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/generate-odr-hash.ll

277 lines
13 KiB
LLVM
Raw Normal View History

Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; REQUIRES: object-emission
Reapply "DebugInfo: Don't put fission type units in comdat sections." This recommits r208930, r208933, and r208975 (by reverting r209338) and reverts r209529 (the FIXME to readd this functionality once the tools were fixed) now that DWP has been fixed to cope with a single section for all fission type units. Original commit message: "Since type units in the dwo file are handled by a debug aware tool, they don't need to leverage the ELF comdat grouping to implement deduplication. Avoid creating all the .group sections for these as a space optimization." llvm-svn: 213956
2014-07-26 01:11:58 +08:00
; RUN: llc < %s -o %t -filetype=obj -O0 -generate-type-units -mtriple=x86_64-unknown-linux-gnu
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; RUN: llvm-dwarfdump %t | FileCheck --check-prefix=CHECK --check-prefix=SINGLE %s
Remove type unit skeletons. GDB no longer needs them & this saves a heap of space. llvm-svn: 216521
2014-08-27 13:04:14 +08:00
; RUN: llvm-readobj -s -t %t | FileCheck --check-prefix=OBJ_SINGLE %s
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
Reapply "DebugInfo: Don't put fission type units in comdat sections." This recommits r208930, r208933, and r208975 (by reverting r209338) and reverts r209529 (the FIXME to readd this functionality once the tools were fixed) now that DWP has been fixed to cope with a single section for all fission type units. Original commit message: "Since type units in the dwo file are handled by a debug aware tool, they don't need to leverage the ELF comdat grouping to implement deduplication. Avoid creating all the .group sections for these as a space optimization." llvm-svn: 213956
2014-07-26 01:11:58 +08:00
; RUN: llc < %s -split-dwarf=Enable -o %t -filetype=obj -O0 -generate-type-units -mtriple=x86_64-unknown-linux-gnu
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; RUN: llvm-dwarfdump %t | FileCheck --check-prefix=CHECK --check-prefix=FISSION %s
Remove type unit skeletons. GDB no longer needs them & this saves a heap of space. llvm-svn: 216521
2014-08-27 13:04:14 +08:00
; RUN: llvm-readobj -s -t %t | FileCheck --check-prefix=OBJ_FISSION %s
DebugInfo: Avoid emitting pubtype entries for type DIEs that just indirect to a type unit. llvm-svn: 195698
2013-11-26 08:22:37 +08:00
DwarfUnit: Include type unit's file strings in the defining compile unit's file_names table There's still one piece missing here, which is adding the DW_AT_stmt_list to the type unit that refer's to the compile unit's line table. Working on that. llvm-svn: 201198
2014-02-12 08:40:47 +08:00
; Generated from bar.cpp:
; #line 1 "bar.h"
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; struct bar {};
DwarfUnit: Include type unit's file strings in the defining compile unit's file_names table There's still one piece missing here, which is adding the DW_AT_stmt_list to the type unit that refer's to the compile unit's line table. Working on that. llvm-svn: 201198
2014-02-12 08:40:47 +08:00
; #line 2 "bar.cpp"
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; struct bar b;
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; void foo(void) {
; struct baz {};
; baz b;
; }
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; namespace echidna {
; namespace capybara {
; namespace mongoose {
; class fluffy {
; int a;
; int b;
; };
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; fluffy animal;
; }
; }
; }
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; namespace {
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
; struct walrus {
; walrus() {}
; };
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; }
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; walrus w;
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
; struct wombat {
; struct {
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
; int a;
; int b;
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
; } a_b;
; };
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
; wombat wom;
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; SINGLE-LABEL: .debug_info contents:
; FISSION-LABEL: .debug_info.dwo contents:
Make these two tests resilient in the face of compile unit size changes. llvm-svn: 196444
2013-12-05 09:00:12 +08:00
; CHECK: Compile Unit: length = [[CU_SIZE:[0-9a-f]+]]
DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers This commit does not complete the type units feature - there are issues around fission support (skeletal type units, pubtypes/pubnames) and hashing of some types including those containing references to types in other type units. Originally committed as r197073 and reverted in r197079. Recommitted as r197197 to reproduce the failure and reverted as r197199 Turns out there was unstable ordering in the type unit dumping code. Fixed by using MapVector in DWARFContext to store the debug_types comdat sections. Recommitted as r197210 with a fix to dumping and reverted as r197211 because I was a bit gun shy and thought I saw a failure that turned out to be unrelated. So here we go - once more with feeling! \o/ llvm-svn: 197275
2013-12-14 05:33:40 +08:00
DebugInfo: Shrink pubnames/pubtypes in the presence of type units by only emitting pub sections for compile units llvm-svn: 203057
2014-03-06 09:42:00 +08:00
; CHECK: [[BAR:^0x........]]: DW_TAG_structure_type
DebugInfo: Flag type unit references as declarations This ensures DWARF consumers don't confuse these references for definitions. I'd argue it might be nice to improve debuggers so we don't need this, but it's just one field in an abbreviation anyway - so it doesn't seem worth the fight. llvm-svn: 200569
2014-02-01 03:52:26 +08:00
; CHECK-NEXT: DW_AT_declaration
DebugInfo: Shrink pubnames/pubtypes in the presence of type units by only emitting pub sections for compile units llvm-svn: 203057
2014-03-06 09:42:00 +08:00
; CHECK-NEXT: DW_AT_signature {{.*}} (0x1d02f3be30cc5688)
; CHECK: [[FLUFFY:^0x........]]: DW_TAG_class_type
DebugInfo: Flag type unit references as declarations This ensures DWARF consumers don't confuse these references for definitions. I'd argue it might be nice to improve debuggers so we don't need this, but it's just one field in an abbreviation anyway - so it doesn't seem worth the fight. llvm-svn: 200569
2014-02-01 03:52:26 +08:00
; CHECK-NEXT: DW_AT_declaration
DebugInfo: Shrink pubnames/pubtypes in the presence of type units by only emitting pub sections for compile units llvm-svn: 203057
2014-03-06 09:42:00 +08:00
; CHECK-NEXT: DW_AT_signature {{.*}} (0xb04af47397402e77)
Revert "Revert "Debug Info: Type Units: Simplify type hashing using IR-provided unique names."" This reverts commit r198398, thus reapplying r198397. I had accidentally introduced an endianness issue when applying the hash to the type unit. Using support::ulittle64_t in the reinterpret_cast in addDwarfTypeUnitType fixes this issue. Original commit message: Debug Info: Type Units: Simplify type hashing using IR-provided unique names. What's good for LTO metadata size problems ought to be good for non-LTO debug info size too, so let's rely on the same uniqueness in both cases. If it's insufficient for non-LTO for whatever reason (since we now won't be uniquing CU-local types or any C types - but these are likely to not be the most significant contributors to type bloat) we should consider a frontend solution that'll help both LTO and non-LTO alike, rather than using DWARF-level DIE-hashing that only helps non-LTO debug info size. It's also much simpler this way and benefits C++ even more since we can deduplicate lexically separate definitions of the same C++ type since they have the same mangled name. llvm-svn: 198436
2014-01-04 02:59:42 +08:00
; Ensure the CU-local type 'walrus' is not placed in a type unit.
