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/X86/2010-05-25-DotDebugLoc.ll

250 lines
16 KiB
LLVM
Raw Normal View History

Force a triple to make this test pass on Darwin. llvm-svn: 132228
2011-05-28 07:12:48 +08:00
; RUN: llc -mtriple=x86_64-pc-linux -O2 < %s | FileCheck %s
; RUN: llc -mtriple=x86_64-pc-linux -O2 -regalloc=basic < %s | FileCheck %s
Use FileCheck llvm-svn: 114475
2010-09-22 04:50:32 +08:00
; Test to check .debug_loc support. This test case emits many debug_loc entries.
DwarfDebug: Emit number of bytes in .debug_loc entry directly Emit the number of bytes in a `.debug_loc` entry directly. The old code created temp labels (expensive), emitted the difference between them, and then emitted one on each side of the relevant bytes. (I'm looking at `llc` memory usage on `verify-uselistorder.lto.opt.bc` (the optimized version of ld64's `-save-temps` when linking the `verify-uselistorder` executable in an LTO bootstrap). I've hacked `MCContext::Allocate()` to just call `malloc()` instead of using the `BumpPtrAllocator` so that the heap profile is easier to read. As far as peak memory is concerned, `MCContext::Allocate()` is equivalent to a leak, since it only gets freed at process teardown. In my heap profile, this patch drops memory usage of `DwarfDebug::emitDebugLoc()` from 132.56 MB (11.4%) down to 29.86 MB (2.7%) at peak memory. Some of that must be noise from `SmallVector` (or other) allocations -- peak memory only dropped from 1160 MB down to 1100 MB -- but this nevertheless shaves 5% off the top.) llvm-svn: 236629
2015-05-07 03:11:20 +08:00
; CHECK: .short 1 # Loc expr size
Use FileCheck llvm-svn: 114475
2010-09-22 04:50:32 +08:00
; CHECK-NEXT: DW_OP_reg
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%0 = type { double }
DI: Reverse direction of subprogram -> function edge. Previously, subprograms contained a metadata reference to the function they described. Because most clients need to get or set a subprogram for a given function rather than the other way around, this created unneeded inefficiency. For example, many passes needed to call the function llvm::makeSubprogramMap() to build a mapping from functions to subprograms, and the IR linker needed to fix up function references in a way that caused quadratic complexity in the IR linking phase of LTO. This change reverses the direction of the edge by storing the subprogram as function-level metadata and removing DISubprogram's function field. Since this is an IR change, a bitcode upgrade has been provided. Fixes PR23367. An upgrade script for textual IR for out-of-tree clients is attached to the PR. Differential Revision: http://reviews.llvm.org/D14265 llvm-svn: 252219
2015-11-06 06:03:56 +08:00
define hidden %0 @__divsc3(float %a, float %b, float %c, float %d) nounwind readnone !dbg !1 {
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
entry:
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %a, i64 0, metadata !0, metadata !DIExpression()), !dbg !DILocation(scope: !1)
tail call void @llvm.dbg.value(metadata float %b, i64 0, metadata !11, metadata !DIExpression()), !dbg !DILocation(scope: !1)
tail call void @llvm.dbg.value(metadata float %c, i64 0, metadata !12, metadata !DIExpression()), !dbg !DILocation(scope: !1)
tail call void @llvm.dbg.value(metadata float %d, i64 0, metadata !13, metadata !DIExpression()), !dbg !DILocation(scope: !1)
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%0 = tail call float @fabsf(float %c) nounwind readnone, !dbg !19 ; <float> [#uses=1]
%1 = tail call float @fabsf(float %d) nounwind readnone, !dbg !19 ; <float> [#uses=1]
%2 = fcmp olt float %0, %1, !dbg !19 ; <i1> [#uses=1]
br i1 %2, label %bb, label %bb1, !dbg !19
bb: ; preds = %entry
