platform/upstream/llvm.git
4 years ago[LLVM/tools][test] - Remove/fix dead check prefixes.
Georgii Rymar [Tue, 14 Apr 2020 15:04:48 +0000 (18:04 +0300)]
[LLVM/tools][test] - Remove/fix dead check prefixes.

We have a few unused/broken FileCheck prefixes in `llvm/test/tools`.
This patch fixes it.

Differential revision: https://reviews.llvm.org/D78110

4 years ago[NFC] clang-format IntrinsicInst.h|cpp
Simon Moll [Wed, 15 Apr 2020 10:05:07 +0000 (12:05 +0200)]
[NFC] clang-format IntrinsicInst.h|cpp

Differential Revision: https://reviews.llvm.org/D78188

4 years ago[gn build] Port 8c11bc0cd06
LLVM GN Syncbot [Wed, 15 Apr 2020 09:38:24 +0000 (09:38 +0000)]
[gn build] Port 8c11bc0cd06

4 years agoIntroduce fix-irreducible pass
Sameer Sahasrabuddhe [Wed, 15 Apr 2020 09:35:51 +0000 (15:05 +0530)]
Introduce fix-irreducible pass

An irreducible SCC is one which has multiple "header" blocks, i.e., blocks
with control-flow edges incident from outside the SCC. This pass converts an
irreducible SCC into a natural loop by introducing a single new header
block and redirecting all the edges on the original headers to this
new block.

This is a useful workaround for a limitation in the structurizer
which, which produces incorrect control flow in the presence of
irreducible regions. The AMDGPU backend provides an option to
enable this pass before the structurizer, which may eventually be
enabled by default.

Reviewed By: nhaehnle

Differential Revision: https://reviews.llvm.org/D77198

This restores commit 2ada8e2525dd2653f30c8696a27162a3b1647d66.

Originally reverted with commit 44e09b59b869a91bf47d76e8bc569d9ee91ad145.

4 years ago[VPlan] Move Load/Store checks out of tryToWiden (NFC).
Florian Hahn [Wed, 15 Apr 2020 09:18:31 +0000 (10:18 +0100)]
[VPlan] Move Load/Store checks out of tryToWiden (NFC).

Handling LoadInst and StoreInst in tryToWiden seems a bit
counter-intuitive, as there is only an assertion for them and in no
case VPWidenRefipes are created for them.

I think it makes sense to move the assertion to handleReplication, where
the non-widened loads and store are handled.

Reviewers: gilr, rengolin, Ayal, hsaito

Reviewed By: Ayal

Differential Revision: https://reviews.llvm.org/D77972

4 years ago[mlir] LLVM dialect: support globals without linkage keyword, assuming 'external'
Alex Zinenko [Tue, 14 Apr 2020 11:01:53 +0000 (13:01 +0200)]
[mlir] LLVM dialect: support globals without linkage keyword, assuming 'external'

Similarly to actual LLVM IR, and to `llvm.mlir.func`, allow the custom syntax
of `llvm.mlir.global` to omit the linkage keyword. If omitted, the linkage is
assumed to be external. This makes the modeling of globals in the LLVM dialect
more consistent, both within the dialect and with LLVM IR.

Differential Revision: https://reviews.llvm.org/D78096

4 years ago[MLIR] Introduce applyOpPatternsAndFold for op local rewrites
Uday Bondhugula [Sun, 5 Apr 2020 02:40:33 +0000 (08:10 +0530)]
[MLIR] Introduce applyOpPatternsAndFold for op local rewrites

Introduce mlir::applyOpPatternsAndFold which applies patterns as well as
any folding only on a specified op (in contrast to
applyPatternsAndFoldGreedily which applies patterns only on the regions
of an op isolated from above).  The caller is made aware of the op being
folded away or erased.

Depends on D77485.

Differential Revision: https://reviews.llvm.org/D77487

4 years ago[nfc] [lldb] Introduce DWARF callbacks
Jan Kratochvil [Wed, 15 Apr 2020 08:36:31 +0000 (10:36 +0200)]
[nfc] [lldb] Introduce DWARF callbacks

As requested by @labath in https://reviews.llvm.org/D73206#1949516
providing DWARF index callbacks refactorization.

Differential Revision: https://reviews.llvm.org/D77327

4 years ago[NFC][test] Mark the section which contains instructions executable
Shengchen Kan [Wed, 15 Apr 2020 08:18:11 +0000 (16:18 +0800)]
[NFC][test] Mark the section which contains instructions executable

As we disscussed in D77971, we haven't confirmed that if putting instructions
in a non-executable section is an undefined behaviour. To make things
easier to go on, we mark these sections executable in test file
align-branch-section-size.s.

4 years ago[VE] Update integer arithmetic instructions
Kazushi (Jam) Marukawa [Wed, 15 Apr 2020 07:47:14 +0000 (09:47 +0200)]
[VE] Update integer arithmetic instructions

Summary:
Changing all mnemonic to match assembly instructions to simplify mnemonic
naming rules.  This time update all fixed-point arithmetic instructions.
This also corrects smax/smin code generations.

Reviewed By: simoll

Differential Revision: https://reviews.llvm.org/D77856

4 years ago[lldb][NFC] Remove FormatterChoiceCriterion
Raphael Isemann [Wed, 15 Apr 2020 05:34:14 +0000 (07:34 +0200)]
[lldb][NFC] Remove FormatterChoiceCriterion

Summary:
The formatters code has a lot of 'reason' or 'why' values that we keep or-ing FormatterChoiceCriterion
enum values into. These values are only read by a single log statement and don't have any functional
purpose. It also seems the implementation is not finished (for example, display names and type
names don't have any dedicated enum values). Also everything is of course not tested or documented.

Let's just remove all of this.

Reviewers: labath, JDevlieghere, jingham, davide, vsk

Reviewed By: labath, vsk

Subscribers: JDevlieghere

Differential Revision: https://reviews.llvm.org/D77968

4 years ago[LV] Fix PR45525: Incorrect assert in blend recipe
Gil Rapaport [Tue, 14 Apr 2020 14:20:41 +0000 (17:20 +0300)]
[LV] Fix PR45525: Incorrect assert in blend recipe

Fix an assert introduced in 41ed5d856c1: a phi with a single predecessor and a
mask is a valid case which is already supported by the code.