DebugInfo: Shrink pubnames/pubtypes in the presence of type units by only emitting pub sections for compile units llvm-svn: 203057
2014-03-06 09:42:00 +08:00
; CHECK: [[WALRUS:^0x........]]: DW_TAG_structure_type
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; CHECK-NEXT: DW_AT_name{{.*}}"walrus"
Revert "Revert "Debug Info: Type Units: Simplify type hashing using IR-provided unique names."" This reverts commit r198398, thus reapplying r198397. I had accidentally introduced an endianness issue when applying the hash to the type unit. Using support::ulittle64_t in the reinterpret_cast in addDwarfTypeUnitType fixes this issue. Original commit message: Debug Info: Type Units: Simplify type hashing using IR-provided unique names. What's good for LTO metadata size problems ought to be good for non-LTO debug info size too, so let's rely on the same uniqueness in both cases. If it's insufficient for non-LTO for whatever reason (since we now won't be uniquing CU-local types or any C types - but these are likely to not be the most significant contributors to type bloat) we should consider a frontend solution that'll help both LTO and non-LTO alike, rather than using DWARF-level DIE-hashing that only helps non-LTO debug info size. It's also much simpler this way and benefits C++ even more since we can deduplicate lexically separate definitions of the same C++ type since they have the same mangled name. llvm-svn: 198436
2014-01-04 02:59:42 +08:00
; CHECK-NEXT: DW_AT_byte_size
; CHECK-NEXT: DW_AT_decl_file
; CHECK-NEXT: DW_AT_decl_line
DebugInfo: Shrink pubnames/pubtypes in the presence of type units by only emitting pub sections for compile units llvm-svn: 203057
2014-03-06 09:42:00 +08:00
; CHECK: [[WOMBAT:^0x........]]: DW_TAG_structure_type
; CHECK-NEXT: DW_AT_declaration
; CHECK-NEXT: DW_AT_signature {{.*}} (0xfd756cee88f8a118)
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; SINGLE-LABEL: .debug_types contents:
Remove type unit skeletons. GDB no longer needs them & this saves a heap of space. llvm-svn: 216521
2014-08-27 13:04:14 +08:00
; FISSION-NOT: .debug_types contents:
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; FISSION-LABEL: .debug_types.dwo contents:
DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers This commit does not complete the type units feature - there are issues around fission support (skeletal type units, pubtypes/pubnames) and hashing of some types including those containing references to types in other type units. Originally committed as r197073 and reverted in r197079. Recommitted as r197197 to reproduce the failure and reverted as r197199 Turns out there was unstable ordering in the type unit dumping code. Fixed by using MapVector in DWARFContext to store the debug_types comdat sections. Recommitted as r197210 with a fix to dumping and reverted as r197211 because I was a bit gun shy and thought I saw a failure that turned out to be unrelated. So here we go - once more with feeling! \o/ llvm-svn: 197275
2013-12-14 05:33:40 +08:00
Revert "DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers" This reverts commit r197197. llvm-svn: 197199
2013-12-13 09:24:54 +08:00
; Check that we generate a hash for bar and the value.
Revert "Revert "Debug Info: Type Units: Simplify type hashing using IR-provided unique names."" This reverts commit r198398, thus reapplying r198397. I had accidentally introduced an endianness issue when applying the hash to the type unit. Using support::ulittle64_t in the reinterpret_cast in addDwarfTypeUnitType fixes this issue. Original commit message: Debug Info: Type Units: Simplify type hashing using IR-provided unique names. What's good for LTO metadata size problems ought to be good for non-LTO debug info size too, so let's rely on the same uniqueness in both cases. If it's insufficient for non-LTO for whatever reason (since we now won't be uniquing CU-local types or any C types - but these are likely to not be the most significant contributors to type bloat) we should consider a frontend solution that'll help both LTO and non-LTO alike, rather than using DWARF-level DIE-hashing that only helps non-LTO debug info size. It's also much simpler this way and benefits C++ even more since we can deduplicate lexically separate definitions of the same C++ type since they have the same mangled name. llvm-svn: 198436
2014-01-04 02:59:42 +08:00
; CHECK-NOT: type_signature
; CHECK-LABEL: type_signature = 0x1d02f3be30cc5688
Revert "DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers" This reverts commit r197197. llvm-svn: 197199
2013-12-13 09:24:54 +08:00
; CHECK: DW_TAG_structure_type
Provide gmlt-like inline scope information in the skeleton CU to facilitate symbolication without needing the .dwo files Clang -gsplit-dwarf self-host -O0, binary increases by 0.0005%, -O2, binary increases by 25%. A large binary inside Google, split-dwarf, -O0, and other internal flags (GDB index, etc) increases by 1.8%, optimized build is 35%. The size impact may be somewhat greater in .o files (I haven't measured that much - since the linked executable -O0 numbers seemed low enough) due to relocations. These relocations could be removed if we taught the llvm-symbolizer to handle indexed addressing in the .o file (GDB can't cope with this just yet, but GDB won't be reading this info anyway). Also debug_ranges could be shared between .o and .dwo, though ideally debug_ranges would get a schema that could used index(+offset) addressing, and move to the .dwo file, then we'd be back to sharing addresses in the address pool again. But for now, these sizes seem small enough to go ahead with this. Verified that no other DW_TAGs are produced into the .o file other than subprograms and inlined_subroutines. llvm-svn: 221306
2014-11-05 06:12:25 +08:00
; FISSION-NEXT: DW_AT_name {{.*}} ( indexed {{.*}} "bar"
; SINGLE-NEXT: DW_AT_name {{.*}} "bar"
DebugInfo: Partial implementation of DWARF type units. Emit DW_TAG_type_units into the debug_info section using compile unit headers. This is bogus/unusable by debuggers, but testable and provides more isolated review. Subsequent patches will include support for type unit headers and emission into the debug_types section, as well as comdat grouping the types based on their hash. Also the CompileUnit type will be renamed 'Unit' and relevant portions pulled out into respective CompileUnit and TypeUnit types. llvm-svn: 195166
2013-11-20 07:08:21 +08:00
DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers This commit does not complete the type units feature - there are issues around fission support (skeletal type units, pubtypes/pubnames) and hashing of some types including those containing references to types in other type units. Originally committed as r197073 and reverted in r197079. Recommitted as r197197 to reproduce the failure and reverted as r197199 Turns out there was unstable ordering in the type unit dumping code. Fixed by using MapVector in DWARFContext to store the debug_types comdat sections. Recommitted as r197210 with a fix to dumping and reverted as r197211 because I was a bit gun shy and thought I saw a failure that turned out to be unrelated. So here we go - once more with feeling! \o/ llvm-svn: 197275
2013-12-14 05:33:40 +08:00
DebugInfo: Partial implementation of DWARF type units. Emit DW_TAG_type_units into the debug_info section using compile unit headers. This is bogus/unusable by debuggers, but testable and provides more isolated review. Subsequent patches will include support for type unit headers and emission into the debug_types section, as well as comdat grouping the types based on their hash. Also the CompileUnit type will be renamed 'Unit' and relevant portions pulled out into respective CompileUnit and TypeUnit types. llvm-svn: 195166
2013-11-20 07:08:21 +08:00
; Check that we generate a hash for fluffy and the value.
