maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/Hexagon/hwloop-dbg.ll

65 lines
4.5 KiB
LLVM
Raw Normal View History

Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
; RUN: llc < %s -march=hexagon -mcpu=hexagonv4 -O2 -disable-lsr | FileCheck %s
; ModuleID = 'hwloop-dbg.o'
target datalayout = "e-p:32:32:32-i64:64:64-i32:32:32-i16:16:16-i1:32:32-f64:64:64-f32:32:32-v64:64:64-v32:32:32-a0:0-n16:32"
target triple = "hexagon"
define void @foo(i32* nocapture %a, i32* nocapture %b) nounwind {
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.value(metadata !{i32* %a}, i64 0, metadata !13, metadata !{metadata !"0x102"}), !dbg !17
tail call void @llvm.dbg.value(metadata !{i32* %b}, i64 0, metadata !14, metadata !{metadata !"0x102"}), !dbg !18
tail call void @llvm.dbg.value(metadata !30, i64 0, metadata !15, metadata !{metadata !"0x102"}), !dbg !19
Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
br label %for.body, !dbg !19
for.body: ; preds = %for.body, %entry
; CHECK: loop0(
; CHECK-NOT: add({{r[0-9]*}}, #
; CHECK: endloop0
%arrayidx.phi = phi i32* [ %a, %entry ], [ %arrayidx.inc, %for.body ]
%i.02 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
%b.addr.01 = phi i32* [ %b, %entry ], [ %incdec.ptr, %for.body ]
%incdec.ptr = getelementptr inbounds i32* %b.addr.01, i32 1, !dbg !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
tail call void @llvm.dbg.value(metadata !{i32* %incdec.ptr}, i64 0, metadata !14, metadata !{metadata !"0x102"}), !dbg !21
TBAA: remove !tbaa from testing cases if not used. This will make it easier to turn on struct-path aware TBAA since the metadata format will change. llvm-svn: 180796
2013-05-01 01:52:57 +08:00
%0 = load i32* %b.addr.01, align 4, !dbg !21
store i32 %0, i32* %arrayidx.phi, align 4, !dbg !21
Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
%inc = add nsw i32 %i.02, 1, !dbg !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 !{i32 %inc}, i64 0, metadata !15, metadata !{metadata !"0x102"}), !dbg !26
Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
%exitcond = icmp eq i32 %inc, 10, !dbg !19
%arrayidx.inc = getelementptr i32* %arrayidx.phi, i32 1
br i1 %exitcond, label %for.end, label %for.body, !dbg !19
for.end: ; preds = %for.body
ret void, !dbg !27
}
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) nounwind readnone
Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
Debug Info Verifier: verify SPs in llvm.dbg.sp. Also always add DIType, DISubprogram and DIGlobalVariable to the list in DebugInfoFinder without checking them, so we can verify them later on. llvm-svn: 187285
2013-07-27 09:26:08 +08:00
!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 = !{!29}
Debug Info Verifier: verify SPs in llvm.dbg.sp. Also always add DIType, DISubprogram and DIGlobalVariable to the list in DebugInfoFinder without checking them, so we can verify them later on. llvm-svn: 187285
2013-07-27 09:26:08 +08:00
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\00QuIC LLVM Hexagon Clang version 6.1-pre-unknown, (git://git-hexagon-aus.quicinc.com/llvm/clang-mainline.git e9382867661454cdf44addb39430741578e9765c) (llvm/llvm-mainline.git 36412bb1fcf03ed426d4437b41198bae066675ac)\001\00\000\00\001", metadata !28, metadata !2, metadata !2, metadata !3, metadata !2, null} ; [ DW_TAG_compile_unit ] [/usr2/kparzysz/s.hex/t/hwloop-dbg.c] [DW_LANG_C99]
DebugInfo: Remove some unneeded conditionals now that DIBuilder no longer emits zero-length arrays as {i32 0} A bunch of test cases needed to be cleaned up for this, many my fault - when implementid imported modules I updated test cases by simply duplicating the prior metadata field - which wasn't always the empty metadata entry. llvm-svn: 200731
2014-02-04 09:23:52 +08:00
!2 = metadata !{}