Differential Revision: https://reviews.llvm.org/D78115

4 years ago[clangd] Fix MSVC compile error, attempt 2
Kadir Cetinkaya [Wed, 15 Apr 2020 07:33:12 +0000 (09:33 +0200)]
[clangd] Fix MSVC compile error, attempt 2

4 years ago[clangd] Fix MSVC builds
Kadir Cetinkaya [Wed, 15 Apr 2020 07:24:58 +0000 (09:24 +0200)]
[clangd] Fix MSVC builds

4 years ago[clangd] Shard preamble symbols in dynamic index
Kadir Cetinkaya [Wed, 8 Apr 2020 14:14:53 +0000 (16:14 +0200)]
[clangd] Shard preamble symbols in dynamic index

Summary:
This reduces memory usage by dynamic index from more than 400MB to 32MB
when all files in clang-tools-extra/clangd/*.cpp are active in clangd.

Reviewers: sammccall

Subscribers: ilya-biryukov, MaskRay, jkorous, arphaman, usaxena95, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D77732

4 years agoDon't print `&` as part of reference template arguments.
Richard Smith [Wed, 15 Apr 2020 06:35:35 +0000 (23:35 -0700)]
Don't print `&` as part of reference template arguments.

In passing, also generalize the mechanism used to allow Decl's printName
functions to override qualified name printing.

4 years agoGeneralize our two different kinds of declaration argument for
Richard Smith [Wed, 15 Apr 2020 06:28:33 +0000 (23:28 -0700)]
Generalize our two different kinds of declaration argument for
attributes to support any kind of declaration.

In preparation for adding a third kind.

4 years ago[gn build] Port 44e09b59b86
LLVM GN Syncbot [Wed, 15 Apr 2020 06:53:11 +0000 (06:53 +0000)]
[gn build] Port 44e09b59b86

4 years agoRevert "Introduce fix-irreducible pass"
Sameer Sahasrabuddhe [Wed, 15 Apr 2020 06:47:11 +0000 (12:17 +0530)]
Revert "Introduce fix-irreducible pass"

This reverts commit 2ada8e2525dd2653f30c8696a27162a3b1647d66.

Buildbots produced compilation errors which I was not able to quickly
reproduce locally. Need more time to investigate.

4 years agoRemove setters from CXXTypeidExpr and CXXUuidofExpr.
Richard Smith [Wed, 15 Apr 2020 02:26:31 +0000 (19:26 -0700)]
Remove setters from CXXTypeidExpr and CXXUuidofExpr.

We generally prefer to have the ASTReader directly set fields rather
than including public setter functions.

4 years ago[gn build] Port 2ada8e2525d
LLVM GN Syncbot [Wed, 15 Apr 2020 06:01:21 +0000 (06:01 +0000)]
[gn build] Port 2ada8e2525d

4 years agoIntroduce fix-irreducible pass
Sameer Sahasrabuddhe [Wed, 15 Apr 2020 05:59:19 +0000 (11:29 +0530)]
Introduce fix-irreducible pass

An irreducible SCC is one which has multiple "header" blocks, i.e., blocks
with control-flow edges incident from outside the SCC. This pass converts an
irreducible SCC into a natural loop by introducing a single new header
block and redirecting all the edges on the original headers to this
new block.

This is a useful workaround for a limitation in the structurizer
which, which produces incorrect control flow in the presence of
irreducible regions. The AMDGPU backend provides an option to
enable this pass before the structurizer, which may eventually be
enabled by default.

Reviewed By: nhaehnle

Differential Revision: https://reviews.llvm.org/D77198

4 years ago[clangd] Color dependent names based on their heuristic target if they have one
Nathan Ridge [Thu, 19 Mar 2020 19:42:10 +0000 (15:42 -0400)]
[clangd] Color dependent names based on their heuristic target if they have one

Summary: Fixes https://github.com/clangd/clangd/issues/297

Subscribers: ilya-biryukov, jkorous, arphaman, kadircet, usaxena95, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D76896

4 years ago[llvm][StringExtras] Add missing include of cctype
River Riddle [Wed, 15 Apr 2020 02:29:54 +0000 (19:29 -0700)]
[llvm][StringExtras] Add missing include of cctype

This fixes build breakages on windows.

4 years ago[NFC][DAGCombine] Change the value of NegatibleCost to make it align with the semantics
QingShan Zhang [Wed, 15 Apr 2020 02:17:36 +0000 (02:17 +0000)]
[NFC][DAGCombine] Change the value of NegatibleCost to make it align with the semantics

This is a minor NFC change to make the code more clear. We have the NegatibleCost that
has cheaper, neutral, and expensive. Typically, the smaller one means the less cost.
It is inverse for current implementation, which makes following code not easy to read.
If (CostX > CostY) negate(X)

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D77993

4 years agoAMDGPU/GlobalISel: Fix selection of scalar f64 G_FABS
Matt Arsenault [Tue, 7 Apr 2020 15:33:26 +0000 (11:33 -0400)]
AMDGPU/GlobalISel: Fix selection of scalar f64 G_FABS

This wasn't covered by existing tablegen patterns, but also suffers
the same issues as G_FNEG. Workaround them by manually selecting, like
G_FNEG.

4 years agoTableGen/GlobalISel: Fix constraining REG_SEQUENCE operands
Matt Arsenault [Tue, 7 Apr 2020 13:32:51 +0000 (09:32 -0400)]
TableGen/GlobalISel: Fix constraining REG_SEQUENCE operands

This was hitting the default instruction constraint code which uses
the register classes in the instruction def, which REG_SEQUENCE does
not have.

Fixes not constraining the register class for AMDGPU fneg/fabs
patterns, which would fail when the use was another generic,
unconstrained instruction.

Another oddity I noticed is that the temporary registers are created
with an unnecessary, but incorrect 16-bit LLT but this shouldn't
matter.

I'm also still unclear why root and sub-instructions have to be
handled differently.

