting a lot of storage in IR, because when an aggregate type is, e.g., SROA'd into all of its n individual members, the IR will contain n copies of the DIVariable, all alike, only differing in the complex address reference at the end. By making the complex address into an extra argument of the dbg.value/dbg.declare intrinsics, all of the pieces can reference the same variable and the complex address expressions can be uniqued across the CU, too. Down the road, this will allow us to move other flags, such as "indirection" out of the DIVariable, too. The new intrinsics look like this: declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr) declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr) This patch adds a new LLVM-local tag to DIExpressions, so we can detect and pretty-print DIExpression metadata nodes. What this patch doesn't do: This patch does not touch the "Indirect" field in DIVariable; but moving that into the expression would be a natural next step. http://reviews.llvm.org/D4919 rdar://problem/17994491 Thanks to dblaikie and dexonsmith for reviewing this patch! Note: I accidentally committed a bogus older version of this patch previously. llvm-svn: 218787
2014-10-02 02:55:02 +08:00
call void @llvm.dbg.declare(metadata !{<{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0* }>* %block}, metadata !86, metadata !110), !dbg !87
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
ret void, !dbg !87
}
define internal void @"__24-[Main initWithContext:]_block_invoke_2"(i8* %.block_descriptor, i8* %object) #0 {
%block = bitcast i8* %.block_descriptor to <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0* }>*, !dbg !103
%block.captured-self = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0* }>* %block, i32 0, i32 5, !dbg !103
Move the complex address expression out of DIVariable and into an extra argument of the llvm.dbg.declare/llvm.dbg.value intrinsics. Previously, DIVariable was a variable-length field that has an optional reference to a Metadata array consisting of a variable number of complex address expressions. In the case of OpPiece expressions this is wasting a lot of storage in IR, because when an aggregate type is, e.g., SROA'd into all of its n individual members, the IR will contain n copies of the DIVariable, all alike, only differing in the complex address reference at the end. By making the complex address into an extra argument of the dbg.value/dbg.declare intrinsics, all of the pieces can reference the same variable and the complex address expressions can be uniqued across the CU, too. Down the road, this will allow us to move other flags, such as "indirection" out of the DIVariable, too. The new intrinsics look like this: declare void @llvm.dbg.declare(metadata %storage, metadata %var, metadata %expr) declare void @llvm.dbg.value(metadata %storage, i64 %offset, metadata %var, metadata %expr) This patch adds a new LLVM-local tag to DIExpressions, so we can detect and pretty-print DIExpression metadata nodes. What this patch doesn't do: This patch does not touch the "Indirect" field in DIVariable; but moving that into the expression would be a natural next step. http://reviews.llvm.org/D4919 rdar://problem/17994491 Thanks to dblaikie and dexonsmith for reviewing this patch! Note: I accidentally committed a bogus older version of this patch previously. llvm-svn: 218787
2014-10-02 02:55:02 +08:00
call void @llvm.dbg.declare(metadata !{<{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0* }>* %block}, metadata !105, metadata !109), !dbg !106
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
ret void, !dbg !106
}
!llvm.dbg.cu = !{!0}
Debug Info: update testing cases to specify the debug info version number. We are going to drop debug info without a version number or with a different version number, to make sure we don't crash when we see bitcode files with different debug info metadata format. llvm-svn: 195504
2013-11-23 05:49:45 +08:00
!llvm.module.flags = !{!108}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!0 = metadata !{metadata !"0x11\0016\00clang version 3.3 \000\00\002\00\000", metadata !107, metadata !2, metadata !4, metadata !23, metadata !15, metadata !15} ; [ DW_TAG_compile_unit ] [llvm/tools/clang/test/CodeGenObjC/debug-info-block-captured-self.m] [DW_LANG_ObjC]
!1 = metadata !{metadata !"0x29", metadata !107} ; [ DW_TAG_file_type ]
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
!2 = metadata !{metadata !3}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!3 = metadata !{metadata !"0x4\00\0020\0032\0032\000\000\000", metadata !107, null, null, metadata !4, null, null, null} ; [ DW_TAG_enumeration_type ] [line 20, size 32, align 32, offset 0] [def] [from ]
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
!4 = metadata !{}