Revert "Revert "Debug Info: Type Units: Simplify type hashing using IR-provided unique names."" This reverts commit r198398, thus reapplying r198397. I had accidentally introduced an endianness issue when applying the hash to the type unit. Using support::ulittle64_t in the reinterpret_cast in addDwarfTypeUnitType fixes this issue. Original commit message: Debug Info: Type Units: Simplify type hashing using IR-provided unique names. What's good for LTO metadata size problems ought to be good for non-LTO debug info size too, so let's rely on the same uniqueness in both cases. If it's insufficient for non-LTO for whatever reason (since we now won't be uniquing CU-local types or any C types - but these are likely to not be the most significant contributors to type bloat) we should consider a frontend solution that'll help both LTO and non-LTO alike, rather than using DWARF-level DIE-hashing that only helps non-LTO debug info size. It's also much simpler this way and benefits C++ even more since we can deduplicate lexically separate definitions of the same C++ type since they have the same mangled name. llvm-svn: 198436
2014-01-04 02:59:42 +08:00
; CHECK-NOT: type_signature
; CHECK-LABEL: type_signature = 0xb04af47397402e77
; CHECK-NOT: DW_AT_GNU_odr_signature [DW_FORM_data8] (0x9a0124d5a0c21c52)
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; CHECK: DW_TAG_namespace
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; CHECK-NEXT: DW_AT_name{{.*}}"echidna"
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; CHECK: DW_TAG_namespace
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; CHECK-NEXT: DW_AT_name{{.*}}"capybara"
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; CHECK: DW_TAG_namespace
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; CHECK-NEXT: DW_AT_name{{.*}}"mongoose"
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; CHECK: DW_TAG_class_type
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; CHECK-NEXT: DW_AT_name{{.*}}"fluffy"
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
Revert "DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers" This reverts commit r197197. llvm-svn: 197199
2013-12-13 09:24:54 +08:00
; Check that we generate a hash for wombat and the value, but not for the
; anonymous type contained within.
Revert "Revert "Debug Info: Type Units: Simplify type hashing using IR-provided unique names."" This reverts commit r198398, thus reapplying r198397. I had accidentally introduced an endianness issue when applying the hash to the type unit. Using support::ulittle64_t in the reinterpret_cast in addDwarfTypeUnitType fixes this issue. Original commit message: Debug Info: Type Units: Simplify type hashing using IR-provided unique names. What's good for LTO metadata size problems ought to be good for non-LTO debug info size too, so let's rely on the same uniqueness in both cases. If it's insufficient for non-LTO for whatever reason (since we now won't be uniquing CU-local types or any C types - but these are likely to not be the most significant contributors to type bloat) we should consider a frontend solution that'll help both LTO and non-LTO alike, rather than using DWARF-level DIE-hashing that only helps non-LTO debug info size. It's also much simpler this way and benefits C++ even more since we can deduplicate lexically separate definitions of the same C++ type since they have the same mangled name. llvm-svn: 198436
2014-01-04 02:59:42 +08:00
; CHECK-NOT: type_signature
; CHECK-LABEL: type_signature = 0xfd756cee88f8a118
; CHECK-NOT: DW_AT_GNU_odr_signature [DW_FORM_data8] (0x685bcc220141e9d7)
Revert "DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers" This reverts commit r197197. llvm-svn: 197199
2013-12-13 09:24:54 +08:00
; CHECK: DW_TAG_structure_type
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; CHECK-NEXT: DW_AT_name{{.*}}"wombat"
DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers This commit does not complete the type units feature - there are issues around fission support (skeletal type units, pubtypes/pubnames) and hashing of some types including those containing references to types in other type units. Originally committed as r197073 and reverted in r197079. Recommitted as r197197 to reproduce the failure and reverted as r197199 Turns out there was unstable ordering in the type unit dumping code. Fixed by using MapVector in DWARFContext to store the debug_types comdat sections. Recommitted as r197210 with a fix to dumping and reverted as r197211 because I was a bit gun shy and thought I saw a failure that turned out to be unrelated. So here we go - once more with feeling! \o/ llvm-svn: 197275
2013-12-14 05:33:40 +08:00
Revert "Revert "Debug Info: Type Units: Simplify type hashing using IR-provided unique names."" This reverts commit r198398, thus reapplying r198397. I had accidentally introduced an endianness issue when applying the hash to the type unit. Using support::ulittle64_t in the reinterpret_cast in addDwarfTypeUnitType fixes this issue. Original commit message: Debug Info: Type Units: Simplify type hashing using IR-provided unique names. What's good for LTO metadata size problems ought to be good for non-LTO debug info size too, so let's rely on the same uniqueness in both cases. If it's insufficient for non-LTO for whatever reason (since we now won't be uniquing CU-local types or any C types - but these are likely to not be the most significant contributors to type bloat) we should consider a frontend solution that'll help both LTO and non-LTO alike, rather than using DWARF-level DIE-hashing that only helps non-LTO debug info size. It's also much simpler this way and benefits C++ even more since we can deduplicate lexically separate definitions of the same C++ type since they have the same mangled name. llvm-svn: 198436
2014-01-04 02:59:42 +08:00
; CHECK-NOT: type_signature
; CHECK-LABEL: type_signature = 0xe94f6d3843e62d6b