4 years ago[llvm][StringExtras] Merge StringExtras from MLIR into LLVM
River Riddle [Wed, 15 Apr 2020 01:54:23 +0000 (18:54 -0700)]
[llvm][StringExtras] Merge StringExtras from MLIR into LLVM

Summary:
This revision adds two utilities currently present in MLIR to LLVM StringExtras:

* convertToSnakeFromCamelCase
Convert a string from a camel case naming scheme, to a snake case scheme

* convertToCamelFromSnakeCase
Convert a string from a snake case naming scheme, to a camel case scheme

Differential Revision: https://reviews.llvm.org/D78167

4 years ago[MS] Fix packed struct layout for arrays of aligned non-record types
Reid Kleckner [Wed, 8 Apr 2020 21:05:54 +0000 (14:05 -0700)]
[MS] Fix packed struct layout for arrays of aligned non-record types

In particular, this affects Clang's vectors. Users encounter this issue
when a struct contains an __m128 type.

Fixes PR45420

Reviewed By: rjmccall

Differential Revision: https://reviews.llvm.org/D77754

4 years agoAllow disabling of vectorization using internal options
Teresa Johnson [Mon, 13 Apr 2020 02:12:38 +0000 (19:12 -0700)]
Allow disabling of vectorization using internal options

Summary:
Currently, the internal options -vectorize-loops, -vectorize-slp, and
-interleave-loops do not have much practical effect. This is because
they are used to initialize the corresponding flags in the pass
managers, and those flags are then unconditionally overwritten when
compiling via clang or via LTO from the linkers. The only exception was
-vectorize-loops via opt because of some special hackery there.

While vectorization could still be disabled when compiling via clang,
using -fno-[slp-]vectorize, this meant that there was no way to disable
it when compiling in LTO mode via the linkers. This only affected
ThinLTO, since for regular LTO vectorization is done during the compile
step for scalability reasons. For ThinLTO it is invoked in the LTO
backends. See also the discussion on PR45434.

This patch makes it so the internal options can actually be used to
disable these optimizations. Ultimately, the best long term solution is
to mark the loops with metadata (similar to the approach used to fix
-fno-unroll-loops in D77058), but this enables a shorter term
workaround, and actually makes these internal options useful.

I constant propagated the initial values of these internal flags into
the pass manager flags (for some reasons vectorize-loops and
interleave-loops were initialized to true, while vectorize-slp was
initialized to false). As mentioned above, they are overwritten
unconditionally so this doesn't have any real impact, and these initial
values aren't particularly meaningful.

I then changed the passes to check the internl values and return without
performing the associated optimization when false (I changed the default
of -vectorize-slp to true so the options behave similarly). I was able
to remove the hackery in opt used to get -vectorize-loops=false to work,
as well as a special option there used to disable SLP vectorization.

Finally, I changed thinlto-slp-vectorize-pm.c to:
a) Only test SLP (moved the loop vectorization checking to a new test).
b) Use code that is slp vectorized when it is enabled, and check that
instead of whether the pass is enabled.
c) Test the new behavior of -vectorize-slp.
d) Test both pass managers.

The loop vectorization (and associated interleaving) testing I moved to
a new thinlto-loop-vectorize-pm.c test, with several changes:
a) Changed the flags on the interleaving testing so that it will
actually interleave, and check that.
b) Test the new behavior of -vectorize-loops and -interleave-loops.
c) Test both pass managers.

Reviewers: fhahn, wmi

Subscribers: hiraditya, steven_wu, dexonsmith, cfe-commits, davezarzycki, llvm-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D77989

4 years ago[test][NFC] Use plain FileCheck in statepoint-stackmap-size.ll
Hubert Tong [Wed, 15 Apr 2020 00:51:50 +0000 (20:51 -0400)]
[test][NFC] Use plain FileCheck in statepoint-stackmap-size.ll

Summary:
The test in question uses a non-portable `grep -A` option in conjunction
with `wc -l`. `FileCheck` can be used to do the check without using
these extra utilities.

Reviewed By: thakis

Differential Revision: https://reviews.llvm.org/D78060

4 years ago[AMDGPU] Update DWARF proposal
Tony [Thu, 20 Feb 2020 07:08:59 +0000 (02:08 -0500)]
[AMDGPU] Update DWARF proposal

- Unify the sections on DWARF expression and location lists.

- Allow a location description to have one or more single location
  descriptions.

- Define context of DWARF expression that includes an initial
  stack. Allow initial stack to be used when evaluating location list
  expression with overlapping PC ranges.

- Reorganize the DWARF proposal in AMDGPUUsage so suitable for
  submission to the DWARF site.

- Replace CFI instruction DW_CFA_LLVM_def_cfa_aspace with
  DW_CFA_def_aspace_cfa and DW_CFA_def_aspace_cfa_sf. This is to avoid
  the problem that DW_CFA_def_cfa and DW_CFA_def_cfa_sf cannot use a
  register that is not the size of an address in the CFA address
  space.

- Clarify DWARF address class and DWARF address space. Define language
  values for DWARF address classes and specify how they are used by
  some common source languages.

- Define rules for accessing registers and derefencing memory when the
  type size and register size or byte size operand do not match.

- Numerous cleanups for consistency.

Differential Revision: https://reviews.llvm.org/D70523

4 years agoBug where insn-based unwind plans on arm64 could be wrong
Jason Molenda [Tue, 14 Apr 2020 23:51:58 +0000 (16:51 -0700)]
Bug where insn-based unwind plans on arm64 could be wrong

Fix a bug where UnwindAssemblyInstEmulation would confuse which
register is used to compute the Canonical Frame Address after it
had branched over a mid-function epilogue (where the CFA reg changes
from $fp to $sp in the process of epiloguing).  Reinstate the
correct CFA register after we forward the unwind rule for branch
targets.  The failure mode was that UnwindAssemblyInstEmulation
would think CFA was set in terms of $sp after one of these epilogues,
and if it sees modifications to $sp after the branch target, it would
change the CFA offset in the unwind rule -- even though the CFA is
defined in terms of $fp and the $sp changes are irrelevant to correct
calculation.

