Nikita Popov [Tue, 26 Mar 2019 18:05:43 +0000 (18:05 +0000)]
[InstCombine] Add tests for ssubo X, C -> saddo X, -C; NFC
Add baseline tests for canonicalization of
ssubo X, C -> saddo X, -C.
Patch by Dan Robertson.
Differential Revision: https://reviews.llvm.org/D59653
llvm-svn: 357013
Sanjay Patel [Tue, 26 Mar 2019 17:50:08 +0000 (17:50 +0000)]
[InstCombine] form uaddsat from add+umin (PR14613)
This is the last step towards solving the examples shown in:
https://bugs.llvm.org/show_bug.cgi?id=14613
With this change, x86 should end up with psubus instructions
when those are available.
All known codegen issues with expanding the saturating intrinsics
were resolved with:
D59006 / rL356855
We also have some early evidence in D58872 that using the intrinsics
will lead to better perf. If some target regresses from this, custom
lowering of the intrinsics (as in the above for x86) may be needed.
llvm-svn: 357012
Heejin Ahn [Tue, 26 Mar 2019 17:46:14 +0000 (17:46 +0000)]
[WebAssembly] Add CFGStacikfied field to WebAssemblyFunctionInfo
Summary:
This adds `CFGStackified` field and its serialization to
WebAssemblyFunctionInfo.
Reviewers: dschuff
Subscribers: sunfish, sbc100, jgravelle-google, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59747
llvm-svn: 357011
Brian Gesiak [Tue, 26 Mar 2019 17:46:06 +0000 (17:46 +0000)]
[coroutines] Add std::experimental::task<T> type
Summary:
Adds the coroutine `std::experimental::task<T>` type described in proposal P1056R0.
See https://wg21.link/P1056R0.
This implementation allows customization of the allocator used to allocate the
coroutine frame by passing std::allocator_arg as the first argument, followed by
the allocator to use.
This supports co_awaiting the same task multiple times. The second and
subsequent times it returns a reference to the already-computed value.
This diff also adds some implementations of other utilities that have potential for
standardization as helpers within the test/... area:
- `sync_wait(awaitable)` - See P1171R0
- `manual_reset_event`
Move the definition of the __aligned_allocation_size helper function
from <experimental/memory_resource> to <experimental/__memory>
so it can be more widely used without pulling in memory_resource.
Outstanding work:
- Use C++14 keywords directly rather than macro versions
eg. use `noexcept` instead of `_NOEXCEPT`).
- Add support for overaligned coroutine frames.
This may need wording in the Coroutines TS to support passing the extra `std::align_val_t`.
- Eliminate use of `if constexpr` if we want it to compile under C++14.
Patch by @lewissbaker (Lewis Baker).
llvm-svn: 357010
Heejin Ahn [Tue, 26 Mar 2019 17:35:35 +0000 (17:35 +0000)]
[WebAssembly] Support WebAssemblyFunctionInfo serialization
Summary:
The framework for supporting target-specific MachineFunctionInfo was
added in r356215. This adds serialization support for
WebAssemblyFunctionInfo on top of that. This patch only adds the
framework and does not actually serialize anything at this point; we
have to add YAML mapping later for the fields in WebAssemblyFunctionInfo
we want to serialize if necessary.
Reviewers: dschuff, arsenm
Subscribers: sunfish, wdng, sbc100, jgravelle-google, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59737
llvm-svn: 357009
Heejin Ahn [Tue, 26 Mar 2019 17:29:55 +0000 (17:29 +0000)]
[WebAssembly] Fix a bug when mixing TRY/LOOP markers
Summary:
When TRY and LOOP markers are in the same BB and END_TRY and END_LOOP
markers are in the same BB, END_TRY should be _before_ END_LOOP, because
LOOP is always before TRY if they are in the same BB. (TRY is placed in
the latest possible position, whereas LOOP is in the earliest possible
position.)
Reviewers: dschuff
Subscribers: sunfish, sbc100, jgravelle-google, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59751
llvm-svn: 357008
Heejin Ahn [Tue, 26 Mar 2019 17:15:55 +0000 (17:15 +0000)]
[WebAssembly] Fix bugs in BLOCK/TRY placement
Summary:
Before we placed all TRY/END_TRY markers before placing BLOCK/END_BLOCK
markers. This couldn't handle this case:
```
bb0:
br bb2
bb1: // nearest common dominator of bb3 and bb4
br_if ... bb3
br bb4
bb2:
...
bb3:
call @foo // unwinds to ehpad
bb4:
call @bar // unwinds to ehpad
ehpad:
catch
...