%3 = fdiv float %c, %d, !dbg !20 ; <float> [#uses=3]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %3, i64 0, metadata !16, metadata !DIExpression()), !dbg !20
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%4 = fmul float %3, %c, !dbg !21 ; <float> [#uses=1]
%5 = fadd float %4, %d, !dbg !21 ; <float> [#uses=2]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %5, i64 0, metadata !14, metadata !DIExpression()), !dbg !21
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%6 = fmul float %3, %a, !dbg !22 ; <float> [#uses=1]
%7 = fadd float %6, %b, !dbg !22 ; <float> [#uses=1]
%8 = fdiv float %7, %5, !dbg !22 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %8, i64 0, metadata !17, metadata !DIExpression()), !dbg !22
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%9 = fmul float %3, %b, !dbg !23 ; <float> [#uses=1]
%10 = fsub float %9, %a, !dbg !23 ; <float> [#uses=1]
%11 = fdiv float %10, %5, !dbg !23 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %11, i64 0, metadata !18, metadata !DIExpression()), !dbg !23
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
br label %bb2, !dbg !23
bb1: ; preds = %entry
%12 = fdiv float %d, %c, !dbg !24 ; <float> [#uses=3]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %12, i64 0, metadata !16, metadata !DIExpression()), !dbg !24
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%13 = fmul float %12, %d, !dbg !25 ; <float> [#uses=1]
%14 = fadd float %13, %c, !dbg !25 ; <float> [#uses=2]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %14, i64 0, metadata !14, metadata !DIExpression()), !dbg !25
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%15 = fmul float %12, %b, !dbg !26 ; <float> [#uses=1]
%16 = fadd float %15, %a, !dbg !26 ; <float> [#uses=1]
%17 = fdiv float %16, %14, !dbg !26 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %17, i64 0, metadata !17, metadata !DIExpression()), !dbg !26
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%18 = fmul float %12, %a, !dbg !27 ; <float> [#uses=1]
%19 = fsub float %b, %18, !dbg !27 ; <float> [#uses=1]
%20 = fdiv float %19, %14, !dbg !27 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %20, i64 0, metadata !18, metadata !DIExpression()), !dbg !27
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
br label %bb2, !dbg !27
bb2: ; preds = %bb1, %bb
%y.0 = phi float [ %11, %bb ], [ %20, %bb1 ] ; <float> [#uses=5]
%x.0 = phi float [ %8, %bb ], [ %17, %bb1 ] ; <float> [#uses=5]
%21 = fcmp uno float %x.0, 0.000000e+00, !dbg !28 ; <i1> [#uses=1]
%22 = fcmp uno float %y.0, 0.000000e+00, !dbg !28 ; <i1> [#uses=1]
%or.cond = and i1 %21, %22 ; <i1> [#uses=1]
br i1 %or.cond, label %bb4, label %bb46, !dbg !28
bb4: ; preds = %bb2
%23 = fcmp une float %c, 0.000000e+00, !dbg !29 ; <i1> [#uses=1]
%24 = fcmp une float %d, 0.000000e+00, !dbg !29 ; <i1> [#uses=1]
%or.cond93 = or i1 %23, %24 ; <i1> [#uses=1]
br i1 %or.cond93, label %bb9, label %bb6, !dbg !29
bb6: ; preds = %bb4
%25 = fcmp uno float %a, 0.000000e+00, !dbg !29 ; <i1> [#uses=1]
%26 = fcmp uno float %b, 0.000000e+00, !dbg !29 ; <i1> [#uses=1]
%or.cond94 = and i1 %25, %26 ; <i1> [#uses=1]
br i1 %or.cond94, label %bb9, label %bb8, !dbg !29
bb8: ; preds = %bb6
%27 = tail call float @copysignf(float 0x7FF0000000000000, float %c) nounwind readnone, !dbg !30 ; <float> [#uses=2]
%28 = fmul float %27, %a, !dbg !30 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %28, i64 0, metadata !17, metadata !DIExpression()), !dbg !30
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%29 = fmul float %27, %b, !dbg !31 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %29, i64 0, metadata !18, metadata !DIExpression()), !dbg !31
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
br label %bb46, !dbg !31
bb9: ; preds = %bb6, %bb4