Revert "DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers" This reverts commit r197197. llvm-svn: 197199
2013-12-13 09:24:54 +08:00
; CHECK: DW_TAG_type_unit
DebugInfo: Implement debug_line.dwo for file names used in type units during -gsplit-dwarf This removes an attribute (and more importantly, a relocation) from skeleton type units and removes some unnecessary file names from the debug_line section that remains in the .o (and linked executable) file. There's still a few places we could shave off some more space here: * use compilation dir of the underlying compilation unit (since all the type units share that compilation dir - though this would be more complicated in LTO cases where they don't (keep a map of compilation dir->line table header?)) * Remove some of the unnecessary header fields from the line table since they're not needed in this situation (about 12 bytes per table). llvm-svn: 204099
2014-03-18 09:17:26 +08:00
; CHECK: DW_AT_stmt_list [DW_FORM_sec_offset] (0x00000000)
DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers This commit does not complete the type units feature - there are issues around fission support (skeletal type units, pubtypes/pubnames) and hashing of some types including those containing references to types in other type units. Originally committed as r197073 and reverted in r197079. Recommitted as r197197 to reproduce the failure and reverted as r197199 Turns out there was unstable ordering in the type unit dumping code. Fixed by using MapVector in DWARFContext to store the debug_types comdat sections. Recommitted as r197210 with a fix to dumping and reverted as r197211 because I was a bit gun shy and thought I saw a failure that turned out to be unrelated. So here we go - once more with feeling! \o/ llvm-svn: 197275
2013-12-14 05:33:40 +08:00
; CHECK-NOT: NULL
; CHECK-NOT: DW_AT_GNU_odr_signature
Revert "DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers" This reverts commit r197197. llvm-svn: 197199
2013-12-13 09:24:54 +08:00
; CHECK: DW_TAG_structure_type
DebugInfo: Move type units into the debug_types section with appropriate comdat grouping and type unit headers This commit does not complete the type units feature - there are issues around fission support (skeletal type units, pubtypes/pubnames) and hashing of some types including those containing references to types in other type units. Originally committed as r197073 and reverted in r197079. Recommitted as r197197 to reproduce the failure and reverted as r197199 Turns out there was unstable ordering in the type unit dumping code. Fixed by using MapVector in DWARFContext to store the debug_types comdat sections. Recommitted as r197210 with a fix to dumping and reverted as r197211 because I was a bit gun shy and thought I saw a failure that turned out to be unrelated. So here we go - once more with feeling! \o/ llvm-svn: 197275
2013-12-14 05:33:40 +08:00
; The signature for the outer 'wombat' type
Revert "Revert "Debug Info: Type Units: Simplify type hashing using IR-provided unique names."" This reverts commit r198398, thus reapplying r198397. I had accidentally introduced an endianness issue when applying the hash to the type unit. Using support::ulittle64_t in the reinterpret_cast in addDwarfTypeUnitType fixes this issue. Original commit message: Debug Info: Type Units: Simplify type hashing using IR-provided unique names. What's good for LTO metadata size problems ought to be good for non-LTO debug info size too, so let's rely on the same uniqueness in both cases. If it's insufficient for non-LTO for whatever reason (since we now won't be uniquing CU-local types or any C types - but these are likely to not be the most significant contributors to type bloat) we should consider a frontend solution that'll help both LTO and non-LTO alike, rather than using DWARF-level DIE-hashing that only helps non-LTO debug info size. It's also much simpler this way and benefits C++ even more since we can deduplicate lexically separate definitions of the same C++ type since they have the same mangled name. llvm-svn: 198436
2014-01-04 02:59:42 +08:00
; CHECK: DW_AT_signature [DW_FORM_ref_sig8] (0xfd756cee88f8a118)
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
; CHECK: DW_TAG_structure_type
DebugInfo: Partial implementation of DWARF type units. Emit DW_TAG_type_units into the debug_info section using compile unit headers. This is bogus/unusable by debuggers, but testable and provides more isolated review. Subsequent patches will include support for type unit headers and emission into the debug_types section, as well as comdat grouping the types based on their hash. Also the CompileUnit type will be renamed 'Unit' and relevant portions pulled out into respective CompileUnit and TypeUnit types. llvm-svn: 195166
2013-11-20 07:08:21 +08:00
; CHECK-NOT: DW_AT_name
; CHECK-NOT: DW_AT_GNU_odr_signature
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
; CHECK: DW_TAG_member
Revert "Revert r198851, "Prototype of skeleton type units for fission"" This reverts commit r198865 which reverts r198851. ASan identified a use-of-uninitialized of the DwarfTypeUnit::Ty variable in skeleton type units. llvm-svn: 198908
2014-01-10 09:38:41 +08:00
; CHECK-NEXT: DW_AT_name{{.*}}"a"
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
DebugInfo: Don't include the name of the CU file in the line table file list when it's unneeded Recommitting r201380 (reverted in r201389) Recommitting r201351 and r201355 (reverted in r201351 and r201355) We weren't emitting the an empty (header only) line table when the line table was empty - this made the DWARF invalid (the compile unit would point to the zero-size debug_lines section where there should've been an empty line table but there was nothing at all). Fix that, and as a consequence this works around/addresses PR18809. Also, we emit a non-empty line table to workaround a darwin linker bug, so XFAILing on darwin too. Also, mark the test as 'REQUIRES: object-emission' because it does. llvm-svn: 201429