<rdar://problem/60300528>

Differential Revision: https://reviews.llvm.org/D78077

4 years ago[mlir] Fix assert on signed integer type in EDSC
Adam Straw [Tue, 14 Apr 2020 19:49:18 +0000 (22:49 +0300)]
[mlir] Fix assert on signed integer type in EDSC

Integer type in Std dialect is signless so we should be checking
for signless integer type instead of signed integer type in EDSC.

Differential Revision: https://reviews.llvm.org/D78144

4 years agoInitiallize property getters as being property accessors (NFC)
Adrian Prantl [Tue, 14 Apr 2020 22:55:00 +0000 (15:55 -0700)]
Initiallize property getters as being property accessors (NFC)

This is a no-op because it is set later on unconditionally again, but
it's far less confusing this way and consistent with how the setters
are initialized.

4 years agollvm-dwarfdump: Don't try to parse a debug_loclist contribution if this CU has no...
David Blaikie [Tue, 14 Apr 2020 22:17:24 +0000 (15:17 -0700)]
llvm-dwarfdump: Don't try to parse a debug_loclist contribution if this CU has no DW_AT_loclists_base

llvm-dwarfdump was trying to parse debug_loclists even in the absence of
a loclists_base if there was a loclists section at all.

4 years ago[www] Turn 'Clang 10' boxes green in C++ status pages to reflect release
Hubert Tong [Tue, 14 Apr 2020 22:36:50 +0000 (18:36 -0400)]
[www] Turn 'Clang 10' boxes green in C++ status pages to reflect release

Summary:
The 'Clang 10' boxes should be green since Clang 10 has been released.

Reviewers: rsmith, aaron.ballman

Reviewed By: aaron.ballman

Differential Revision: https://reviews.llvm.org/D78068

4 years ago[gn build] Port ebf190fcda0
LLVM GN Syncbot [Tue, 14 Apr 2020 22:33:30 +0000 (22:33 +0000)]
[gn build] Port ebf190fcda0

4 years ago[gn build] Port 8cbe371c28a
LLVM GN Syncbot [Tue, 14 Apr 2020 22:33:30 +0000 (22:33 +0000)]
[gn build] Port 8cbe371c28a

4 years ago[gn build] Port 204c3b55162
LLVM GN Syncbot [Tue, 14 Apr 2020 22:33:29 +0000 (22:33 +0000)]
[gn build] Port 204c3b55162

4 years ago[SelectionDAG] Always preserve offset in MachinePointerInfo
Eli Friedman [Tue, 7 Apr 2020 21:05:29 +0000 (14:05 -0700)]
[SelectionDAG] Always preserve offset in MachinePointerInfo

Previously, getWithOffset() would drop the offset if the base was null.
Because of this, MachineMemOperand would return the wrong result from
getAlign() in these cases.  MachineMemOperand stores the alignment of
the pointer without the offset.

A bunch of MIR tests changed because we print the offset now.

Split off from D77687.

Differential Revision: https://reviews.llvm.org/D78049

4 years ago[llvm-objdump][Wasm][NFC] Create WasmDump.h
Hubert Tong [Tue, 14 Apr 2020 22:24:37 +0000 (18:24 -0400)]
[llvm-objdump][Wasm][NFC] Create WasmDump.h

Summary:
Continuing from D77285, the external interfaces implemented by
`WasmDump.cpp` are now declared in `WasmDump.h` and moved into the
`llvm::objdump` namespace.

Reviewers: jhenderson, MaskRay, DiggerLin, jasonliu, daltenty

Reviewed By: jhenderson, MaskRay

Differential Revision: https://reviews.llvm.org/D77990

4 years ago[AVR] Define __ELF__
Ayke van Laethem [Tue, 14 Apr 2020 15:30:24 +0000 (17:30 +0200)]
[AVR] Define __ELF__

This symbol is defined in avr-gcc. Because AVR normally uses the ELF
format, define the symbol unconditionally.

This patch is needed to get Clang to compile compiler-rt.

Differential Revision: https://reviews.llvm.org/D78117

4 years ago[mlir][NFC] Remove the STLExtras.h header file now that it has been merged into LLVM.
River Riddle [Tue, 14 Apr 2020 21:54:15 +0000 (14:54 -0700)]
[mlir][NFC] Remove the STLExtras.h header file now that it has been merged into LLVM.

Now that no more utilities exist within, this file can be deleted.

Differential Revision: https://reviews.llvm.org/D78079

4 years ago[llvm][ADT] Move TypeSwitch class from MLIR to LLVM
River Riddle [Tue, 14 Apr 2020 21:53:50 +0000 (14:53 -0700)]
[llvm][ADT] Move TypeSwitch class from MLIR to LLVM

This class implements a switch-like dispatch statement for a value of 'T' using dyn_cast functionality. Each `Case<T>` takes a callable to be invoked if the root value isa<T>, the callable is invoked with the result of dyn_cast<T>() as a parameter.

Differential Revision: https://reviews.llvm.org/D78070

4 years ago[llvm][STLExtras] Move the algorithm `interleave*` methods from MLIR to LLVM
River Riddle [Tue, 14 Apr 2020 21:53:28 +0000 (14:53 -0700)]
[llvm][STLExtras] Move the algorithm `interleave*` methods from MLIR to LLVM

These have proved incredibly useful for interleaving values between a range w.r.t to streams. After this revision, the mlir/Support/STLExtras.h is empty. A followup revision will remove it from the tree.

Differential Revision: https://reviews.llvm.org/D78067

4 years ago[llvm][STLExtras] Move various iterator/range utilities from MLIR to LLVM
River Riddle [Tue, 14 Apr 2020 21:53:07 +0000 (14:53 -0700)]
[llvm][STLExtras] Move various iterator/range utilities from MLIR to LLVM

This revision moves the various range utilities present in MLIR to LLVM to enable greater reuse. This revision moves the following utilities:

* indexed_accessor_*
This is set of utility iterator/range base classes that allow for building a range class where the iterators are represented by an object+index pair.

* make_second_range
Given a range of pairs, returns a range iterating over the `second` elements.

* hasSingleElement
Returns if the given range has 1 element. size() == 1 checks end up being very common, but size() is not always O(1) (e.g., ilist). This method provides O(1) checks for those cases.

