[X86][AVX512] Add support for VBMI VPERMV3 target shuffle combines

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287496 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX512] Add support for VBMI VPERMV target shuffle combines

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287495 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX512] Add some initial VBMI target shuffle combine tests

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287494 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX512VL] Removed duplicate operation action

Basic AVX512F already declared uint_to_fp v4i32 as legal

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287493 91177308-0d34-0410-b5e6-96231b3b80d8

Strip trailing whitespace

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287492 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX512F] Add support for uint_to_fp v2i32 to v2f64 on AVX512F-only targets

Use 512-bit instructions (we already do something similar for uint_to_fp v4i32 to v4f64)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287491 91177308-0d34-0410-b5e6-96231b3b80d8

Fix comment typos. NFC.

Identified by Pedro Giffuni in PR27636.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287490 91177308-0d34-0410-b5e6-96231b3b80d8

Fix spelling mistakes in Tools/Tests comments. NFC.

Identified by Pedro Giffuni in PR27636.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287489 91177308-0d34-0410-b5e6-96231b3b80d8

Fix spelling mistakes in Transforms comments. NFC.

Identified by Pedro Giffuni in PR27636.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287488 91177308-0d34-0410-b5e6-96231b3b80d8

Fix spelling mistakes in SelectionDAG comments. NFC.

Identified by Pedro Giffuni in PR27636.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287487 91177308-0d34-0410-b5e6-96231b3b80d8

Fix comment typos. NFC.

Identified by Pedro Giffuni in PR27636.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287486 91177308-0d34-0410-b5e6-96231b3b80d8

[X86] RegCall - Handling long double arguments

The change is part of RegCall calling convention support for LLVM.
Long double (f80) requires special treatment as the first f80 parameter is saved in FP0 (floating point stack).
This review present the change and the corresponding tests.

Differential Revision: https://reviews.llvm.org/D26151

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287485 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][InlineAsm]Test commit.
Fixing a wrong comment on X86AsmParser.cpp::ParseZ: "true" --> "false"

Differential Revision: https://reviews.llvm.org/D26797

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287484 91177308-0d34-0410-b5e6-96231b3b80d8

Fix file name resolution in nested response files

If a response file in construct `@file` was specified by relative name,
constructs `@file` nested within it were resolved incorrectly if the
flag RelativeNames in call to ExpandResponseFile was set to true.
This feature is used in configuration files, tests for it are in
respective change (see D24933).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287482 91177308-0d34-0410-b5e6-96231b3b80d8

ExceptionDemo: remove some undefined behaviour

The casting based reading of the LSDA could attempt to read unsuitably aligned
data.  Avoid that case by explicitly using a memcpy.  A similar approach is used
in libc++abi to address the same UB.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287479 91177308-0d34-0410-b5e6-96231b3b80d8

ExceptionDemo: prefer headers over redeclarations

Rather than redeclaring the interfaces for exceptions, prefer using the
`unwind.h` header.  This is vended by at least gcc and clang, and can also be
found by an external unwinding library (e.g. libunwind).  Doing this simplifies
the example to the exception handling itself.  Minor tweaks are the result of
_Unwind_Context_t not being defined, which is just a typedef for struct
_Unwind_Context *.  NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287478 91177308-0d34-0410-b5e6-96231b3b80d8

[bpf] add BPF disassembler

add BPF disassembler, so tools like llvm-objdump can be used:
$ llvm-objdump -d -no-show-raw-insn ./sockex1_kern.o

./sockex1_kern.o: file format ELF64-BPF

Disassembly of section socket1:
bpf_prog1:
       0: r6 = r1
       8: r0 = *(u8 *)skb[23]
      10: *(u32 *)(r10 - 4) = r0
      18: r1 = *(u32 *)(r6 + 4)
      20: if r1 != 4 goto 8
      28: r2 = r10
      30: r2 += -4

ld_imm64 (the only 16-byte insn) and special ld_abs/ld_ind instructions
had to be treated in a special way. The decoders for the rest of the insns
are automatically generated.

Add tests to cover new functionality.