Debug Info Verifier: verify SPs in llvm.dbg.sp. Also always add DIType, DISubprogram and DIGlobalVariable to the list in DebugInfoFinder without checking them, so we can verify them later on. llvm-svn: 187285
2013-07-27 09:26:08 +08:00
!3 = metadata !{metadata !5}
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
!5 = metadata !{metadata !"0x2e\00foo\00foo\00\001\000\001\000\006\00256\001\001", metadata !28, null, metadata !7, null, void (i32*, i32*)* @foo, null, null, metadata !11} ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
!6 = metadata !{metadata !"0x29", metadata !28} ; [ DW_TAG_file_type ]
!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 ]
Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
!8 = metadata !{null, metadata !9, 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 !"0xf\00\000\0032\0032\000\000", null, null, metadata !10} ; [ DW_TAG_pointer_type ] [line 0, size 32, align 32, offset 0] [from int]
!10 = 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]
DebugInfo: Sure up subprogram variable list handling with more assertions and fewer conditionals. Many old tests using prior schemas still had some brokenness here (both indirect arrays and arrays with single bogus elements). Fixed those up so they don't hit the new assertions. Also reduced nesting in some places, etc. llvm-svn: 208817
2014-05-15 05:52:46 +08:00
!11 = metadata !{metadata !13, metadata !14, metadata !15}
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
!13 = metadata !{metadata !"0x101\00a\0016777217\000", metadata !5, metadata !6, metadata !9} ; [ DW_TAG_arg_variable ] [a] [line 1]
!14 = metadata !{metadata !"0x101\00b\0033554433\000", metadata !5, metadata !6, metadata !9} ; [ DW_TAG_arg_variable ] [b] [line 1]
!15 = metadata !{metadata !"0x100\00i\002\000", metadata !16, metadata !6, metadata !10} ; [ DW_TAG_auto_variable ] [i] [line 2]
!16 = metadata !{metadata !"0xb\001\0026\000", metadata !28, metadata !5} ; [ DW_TAG_lexical_block ] [/usr2/kparzysz/s.hex/t/hwloop-dbg.c]
Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
!17 = metadata !{i32 1, i32 15, metadata !5, null}
!18 = metadata !{i32 1, i32 23, metadata !5, null}
!19 = metadata !{i32 3, i32 8, metadata !20, 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
!20 = metadata !{metadata !"0xb\003\003\001", metadata !28, metadata !16} ; [ DW_TAG_lexical_block ] [/usr2/kparzysz/s.hex/t/hwloop-dbg.c]
Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
!21 = metadata !{i32 4, i32 5, metadata !22, 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
!22 = metadata !{metadata !"0xb\003\0028\002", metadata !28, metadata !20} ; [ DW_TAG_lexical_block ] [/usr2/kparzysz/s.hex/t/hwloop-dbg.c]
Extend Hexagon hardware loop generation to handle various additional cases: - variety of compare instructions, - loops with no preheader, - arbitrary lower and upper bounds. llvm-svn: 174904
2013-02-12 05:37:55 +08:00
!26 = metadata !{i32 3, i32 23, metadata !20, null}
!27 = metadata !{i32 6, i32 1, metadata !16, null}
Debug Info Verifier: verify SPs in llvm.dbg.sp. Also always add DIType, DISubprogram and DIGlobalVariable to the list in DebugInfoFinder without checking them, so we can verify them later on. llvm-svn: 187285
2013-07-27 09:26:08 +08:00
!28 = metadata !{metadata !"hwloop-dbg.c", metadata !"/usr2/kparzysz/s.hex/t"}
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
!29 = metadata !{i32 1, metadata !"Debug Info Version", i32 2}
DebugInfo: Remove some unneeded conditionals now that DIBuilder no longer emits zero-length arrays as {i32 0} A bunch of test cases needed to be cleaned up for this, many my fault - when implementid imported modules I updated test cases by simply duplicating the prior metadata field - which wasn't always the empty metadata entry. llvm-svn: 200731
2014-02-04 09:23:52 +08:00
!30 = metadata !{i32 0}