Changes in fp-stack.ll

a48e0503 - [DebugInfo][InstrRef] "final final" test cleanups for x86 tests

Tue, 30 Nov 2021 10:21:31 +0000

[DebugInfo][InstrRef] "final final" test cleanups for x86 testsTwo "totally definitely the last ones" instruction referencing testupdates: * fp-stack.ll: this test targets i686, and so it won't be getting instruction referencing, or at least not right now, * X86/live-debug-values.ll: instruction referencing will produce entry values in this test, add check lines to account for this. It's not clear what the test is supposed to be testing anyway, but the entry values appear to be correct.Differential Revision: https://reviews.llvm.org/D114626 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

1dc0e47c - [DebugInfo][NFC] Force some tests to not use instruction-referencing

Wed, 17 Nov 2021 11:35:58 +0000

[DebugInfo][NFC] Force some tests to not use instruction-referencingThere are various tests that need to be adjusted to test the rightthing with instruction referencing -- usually because the internalrepresentation of variables is different, sometimes that location listschange. This patch makes a bunch of tests explicitly not useinstruction referencing, so that a check-llvm test with instructionreferencing on for x86_64 doesn't fail. I'll then convert the teststo have instr-ref CHECK lines, and similar.Differential Revision: https://reviews.llvm.org/D113194 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

da60fc81 - Changing representation of .cv_def_range directives in Codeview debug info assembly format for better readability

Mon, 05 Aug 2019 14:16:58 +0000

Changing representation of .cv_def_range directives in Codeview debug info assembly format for better readabilityllvm-svn: 367867 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

b5e4d7de - Revert "Changing representation of .cv_def_range directives in Codeview debug info assembly format for better readability"

Mon, 05 Aug 2019 13:55:21 +0000

Revert "Changing representation of .cv_def_range directives in Codeview debug info assembly format for better readability"This reverts commit a885afa9fa8cab3b34f1ddf3d21535f88b662881.llvm-svn: 367861 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

a885afa9 - Changing representation of .cv_def_range directives in Codeview debug info assembly format for better readability

Mon, 05 Aug 2019 13:11:51 +0000

Changing representation of .cv_def_range directives in Codeview debug info assembly format for better readabilityllvm-svn: 367850 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

5fb56b19 - Temporarily Revert "Changing representation of cv_def_range directives in Codeview debug info assembly format for better readability"

Fri, 02 Aug 2019 19:10:37 +0000

Temporarily Revert "Changing representation of cv_def_range directives in Codeview debug info assembly format for better readability"This is breaking bots and the author asked me to revert.This reverts commit 367704.llvm-svn: 367707 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

1c675215 - Changing representation of cv_def_range directives in Codeview debug info assembly format for better readability

Fri, 02 Aug 2019 18:44:39 +0000

Changing representation of cv_def_range directives in Codeview debug info assembly format for better readabilityllvm-svn: 367704 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

e29e30b1 - [llvm-readobj] Change -long-option to --long-option in tests. NFC

Wed, 01 May 2019 05:27:20 +0000

[llvm-readobj] Change -long-option to --long-option in tests. NFCWe use both -long-option and --long-option in tests. Switch to --long-option for consistency.In the "llvm-readelf" mode, -long-option is discouraged as it conflicts with grouped short options and it is not accepted by GNU readelf.While updating the tests, change llvm-readobj -s to llvm-readobj -S to reduce confusion ("s" is --section-headers in llvm-readobj but --symbols in llvm-readelf).llvm-svn: 359649 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

9dbeca3d - [DebugInfo] Stop changing labels for register-described parameter DBG_VALUEs