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287477 91177308-0d34-0410-b5e6-96231b3b80d8

Attempt to fix big-endian buildbots.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287476 91177308-0d34-0410-b5e6-96231b3b80d8

Style fix. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287475 91177308-0d34-0410-b5e6-96231b3b80d8

Fix buildbot.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287474 91177308-0d34-0410-b5e6-96231b3b80d8

SHA1: unroll loop in hashBlock.

This code is taken from public domain.
https://github.com/jsonn/src/blob/trunk/common/lib/libc/hash/sha1/sha1.c

I wrote a sha1 command and ran it on my Xeon E5-2680 v2 2.80GHz machine.
Here is a result. The new hash function is 37% faster than before.

 Performance counter stats for './llvm-sha1-old /ssd/build/bin/lld' (10 runs):

       6640.503687 task-clock (msec)         #    1.001 CPUs utilized            ( +-  0.03% )
                54 context-switches          #    0.008 K/sec                    ( +-  5.03% )
                 5 cpu-migrations            #    0.001 K/sec                    ( +- 31.73% )
           183,803 page-faults               #    0.028 M/sec                    ( +-  0.00% )
    18,527,954,113 cycles                    #    2.790 GHz                      ( +-  0.03% )
     4,993,237,485 stalled-cycles-frontend   #   26.95% frontend cycles idle     ( +-  0.11% )
   <not supported> stalled-cycles-backend
    50,217,149,423 instructions              #    2.71  insns per cycle
                                             #    0.10  stalled cycles per insn  ( +-  0.00% )
     6,094,322,337 branches                  #  917.750 M/sec                    ( +-  0.00% )
        11,778,239 branch-misses             #    0.19% of all branches          ( +-  0.01% )

       6.634017401 seconds time elapsed                                          ( +-  0.03% )

 Performance counter stats for './llvm-sha1-new /ssd/build/bin/lld' (10 runs):

       4167.062720 task-clock (msec)         #    1.001 CPUs utilized            ( +-  0.02% )
                52 context-switches          #    0.012 K/sec                    ( +- 16.45% )
                 7 cpu-migrations            #    0.002 K/sec                    ( +- 32.20% )
           183,804 page-faults               #    0.044 M/sec                    ( +-  0.00% )
    11,626,611,958 cycles                    #    2.790 GHz                      ( +-  0.02% )
     4,491,897,976 stalled-cycles-frontend   #   38.63% frontend cycles idle     ( +-  0.05% )
   <not supported> stalled-cycles-backend
    24,320,180,617 instructions              #    2.09  insns per cycle
                                             #    0.18  stalled cycles per insn  ( +-  0.00% )
     1,574,674,576 branches                  #  377.886 M/sec                    ( +-  0.00% )
        11,769,693 branch-misses             #    0.75% of all branches          ( +-  0.00% )

       4.163251552 seconds time elapsed                                          ( +-  0.02% )

Differential Revision: https://reviews.llvm.org/D26890

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287473 91177308-0d34-0410-b5e6-96231b3b80d8

Demangle: remove references to allocator for default allocator

The demangler had stopped using a custom allocator but had not been updated to
remove the use of the explicit allocator passing.  This removes that as we do
not need to do anything special here anymore.  This just makes the code more
compact.  NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287472 91177308-0d34-0410-b5e6-96231b3b80d8

Demangle: remove unnecessary typedef for std::vector

We could create a local typedef for std::vector called Vector.  Inline the use
of std::vector rather than use the typedef.  NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287471 91177308-0d34-0410-b5e6-96231b3b80d8

Demangle: replace custom typedef for std::string with std::string

We created a local typedef for `std::basic_string<char, std::char_traits<char>>`
which is just `std::string`.  Remove the local typedef and propagate the type
information through the rest of the demangler.  NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287470 91177308-0d34-0410-b5e6-96231b3b80d8

Demangle: use direct member initialization (NFC)

Prefer direct member initialization over the explicit out-of-line initialization
for the construction of the local type.  NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287469 91177308-0d34-0410-b5e6-96231b3b80d8

Give some helper classes/functions internal linkage. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287462 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][SSE] Improve PSHUFB lowering from either input

Canonicalization may leave the zeroable vector in the first input.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287461 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX512] Add VPERMV/VPERMV3 v64i8 byte shuffles on avx512vbmi targets

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287459 91177308-0d34-0410-b5e6-96231b3b80d8

[ThinLTO] Fix crash when importing an opaque type

It seems that because ThinLTO does not import the full module,
some invariant of the type mapper are broken.