%30 = fcmp ord float %a, 0.000000e+00 ; <i1> [#uses=1]
%31 = fsub float %a, %a, !dbg !32 ; <float> [#uses=3]
%32 = fcmp uno float %31, 0.000000e+00 ; <i1> [#uses=1]
%33 = and i1 %30, %32, !dbg !32 ; <i1> [#uses=2]
br i1 %33, label %bb14, label %bb11, !dbg !32
bb11: ; preds = %bb9
%34 = fcmp ord float %b, 0.000000e+00 ; <i1> [#uses=1]
%35 = fsub float %b, %b, !dbg !32 ; <float> [#uses=1]
%36 = fcmp uno float %35, 0.000000e+00 ; <i1> [#uses=1]
%37 = and i1 %34, %36, !dbg !32 ; <i1> [#uses=1]
br i1 %37, label %bb14, label %bb27, !dbg !32
bb14: ; preds = %bb11, %bb9
%38 = fsub float %c, %c, !dbg !32 ; <float> [#uses=1]
%39 = fcmp ord float %38, 0.000000e+00 ; <i1> [#uses=1]
br i1 %39, label %bb15, label %bb27, !dbg !32
bb15: ; preds = %bb14
%40 = fsub float %d, %d, !dbg !32 ; <float> [#uses=1]
%41 = fcmp ord float %40, 0.000000e+00 ; <i1> [#uses=1]
br i1 %41, label %bb16, label %bb27, !dbg !32
bb16: ; preds = %bb15
%iftmp.0.0 = select i1 %33, float 1.000000e+00, float 0.000000e+00 ; <float> [#uses=1]
%42 = tail call float @copysignf(float %iftmp.0.0, float %a) nounwind readnone, !dbg !33 ; <float> [#uses=2]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %42, i64 0, metadata !0, metadata !DIExpression()), !dbg !33
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%43 = fcmp ord float %b, 0.000000e+00 ; <i1> [#uses=1]
%44 = fsub float %b, %b, !dbg !34 ; <float> [#uses=1]
%45 = fcmp uno float %44, 0.000000e+00 ; <i1> [#uses=1]
%46 = and i1 %43, %45, !dbg !34 ; <i1> [#uses=1]
%iftmp.1.0 = select i1 %46, float 1.000000e+00, float 0.000000e+00 ; <float> [#uses=1]
%47 = tail call float @copysignf(float %iftmp.1.0, float %b) nounwind readnone, !dbg !34 ; <float> [#uses=2]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %47, i64 0, metadata !11, metadata !DIExpression()), !dbg !34
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%48 = fmul float %42, %c, !dbg !35 ; <float> [#uses=1]
%49 = fmul float %47, %d, !dbg !35 ; <float> [#uses=1]
%50 = fadd float %48, %49, !dbg !35 ; <float> [#uses=1]
%51 = fmul float %50, 0x7FF0000000000000, !dbg !35 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %51, i64 0, metadata !17, metadata !DIExpression()), !dbg !35
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%52 = fmul float %47, %c, !dbg !36 ; <float> [#uses=1]
%53 = fmul float %42, %d, !dbg !36 ; <float> [#uses=1]
%54 = fsub float %52, %53, !dbg !36 ; <float> [#uses=1]
%55 = fmul float %54, 0x7FF0000000000000, !dbg !36 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %55, i64 0, metadata !18, metadata !DIExpression()), !dbg !36
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
br label %bb46, !dbg !36
bb27: ; preds = %bb15, %bb14, %bb11
%56 = fcmp ord float %c, 0.000000e+00 ; <i1> [#uses=1]
%57 = fsub float %c, %c, !dbg !37 ; <float> [#uses=1]
%58 = fcmp uno float %57, 0.000000e+00 ; <i1> [#uses=1]
%59 = and i1 %56, %58, !dbg !37 ; <i1> [#uses=2]
br i1 %59, label %bb33, label %bb30, !dbg !37
bb30: ; preds = %bb27
%60 = fcmp ord float %d, 0.000000e+00 ; <i1> [#uses=1]
%61 = fsub float %d, %d, !dbg !37 ; <float> [#uses=1]
%62 = fcmp uno float %61, 0.000000e+00 ; <i1> [#uses=1]
%63 = and i1 %60, %62, !dbg !37 ; <i1> [#uses=1]
%64 = fcmp ord float %31, 0.000000e+00 ; <i1> [#uses=1]
%or.cond95 = and i1 %63, %64 ; <i1> [#uses=1]
br i1 %or.cond95, label %bb34, label %bb46, !dbg !37
bb33: ; preds = %bb27
%.old = fcmp ord float %31, 0.000000e+00 ; <i1> [#uses=1]
br i1 %.old, label %bb34, label %bb46, !dbg !37
bb34: ; preds = %bb33, %bb30
%65 = fsub float %b, %b, !dbg !37 ; <float> [#uses=1]
%66 = fcmp ord float %65, 0.000000e+00 ; <i1> [#uses=1]
br i1 %66, label %bb35, label %bb46, !dbg !37
bb35: ; preds = %bb34
%iftmp.2.0 = select i1 %59, float 1.000000e+00, float 0.000000e+00 ; <float> [#uses=1]