Wed, 13 Feb 2019 09:34:07 +0000

[DebugInfo] Stop changing labels for register-described parameter DBG_VALUEsSummary:This is a follow-up to D57510. This patch stops DebugHandlerBase fromchanging the starting label for the first non-overlapping,register-described parameter DBG_VALUEs to the beginning of thefunction. That code did not consider what defined the registers, whichcould result in the ranges for the debug values starting before theirdefining instructions. We currently do not emit debug values forconstant values directly at the start of the function, so this code isstill useful for such values, but my intention is to remove the codefrom DebugHandlerBase completely when we get there. One reason forremoving it is that the code violates the history map's ranges, which Ithink can make it quite confusing when troubleshooting.In D57510, PrologEpilogInserter was amended so that parameter DBG_VALUEsnow are kept at the start of the entry block, even after emission ofprologue code. That was done to reduce the degradation of debugcompleteness from this patch. PR40638 is another example, where thelexical-scope trimming that LDV does, in combination with scheduling,results in instructions after the prologue being left without locations.There might be other cases where the DBG_VALUEs are pushed further down,for which the DebugHandlerBase code may be helpful, but as it now quiteoften result in incorrect locations, even after the prologue, it seemsbetter to remove that code, and try to work our way up with accuratelocations.In the long run we should maybe not aim to provide accurate locationsinside the prologue. Some single location descriptions, at least thosereferring to stack values, generate inaccurate values inside theepilogue, so we maybe should not aim to achieve accuracy for locationlists. However, it seems that we now emit line number programs that canresult in GDB and LLDB stopping inside the prologue when doing linenumber stepping into functions. See PR40188 for more information.A summary of some of the changed test cases is available in PR40188#c2.Reviewers: aprantl, dblaikie, rnk, jmorseReviewed By: aprantlSubscribers: jdoerfert, jholewinski, jvesely, javed.absar, llvm-commitsTags: #debug-info, #llvmDifferential Revision: https://reviews.llvm.org/D57511llvm-svn: 353928 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

bd5d7122 - [codeview] Use push_macro to avoid conflicts instead of a prefix

Thu, 16 Aug 2018 17:34:31 +0000

[codeview] Use push_macro to avoid conflicts instead of a prefixSummary:This prefix was added in r333421, and it changed our dumper output tosay things like "CVRegEAX" instead of just "EAX". That's a functionalchange that I'd rather avoid.I tested GCC, Clang, and MSVC, and all of them support #pragmapush_macro. They don't issue warnings whem the macro is not definedeither.I don't have a Mac so I can't test the real termios.h header, but Ilooked at the termios.h sources online and looked for other conflicts.I saw only the CR* macros, so those are the ones we work around.Reviewers: zturner, JDevlieghereSubscribers: hiraditya, llvm-commitsDifferential Revision: https://reviews.llvm.org/D50851llvm-svn: 339907 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

43dce3ed - [CodeView] Add prefix to CodeView registers.

Tue, 29 May 2018 14:35:34 +0000

[CodeView] Add prefix to CodeView registers.Adds CVReg to CodeView register names to prevent a duplicate symbol withCR3 defined in termios.h, as suggested by Zachary on the mailing list.http://lists.llvm.org/pipermail/llvm-dev/2018-May/123372.htmlDifferential revision: https://reviews.llvm.org/D47478rdar://39863705llvm-svn: 333421 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

2c864551 - [DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label.

Wed, 09 May 2018 02:40:45 +0000

[DebugInfo] Add DILabel metadata and intrinsic llvm.dbg.label.In order to set breakpoints on labels and list source code aroundlabels, we need collect debug information for labels, i.e., labelname, the function label belong, line number in the file, and theaddress label located. In order to keep these information in LLVMIR and to allow backend to generate debug information correctly.We create a new kind of metadata for labels, DILabel. The formatof DILabel is!DILabel(scope: !1, name: "foo", file: !2, line: 3)We hope to keep debug information as much as possible even thecode is optimized. So, we create a new kind of intrinsic for labelmetadata 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 isllvm.dbg.label(metadata !1)It has only one argument, that is the DILabel metadata. Theintrinsic will follow the label immediately. Backend could get thelabel 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 useinsertLabel() to insert llvm.dbg.label intrinsic in LLVM IR.Differential Revision: https://reviews.llvm.org/D45024Patch by Hsiangkai Wang.llvm-svn: 331841 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