In Monolithic LTO, we import every globals: when calling
IRLinker::copyFunctionProto() on @foo(), we end-up calling
TypeMapTy::get(FTy) on the type of @foo(), which will map
%0 and record the destination as opaque.

ThinLTO skips this because @foo is not imported and goes directly
to the next stage.

Next we call computeTypeMapping() that map the types for each
globals, and ends up checking for type isomorphism, and may add
type mapping. However it doesn't record if there was an opaque
destination type that was resolved.

Instead of lazily "discovering" opaque type in the destination
module on the go, we change the TypeFinder to eagerly record all
types and not only the named ones.

Differential Revision: https://reviews.llvm.org/D26840

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287453 91177308-0d34-0410-b5e6-96231b3b80d8

[ThinLTO] Implement -pass-remarks-output in ThinLTOCodeGenerator

Summary:
This will also be added to the LTO API, right now this will
bring ThinLTO on par with Monolithic LTO on Darwin.

Reviewers: anemet

Subscribers: tejohnson, llvm-commits

Differential Revision: https://reviews.llvm.org/D26886

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287450 91177308-0d34-0410-b5e6-96231b3b80d8

Change setDiagnosticsOutputFile to take a unique_ptr from a raw pointer (NFC)

Summary:
This makes it explicit that ownership is taken. Also replace all `new`
with make_unique<> at call sites.

Reviewers: anemet

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D26884

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287449 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX512] Add avx512vbmi tests

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287447 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX512] Added some more complex v64i8 shuffles

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287444 91177308-0d34-0410-b5e6-96231b3b80d8

[X86] Simplify some code a little by removing a dulicate variable and combinining two if statements. NFCI

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287443 91177308-0d34-0410-b5e6-96231b3b80d8

Try again to fix unused variable warning on lld-x86_64-darwin13 after r287439.

The previous attempt didn't work. I assume LLVM_ATTRIBUTE_UNUSED isn't
available on that machine.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287442 91177308-0d34-0410-b5e6-96231b3b80d8

Try to fix unused variable warning on lld-x86_64-darwin13 after r287439.

Whether the variable is used or not depends on NDEBUG.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287440 91177308-0d34-0410-b5e6-96231b3b80d8

Check that emitted instructions meet their predicates on all targets except ARM, Mips, and X86.

Summary:
* ARM is omitted from this patch because this check appears to expose bugs in this target.
* Mips is omitted from this patch because this check either detects bugs or deliberate
  emission of instructions that don't satisfy their predicates. One deliberate
  use is the SYNC instruction where the version with an operand is correctly
  defined as requiring MIPS32 while the version without an operand is defined
  as an alias of 'SYNC 0' and requires MIPS2.
* X86 is omitted from this patch because it doesn't use the tablegen-erated
  MCCodeEmitter infrastructure.

Patches for ARM and Mips will follow.

Depends on D25617

Reviewers: tstellarAMD, jmolloy

Subscribers: wdng, jmolloy, aemerson, rengolin, arsenm, jyknight, nemanjai, nhaehnle, tstellarAMD, llvm-commits

Differential Revision: https://reviews.llvm.org/D25618

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287439 91177308-0d34-0410-b5e6-96231b3b80d8

[tablegen] Merge duplicate definitions of getMinimalTypeForRange. NFC.