```
When we placed TRY markers, we placed it in bb1 because it is the
nearest common dominator of bb3 and bb4. But because bb0 jumps to bb2,
when we placed block markers, we ended up with interleaved scopes like
```
block
try
end_block
catch
end_try
```
which was not correct.
This patch fixes the bug by placing BLOCK and TRY markers in one pass
while iterating BBs in a function. This also adds some more routines to
`placeTryMarkers`, because we now have to assume that there can be
previously placed BLOCK and END_BLOCK.
Reviewers: dschuff
Subscribers: sunfish, sbc100, jgravelle-google, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59739
llvm-svn: 357007
Davide Italiano [Tue, 26 Mar 2019 16:43:58 +0000 (16:43 +0000)]
[ScriptInterpreterPython] Try to make the sanitizer bot green again.
Removing a use-after-free error.
llvm-svn: 357006
Sanjay Patel [Tue, 26 Mar 2019 16:19:13 +0000 (16:19 +0000)]
[InstCombine] add tests for uaddsat using min; NFC
llvm-svn: 357005
Sanjay Patel [Tue, 26 Mar 2019 15:58:33 +0000 (15:58 +0000)]
[InstCombine] update tests to use FileCheck; NFC
llvm-svn: 357004
Clement Courbet [Tue, 26 Mar 2019 15:44:57 +0000 (15:44 +0000)]
[llvm-exegesis] Allow the target to disable the selection of some registers.
Summary:
This prevents "Cannot encode high byte register in REX-prefixed instruction"
from happening on instructions that require REX encoding when AH & co
get selected.
On the down side, these 4 registers can no longer be selected
automatically, but this avoids having to expose all the X86 encoding
complexity.
Reviewers: gchatelet
Subscribers: tschuett, jdoerfert, llvm-commits, bdb
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59821
llvm-svn: 357003
Luis Marques [Tue, 26 Mar 2019 15:41:45 +0000 (15:41 +0000)]
[RISCV] Update setcc-logic.ll codegen test
This should have been updated as part of D59753.
llvm-svn: 357002
Fangrui Song [Tue, 26 Mar 2019 15:39:45 +0000 (15:39 +0000)]
[CodeGen] Delete never used LValueAlign
It was added by rC176658 but never used since then.
llvm-svn: 357001
Andrea Di Biagio [Tue, 26 Mar 2019 15:38:37 +0000 (15:38 +0000)]
[MCA] Correctly update the UsedResourceGroups mask in the InstrBuilder.
Found by inspection when looking at the debug output of MCA.
This problem was latent, and none of the upstream models were affected by it.
No functional change intended.
llvm-svn: 357000
Ronald Wampler [Tue, 26 Mar 2019 15:26:15 +0000 (15:26 +0000)]
Test commit: fix typo
llvm-svn: 356999
Michael Kruse [Tue, 26 Mar 2019 15:19:15 +0000 (15:19 +0000)]
[libomptarget] Introduce LIBOMPTARGET_ENABLE_DEBUG cmake option.
At the moment, support for runtime debug output using the
OMPTARGET_DEBUG=1 environment variable is only available with
CMAKE_BUILD_TYPE=Debug builds. The patch allows setting it independently
using the LIBOMPTARGET_ENABLE_DEBUG option, which is enabled by default
depending on CMAKE_BUILD_TYPE. That is, unless this option is set
explicitly, nothing changes. This is the same mechanism used by LLVM for
LLVM_ENABLE_ASSERTIONS.
This patch also removes adding -g -O0 in debug builds, it should be
handled by cmake's CMAKE_{C|CXX}_FLAGS_DEBUG configuration option.
Idea by Hal Finkel
Differential Revision: https://reviews.llvm.org/D55952
llvm-svn: 356998
Jonas Paulsson [Tue, 26 Mar 2019 15:13:48 +0000 (15:13 +0000)]
[SystemZ] Remove LRMux pseudo instruction.
This instruction is unused and not needed.
Review: Ulrich Weigand.
llvm-svn: 356997
Nirav Dave [Tue, 26 Mar 2019 15:08:14 +0000 (15:08 +0000)]
[DAG] Avoid smart constructor-based dangling nodes.
Various SelectionDAG non-combine operations (e.g. the getNode smart
constructor and legalization) may leave dangling nodes by applying
optimizations or not fully pruning unused result values. This can
result in nodes that are never added to the worklist and therefore can
not be pruned.
Add a node inserter as the current node deleter to make sure such
nodes have the chance of being pruned.