DebugInfo: Remove some unneeded conditionals now that DIBuilder no longer emits zero-length arrays as {i32 0} A bunch of test cases needed to be cleaned up for this, many my fault - when implementid imported modules I updated test cases by simply duplicating the prior metadata field - which wasn't always the empty metadata entry. llvm-svn: 200731
2014-02-04 09:23:52 +08:00
!15 = metadata !{}
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
!23 = metadata !{metadata !38, metadata !42}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!27 = metadata !{metadata !"0x16\00id\0031\000\000\000\000", metadata !107, null, metadata !28} ; [ DW_TAG_typedef ] [id] [line 31, size 0, align 0, offset 0] [from ]
!28 = metadata !{metadata !"0xf\00\000\0064\0064\000\000", null, null, metadata !29} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from objc_object]
!29 = metadata !{metadata !"0x13\00objc_object\000\000\000\000\000\000", metadata !107, null, null, metadata !30, null, null, null} ; [ DW_TAG_structure_type ] [objc_object] [line 0, size 0, align 0, offset 0] [def] [from ]
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
!30 = metadata !{metadata !31}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!31 = metadata !{metadata !"0xd\00isa\000\0064\000\000\000", metadata !107, metadata !29, metadata !32} ; [ DW_TAG_member ] [isa] [line 0, size 64, align 0, offset 0] [from ]
!32 = metadata !{metadata !"0xf\00\000\0064\000\000\000", null, null, metadata !33} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 0, offset 0] [from objc_class]
!33 = metadata !{metadata !"0x13\00objc_class\000\000\000\000\004\000", metadata !107, null, null, null, null, null, null} ; [ DW_TAG_structure_type ] [objc_class] [line 0, size 0, align 0, offset 0] [decl] [from ]
!34 = metadata !{metadata !"0x13\00Main\0023\000\000\000\001092\0016", metadata !107, null, null, i32 0, null, null, null} ; [ DW_TAG_structure_type ] [Main] [line 23, size 0, align 0, offset 0] [artificial] [decl] [from ]
!38 = metadata !{metadata !"0x2e\00__24-[Main initWithContext:]_block_invoke\00__24-[Main initWithContext:]_block_invoke\00\0033\001\001\000\006\00256\000\0033", metadata !1, metadata !1, metadata !39, null, void (i8*, i8*)* @"__24-[Main initWithContext:]_block_invoke", null, null, metadata !15} ; [ DW_TAG_subprogram ] [line 33] [local] [def] [__24-[Main initWithContext:]_block_invoke]
!39 = metadata !{metadata !"0x15\00\000\000\000\000\000\000", null, null, null, metadata !40, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
!40 = metadata !{null, metadata !41, metadata !27}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!41 = metadata !{metadata !"0xf\00\000\0064\0064\000\000", null, null, null} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from ]
!42 = metadata !{metadata !"0x2e\00__24-[Main initWithContext:]_block_invoke_2\00__24-[Main initWithContext:]_block_invoke_2\00\0035\001\001\000\006\00256\000\0035", metadata !1, metadata !1, metadata !39, null, void (i8*, i8*)* @"__24-[Main initWithContext:]_block_invoke_2", null, null, metadata !15} ; [ DW_TAG_subprogram ] [line 35] [local] [def] [__24-[Main initWithContext:]_block_invoke_2]
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
!84 = metadata !{i32 33, i32 0, metadata !38, null}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!86 = metadata !{metadata !"0x100\00self\0041\000", metadata !38, metadata !1, metadata !34} ; [ DW_TAG_auto_variable ] [self] [line 41]
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
!87 = metadata !{i32 41, i32 0, metadata !38, null}
!103 = metadata !{i32 35, i32 0, metadata !42, null}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!105 = metadata !{metadata !"0x100\00self\0040\000", metadata !42, metadata !1, metadata !34} ; [ DW_TAG_auto_variable ] [self] [line 40]
Test that we emit a DW_AT_location for self captured by a block. This is the backend part of a CFE test with the same name. llvm-svn: 177087
2013-03-15 01:54:13 +08:00
!106 = metadata !{i32 40, i32 0, metadata !42, null}
Split out filename & directory from DIFile to start generalizing over DIScopes This is the first step to making all DIScopes have a common metadata prefix (so that things (using directives, for example) that can appear in any scope can be added to that common prefix). DIFile is itself a DIScope so the common prefix of all DIScopes cannot be a DIFile - instead it's the raw filename/directory name pair. llvm-svn: 177239
2013-03-18 05:13:55 +08:00
!107 = metadata !{metadata !"llvm/tools/clang/test/CodeGenObjC/debug-info-block-captured-self.m", metadata !""}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!108 = metadata !{i32 1, metadata !"Debug Info Version", i32 2}
!109 = metadata !{metadata !"0x102\0034\0032"} ; [ DW_TAG_expression ] [DW_OP_plus 32]
!110 = metadata !{metadata !"0x102\0034\0032"} ; [ DW_TAG_expression ] [DW_OP_plus 32]