%67 = tail call float @copysignf(float %iftmp.2.0, float %c) nounwind readnone, !dbg !38 ; <float> [#uses=2]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %67, i64 0, metadata !12, metadata !DIExpression()), !dbg !38
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%68 = fcmp ord float %d, 0.000000e+00 ; <i1> [#uses=1]
%69 = fsub float %d, %d, !dbg !39 ; <float> [#uses=1]
%70 = fcmp uno float %69, 0.000000e+00 ; <i1> [#uses=1]
%71 = and i1 %68, %70, !dbg !39 ; <i1> [#uses=1]
%iftmp.3.0 = select i1 %71, float 1.000000e+00, float 0.000000e+00 ; <float> [#uses=1]
%72 = tail call float @copysignf(float %iftmp.3.0, float %d) nounwind readnone, !dbg !39 ; <float> [#uses=2]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %72, i64 0, metadata !13, metadata !DIExpression()), !dbg !39
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%73 = fmul float %67, %a, !dbg !40 ; <float> [#uses=1]
%74 = fmul float %72, %b, !dbg !40 ; <float> [#uses=1]
%75 = fadd float %73, %74, !dbg !40 ; <float> [#uses=1]
%76 = fmul float %75, 0.000000e+00, !dbg !40 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %76, i64 0, metadata !17, metadata !DIExpression()), !dbg !40
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
%77 = fmul float %67, %b, !dbg !41 ; <float> [#uses=1]
%78 = fmul float %72, %a, !dbg !41 ; <float> [#uses=1]
%79 = fsub float %77, %78, !dbg !41 ; <float> [#uses=1]
%80 = fmul float %79, 0.000000e+00, !dbg !41 ; <float> [#uses=1]
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
tail call void @llvm.dbg.value(metadata float %80, i64 0, metadata !18, metadata !DIExpression()), !dbg !41
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
br label %bb46, !dbg !41
bb46: ; preds = %bb35, %bb34, %bb33, %bb30, %bb16, %bb8, %bb2
%y.1 = phi float [ %80, %bb35 ], [ %y.0, %bb34 ], [ %y.0, %bb33 ], [ %y.0, %bb30 ], [ %55, %bb16 ], [ %29, %bb8 ], [ %y.0, %bb2 ] ; <float> [#uses=2]
%x.1 = phi float [ %76, %bb35 ], [ %x.0, %bb34 ], [ %x.0, %bb33 ], [ %x.0, %bb30 ], [ %51, %bb16 ], [ %28, %bb8 ], [ %x.0, %bb2 ] ; <float> [#uses=1]
%81 = fmul float %y.1, 0.000000e+00, !dbg !42 ; <float> [#uses=1]
%82 = fadd float %y.1, 0.000000e+00, !dbg !42 ; <float> [#uses=1]
%tmpr = fadd float %x.1, %81, !dbg !42 ; <float> [#uses=1]
%tmp89 = bitcast float %tmpr to i32 ; <i32> [#uses=1]
%tmp90 = zext i32 %tmp89 to i64 ; <i64> [#uses=1]
%tmp85 = bitcast float %82 to i32 ; <i32> [#uses=1]
%tmp86 = zext i32 %tmp85 to i64 ; <i64> [#uses=1]
%tmp87 = shl i64 %tmp86, 32 ; <i64> [#uses=1]
%ins = or i64 %tmp90, %tmp87 ; <i64> [#uses=1]
%tmp84 = bitcast i64 %ins to double ; <double> [#uses=1]
%mrv75 = insertvalue %0 undef, double %tmp84, 0, !dbg !42 ; <%0> [#uses=1]
ret %0 %mrv75, !dbg !42
}
declare float @fabsf(float)
declare float @copysignf(float, float) 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) nounwind readnone
First cut at supporting .debug_loc section. This is used to track variable information. llvm-svn: 104649
2010-05-26 07:40:22 +08:00
Upgrade tests to the latest debug info format. Mostly this is just changing the named metadata (llvm.dbg.sp, llvm.dbg.gv, llvm.dbg.<func>.lv, etc -> llvm.dbg.cu), adding a few fields to older records (DIVariable: flags/inlined-at, DICompileUnit: sp/gv/types, DISubprogram: local variables list) The tests to update were discovered by a change I'm working on to remove debug info version support - so any tests using old debug info versions I haven't updated probably are bad tests or just not actually designed to test debug info. llvm-svn: 176671
2013-03-08 08:23:31 +08:00
!llvm.dbg.cu = !{!3}
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 = !{!48}