Many minor changes, mostly positive.
llvm-svn: 356996
Serge Guelton [Tue, 26 Mar 2019 14:46:15 +0000 (14:46 +0000)]
python 2/3 compat: commands vs subprocess
Differential Revision: https://reviews.llvm.org/D59584
llvm-svn: 356995
Mikhail R. Gadelha [Tue, 26 Mar 2019 14:25:12 +0000 (14:25 +0000)]
Moved body of methods dump to .cpp file to fix compilation when modules
are enabled
llvm-svn: 356994
Pavel Labath [Tue, 26 Mar 2019 13:35:54 +0000 (13:35 +0000)]
Remove the TypePair class
Summary:
After D59297, the TypePair class kind of lost its purpose as it was no
longer a "pair". This finishes the job started in that patch and deletes
the class altogether. All usages have been updated to use CompilerType
class directly.
Reviewers: clayborg, jingham, zturner
Subscribers: mehdi_amini, dexonsmith, jdoerfert, lldb-commits
Differential Revision: https://reviews.llvm.org/D59414
llvm-svn: 356993
Pavel Labath [Tue, 26 Mar 2019 13:23:01 +0000 (13:23 +0000)]
Minidump: Use minidump types defined in llvm
This is the next step in moving the minidump parsing into llvm. I remove
the minidump structures already defined in the llvm Object library and
convert our parser to use those. NFC.
llvm-svn: 356992
Luis Marques [Tue, 26 Mar 2019 12:55:00 +0000 (12:55 +0000)]
[RISCV] Improve codegen for icmp {ne,eq} with a constant
Adds two patterns to improve the codegen of GPR value comparisons with small
constants. Instead of first loading the constant into another register and then
doing an XOR of those registers, these patterns directly use the constant as an
XORI immediate.
llvm-svn: 356990
Simon Pilgrim [Tue, 26 Mar 2019 12:32:01 +0000 (12:32 +0000)]
[TargetLowering] Add SimplifyDemandedBits support for ISD::INSERT_VECTOR_ELT
This helps us relax the extension of a lot of scalar elements before they are inserted into a vector.
Its exposes an issue in DAGCombiner::convertBuildVecZextToZext as some/all the zero-extensions may be relaxed to ANY_EXTEND, so we need to handle that case to avoid a couple of AVX2 VPMOVZX test regressions.
Once this is in it should be easier to fix a number of remaining failures to fold loads into VBROADCAST nodes.
Differential Revision: https://reviews.llvm.org/D59484
llvm-svn: 356989
Yi Kong [Tue, 26 Mar 2019 12:18:08 +0000 (12:18 +0000)]
Fix nondeterminism introduced in r353954
DenseMap iteration order is not guaranteed, use MapVector instead.
Fix provided by srhines.
Differential Revision: https://reviews.llvm.org/D59807
llvm-svn: 356988
Anastasia Stulova [Tue, 26 Mar 2019 11:22:37 +0000 (11:22 +0000)]
[OpenCL] Allow variadic macros as Clang feature.
llvm-svn: 356987
Javed Absar [Tue, 26 Mar 2019 11:16:01 +0000 (11:16 +0000)]
[TableGen] Let list elements have a trailing comma
Let lists have an trailing comma to allow cleaner diffs e.g:
def : Features<[FeatureA,
FeatureB,
]>;
Reviewed By: hfinkel
Differential Revision: https://reviews.llvm.org/D59247
llvm-svn: 356986
Javed Absar [Tue, 26 Mar 2019 10:49:09 +0000 (10:49 +0000)]
[TableGen] Give meaningful msg for def use in multiclass
When one mistakenly specifies 'def' instead of using 'defm',
the error message is quite misleading: 'Couldn't find class..'
Instead, it should recommend using defm if the multiclass of
same name exists.
Reviewed By: hfinkel
Differential Revision: https://reviews.llvm.org/D59294
llvm-svn: 356985
Oliver Stannard [Tue, 26 Mar 2019 10:24:03 +0000 (10:24 +0000)]
[ARM][Asm] Accept upper case coprocessor number and registers
Differential revision: https://reviews.llvm.org/D59760
llvm-svn: 356984
Sylvestre Ledru [Tue, 26 Mar 2019 09:48:23 +0000 (09:48 +0000)]
update the release notes after the change of 'clang -dumpversion'
llvm-svn: 356983
Martin Storsjo [Tue, 26 Mar 2019 09:02:44 +0000 (09:02 +0000)]
[llvm-dlltool] Set a proper machine type for weak symbol object files
This makes GNU binutils not reject the libraries outright.