Differential Revision: https://reviews.llvm.org/D78064

4 years ago[llvm][STLExtras] Add various type_trait utilities currently present in MLIR
River Riddle [Tue, 14 Apr 2020 21:52:52 +0000 (14:52 -0700)]
[llvm][STLExtras] Add various type_trait utilities currently present in MLIR

This revision moves several type_trait utilities from MLIR into LLVM. Namely, this revision adds:
is_detected - This matches the experimental std::is_detected
is_invocable - This matches the c++17 std::is_invocable
function_traits - A utility traits class for getting the argument and result types of a callable type

Differential Revision: https://reviews.llvm.org/D78059

4 years ago[llvm][DenseMapInfo] Add an info specialization for std::tuple
River Riddle [Tue, 14 Apr 2020 21:52:42 +0000 (14:52 -0700)]
[llvm][DenseMapInfo] Add an info specialization for std::tuple

This revision adds a DenseMapInfo overload for std::tuples whose elements all have a DenseMapInfo. The implementation is similar to that of std::pair, and has been used within MLIR for over a year.

Differential Revision: https://reviews.llvm.org/D78057

4 years agoUse index for shape.to_extent_tensor
Sean Silva [Tue, 14 Apr 2020 20:30:44 +0000 (13:30 -0700)]
Use index for shape.to_extent_tensor

Summary:
Also,
- add IndexTensor to OpBase.td
- fix typo in the op name. It was mistakenly `to_tensor` instead of
  `to_extent_tensor`.

Differential Revision: https://reviews.llvm.org/D78149

4 years agoEnable new passmanager plugin support for LTO.
Eli Friedman [Thu, 26 Mar 2020 17:09:13 +0000 (10:09 -0700)]
Enable new passmanager plugin support for LTO.

This should make both static and dynamic NewPM plugins work with LTO.
And as a bonus, it makes static linking of OldPM plugins more reliable
for plugins with both an OldPM and NewPM interface.

I only implemented the command-line flag to specify NewPM plugins in
llvm-lto2, to show it works. Support can be added for other tools later.

Differential Revision: https://reviews.llvm.org/D76866

4 years ago[lldb/Test] s/lldb-repo/lldb-repro/ (NFC)
Jonas Devlieghere [Tue, 14 Apr 2020 21:51:35 +0000 (14:51 -0700)]
[lldb/Test] s/lldb-repo/lldb-repro/ (NFC)

4 years ago[llvm][NFC][CallSite] Remove Implementation uses of CallSite
Mircea Trofin [Tue, 14 Apr 2020 19:33:49 +0000 (12:33 -0700)]
[llvm][NFC][CallSite] Remove Implementation uses of CallSite

Reviewers: dblaikie, davidxl, craig.topper

Subscribers: arsenm, dschuff, nemanjai, jvesely, nhaehnle, sbc100, jgravelle-google, hiraditya, aheejin, kbarton, kerbowa, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D78142

4 years ago[lldb/Test] Use RAII for reproducer instrumentation data in unittest.
Jonas Devlieghere [Mon, 13 Apr 2020 17:23:35 +0000 (10:23 -0700)]
[lldb/Test] Use RAII for reproducer instrumentation data in unittest.

Use a RAII object to manage the lifetime of the reproducer
instrumentation data.

4 years agoReapply "llvm-dwarfdump: Report errors when failing to parse loclist/debug_loc entries""
David Blaikie [Tue, 14 Apr 2020 21:21:33 +0000 (14:21 -0700)]
Reapply "llvm-dwarfdump: Report errors when failing to parse loclist/debug_loc entries""

Originally committed as 416fa7720e30750939c53935051c6c750dfad2c2
Reverted (due to buildbot failure - breaking lldb) in 7a45aeacf3a23449039ef2efcf476995ae1c7007.

I still can't seem to build lldb locally, but Pavel Labath has kindly
provided a potential fix to preserve the old behavior in lldb by
registering a simple recoverable error handler there that prints to the
desired stream in lldb, rather than stderr.

4 years ago[mlir][Linalg] Handle null affine map returns from inversePermutation.
MaheshRavishankar [Tue, 14 Apr 2020 18:54:15 +0000 (11:54 -0700)]
[mlir][Linalg] Handle null affine map returns from inversePermutation.

The inversePermutation method returns a null map on failure. Update
uses of this method within Linalg to handle this. In LinalgToLoops the
null return value was used to emit scalar code. Modify that to return
failure, and emit scalar implementation when affine map is "empty",
i.e. 1 dims, 0 symbols and no result exprs.

Differential Revision: https://reviews.llvm.org/D77964

4 years ago[CodeGen] clarify a comment; NFC
George Burgess IV [Tue, 14 Apr 2020 21:29:29 +0000 (14:29 -0700)]
[CodeGen] clarify a comment; NFC

Prompted by discussion on https://reviews.llvm.org/D78148.

4 years ago[SVE] Remove calls to getBitWidth from Transforms
Christopher Tetreault [Tue, 14 Apr 2020 20:46:27 +0000 (13:46 -0700)]
[SVE] Remove calls to getBitWidth from Transforms

Reviewers: efriedma, sdesmalen, spatel, eugenis, chandlerc

Reviewed By: efriedma

Subscribers: tschuett, hiraditya, rkruppe, psnobl, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D77896

4 years ago[ValueTracking] Implement canCreatePoison
Juneyoung Lee [Tue, 14 Apr 2020 01:28:45 +0000 (10:28 +0900)]
[ValueTracking] Implement canCreatePoison

Summary:
This PR adds `canCreatePoison(Instruction *I)` which returns true if `I` can generate poison from non-poison
operands.

Reviewers: spatel, nikic, lebedev.ri

Reviewed By: spatel

Subscribers: hiraditya, llvm-commits, regehr, nlopes

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D77890

4 years ago[lld][WebAssembly] Do not require --shared-memory with --relocatable
Thomas Lively [Tue, 14 Apr 2020 02:16:20 +0000 (19:16 -0700)]
[lld][WebAssembly] Do not require --shared-memory with --relocatable

Summary:
wasm-ld requires --shared-memory to be passed when the atomics feature
is enabled because historically atomic operations were only valid with
shared memories. This change relaxes that requirement for when
building relocatable objects because their memories are not
meaningful. This technically maintains the validity of object files
because the threads spec now allows atomic operations with unshared
memories, although we don't support that elsewhere in the tools yet.