Upgrade tests to the latest debug info format. Mostly this is just changing the named metadata (llvm.dbg.sp, llvm.dbg.gv, llvm.dbg.<func>.lv, etc -> llvm.dbg.cu), adding a few fields to older records (DIVariable: flags/inlined-at, DICompileUnit: sp/gv/types, DISubprogram: local variables list) The tests to update were discovered by a change I'm working on to remove debug info version support - so any tests using old debug info versions I haven't updated probably are bad tests or just not actually designed to test debug info. llvm-svn: 176671
2013-03-08 08:23:31 +08:00
DI: Remove DW_TAG_arg_variable and DW_TAG_auto_variable Remove the fake `DW_TAG_auto_variable` and `DW_TAG_arg_variable` tags, using `DW_TAG_variable` in their place Stop exposing the `tag:` field at all in the assembly format for `DILocalVariable`. Most of the testcase updates were generated by the following sed script: find test/ -name "*.ll" -o -name "*.mir" | xargs grep -l 'DILocalVariable' | xargs sed -i '' \ -e 's/tag: DW_TAG_arg_variable, //' \ -e 's/tag: DW_TAG_auto_variable, //' There were only a handful of tests in `test/Assembly` that I needed to update by hand. (Note: a follow-up could change `DILocalVariable::DILocalVariable()` to set the tag to `DW_TAG_formal_parameter` instead of `DW_TAG_variable` (as appropriate), instead of having that logic magically in the backend in `DbgVariable`. I've added a FIXME to that effect.) llvm-svn: 243774
2015-08-01 02:58:39 +08:00
!0 = !DILocalVariable(name: "a", line: 1921, arg: 1, scope: !1, file: !2, type: !9)
[DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label. In order to set breakpoints on labels and list source code around labels, we need collect debug information for labels, i.e., label name, the function label belong, line number in the file, and the address label located. In order to keep these information in LLVM IR and to allow backend to generate debug information correctly. We create a new kind of metadata for labels, DILabel. The format of DILabel is !DILabel(scope: !1, name: "foo", file: !2, line: 3) We hope to keep debug information as much as possible even the code is optimized. So, we create a new kind of intrinsic for label metadata to avoid the metadata is eliminated with basic block. The intrinsic will keep existing if we keep it from optimized out. The format of the intrinsic is llvm.dbg.label(metadata !1) It has only one argument, that is the DILabel metadata. The intrinsic will follow the label immediately. Backend could get the label metadata through the intrinsic's parameter. We also create DIBuilder API for labels to be used by Frontend. Frontend could use createLabel() to allocate DILabel objects, and use insertLabel() to insert llvm.dbg.label intrinsic in LLVM IR. Differential Revision: https://reviews.llvm.org/D45024 Patch by Hsiangkai Wang. llvm-svn: 331841
2018-05-09 10:40:45 +08:00
!1 = distinct !DISubprogram(name: "__divsc3", linkageName: "__divsc3", line: 1922, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: true, unit: !3, scopeLine: 1922, file: !45, scope: !2, type: !4, retainedNodes: !43)
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
!2 = !DIFile(filename: "libgcc2.c", directory: "/Users/yash/clean/LG.D/gcc/../../llvmgcc/gcc")
[PR27284] Reverse the ownership between DICompileUnit and DISubprogram. Currently each Function points to a DISubprogram and DISubprogram has a scope field. For member functions the scope is a DICompositeType. DIScopes point to the DICompileUnit to facilitate type uniquing. Distinct DISubprograms (with isDefinition: true) are not part of the type hierarchy and cannot be uniqued. This change removes the subprograms list from DICompileUnit and instead adds a pointer to the owning compile unit to distinct DISubprograms. This would make it easy for ThinLTO to strip unneeded DISubprograms and their transitively referenced