2014-02-15 03:51:35 +08:00
; CHECK-LABEL: .debug_line contents:
; CHECK: Line table prologue
; CHECK-NOT: file_names[
DebugInfo: Implement debug_line.dwo for file names used in type units during -gsplit-dwarf This removes an attribute (and more importantly, a relocation) from skeleton type units and removes some unnecessary file names from the debug_line section that remains in the .o (and linked executable) file. There's still a few places we could shave off some more space here: * use compilation dir of the underlying compilation unit (since all the type units share that compilation dir - though this would be more complicated in LTO cases where they don't (keep a map of compilation dir->line table header?)) * Remove some of the unnecessary header fields from the line table since they're not needed in this situation (about 12 bytes per table). llvm-svn: 204099
2014-03-18 09:17:26 +08:00
; SINGLE: file_names{{.*}} bar.h
DebugInfo: Don't include the name of the CU file in the line table file list when it's unneeded Recommitting r201380 (reverted in r201389) Recommitting r201351 and r201355 (reverted in r201351 and r201355) We weren't emitting the an empty (header only) line table when the line table was empty - this made the DWARF invalid (the compile unit would point to the zero-size debug_lines section where there should've been an empty line table but there was nothing at all). Fix that, and as a consequence this works around/addresses PR18809. Also, we emit a non-empty line table to workaround a darwin linker bug, so XFAILing on darwin too. Also, mark the test as 'REQUIRES: object-emission' because it does. llvm-svn: 201429
2014-02-15 03:51:35 +08:00
; CHECK: file_names{{.*}} bar.cpp
; CHECK-NOT: file_names[
DebugInfo: Implement debug_line.dwo for file names used in type units during -gsplit-dwarf This removes an attribute (and more importantly, a relocation) from skeleton type units and removes some unnecessary file names from the debug_line section that remains in the .o (and linked executable) file. There's still a few places we could shave off some more space here: * use compilation dir of the underlying compilation unit (since all the type units share that compilation dir - though this would be more complicated in LTO cases where they don't (keep a map of compilation dir->line table header?)) * Remove some of the unnecessary header fields from the line table since they're not needed in this situation (about 12 bytes per table). llvm-svn: 204099
2014-03-18 09:17:26 +08:00
; CHECK-LABEL: .debug_line.dwo contents:
; FISSION: Line table prologue
DebugInfo: Avoid emitting standard opcode lengths in debug_line.dwo headers where opcodes are never used anyway Introduce a slightly tighter wrapper around the header structure that handles this use case. (MCDwarfDwoLineTable) llvm-svn: 204101
2014-03-18 10:13:23 +08:00
; FISSION: opcode_base: 1
; FISSION-NOT: standard_opcode_lengths
DebugInfo: Use the comp_dir of the referencing type units when building debug_line.dwo This isn't a complete fix - it falls back to non-comp_dir when multiple compile units are in play. Adding a map of comp_dir to table is part of the more general solution, but I gave up (in the short term) when I realized I'd also have to calculate the size of each type unit so as to produce correct DW_AT_stmt_list attributes. llvm-svn: 204202
2014-03-19 08:11:28 +08:00
; FISSION-NOT: include_directories
DebugInfo: Implement debug_line.dwo for file names used in type units during -gsplit-dwarf This removes an attribute (and more importantly, a relocation) from skeleton type units and removes some unnecessary file names from the debug_line section that remains in the .o (and linked executable) file. There's still a few places we could shave off some more space here: * use compilation dir of the underlying compilation unit (since all the type units share that compilation dir - though this would be more complicated in LTO cases where they don't (keep a map of compilation dir->line table header?)) * Remove some of the unnecessary header fields from the line table since they're not needed in this situation (about 12 bytes per table). llvm-svn: 204099
2014-03-18 09:17:26 +08:00
; FISSION-NOT: file_names[
; FISSION: file_names{{.*}} bar.h
; FISSION: file_names{{.*}} bar.cpp
; FISSION-NOT: file_names[
DebugInfo: Don't include the name of the CU file in the line table file list when it's unneeded Recommitting r201380 (reverted in r201389) Recommitting r201351 and r201355 (reverted in r201351 and r201355) We weren't emitting the an empty (header only) line table when the line table was empty - this made the DWARF invalid (the compile unit would point to the zero-size debug_lines section where there should've been an empty line table but there was nothing at all). Fix that, and as a consequence this works around/addresses PR18809. Also, we emit a non-empty line table to workaround a darwin linker bug, so XFAILing on darwin too. Also, mark the test as 'REQUIRES: object-emission' because it does. llvm-svn: 201429
2014-02-15 03:51:35 +08:00
; CHECK-LABEL: .debug_str contents:
DwarfDebug: Include type units in accelerator tables. Since type units aren't in the CUMap, use the DwarfUnits list to iterate over units for tasks such as accelerator table building. llvm-svn: 195776
2013-11-27 03:14:34 +08:00
; Use the unit size as a rough hash/identifier for the unit we're dealing with
; it happens to be unambiguous at the moment, but it's hardly ideal.
DebugInfo: Avoid emitting pubtype entries for type DIEs that just indirect to a type unit. llvm-svn: 195698
2013-11-26 08:22:37 +08:00
; CHECK-LABEL: .debug_pubtypes contents:
DwarfDebug: Include type units in accelerator tables. Since type units aren't in the CUMap, use the DwarfUnits list to iterate over units for tasks such as accelerator table building. llvm-svn: 195776
2013-11-27 03:14:34 +08:00
; Don't emit pubtype entries for type DIEs in the compile unit that just indirect to a type unit.