This fixes and Emscripten build issue reported at
https://bugs.chromium.org/p/webp/issues/detail?id=463.

Reviewers: sbc100

Subscribers: dschuff, jgravelle-google, aheejin, sunfish, jfb, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D78072

4 years ago[SVE] Remove calls to getBitWidth from IR
Christopher Tetreault [Tue, 14 Apr 2020 20:30:18 +0000 (13:30 -0700)]
[SVE] Remove calls to getBitWidth from IR

Reviewers: efriedma, sdesmalen, RKSimon, majnemer

Reviewed By: majnemer

Subscribers: tschuett, hiraditya, rkruppe, psnobl, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D77897

4 years agoEmit Objective-C constructors as writable
Joerg Sonnenberger [Tue, 14 Apr 2020 13:13:13 +0000 (15:13 +0200)]
Emit Objective-C constructors as writable

They end up as .init_array sections and those need to be writable,
otherwise bad merging will happen.

4 years agoNFC: update of ARM llvm regr test, follow up of 9633fc14aef7ee3da3b.
Sjoerd Meijer [Tue, 14 Apr 2020 20:04:24 +0000 (21:04 +0100)]
NFC: update of ARM llvm regr test, follow up of 9633fc14aef7ee3da3b.

4 years ago[SVE] Remove calls to getBitWidth from clang
Christopher Tetreault [Tue, 14 Apr 2020 19:55:48 +0000 (12:55 -0700)]
[SVE] Remove calls to getBitWidth from clang

Reviewers: efriedma

Reviewed By: efriedma

Subscribers: tschuett, rkruppe, psnobl, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D77903

4 years ago[WebAssembly] Emit .llvmcmd and .llvmbc as custom sections
Sam Clegg [Tue, 31 Mar 2020 00:37:01 +0000 (17:37 -0700)]
[WebAssembly] Emit .llvmcmd and .llvmbc as custom sections

Fixes: https://bugs.llvm.org/show_bug.cgi?id=45362

Differential Revision: https://reviews.llvm.org/D77115

4 years ago[libc++] run.py: Create directory hierarchy if it does not exist
Louis Dionne [Tue, 14 Apr 2020 20:20:00 +0000 (16:20 -0400)]
[libc++] run.py: Create directory hierarchy if it does not exist

This is useful when replaying tests after the fact and the directories
don't exist anymore.

4 years ago[InstCombine][SVE] Fix visitGetElementPtrInst for scalable type.
Huihui Zhang [Tue, 14 Apr 2020 19:38:03 +0000 (12:38 -0700)]
[InstCombine][SVE] Fix visitGetElementPtrInst for scalable type.

Summary:
This patch fix the following issues in InstCombiner::visitGetElementPtrInst

    1. Skip for scalable type if transformation requires fixed size number of
    vector element.
    2. Skip for scalable type if transformation relies on compile-time known type
    alloc size.
    3. Use VectorType::getElementCount when scalable property is used to construct
    new VectorType.
    4. Use TypeSize::getKnownMinSize when minimal size of a scalable type is valid to determine GEP 'inbounds'.
    5. Explicitly call TypeSize::getFixedSize to avoid implicit type conversion to uint64_t.

Reviewers: sdesmalen, efriedma, spatel, ctetreau

Reviewed By: efriedma

Subscribers: tschuett, hiraditya, rkruppe, psnobl, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D78081

4 years ago[lldb] Fix a typo in a test name
Walter Erquinigo [Fri, 10 Apr 2020 15:23:42 +0000 (08:23 -0700)]
[lldb] Fix a typo in a test name

Summary:
Removing the Test prefix from the file name and its usages. The standard is using only Test as a suffix.
This was correctly pointed out in https://reviews.llvm.org/D77444.

Reviewers: labath, clayborg

Subscribers: mgorny, lldb-commits

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D77878

4 years ago[mlir][Linalg] NFC: Add missing comments from earlier commit (D77965)
MaheshRavishankar [Tue, 14 Apr 2020 19:13:27 +0000 (12:13 -0700)]
[mlir][Linalg] NFC: Add missing comments from earlier commit (D77965)

Differential Revision: https://reviews.llvm.org/D78140

4 years ago[libc++] CI: Allow pointing to specific dylib roots during backdeployment testing
Louis Dionne [Tue, 14 Apr 2020 19:07:38 +0000 (15:07 -0400)]
[libc++] CI: Allow pointing to specific dylib roots during backdeployment testing

This can be useful when the CI jobs don't have internet access.

4 years ago[InstCombine] try to reduce more shuffles with bitcasted operand
Sanjay Patel [Tue, 14 Apr 2020 18:41:35 +0000 (14:41 -0400)]
[InstCombine] try to reduce more shuffles with bitcasted operand

This is the widen mask element sibling to D76844.

shuf (bitcast X), undef, Mask --> bitcast X'

http://volta.cs.utah.edu:8080/z/4dt3V8

4 years ago[libc++] Add .verify.cpp tests to the new format
Louis Dionne [Tue, 14 Apr 2020 16:51:39 +0000 (12:51 -0400)]
[libc++] Add .verify.cpp tests to the new format

With this patch, .verify.cpp tests explicitly require clang-verify, but
no other test types require clang-verify out of the box. This will allow
making several .fail.cpp tests that don't have any clang-verify markup
to be just .compile.fail.cpp tests, which in turn should allow removing
a long standing workaround that requires parsing tests to detect whether
they have any clang-verify markup in them.

4 years ago[AntidepBreaker] Move AntiDepBreaker to include folder.
Thomas Raoux [Mon, 13 Apr 2020 19:26:44 +0000 (12:26 -0700)]
[AntidepBreaker] Move AntiDepBreaker to include folder.

This allows AntiDepBreaker to be used in target specific postRA
scheduler.