debug info. Motivation ---------- Materializing DISubprograms is currently the most expensive operation when doing a ThinLTO build of clang. We want the DISubprogram to be stored in a separate Bitcode block (or the same block as the function body) so we can avoid having to expensively deserialize all DISubprograms together with the global metadata. If a function has been inlined into another subprogram we need to store a reference the block containing the inlined subprogram. Attached to https://llvm.org/bugs/show_bug.cgi?id=27284 is a python script that updates LLVM IR testcases to the new format. http://reviews.llvm.org/D19034 <rdar://problem/25256815> llvm-svn: 266446
2016-04-15 23:57:41 +08:00
!3 = distinct !DICompileUnit(language: DW_LANG_C89, producer: "4.2.1 (Based on Apple Inc. build 5658) (LLVM build)", isOptimized: true, emissionKind: FullDebug, file: !45, enums: !47, retainedTypes: !47, imports: null)
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
!4 = !DISubroutineType(types: !5)
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
!5 = !{!6, !9, !9, !9, !9}
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
!6 = !DIDerivedType(tag: DW_TAG_typedef, name: "SCtype", line: 170, file: !46, scope: !7, baseType: !8)
!7 = !DIFile(filename: "libgcc2.h", directory: "/Users/yash/clean/LG.D/gcc/../../llvmgcc/gcc")
!8 = !DIBasicType(tag: DW_TAG_base_type, name: "complex float", size: 64, align: 32, encoding: DW_ATE_complex_float)
!9 = !DIDerivedType(tag: DW_TAG_typedef, name: "SFtype", line: 167, file: !46, scope: !7, baseType: !10)
!10 = !DIBasicType(tag: DW_TAG_base_type, name: "float", size: 32, align: 32, encoding: DW_ATE_float)
DI: Remove DW_TAG_arg_variable and DW_TAG_auto_variable Remove the fake `DW_TAG_auto_variable` and `DW_TAG_arg_variable` tags, using `DW_TAG_variable` in their place Stop exposing the `tag:` field at all in the assembly format for `DILocalVariable`. Most of the testcase updates were generated by the following sed script: find test/ -name "*.ll" -o -name "*.mir" | xargs grep -l 'DILocalVariable' | xargs sed -i '' \ -e 's/tag: DW_TAG_arg_variable, //' \ -e 's/tag: DW_TAG_auto_variable, //' There were only a handful of tests in `test/Assembly` that I needed to update by hand. (Note: a follow-up could change `DILocalVariable::DILocalVariable()` to set the tag to `DW_TAG_formal_parameter` instead of `DW_TAG_variable` (as appropriate), instead of having that logic magically in the backend in `DbgVariable`. I've added a FIXME to that effect.) llvm-svn: 243774
2015-08-01 02:58:39 +08:00
!11 = !DILocalVariable(name: "b", line: 1921, arg: 2, scope: !1, file: !2, type: !9)
!12 = !DILocalVariable(name: "c", line: 1921, arg: 3, scope: !1, file: !2, type: !9)
!13 = !DILocalVariable(name: "d", line: 1921, arg: 4, scope: !1, file: !2, type: !9)
!14 = !DILocalVariable(name: "denom", line: 1923, scope: !15, file: !2, type: !9)
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
!15 = distinct !DILexicalBlock(line: 1922, column: 0, file: !45, scope: !1)
DI: Remove DW_TAG_arg_variable and DW_TAG_auto_variable Remove the fake `DW_TAG_auto_variable` and `DW_TAG_arg_variable` tags, using `DW_TAG_variable` in their place Stop exposing the `tag:` field at all in the assembly format for `DILocalVariable`. Most of the testcase updates were generated by the following sed script: find test/ -name "*.ll" -o -name "*.mir" | xargs grep -l 'DILocalVariable' | xargs sed -i '' \ -e 's/tag: DW_TAG_arg_variable, //' \ -e 's/tag: DW_TAG_auto_variable, //' There were only a handful of tests in `test/Assembly` that I needed to update by hand. (Note: a follow-up could change `DILocalVariable::DILocalVariable()` to set the tag to `DW_TAG_formal_parameter` instead of `DW_TAG_variable` (as appropriate), instead of having that logic magically in the backend in `DbgVariable`. I've added a FIXME to that effect.) llvm-svn: 243774