Make these two tests resilient in the face of compile unit size changes. llvm-svn: 196444
2013-12-05 09:00:12 +08:00
; CHECK-NEXT: unit_size = [[CU_SIZE]]
DwarfDebug: Include type units in accelerator tables. Since type units aren't in the CUMap, use the DwarfUnits list to iterate over units for tasks such as accelerator table building. llvm-svn: 195776
2013-11-27 03:14:34 +08:00
; CHECK-NEXT: Offset Name
DebugInfo: pubnames: include file-local (static or anonymous namespace) variables and anonymous namespaces themselves. Still some issues with name qualification, FIXMEs added to test cases and fixes will come next. llvm-svn: 210378
2014-06-07 06:16:56 +08:00
; CHECK-DAG: [[BAR]] "bar"
; CHECK-DAG: [[WALRUS]] "(anonymous namespace)::walrus"
; CHECK-DAG: [[WOMBAT]] "wombat"
; CHECK-DAG: [[FLUFFY]] "echidna::capybara::mongoose::fluffy"
DebugInfo: Avoid emitting pubtype entries for type DIEs that just indirect to a type unit. llvm-svn: 195698
2013-11-26 08:22:37 +08:00
Reapply "DebugInfo: Don't put fission type units in comdat sections." This recommits r208930, r208933, and r208975 (by reverting r209338) and reverts r209529 (the FIXME to readd this functionality once the tools were fixed) now that DWP has been fixed to cope with a single section for all fission type units. Original commit message: "Since type units in the dwo file are handled by a debug aware tool, they don't need to leverage the ELF comdat grouping to implement deduplication. Avoid creating all the .group sections for these as a space optimization." llvm-svn: 213956
2014-07-26 01:11:58 +08:00
; Make sure debug_types are in comdat groups. This could be more rigid to check
; that they're the right comdat groups (each type in a separate comdat group,
; etc)
Remove type unit skeletons. GDB no longer needs them & this saves a heap of space. llvm-svn: 216521
2014-08-27 13:04:14 +08:00
; OBJ_SINGLE: Name: .debug_types (
; OBJ_SINGLE-NOT: }
; OBJ_SINGLE: SHF_GROUP
Reapply "DebugInfo: Don't put fission type units in comdat sections." This recommits r208930, r208933, and r208975 (by reverting r209338) and reverts r209529 (the FIXME to readd this functionality once the tools were fixed) now that DWP has been fixed to cope with a single section for all fission type units. Original commit message: "Since type units in the dwo file are handled by a debug aware tool, they don't need to leverage the ELF comdat grouping to implement deduplication. Avoid creating all the .group sections for these as a space optimization." llvm-svn: 213956
2014-07-26 01:11:58 +08:00
; Fission type units don't go in comdat groups, since their linker is debug
; aware it's handled using the debug info semantics rather than raw ELF object
; semantics.
; OBJ_FISSION: Name: .debug_types.dwo (
; OBJ_FISSION-NOT: SHF_GROUP
; OBJ_FISSION: }
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
%struct.bar = type { i8 }
%"class.echidna::capybara::mongoose::fluffy" = type { i32, i32 }
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
%"struct.<anonymous namespace>::walrus" = type { i8 }
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
%struct.wombat = type { %struct.anon }
%struct.anon = type { i32, i32 }
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
%struct.baz = type { i8 }
@b = global %struct.bar zeroinitializer, align 1
@_ZN7echidna8capybara8mongoose6animalE = global %"class.echidna::capybara::mongoose::fluffy" zeroinitializer, align 4
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
@w = internal global %"struct.<anonymous namespace>::walrus" zeroinitializer, align 1
Make sure we don't emit an ODR hash for types with no name and make sure the comments for each testcase are a bit easier to distinguish. llvm-svn: 187392
2013-07-30 07:53:05 +08:00
@wom = global %struct.wombat zeroinitializer, align 4
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
@llvm.global_ctors = appending global [1 x { i32, void ()* }] [{ i32, void ()* } { i32 65535, void ()* @_GLOBAL__I_a }]
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
; Function Attrs: nounwind uwtable
define void @_Z3foov() #0 {
entry:
%b = alloca %struct.baz, align 1
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
call void @llvm.dbg.declare(metadata %struct.baz* %b, metadata !46, metadata !MDExpression()), !dbg !48
DwarfUnit: Include type unit's file strings in the defining compile unit's file_names table There's still one piece missing here, which is adding the DW_AT_stmt_list to the type unit that refer's to the compile unit's line table. Working on that. llvm-svn: 201198
2014-02-12 08:40:47 +08:00
ret void, !dbg !49
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
}
; Function Attrs: nounwind readnone
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
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
define internal void @__cxx_global_var_init() section ".text.startup" {
entry:
DwarfUnit: Include type unit's file strings in the defining compile unit's file_names table There's still one piece missing here, which is adding the DW_AT_stmt_list to the type unit that refer's to the compile unit's line table. Working on that. llvm-svn: 201198
2014-02-12 08:40:47 +08:00
call void @_ZN12_GLOBAL__N_16walrusC2Ev(%"struct.<anonymous namespace>::walrus"* @w), !dbg !50
ret void, !dbg !50
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
}
; Function Attrs: nounwind uwtable
define internal void @_ZN12_GLOBAL__N_16walrusC2Ev(%"struct.<anonymous namespace>::walrus"* %this) unnamed_addr #0 align 2 {
entry:
%this.addr = alloca %"struct.<anonymous namespace>::walrus"*, align 8
store %"struct.<anonymous namespace>::walrus"* %this, %"struct.<anonymous namespace>::walrus"** %this.addr, align 8
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
call void @llvm.dbg.declare(metadata %"struct.<anonymous namespace>::walrus"** %this.addr, metadata !51, metadata !MDExpression()), !dbg !53
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%this1 = load %"struct.<anonymous namespace>::walrus"*, %"struct.<anonymous namespace>::walrus"** %this.addr
DwarfUnit: Include type unit's file strings in the defining compile unit's file_names table There's still one piece missing here, which is adding the DW_AT_stmt_list to the type unit that refer's to the compile unit's line table. Working on that. llvm-svn: 201198
2014-02-12 08:40:47 +08:00
ret void, !dbg !54
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
}
define internal void @_GLOBAL__I_a() section ".text.startup" {
entry:
DwarfUnit: Include type unit's file strings in the defining compile unit's file_names table There's still one piece missing here, which is adding the DW_AT_stmt_list to the type unit that refer's to the compile unit's line table. Working on that. llvm-svn: 201198