Summary: Depends on D25614

Reviewers: qcolombet

Subscribers: qcolombet, beanz, mgorny, llvm-commits

Differential Revision: https://reviews.llvm.org/D25617

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287438 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-lto2 depends on intrinsics_gen

llvm-lto2.cpp has the following include chain:

llvm/LTO/Caching.h
llvm/LTO/LTO.h
llvm/CodeGen/Analysis.h
llvm/IR/CallSite.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-lto2 needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287434 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] opt depends on intrinsics_gen

AnalysisWrappers.cpp has the following include chain:

llvm/Analysis/CallGraph.h
llvm/IR/CallSite.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means opt needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287433 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-nm depends on intrinsics_gen

llvm-nm.cpp has the following include chain:

llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-nm needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287432 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-link depends on intrinsics_gen

llvm-link.cpp has the following include chain:

llvm/Bitcode/BitcodeWriter.h
llvm/IR/ModuleSummaryIndex.h
llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-link needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287431 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-extract depends on intrinsics_gen

llvm-extract.cpp has the following include chain:

llvm/Bitcode/BitcodeWriterPass.h
llvm/IR/PassManager.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-extract needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287430 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-dwp depends on intrinsics_gen

llvm-dwp.cpp has the following include chain:

llvm/CodeGen/AsmPrinter.h
llvm/CodeGen/MachineFunctionPass.h
llvm/CodeGen/MachineFunction.h
llvm/CodeGen/MachineBasicBlock.h
llvm/CodeGen/MachineInstr.h
llvm/Analysis/AliasAnalysis.h
llvm/IR/CallSite.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-dwp needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287429 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-dis depends on intrinsics_gen

llvm-dis.cpp has the following include chain:

llvm/Bitcode/BitcodeReader.h
llvm/IR/ModuleSummaryIndex.h
llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-dis needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287428 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-diff depends on intrinsics_gen

llvm-diff.cpp has the following include chain:

llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-diff needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287427 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-stress depends on intrinsics_gen

llvm-stress.cpp has the following include chain:

llvm/Analysis/CallGraphSCCPass.h
llvm/Analysis/CallGraph.h
llvm/IR/CallSite.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-stress needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287426 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] bugpoint-passes depends on intrinsics_gen

TestPasses.cpp has the following include chain:

llvm/IR/InstVisitor.h
llvm/IR/CallSite.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means bugpoint-passes needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287425 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-bcanalyzer depends on intrinsics_gen

llvm-bcanalyzer.cpp has the following include chain:

llvm/Bitcode/BitcodeReader.h
llvm/IR/ModuleSummaryIndex.h
llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-bcanalyzer needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287424 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-as depends on intrinsics_gen

llvm-as.cpp has the following include chain:

llvm/Bitcode/BitcodeWriter.h
llvm/IR/ModuleSummaryIndex.h
llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-as needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287423 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llc depends on intrinsics_gen

llc.cpp has the following include chain:

llvm/Analysis/TargetLibraryInfo.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llc needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287422 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] lli-child-target depends on intrinsics gen

Messed up in r287420, it isn't just lli, but also but lli-child-target that need to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287421 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] lli depends on intrinsics_gen

ChildTarget.cpp has the following include chain:

llvm/ExecutionEngine/Orc/OrcABISupport.h
llvm/ExecutionEngine/Orc/IndirectionUtils.h
llvm/IR/IRBuilder.h
llvm/IR/ConstantFolder.h
llvm/IR/InstrTypes.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means lli needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287420 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-dsymutil depends on intrinsics_gen

DwarfLinker.cpp has the following include chain:

llvm/CodeGen/AsmPrinter.h
llvm/CodeGen/MachineFunctionPass.h
llvm/CodeGen/MachineFunction.h
llvm/CodeGen/MachineBasicBlock.h
llvm/CodeGen/MachineInstr.h
llvm/Analysis/AliasAnalysis.h
llvm/IR/CallSite.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-dsymutil needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287419 91177308-0d34-0410-b5e6-96231b3b80d8

[AVR] Remove a bunch of unused variables

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287416 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] Apply sandbox profile to target not directory

When LLVM_DEPENDENCY_DEBUGGING=On we should apply the sandbox only on the target, not the directory. This is important for directories that create more than one target, or for nested directories.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287415 91177308-0d34-0410-b5e6-96231b3b80d8

