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/ARM/lowerbdgdeclare_vla.ll

104 lines
6.4 KiB
LLVM
Raw Normal View History

Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
; RUN: opt -instcombine %s -S | FileCheck %s
;
; Generate me from:
; clang -cc1 -triple thumbv7-apple-ios7.0.0 -S -target-abi apcs-gnu -gdwarf-2 -Os test.c -o test.ll -emit-llvm
; void run(float r)
; {
; int count = r;
; float vla[count];
; vla[0] = r;
; for (int i = 0; i < count; i++)
; vla[i] /= r;
; }
; rdar://problem/15464571
;
; ModuleID = 'test.c'
target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:32:64-v128:32:128-a0:0:32-n32-S32"
target triple = "thumbv7-apple-ios8.0.0"
; Function Attrs: nounwind optsize readnone
define void @run(float %r) #0 {
entry:
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
tail call void @llvm.dbg.declare(metadata !{float %r}, metadata !11, metadata !{metadata !"0x102"}), !dbg !22
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
%conv = fptosi float %r to i32, !dbg !23
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
tail call void @llvm.dbg.declare(metadata !{i32 %conv}, metadata !12, metadata !{metadata !"0x102"}), !dbg !23
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
%vla = alloca float, i32 %conv, align 4, !dbg !24
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
tail call void @llvm.dbg.declare(metadata !{float* %vla}, metadata !14, metadata !{metadata !"0x102"}), !dbg !24
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
; The VLA alloca should be described by a dbg.declare:
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
; CHECK: call void @llvm.dbg.declare(metadata !{float* %vla}, metadata ![[VLA:.*]], metadata {{.*}})
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
; The VLA alloca and following store into the array should not be lowered to like this:
; CHECK-NOT: call void @llvm.dbg.value(metadata !{float %r}, i64 0, metadata ![[VLA]])
; the backend interprets this as "vla has the location of %r".
store float %r, float* %vla, align 4, !dbg !25, !tbaa !26
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
tail call void @llvm.dbg.value(metadata !2, i64 0, metadata !18, metadata !{metadata !"0x102"}), !dbg !30
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
%cmp8 = icmp sgt i32 %conv, 0, !dbg !30
br i1 %cmp8, label %for.body, label %for.end, !dbg !30
for.body: ; preds = %entry, %for.body.for.body_crit_edge
%0 = phi float [ %.pre, %for.body.for.body_crit_edge ], [ %r, %entry ]
%i.09 = phi i32 [ %inc, %for.body.for.body_crit_edge ], [ 0, %entry ]
%arrayidx2 = getelementptr inbounds float* %vla, i32 %i.09, !dbg !31
%div = fdiv float %0, %r, !dbg !31
store float %div, float* %arrayidx2, align 4, !dbg !31, !tbaa !26
%inc = add nsw i32 %i.09, 1, !dbg !30
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
tail call void @llvm.dbg.value(metadata !{i32 %inc}, i64 0, metadata !18, metadata !{metadata !"0x102"}), !dbg !30
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
%exitcond = icmp eq i32 %inc, %conv, !dbg !30
br i1 %exitcond, label %for.end, label %for.body.for.body_crit_edge, !dbg !30
for.body.for.body_crit_edge: ; preds = %for.body
%arrayidx2.phi.trans.insert = getelementptr inbounds float* %vla, i32 %inc
%.pre = load float* %arrayidx2.phi.trans.insert, align 4, !dbg !31, !tbaa !26
br label %for.body, !dbg !30
for.end: ; preds = %for.body, %entry
ret void, !dbg !32
}
; 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
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +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.value(metadata, i64, metadata, metadata) #1
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
attributes #0 = { nounwind optsize readnone "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "no-realign-stack" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
attributes #1 = { nounwind readnone }
!llvm.dbg.cu = !{!0}
Debug Info: update testing cases to specify the debug info version number. We are going to drop debug info without a version number or with a different version number, to make sure we don't crash when we see bitcode files with different debug info metadata format. llvm-svn: 195504
2013-11-23 05:49:45 +08:00
!llvm.module.flags = !{!20, !33}
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
!llvm.ident = !{!21}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!0 = metadata !{metadata !"0x11\0012\00clang version 3.4 \001\00\000\00\000", metadata !1, metadata !2, metadata !2, metadata !3, metadata !2, metadata !2} ; [ DW_TAG_compile_unit ] [/Volumes/Data/radar/15464571/<unknown>] [DW_LANG_C99]