Differential Revision: https://reviews.llvm.org/D78047

4 years ago[clang] Add missing FileCheck colons
Jon Roelofs [Tue, 14 Apr 2020 16:46:21 +0000 (10:46 -0600)]
[clang] Add missing FileCheck colons

4 years ago[OpenMP] Sync writes to child thread's data before reduction
Bryan Chan [Mon, 6 Apr 2020 22:43:35 +0000 (18:43 -0400)]
[OpenMP] Sync writes to child thread's data before reduction

On systems with weak memory consistency, this patch fixes an intermittent crash
in the reduction function called by __kmp_hyper_barrier_gather, which suffers
from a race on a child thread's data.

Reviewed-By: AndreyChurbanov
Differential Revision: https://reviews.llvm.org/D77603

4 years ago[scudo][standalone] Split logs on Android
Kostya Kortchinsky [Mon, 13 Apr 2020 14:25:35 +0000 (07:25 -0700)]
[scudo][standalone] Split logs on Android

Summary:
The function used to log on Android will cut the message past
a certain amount of characters, which mostly materializes when
dumping the size class map on OOM.

This change splits the log message at newline boundaries.

Reviewers: pcc, cferris, hctim, eugenis

Subscribers: #sanitizers, llvm-commits

Tags: #sanitizers

Differential Revision: https://reviews.llvm.org/D78018

4 years ago[DWARF5]: Added support for DW_MACRO_import form in llvm-dwarfdump
Sourabh Singh Tomar [Sun, 12 Apr 2020 20:05:05 +0000 (01:35 +0530)]
[DWARF5]: Added support for DW_MACRO_import form in llvm-dwarfdump

GCC emits this new form along with others forms(supported in llvm-dwardump)
and since it's support was missing in llvm-dwarfdump, it was not
able to correctly dump the content a debug_macro section for GCC
generated binaries.

This patch extends llvm-dwarfdump to support this form,
now GCC generated debug_macro section can be correctly dumped
using llvm-dwarfdump.

Reviewed By: dblaikie

Differential Revision: https://reviews.llvm.org/D78006

4 years ago[Fixed Point] [AST] Add an AST serialization code for fixed-point literals.
Vince Bridgers [Tue, 7 Apr 2020 19:46:08 +0000 (14:46 -0500)]
[Fixed Point] [AST] Add an AST serialization code for fixed-point literals.

Summary:
This patch adds the EXPR_FIXEDPOINT_LITERAL AST
code to serialize FixedPointLiterals. They were previously
being serialized with the code for integer literals, which
doesn't work properly.

Reviewers: leonardchan, rjmccall

Reviewed By: leonardchan, rjmccall

Subscribers: vabridgers, JesperAntonsson, bjope, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D57226

4 years ago[SVE] Remove calls to getBitWidth from Hexagon
Christopher Tetreault [Tue, 14 Apr 2020 17:58:39 +0000 (10:58 -0700)]
[SVE] Remove calls to getBitWidth from Hexagon

Reviewers: efriedma, sdesmalen, kparzysz

Reviewed By: kparzysz

Subscribers: tschuett, hiraditya, rkruppe, psnobl, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D77899

4 years ago[STLExtras] Make const the * operator for mapped_iterator.
Alina Sbirlea [Tue, 14 Apr 2020 01:51:20 +0000 (18:51 -0700)]
[STLExtras] Make const the * operator for mapped_iterator.

Summary:
The current non-const * operator shadows the const operator in
iterator_adaptor_base.

Reviewers: mehdi_amini, rriddle!, dblaikie, timshen

Subscribers: dexonsmith, jpienaar, burmako, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, Joonsoo, grosul1, frgossen, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D78073

4 years ago[InstCombine] add tests for bitcasted shuffle operand; NFC
Sanjay Patel [Tue, 14 Apr 2020 17:38:21 +0000 (13:38 -0400)]
[InstCombine] add tests for bitcasted shuffle operand; NFC

Similar to D76844, but this is casted from wider element type.
See D77881.

4 years ago[InstSimplify] add test for select that should not be simplified; NFC
Sanjay Patel [Mon, 13 Apr 2020 19:31:59 +0000 (15:31 -0400)]
[InstSimplify] add test for select that should not be simplified; NFC

See discussion in D77868

4 years ago[SVE] Remove calls to getBitWidth from ARM
Christopher Tetreault [Tue, 14 Apr 2020 17:34:35 +0000 (10:34 -0700)]
[SVE] Remove calls to getBitWidth from ARM

Reviewers: efriedma

Reviewed By: efriedma

Subscribers: tschuett, kristof.beyls, hiraditya, rkruppe, psnobl, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D77904

4 years agoFix setting Python3_ROOT_DIR on Windows
Isuru Fernando [Fri, 10 Apr 2020 00:44:56 +0000 (19:44 -0500)]
Fix setting Python3_ROOT_DIR on Windows

Summary:
Previously the value of Python3_ROOT_DIR was set to the string
"PYTHON_HOME" instead of the value of the variable named
PYTHON_HOME. This commit fixes that as CMake expects
a path as the value of Python3_ROOT_DIR

Reviewers: #lldb, JDevlieghere, teemperor

Reviewed By: #lldb, JDevlieghere, teemperor

Subscribers: teemperor, JDevlieghere, mgorny, lldb-commits

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D77842

4 years ago[X86] Add tests to clang Driver to ensure that SLH/Retpoline features are not enabled...
Scott Constable [Tue, 14 Apr 2020 17:32:12 +0000 (10:32 -0700)]
[X86] Add tests to clang Driver to ensure that SLH/Retpoline features are not enabled with LVI-CFI

Differential Revision: https://reviews.llvm.org/D77427

4 years ago[builtins] Make __umodsi3/__udivdi3/__umoddi3 standalone (shift and subtract)
Fangrui Song [Fri, 10 Apr 2020 22:14:52 +0000 (15:14 -0700)]
[builtins] Make __umodsi3/__udivdi3/__umoddi3 standalone (shift and subtract)

@kamleshbhalui reported that when the Standard Extension M
(Multiplication and Division) is disabled for RISC-V,
`__udivdi3` will call __udivmodti4 which will in turn calls `__udivdi3`.