2014-02-12 08:40:47 +08:00
call void @__cxx_global_var_init(), !dbg !55
ret void, !dbg !55
Fix a truly egregious thinko in anonymous namespace check, update testcase to make sure we generate debug info for walrus by adding a non-trivial constructor and verify that we don't emit an ODR signature for the type. llvm-svn: 187393
2013-07-30 07:53:08 +08:00
}
DebugInfo: regenerate test case from Clang to adjust for fixes/improvements I hit some problems with future work due to the member subprogram of 'a_b's type having a subprogram (an implicit default ctor, !52 in the pre-commit source) with no name. Clang now generates a name for such a function but in this case doesn't even emit debug info for it as it is unused (Clang never emits the body of the ctor, instead just emitting memset if needed). llvm-svn: 193892
2013-11-02 06:29:28 +08:00
attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
attributes #1 = { nounwind readnone }
!llvm.dbg.cu = !{!0}
DwarfUnit: Include type unit's file strings in the defining compile unit's file_names table There's still one piece missing here, which is adding the DW_AT_stmt_list to the type unit that refer's to the compile unit's line table. Working on that. llvm-svn: 201198
2014-02-12 08:40:47 +08:00
!llvm.module.flags = !{!43, !44}
!llvm.ident = !{!45}
Add preliminary support for hashing DIEs and breaking them into type units. Initially this support is used in the computation of an ODR checker for C++. For now we're attaching it to the DIE, but in the future it will be attached to the type unit. This also starts breaking out types into the separation for type units, but without actually splitting the DIEs. In preparation for hashing the DIEs this adds a DIEString type that contains a StringRef with the string contained at the label. llvm-svn: 187213
2013-07-27 01:02:41 +08:00
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!0 = !MDCompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5 ", isOptimized: false, splitDebugFilename: "bar.dwo", emissionKind: 0, file: !1, enums: !2, retainedTypes: !3, subprograms: !21, globals: !38, imports: !2)
!1 = !MDFile(filename: "bar.cpp", directory: "/tmp/dbginfo")
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!2 = !{}
!3 = !{!4, !6, !14, !17}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!4 = !MDCompositeType(tag: DW_TAG_structure_type, name: "bar", line: 1, size: 8, align: 8, file: !5, elements: !2, identifier: "_ZTS3bar")
!5 = !MDFile(filename: "bar.h", directory: "/tmp/dbginfo")
!6 = !MDCompositeType(tag: DW_TAG_class_type, name: "fluffy", line: 13, size: 64, align: 32, file: !1, scope: !7, elements: !10, identifier: "_ZTSN7echidna8capybara8mongoose6fluffyE")
!7 = !MDNamespace(name: "mongoose", line: 12, file: !1, scope: !8)
!8 = !MDNamespace(name: "capybara", line: 11, file: !1, scope: !9)
!9 = !MDNamespace(name: "echidna", line: 10, file: !1, scope: null)
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!10 = !{!11, !13}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!11 = !MDDerivedType(tag: DW_TAG_member, name: "a", line: 14, size: 32, align: 32, flags: DIFlagPrivate, file: !1, scope: !"_ZTSN7echidna8capybara8mongoose6fluffyE", baseType: !12)
!12 = !MDBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
!13 = !MDDerivedType(tag: DW_TAG_member, name: "b", line: 15, size: 32, align: 32, offset: 32, flags: DIFlagPrivate, file: !1, scope: !"_ZTSN7echidna8capybara8mongoose6fluffyE", baseType: !12)
!14 = !MDCompositeType(tag: DW_TAG_structure_type, name: "wombat", line: 31, size: 64, align: 32, file: !1, elements: !15, identifier: "_ZTS6wombat")
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!15 = !{!16}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!16 = !MDDerivedType(tag: DW_TAG_member, name: "a_b", line: 35, size: 64, align: 32, file: !1, scope: !"_ZTS6wombat", baseType: !"_ZTSN6wombatUt_E")
!17 = !MDCompositeType(tag: DW_TAG_structure_type, line: 32, size: 64, align: 32, file: !1, scope: !"_ZTS6wombat", elements: !18, identifier: "_ZTSN6wombatUt_E")
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!18 = !{!19, !20}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!19 = !MDDerivedType(tag: DW_TAG_member, name: "a", line: 33, size: 32, align: 32, file: !1, scope: !"_ZTSN6wombatUt_E", baseType: !12)
!20 = !MDDerivedType(tag: DW_TAG_member, name: "b", line: 34, size: 32, align: 32, offset: 32, file: !1, scope: !"_ZTSN6wombatUt_E", baseType: !12)
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!21 = !{!22, !26, !27, !36}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!22 = !MDSubprogram(name: "foo", linkageName: "_Z3foov", line: 5, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 5, file: !1, scope: !23, type: !24, function: void ()* @_Z3foov, variables: !2)
!23 = !MDFile(filename: "bar.cpp", directory: "/tmp/dbginfo")
!24 = !MDSubroutineType(types: !25)
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!25 = !{null}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!26 = !MDSubprogram(name: "__cxx_global_var_init", line: 29, isLocal: true, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 29, file: !1, scope: !23, type: !24, function: void ()* @__cxx_global_var_init, variables: !2)
!27 = !MDSubprogram(name: "walrus", linkageName: "_ZN12_GLOBAL__N_16walrusC2Ev", line: 25, isLocal: true, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 25, file: !1, scope: !28, type: !32, function: void (%"struct.<anonymous namespace>::walrus"*)* @_ZN12_GLOBAL__N_16walrusC2Ev, declaration: !31, variables: !2)
!28 = !MDCompositeType(tag: DW_TAG_structure_type, name: "walrus", line: 24, size: 8, align: 8, file: !1, scope: !29, elements: !30)
!29 = !MDNamespace(line: 23, file: !1, scope: null)
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!30 = !{!31}
DebugInfo: Fix testcases with invalid MDSubprogram nodes Fix testcases that don't pass the verifier after a WIP patch to check `MDSubprogram` operands more effectively. I found the following issues: - When `isDefinition: false`, the `variables:` field might point at `!{i32 786468}`, or at a tuple that pointed at an empty tuple with the comment "previously: invalid DW_TAG_base_type" (I vaguely recall adding those comments during an upgrade script). In these cases, I just dropped the array. - The `variables:` field might point at something like `!{!{!8}}`, where `!8` was an `MDLocation`. I removed the extra layer of indirection. - Invalid `type:` (not an `MDSubroutineType`). llvm-svn: 233466
2015-03-28 10:26:45 +08:00
!31 = !MDSubprogram(name: "walrus", line: 25, isLocal: false, isDefinition: false, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 25, file: !1, scope: !28, type: !32)