66053108 - CodeView: Provide a .def file with the register ids

Tue, 03 Oct 2017 18:27:22 +0000

CodeView: Provide a .def file with the register idsThe list of register ids was previously written out in a couple of dirrentplaces. This puts it in a .def file and also adds a few more registers (e.g.the x87 regs) which should lead to more readable dumps, but I didn't includethe whole list since that seems unnecessary.X86_MC::initLLVMToSEHAndCVRegMapping is pretty ugly, but at least it's notrelying on magic constants anymore. The TODO of using tablegen still stands.Differential revision: https://reviews.llvm.org/D38480llvm-svn: 314821 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

ea89ff7c - CodeView symbol dumper: use symbolic names for registers

Mon, 02 Oct 2017 17:44:47 +0000

CodeView symbol dumper: use symbolic names for registershttps://reviews.llvm.org/D38469llvm-svn: 314690 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

abe04759 - Remove the obsolete offset parameter from @llvm.dbg.value

Fri, 28 Jul 2017 20:21:02 +0000

Remove the obsolete offset parameter from @llvm.dbg.valueThere is no situation where this rarely-used argument cannot besubstituted with a DIExpression and removing it allows us to simplifythe DWARF backend. Note that this patch does not yet remove any ofthe newly dead code.rdar://problem/33580047Differential Revision: https://reviews.llvm.org/D35951llvm-svn: 309426 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

8d8888ff - [codeview] Change readobj symbol dumping format

Tue, 11 Jul 2017 23:41:41 +0000

[codeview] Change readobj symbol dumping formatAvoid duplicating DictScope with hand-written names everywhere. Printthe S_-prefixed symbol kind for every record. This should make it easierto search for certain kinds of records when debugging PDB linking.llvm-svn: 307732 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

75819aed - [PR27284] Reverse the ownership between DICompileUnit and DISubprogram.

Fri, 15 Apr 2016 15:57:41 +0000

[PR27284] Reverse the ownership between DICompileUnit and DISubprogram.Currently each Function points to a DISubprogram and DISubprogram has ascope field. For member functions the scope is a DICompositeType. DIScopespoint to the DICompileUnit to facilitate type uniquing.Distinct DISubprograms (with isDefinition: true) are not part of the typehierarchy and cannot be uniqued. This change removes the subprogramslist from DICompileUnit and instead adds a pointer to the owning compileunit to distinct DISubprograms. This would make it easy for ThinLTO tostrip unneeded DISubprograms and their transitively referenced debug info.Motivation----------Materializing DISubprograms is currently the most expensive operation whendoing a ThinLTO build of clang.We want the DISubprogram to be stored in a separate Bitcode block (or thesame block as the function body) so we can avoid having to expensivelydeserialize all DISubprograms together with the global metadata. If afunction has been inlined into another subprogram we need to store areference the block containing the inlined subprogram.Attached to https://llvm.org/bugs/show_bug.cgi?id=27284 is a python scriptthat updates LLVM IR testcases to the new format.http://reviews.llvm.org/D19034llvm-svn: 266446 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

b8089516 - testcase gardening: update the emissionKind enum to the new syntax. (NFC)

Fri, 01 Apr 2016 00:16:49 +0000

testcase gardening: update the emissionKind enum to the new syntax. (NFC)llvm-svn: 265081 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll

e2ad7375 - [CodeView] Describe variables live in x87 registers

Wed, 24 Feb 2016 10:01:24 +0000

[CodeView] Describe variables live in x87 registersWe didn't have a mapping from LLVM's x87 floating point registers toCodeView's encoding.llvm-svn: 261730 List of files: /llvm-project-15.0.7/llvm/test/DebugInfo/COFF/fp-stack.ll