This patch moves __udivsi3 (shift and subtract) to int_div_impl.inc
`__udivXi3`, optimize a bit, add a `__umodXi3`, and use `__udivXi3` and
`__umodXi3` to define `__udivsi3` `__umodsi3` `__udivdi3` `__umoddi3`.

Reviewed By: kamleshbhalui

Differential Revision: https://reviews.llvm.org/D77912

4 years ago[FileCheck] Add missing include in FileCheckImpl.h
Thomas Preud'homme [Tue, 14 Apr 2020 17:24:32 +0000 (18:24 +0100)]
[FileCheck] Add missing include in FileCheckImpl.h

FileCheckImpl.h internal header uses type defined in the public
FileCheck.h header but fails to include it. This commit fixes that.

Test Plan: Built it locally successfully.

4 years ago[SVE] Remove calls to getBitWidth from AArch64
Christopher Tetreault [Tue, 14 Apr 2020 17:06:23 +0000 (10:06 -0700)]
[SVE] Remove calls to getBitWidth from AArch64

Reviewers: efriedma

Reviewed By: efriedma

Subscribers: danielkiss, tschuett, kristof.beyls, hiraditya, rkruppe, psnobl, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D77905

4 years ago[lldb/Reproducers] Make static methods go through the invoke wrapper.
Jonas Devlieghere [Tue, 14 Apr 2020 17:18:50 +0000 (10:18 -0700)]
[lldb/Reproducers] Make static methods go through the invoke wrapper.

They don't actually have to, but it makes it easier to unify the
corresponding code and macros. NFC.

4 years ago[AArch64][SVE] Refine node definitions for ff & nf loads/stores (NFC)
Andrzej Warzynski [Wed, 8 Apr 2020 10:15:48 +0000 (11:15 +0100)]
[AArch64][SVE] Refine node definitions for ff & nf loads/stores (NFC)

Summary:
Only first-faulting and non-faulting loads read/update the FFR register
and hence only the corresponding SDNodes should be decorated with
`SDNPOptInGlue` and `SDNPOutGlue`. This patch:
  * removes SDNPOptInGlue from regular loads stores (FFR is not read)
  * adds SDNPOutGlue to first-faulting and non-faulting loads (FFR is
    both read and updated)

Differential Revision: https://reviews.llvm.org/D77724

4 years ago[flang] Upstream patch to enable the fir-ops test. bug 45487
Eric Schweitz [Tue, 14 Apr 2020 16:50:50 +0000 (09:50 -0700)]
[flang] Upstream patch to enable the fir-ops test. bug 45487

Differential revision: https://reviews.llvm.org/D77834

4 years ago[gn build] (manually) port c8a5b30bac6
Nico Weber [Tue, 14 Apr 2020 16:51:58 +0000 (12:51 -0400)]
[gn build] (manually) port c8a5b30bac6

4 years ago[lldb/Reproducers] Capture reproducers from the API test suite.
Jonas Devlieghere [Tue, 14 Apr 2020 15:41:50 +0000 (08:41 -0700)]
[lldb/Reproducers] Capture reproducers from the API test suite.

Make it possible to capture reproducers from the API test suite. Given
the symmetry between capture and replay, this patch also adds the
necessary code for replay. For now this is a NO-OP until the
corresponding reproducer instrumentation changes land.

For more info please refer to the RFC on lldb-dev:
http://lists.llvm.org/pipermail/lldb-dev/2020-April/016100.html

Differential revision: https://reviews.llvm.org/D77588

4 years ago[AbstractCallSite] Check that callback callee index is within call arguments
Sergey Dmitriev [Tue, 14 Apr 2020 16:12:34 +0000 (09:12 -0700)]
[AbstractCallSite] Check that callback callee index is within call arguments

Summary:
AbstractCallSite::getCallbackUses() does not check that callback callee index from
the callback metadata does not exceed the total number of call arguments. This patch
add such validation check.

Reviewers: jdoerfert, sstefan1

Reviewed By: jdoerfert

Subscribers: hiraditya, arphaman, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D78112

4 years ago[libc++] Remove support for .fail.mm tests
Louis Dionne [Tue, 14 Apr 2020 16:16:58 +0000 (12:16 -0400)]
[libc++] Remove support for .fail.mm tests

There are no such tests in the libc++ test suite, and I want to move
away from `.fail.cpp` tests (in favour of something else) too, which
require a workaround.

4 years ago[ADT] Mix the bit width into APInt's hash_value
Benjamin Kramer [Tue, 14 Apr 2020 16:13:49 +0000 (18:13 +0200)]
[ADT] Mix the bit width into APInt's hash_value

Otherwise all zero values from i1 to i64 collide.

4 years ago[SveEmitter] Add range checks for immediates and predicate patterns.
Sander de Smalen [Tue, 14 Apr 2020 14:56:36 +0000 (15:56 +0100)]
[SveEmitter] Add range checks for immediates and predicate patterns.

Summary:
This patch adds a mechanism to easily add range checks for a builtin's
immediate operands. This patch is tested with the qdech intrinsic, which takes
both an enum for the predicate pattern, as well as an immediate for the
multiplier.

Reviewers: efriedma, SjoerdMeijer, rovka

Reviewed By: efriedma, SjoerdMeijer

Subscribers: mgorny, tschuett, mgrang, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D76678

4 years ago[lldb/Docs] Document active and passive replay.
Jonas Devlieghere [Tue, 14 Apr 2020 15:39:46 +0000 (08:39 -0700)]
[lldb/Docs] Document active and passive replay.

Document reproducer replay and its two modes.

Differential revision: https://reviews.llvm.org/D77771

4 years ago[FunctionAttrs] Don't copy all the nodes where a reference is fine.
Benjamin Kramer [Tue, 14 Apr 2020 14:52:26 +0000 (16:52 +0200)]
[FunctionAttrs] Don't copy all the nodes where a reference is fine.

4 years ago[PowerPC][NFC] Remove spurious incorrect CHECKNEXT directive from test
Nemanja Ivanovic [Tue, 14 Apr 2020 15:11:13 +0000 (10:11 -0500)]
[PowerPC][NFC] Remove spurious incorrect CHECKNEXT directive from test

The directive was a typo when I first wrote the test case, then
decided to use the script and the script didn't remove the line
with the typo.