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!32 = !MDSubroutineType(types: !33)
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!33 = !{null, !34}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!34 = !MDDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, flags: DIFlagArtificial | DIFlagObjectPointer, baseType: !28)
!36 = !MDSubprogram(name: "", linkageName: "_GLOBAL__I_a", line: 25, isLocal: true, isDefinition: true, virtualIndex: 6, flags: DIFlagArtificial, isOptimized: false, scopeLine: 25, file: !1, scope: !23, type: !37, function: void ()* @_GLOBAL__I_a, variables: !2)
!37 = !MDSubroutineType(types: !2)
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!38 = !{!39, !40, !41, !42}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!39 = !MDGlobalVariable(name: "b", line: 3, isLocal: false, isDefinition: true, scope: null, file: !23, type: !4, variable: %struct.bar* @b)
!40 = !MDGlobalVariable(name: "animal", linkageName: "_ZN7echidna8capybara8mongoose6animalE", line: 18, isLocal: false, isDefinition: true, scope: !7, file: !23, type: !6, variable: %"class.echidna::capybara::mongoose::fluffy"* @_ZN7echidna8capybara8mongoose6animalE)
!41 = !MDGlobalVariable(name: "w", line: 29, isLocal: true, isDefinition: true, scope: null, file: !23, type: !28, variable: %"struct.<anonymous namespace>::walrus"* @w)
!42 = !MDGlobalVariable(name: "wom", line: 38, isLocal: false, isDefinition: true, scope: null, file: !23, type: !14, variable: %struct.wombat* @wom)
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!43 = !{i32 2, !"Dwarf Version", i32 4}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!44 = !{i32 1, !"Debug Info Version", i32 3}
IR: Make metadata typeless in assembly Now that `Metadata` is typeless, reflect that in the assembly. These are the matching assembly changes for the metadata/value split in r223802. - Only use the `metadata` type when referencing metadata from a call intrinsic -- i.e., only when it's used as a `Value`. - Stop pretending that `ValueAsMetadata` is wrapped in an `MDNode` when referencing it from call intrinsics. So, assembly like this: define @foo(i32 %v) { call void @llvm.foo(metadata !{i32 %v}, metadata !0) call void @llvm.foo(metadata !{i32 7}, metadata !0) call void @llvm.foo(metadata !1, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{metadata !3}, metadata !0) ret void, !bar !2 } !0 = metadata !{metadata !2} !1 = metadata !{i32* @global} !2 = metadata !{metadata !3} !3 = metadata !{} turns into this: define @foo(i32 %v) { call void @llvm.foo(metadata i32 %v, metadata !0) call void @llvm.foo(metadata i32 7, metadata !0) call void @llvm.foo(metadata i32* @global, metadata !0) call void @llvm.foo(metadata !3, metadata !0) call void @llvm.foo(metadata !{!3}, metadata !0) ret void, !bar !2 } !0 = !{!2} !1 = !{i32* @global} !2 = !{!3} !3 = !{} I wrote an upgrade script that handled almost all of the tests in llvm and many of the tests in cfe (even handling many `CHECK` lines). I've attached it (or will attach it in a moment if you're speedy) to PR21532 to help everyone update their out-of-tree testcases. This is part of PR21532. llvm-svn: 224257
2014-12-16 03:07:53 +08:00
!45 = !{!"clang version 3.5 "}
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!46 = !MDLocalVariable(tag: DW_TAG_auto_variable, name: "b", line: 7, scope: !22, file: !23, type: !47)
!47 = !MDCompositeType(tag: DW_TAG_structure_type, name: "baz", line: 6, size: 8, align: 8, file: !1, scope: !22, elements: !2)
IR: Move MDLocation into place This commit moves `MDLocation`, finishing off PR21433. There's an accompanying clang commit for frontend testcases. I'll attach the testcase upgrade script I used to PR21433 to help out-of-tree frontends/backends. This changes the schema for `DebugLoc` and `DILocation` from: !{i32 3, i32 7, !7, !8} to: !MDLocation(line: 3, column: 7, scope: !7, inlinedAt: !8) Note that empty fields (line/column: 0 and inlinedAt: null) don't get printed by the assembly writer. llvm-svn: 226048
2015-01-15 06:27:36 +08:00
!48 = !MDLocation(line: 7, scope: !22)
!49 = !MDLocation(line: 8, scope: !22)
!50 = !MDLocation(line: 29, scope: !26)
DebugInfo: Move new hierarchy into place Move the specialized metadata nodes for the new debug info hierarchy into place, finishing off PR22464. I've done bootstraps (and all that) and I'm confident this commit is NFC as far as DWARF output is concerned. Let me know if I'm wrong :). The code changes are fairly mechanical: - Bumped the "Debug Info Version". - `DIBuilder` now creates the appropriate subclass of `MDNode`. - Subclasses of DIDescriptor now expect to hold their "MD" counterparts (e.g., `DIBasicType` expects `MDBasicType`). - Deleted a ton of dead code in `AsmWriter.cpp` and `DebugInfo.cpp` for printing comments. - Big update to LangRef to describe the nodes in the new hierarchy. Feel free to make it better. Testcase changes are enormous. There's an accompanying clang commit on its way. If you have out-of-tree debug info testcases, I just broke your build. - `upgrade-specialized-nodes.sh` is attached to PR22564. I used it to update all the IR testcases. - Unfortunately I failed to find way to script the updates to CHECK lines, so I updated all of these by hand. This was fairly painful, since the old CHECKs are difficult to reason about. That's one of the benefits of the new hierarchy. This work isn't quite finished, BTW. The `DIDescriptor` subclasses are almost empty wrappers, but not quite: they still have loose casting checks (see the `RETURN_FROM_RAW()` macro). Once they're completely gutted, I'll rename the "MD" classes to "DI" and kill the wrappers. I also expect to make a few schema changes now that it's easier to reason about everything. llvm-svn: 231082
2015-03-04 01:24:31 +08:00
!51 = !MDLocalVariable(tag: DW_TAG_arg_variable, name: "this", arg: 1, flags: DIFlagArtificial | DIFlagObjectPointer, scope: !27, type: !52)
!52 = !MDDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, baseType: !28)
IR: Move MDLocation into place This commit moves `MDLocation`, finishing off PR21433. There's an accompanying clang commit for frontend testcases. I'll attach the testcase upgrade script I used to PR21433 to help out-of-tree frontends/backends. This changes the schema for `DebugLoc` and `DILocation` from: !{i32 3, i32 7, !7, !8} to: !MDLocation(line: 3, column: 7, scope: !7, inlinedAt: !8) Note that empty fields (line/column: 0 and inlinedAt: null) don't get printed by the assembly writer. llvm-svn: 226048
2015-01-15 06:27:36 +08:00
!53 = !MDLocation(line: 0, scope: !27)
!54 = !MDLocation(line: 25, scope: !27)
!55 = !MDLocation(line: 25, scope: !36)