2015-08-01 02:58:39 +08:00
!16 = !DILocalVariable(name: "ratio", line: 1923, scope: !15, file: !2, type: !9)
!17 = !DILocalVariable(name: "x", line: 1923, scope: !15, file: !2, type: !9)
!18 = !DILocalVariable(name: "y", line: 1923, scope: !15, file: !2, type: !9)
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
!19 = !DILocation(line: 1929, scope: !15)
!20 = !DILocation(line: 1931, scope: !15)
!21 = !DILocation(line: 1932, scope: !15)
!22 = !DILocation(line: 1933, scope: !15)
!23 = !DILocation(line: 1934, scope: !15)
!24 = !DILocation(line: 1938, scope: !15)
!25 = !DILocation(line: 1939, scope: !15)
!26 = !DILocation(line: 1940, scope: !15)
!27 = !DILocation(line: 1941, scope: !15)
!28 = !DILocation(line: 1946, scope: !15)
!29 = !DILocation(line: 1948, scope: !15)
!30 = !DILocation(line: 1950, scope: !15)
!31 = !DILocation(line: 1951, scope: !15)
!32 = !DILocation(line: 1953, scope: !15)
!33 = !DILocation(line: 1955, scope: !15)
!34 = !DILocation(line: 1956, scope: !15)
!35 = !DILocation(line: 1957, scope: !15)
!36 = !DILocation(line: 1958, scope: !15)
!37 = !DILocation(line: 1960, scope: !15)
!38 = !DILocation(line: 1962, scope: !15)
!39 = !DILocation(line: 1963, scope: !15)
!40 = !DILocation(line: 1964, scope: !15)
!41 = !DILocation(line: 1965, scope: !15)
!42 = !DILocation(line: 1969, scope: !15)
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 = !{!0, !11, !12, !13, !14, !16, !17, !18}
IR: Give 'DI' prefix to debug info metadata Finish off PR23080 by renaming the debug info IR constructs from `MD*` to `DI*`. The last of the `DIDescriptor` classes were deleted in r235356, and the last of the related typedefs removed in r235413, so this has all baked for about a week. Note: If you have out-of-tree code (like a frontend), I recommend that you get everything compiling and tests passing with the *previous* commit before updating to this one. It'll be easier to keep track of what code is using the `DIDescriptor` hierarchy and what you've already updated, and I think you're extremely unlikely to insert bugs. YMMV of course. Back to *this* commit: I did this using the rename-md-di-nodes.sh upgrade script I've attached to PR23080 (both code and testcases) and filtered through clang-format-diff.py. I edited the tests for test/Assembler/invalid-generic-debug-node-*.ll by hand since the columns were off-by-three. It should work on your out-of-tree testcases (and code, if you've followed the advice in the previous paragraph). Some of the tests are in badly named files now (e.g., test/Assembler/invalid-mdcompositetype-missing-tag.ll should be 'dicompositetype'); I'll come back and move the files in a follow-up commit. llvm-svn: 236120
2015-04-30 00:38:44 +08:00
!45 = !DIFile(filename: "libgcc2.c", directory: "/Users/yash/clean/LG.D/gcc/../../llvmgcc/gcc")
!46 = !DIFile(filename: "libgcc2.h", directory: "/Users/yash/clean/LG.D/gcc/../../llvmgcc/gcc")
DebugInfo: Fix bad debug info for compile units and types Fix debug info in these tests, which started failing with a WIP patch to verify compile units and types. The problems look like they were all caused by bitrot. They fell into these categories: - Using `!{i32 0}` instead of `!{}`. - Using `!{null}` instead of `!{}`. - Using `!MDExpression()` instead of `!{}`. - Using `!8` instead of `!{!8}`. - `file:` references that pointed at `MDCompileUnit`s instead of the same `MDFile` as the compile unit. - `file:` references that were numerically off-by-one or (off-by-ten). llvm-svn: 233415
2015-03-28 04:46:33 +08:00
!47 = !{}
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
!48 = !{i32 1, !"Debug Info Version", i32 3}