[AVR] Remove a variable which was unused in release mode

In release mode where assertions are not enabled, this caused an 'unused
variable' warning.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287414 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] verify-uselistorder depends on intrinsics_gen

verify-uselistorder.cpp has the following include chain:

llvm/Bitcode/BitcodeReader.h
llvm/IR/ModuleSummaryIndex.h
llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means verify-uselistorder needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287405 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] sanstats depends on intrinsics_gen

sanstats.cpp has the following include chain:

llvm/Transforms/Utils/SanitizerStats.h
llvm/IR/IRBuilder.h
llvm/IR/ConstantFolder.h
llvm/IR/InstrTypes.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means sanstats needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287404 91177308-0d34-0410-b5e6-96231b3b80d8

[lit] When setting SDKROOT on Darwin, use '--sdk macosx' to find the right SDK path.

This will make sure that we find an actual path in case you have Command Line Tools installed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287403 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] bugpoint depends on intrinsics_gen

CrashDebugger.cpp has the following include chain:

llvm/Analysis/TargetTransformInfo.h
llvm/IR/IntrinsicInst.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means bugpoint needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287402 91177308-0d34-0410-b5e6-96231b3b80d8

[InstCombine] add tests to show likely unwanted select widening; NFC

This is a prerequisite patch for D26556:
https://reviews.llvm.org/D26556

...because there was no direct coverage for these folds (which in some cases are adding instructions).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287400 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-split depends on intrinsics_gen

llvm-split.cpp has the following include chain:

llvm/Bitcode/BitcodeWriter.h
llvm/IR/ModuleSummaryIndex.h
llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-split needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287399 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-lto depends on intrinsics_gen

llvm-lto.cpp has the following include chain:

llvm/Bitcode/BitcodeReader.h
llvm/IR/ModuleSummaryIndex.h
llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-lto needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287398 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-ar depends on intrinsics_gen

llvm-ar.cpp has the following include chain:

llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-ar needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287395 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] llvm-profdata depends on intrinsics_gen

llvm-profdata.cpp has the following include chain:

llvm/ProfileData/SampleProfReader.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means llvm-profdata needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287394 91177308-0d34-0410-b5e6-96231b3b80d8

[CMake] LTO depends on intrinsics_gen

lto.cpp has the following include chain:

llvm/Bitcode/BitcodeReader.h
llvm/IR/ModuleSummaryIndex.h
llvm/IR/Module.h
llvm/IR/Function.h
llvm/IR/Argument.h
llvm/IR/Attributes.h
llvm/IR/Attributes.gen

This means LTO needs to depend on intrinsics_gen.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287393 91177308-0d34-0410-b5e6-96231b3b80d8

[AMDGPU] Change frexp.exp intrinsic to return i16 for f16 input

Differential Revision: https://reviews.llvm.org/D26862

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287389 91177308-0d34-0410-b5e6-96231b3b80d8

[SelectionDAG] Add knowbits support for CONCAT_VECTOR opcode

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287387 91177308-0d34-0410-b5e6-96231b3b80d8

[X86] Add knownbits concat_vector test

Support coming in a future patch

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287385 91177308-0d34-0410-b5e6-96231b3b80d8

[Examples] Fix some Clang-tidy modernize-use-default and Include What You Use warnings; other minor fixes.

Differential revision: https://reviews.llvm.org/D26433

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287384 91177308-0d34-0410-b5e6-96231b3b80d8

[LoopSimplify] Preserve LCSSA when removing edges from unreachable blocks.

This fixes PR30454.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287379 91177308-0d34-0410-b5e6-96231b3b80d8

[MIRPrinter] XFAIL test for powerpc

This test introduced in r287368 is failing on powerpc for reasons
unrelated to branch probabilities.  See PR31062.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287375 91177308-0d34-0410-b5e6-96231b3b80d8

Revert "Add link-time detection of LLVM_ABI_BREAKING_CHECKS mismatch"

This reverts commit r287352, LLDB CI is broken.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287374 91177308-0d34-0410-b5e6-96231b3b80d8

Statistic/Timer: Include timers in PrintStatisticsJSON().

Differential Revision: https://reviews.llvm.org/D25588

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287370 91177308-0d34-0410-b5e6-96231b3b80d8

Timer: Track name and description.