GNU ld handles weak externals slightly differently though, so it
can't use them for aliases in import libraries, but this makes GNU
ld able to use the rest of the import libraries.
LLD accepted object files with machine type 0 aka
IMAGE_FILE_MACHINE_UNKNOWN.
Differential Revision: https://reviews.llvm.org/D59742
llvm-svn: 356982
Roger Ferrer Ibanez [Tue, 26 Mar 2019 08:01:18 +0000 (08:01 +0000)]
[RISCV] Pass -target-abi to -cc1as
The RISC-V assembler needs the target ABI because it defines a flag of the ELF
file, as described in [1].
Make clang (the driver) to pass the target ABI to -cc1as in exactly the same
way it does for -cc1.
Currently -cc1as knows about -target-abi but is not handling it. Handle it and
pass it to the MC layer via MCTargetOptions.
[1] https://github.com/riscv/riscv-elf-psabi-doc/blob/master/riscv-elf.md#file-header
Differential Revision: https://reviews.llvm.org/D59298
llvm-svn: 356981
Fangrui Song [Tue, 26 Mar 2019 05:33:52 +0000 (05:33 +0000)]
[ADT] Update SmallVectorTest.EmplaceBack tests after rL356312
rL356312 changed the return type of emplace_back from void to reference.
Update the tests to check the behavior.
Reviewers: dblaikie
Reviewed By: dblaikie
Subscribers: dexonsmith, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59809
llvm-svn: 356980
Craig Topper [Tue, 26 Mar 2019 05:31:32 +0000 (05:31 +0000)]
[X86] In matchBitExtract, place all of the new nodes before Node's position in the DAG for the topological sort.
We were using OrigNBits, but that put all the nodes before the node we used to start the control computation. This caused some node earlier than the sequence we inserted to be selected before the sequence we created. We want our new sequence to be selected first since it depends on OrigNBits.
I don't have a test case. Found by reviewing the code.
llvm-svn: 356979
Craig Topper [Tue, 26 Mar 2019 05:12:23 +0000 (05:12 +0000)]
[X86] In matchBitExtract, if we need to truncate the BEXTR make sure we put the BEXTR at Node's position in the DAG for the topological sort.
We were using OrigNBits, but that doesn't guarantee that it will be selected before the nodes that make up X.
llvm-svn: 356978
Craig Topper [Tue, 26 Mar 2019 05:12:21 +0000 (05:12 +0000)]
[X86] Remove unneeded FIXME. NFC
We do fold loads right below this.
llvm-svn: 356977
Thomas Lively [Tue, 26 Mar 2019 04:11:05 +0000 (04:11 +0000)]
Reland "[WebAssembly] Add linker options to control feature checking"
Do not pipe binary data between processes in lit tests this time,
since it turns out that can break on Windows.
This reverts commit
84c8652fc3085155d0f9c355455e5a797c6d9db6.
llvm-svn: 356975
Jan Korous [Tue, 26 Mar 2019 03:48:25 +0000 (03:48 +0000)]
[clangd][xpc][cmake] Respect explicit value of CLANGD_BUILD_XPC
We shouldn't prevent user from disabling XPC framework build on Darwin.
However, by keeping it on by default our CI systems also test
it by default on macOS.
Based on user request:
http://lists.llvm.org/pipermail/cfe-dev/2019-March/061778.html
Differential Revision: https://reviews.llvm.org/D59808
llvm-svn: 356974
Craig Topper [Tue, 26 Mar 2019 03:12:43 +0000 (03:12 +0000)]
X86Parser: Fix potential reference to deleted object
Within the MatchFPUWaitAlias function, Operands[0] is potentially overwritten leading to &Op referencing a deleted object. To fix this, assign the reference after the function.
Differential Revision: https://reviews.llvm.org/D57376
llvm-svn: 356973
Craig Topper [Tue, 26 Mar 2019 03:12:41 +0000 (03:12 +0000)]
X86AsmParser: Do not process a non-existent token
This error can only happen if an unfinished operation is at Eof.
Patch by Brandon Jones
Differential Revision: https://reviews.llvm.org/D57379
llvm-svn: 356972
Jonas Devlieghere [Tue, 26 Mar 2019 01:11:15 +0000 (01:11 +0000)]
[Python] Define empty SWIG wrapper for unit testin"
The python plugin uses wrappers generated by swig. For the symbols to be
available, we'd need to link against liblldb, which is not an option
because the symbols could conflict with the static library we are
testing. Instead we define the symbols ourselves in the unit test.
llvm-svn: 356971
Petr Hosek [Tue, 26 Mar 2019 00:43:29 +0000 (00:43 +0000)]
[CMake][Fuchsia] Merge static libunwind and libc++abi into libc++ on Darwin
We want to distribute only a single archive so -lc++ does the right
thing and users don't have to explicitly link -lc++abi and -lunwind.