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
!1 = metadata !{metadata !"<unknown>", metadata !"/Volumes/Data/radar/15464571"}
!2 = metadata !{i32 0}
!3 = metadata !{metadata !4}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!4 = metadata !{metadata !"0x2e\00run\00run\00\001\000\001\000\006\00256\001\002", metadata !5, metadata !6, metadata !7, null, void (float)* @run, null, null, metadata !10} ; [ DW_TAG_subprogram ] [line 1] [def] [scope 2] [run]
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
!5 = metadata !{metadata !"test.c", metadata !"/Volumes/Data/radar/15464571"}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!6 = metadata !{metadata !"0x29", metadata !5} ; [ DW_TAG_file_type ] [/Volumes/Data/radar/15464571/test.c]
!7 = metadata !{metadata !"0x15\00\000\000\000\000\000\000", i32 0, null, null, metadata !8, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
!8 = metadata !{null, metadata !9}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!9 = metadata !{metadata !"0x24\00float\000\0032\0032\000\000\004", null, null} ; [ DW_TAG_base_type ] [float] [line 0, size 32, align 32, offset 0, enc DW_ATE_float]
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
!10 = metadata !{metadata !11, metadata !12, metadata !14, metadata !18}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!11 = metadata !{metadata !"0x101\00r\0016777217\000", metadata !4, metadata !6, metadata !9} ; [ DW_TAG_arg_variable ] [r] [line 1]
!12 = metadata !{metadata !"0x100\00count\003\000", metadata !4, metadata !6, metadata !13} ; [ DW_TAG_auto_variable ] [count] [line 3]
!13 = metadata !{metadata !"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
!14 = metadata !{metadata !"0x100\00vla\004\008192", metadata !4, metadata !6, metadata !15} ; [ DW_TAG_auto_variable ] [vla] [line 4]
!15 = metadata !{metadata !"0x1\00\000\000\0032\000\000", null, null, metadata !9, metadata !16, i32 0, null, null, null} ; [ DW_TAG_array_type ] [line 0, size 0, align 32, offset 0] [from float]
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
!16 = metadata !{metadata !17}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!17 = metadata !{metadata !"0x21\000\00-1"} ; [ DW_TAG_subrange_type ] [unbounded]
!18 = metadata !{metadata !"0x100\00i\006\000", metadata !19, metadata !6, metadata !13} ; [ DW_TAG_auto_variable ] [i] [line 6]
!19 = metadata !{metadata !"0xb\006\000\000", metadata !5, metadata !4} ; [ DW_TAG_lexical_block ] [/Volumes/Data/radar/15464571/test.c]
Debug info: Let LowerDbgDeclare perfom the dbg.declare -> dbg.value lowering only for load/stores to scalar allocas. The resulting values confuse the backend and don't add anything because we can describe array-allocas with a dbg.declare intrinsic just fine. rdar://problem/15464571 llvm-svn: 195052
2013-11-19 07:04:38 +08:00
!20 = metadata !{i32 2, metadata !"Dwarf Version", i32 2}
!21 = metadata !{metadata !"clang version 3.4 "}
!22 = metadata !{i32 1, i32 0, metadata !4, null}
!23 = metadata !{i32 3, i32 0, metadata !4, null}
!24 = metadata !{i32 4, i32 0, metadata !4, null}
!25 = metadata !{i32 5, i32 0, metadata !4, null}
!26 = metadata !{metadata !27, metadata !27, i64 0}
!27 = metadata !{metadata !"float", metadata !28, i64 0}
!28 = metadata !{metadata !"omnipotent char", metadata !29, i64 0}
!29 = metadata !{metadata !"Simple C/C++ TBAA"}
!30 = metadata !{i32 6, i32 0, metadata !19, null}
!31 = metadata !{i32 7, i32 0, metadata !19, null}
Revert "Revert "DI: Fold constant arguments into a single MDString"" This reverts commit r218918, effectively reapplying r218914 after fixing an Ocaml bindings test and an Asan crash. The root cause of the latter was a tightened-up check in `DILexicalBlock::Verify()`, so I'll file a PR to investigate who requires the loose check (and why). Original commit message follows. -- This patch addresses the first stage of PR17891 by folding constant arguments together into a single MDString. Integers are stringified and a `\0` character is used as a separator. Part of PR17891. Note: I've attached my testcases upgrade scripts to the PR. If I've just broken your out-of-tree testcases, they might help. llvm-svn: 219010
2014-10-04 04:01:09 +08:00
!32 = metadata !{i32 8, i32 0, metadata !4, null}
!33 = metadata !{i32 1, metadata !"Debug Info Version", i32 2}