The previously used "names" are rather descriptions (they use multiple
words and contain spaces), use short programming language identifier
like strings for the "names" which should be used when exporting to
machine parseable formats.

Also removed a unused TimerGroup from Hexxagon.

Differential Revision: https://reviews.llvm.org/D25583

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287369 91177308-0d34-0410-b5e6-96231b3b80d8

[MIRPrinter] Print raw branch probabilities as expected by MIRParser

Fixes PR28751.

Reviewers: MatzeB, qcolombet

Subscribers: mcrosier, llvm-commits

Differential Revision: https://reviews.llvm.org/D26775

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287368 91177308-0d34-0410-b5e6-96231b3b80d8

AMDGPU: Fix unused variable warning

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287362 91177308-0d34-0410-b5e6-96231b3b80d8

Fix test from r287353: don't use /dev/null

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287360 91177308-0d34-0410-b5e6-96231b3b80d8

[LTO] Add option to generate optimization records

It is used to drive this from the clang driver via -mllvm.

Same option name is used as in opt.

Differential Revision: https://reviews.llvm.org/D26832

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287356 91177308-0d34-0410-b5e6-96231b3b80d8

[DebugInfo] Fix some Clang-tidy modernize-use-default, modernize-use-equal-delete and Include What You Use warnings; other minor fixes (NFC).

Per Zachary Turner and Mehdi Amini suggestion to make only post-commit reviews.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287355 91177308-0d34-0410-b5e6-96231b3b80d8

IRMover: Avoid accidentally mapping types from the destination module (PR30799)

During Module linking, it's possible for SrcM->getIdentifiedStructTypes();
to return types that are actually defined in the destination module
(DstM). Depending on how the bitcode file was read,
getIdentifiedStructTypes() might do a walk over all values, including
metadata nodes, looking for types. In my case, a debug info metadata
node was shared between the two modules, and it referred to a type
defined in the destination module (see test case).

Differential Revision: https://reviews.llvm.org/D26212

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287353 91177308-0d34-0410-b5e6-96231b3b80d8

Add link-time detection of LLVM_ABI_BREAKING_CHECKS mismatch

Summary:
LLVM will define a symbol, either EnableABIBreakingChecks or
DisableABIBreakingChecks depending on the configuration setting for
LLVM_ABI_BREAKING_CHECKS.

The llvm-config.h header will add weak references to these symbols in
every clients that includes this header. This should ensure that
a mismatch triggers a link failure (or a load time failure for DSO).

On MSVC, the pragma "detect_mismatch" is used instead.

Reviewers: rnk, jroelofs

Subscribers: llvm-commits, mgorny

Differential Revision: https://reviews.llvm.org/D26841

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287352 91177308-0d34-0410-b5e6-96231b3b80d8

[PPC] limit line width to 80 characters

NFC. Forgot to fix this in the original commit.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287350 91177308-0d34-0410-b5e6-96231b3b80d8

[mips][msa] Implement f16 support

The MIPS MSA ASE provides instructions to convert to and from half precision
floating point. This patch teaches the MIPS backend to treat f16 as a legal
type and how to promote such values to f32 for the usual set of operations.

As a result of this, the fexup[lr].w intrinsics no longer crash LLVM during
type legalization.

Reviewers: zoran.jovanvoic, vkalintiris

Differential Revision: https://reviews.llvm.org/D26398

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287349 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX512] Split AVX512F/AVX512VL tests to demonstrate missed int2fp opportunities without AVX512VL

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287348 91177308-0d34-0410-b5e6-96231b3b80d8

GlobalISel: Fix unconditional fallback with global isel abort is disabled

Reviewers: t.p.northover, ab, qcolombet

Subscribers: mehdi_amini, vkalintiris, wdng, dberris, llvm-commits, rovka

Differential Revision: https://reviews.llvm.org/D26765

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287344 91177308-0d34-0410-b5e6-96231b3b80d8

AMDGPU/SI: Remove zero_extend patterns for i16 ops selected to 32-bit insts

Summary:
The 32-bit instructions don't zero the high 16-bits like the 16-bit
instructions do.

Reviewers: arsenm

Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, llvm-commits, tony-tye

Differential Revision: https://reviews.llvm.org/D26828

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287342 91177308-0d34-0410-b5e6-96231b3b80d8

[simplifycfg][loop-simplify] Preserve loop metadata in 2 transformations.

insertUniqueBackedgeBlock in lib/Transforms/Utils/LoopSimplify.cpp now
propagates existing llvm.loop metadata to newly the added backedge.

llvm::TryToSimplifyUncondBranchFromEmptyBlock in lib/Transforms/Utils/Local.cpp
now propagates existing llvm.loop metadata to the branch instructions in the
predecessor blocks of the empty block that is removed.

Differential Revision: https://reviews.llvm.org/D26495

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287341 91177308-0d34-0410-b5e6-96231b3b80d8

Cleanup function with clang-format. NFCI.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287340 91177308-0d34-0410-b5e6-96231b3b80d8

AMDGPU: Fix legalization of MUBUF instructions in shaders

Summary:
The addr64-based legalization is incorrect for MUBUF instructions with idxen
set as well as for BUFFER_LOAD/STORE_FORMAT_* instructions.  This affects
e.g.  shaders that access buffer textures.

Since we never actually need the addr64-legalization in shaders, this patch
takes the easy route and keys off the calling convention.  If this ever
affects (non-OpenGL) compute, the type of legalization needs to be chosen
based on some TSFlag.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98664

Reviewers: arsenm, tstellarAMD

Subscribers: kzhuravl, wdng, yaxunl, tony-tye, llvm-commits

Differential Revision: https://reviews.llvm.org/D26747

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287339 91177308-0d34-0410-b5e6-96231b3b80d8

Fix spelling mistakes in MIPS target comments. NFC.

Identified by Pedro Giffuni in PR27636.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287338 91177308-0d34-0410-b5e6-96231b3b80d8

[Power9] Add patterns for vnegd, vnegw

Exploit new instructions by adding patterns to .td file.
https://reviews.llvm.org/D26551

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287334 91177308-0d34-0410-b5e6-96231b3b80d8

Fix spelling mistakes in AMDGPU target comments. NFC.

Identified by Pedro Giffuni in PR27636.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287333 91177308-0d34-0410-b5e6-96231b3b80d8

[X86][AVX2] Add v8i32->v8i64 mul test (PR30845)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287332 91177308-0d34-0410-b5e6-96231b3b80d8

Fix typo in comment. NFC.

Identified by Pedro Giffuni in PR27636.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287331 91177308-0d34-0410-b5e6-96231b3b80d8

[PPC][DAGCombine] Convert SETCC to subtract when the result is zero extended

When we see a SETCC whose only users are zero extend operations, we can replace
it with a subtraction. This results in doing all calculations in GPRs and
avoids CR use.

Currently we do this only for ULT, ULE, UGT and UGE condition codes. There are
ways that this can be extended. For example for signed condition codes. In that
case we will be introducing additional sign extend instructions, so more careful
profitability analysis may be required.

Another direction to extend this is for equal, not equal conditions. Also when
users of SETCC are any_ext or sign_ext, we might be able to do something
similar.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287329 91177308-0d34-0410-b5e6-96231b3b80d8

Fix go binding to adapt the new attribute API

https://reviews.llvm.org/D26339

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287328 91177308-0d34-0410-b5e6-96231b3b80d8

[InstCombine][AVX-512] Teach InstCombineCalls how to handle the intrinsics for variable shift with 16-bit elements.

This is a straightforward extension of the existing support for 32/64-bit element types. Just needed to add the additional instrinsics to the switches.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287316 91177308-0d34-0410-b5e6-96231b3b80d8

[AVX-512] Replace masked 16-bit element variable shift intrinsics with new unmasked versions and selects.

The same thing was done to 32-bit and 64-bit element sizes previously.

This will allow us to support these shuffls in InstCombineCalls along with the other variable shift intrinsics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@287312 91177308-0d34-0410-b5e6-96231b3b80d8