Differential Revision: https://reviews.llvm.org/D59803
llvm-svn: 356970
Artem Dergachev [Tue, 26 Mar 2019 00:36:53 +0000 (00:36 +0000)]
[CFG] [analyzer] pr41142: C++17: Skip transparent InitListExprs in ExprEngine.
r356634 didn't fix all the problems caused by r356222 - even though simple
constructors involving transparent init-list expressions are now evaluated
precisely, many more complicated constructors aren't, for other reasons.
The attached test case is an example of a constructor that will never be
evaluated precisely - simply because there isn't a constructor there (instead,
the program invokes run-time undefined behavior by returning without a return
statement that should have constructed the return value).
Fix another part of the problem for such situations: evaluate transparent
init-list expressions transparently, so that to avoid creating ill-formed
"transparent" nonloc::CompoundVals.
Differential Revision: https://reviews.llvm.org/D59622
llvm-svn: 356969
Fangrui Song [Mon, 25 Mar 2019 23:46:19 +0000 (23:46 +0000)]
[ELF] Change GOT*_FROM_END (relative to end(.got)) to GOTPLT* (start(.got.plt))
Summary:
This should address remaining issues discussed in PR36555.
Currently R_GOT*_FROM_END are exclusively used by x86 and x86_64 to
express relocations types relative to the GOT base. We have
_GLOBAL_OFFSET_TABLE_ (GOT base) = start(.got.plt) but end(.got) !=
start(.got.plt)
This can have problems when _GLOBAL_OFFSET_TABLE_ is used as a symbol, e.g.
glibc dl_machine_dynamic assumes _GLOBAL_OFFSET_TABLE_ is start(.got.plt),
which is not true.
extern const ElfW(Addr) _GLOBAL_OFFSET_TABLE_[] attribute_hidden;
return _GLOBAL_OFFSET_TABLE_[0]; // R_X86_64_GOTPC32
In this patch, we
* Change all GOT*_FROM_END to GOTPLT* to fix the problem.
* Add HasGotPltOffRel to denote whether .got.plt should be kept even if
the section is empty.
* Simplify GotSection::empty and GotPltSection::empty by setting
HasGotOffRel and HasGotPltOffRel according to GlobalOffsetTable early.
The change of R_386_GOTPC makes X86::writePltHeader simpler as we don't
have to compute the offset start(.got.plt) - Ebx (it is constant 0).
We still diverge from ld.bfd (at least in most cases) and gold in that
.got.plt and .got are not adjacent, but the advantage doing that is
unclear.
Reviewers: ruiu, sivachandra, espindola
Subscribers: emaste, mehdi_amini, arichardson, dexonsmith, jdoerfert, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59594
llvm-svn: 356968
Jonas Devlieghere [Mon, 25 Mar 2019 23:36:12 +0000 (23:36 +0000)]
[Python] Move SWIG wrapper dependency into the plugin
This should fix the Windows bot (fingers crossed).
llvm-svn: 356967
Rui Ueyama [Mon, 25 Mar 2019 23:28:47 +0000 (23:28 +0000)]
Use a class instead of lambda-based callbacks to organize garbage collector.
lld's mark-sweep garbage collector was written in the visitor pattern.
There are functions that traverses a given graph, and the functions calls
callback functions to dispatch according to node type.
The code was originaly pretty simple, and lambdas worked pretty
well. However, as we add more features to the garbage collector, that became
more like a callback hell. We now have a callback function that wraps
another callback function, for example. It is not easy to follow the flow of
the control.
This patch rewrites it as a regular class. What was once a lambda is now a
regular class member function. I think this change fixes the readability
issue.
No functionality change intended.
Differential Revision: https://reviews.llvm.org/D59800
llvm-svn: 356966
Michael Kruse [Mon, 25 Mar 2019 23:26:59 +0000 (23:26 +0000)]
Moving ManagedMemoryRewritePass when hybrid option is selected
Compiling with -polly-target=hybrid was causing Polly to occur two times
in the pipeline. The reason was how the ManagedMemoryRewritePass was
registered in the pass manager. ManagedMemoryRewritePass being a
ModulePass was forcing all previous passes to get recomputed. This
commit avoids Polly to appear two times in the pipeline registering the
ManagedMemoryRewritePass later in the pass manager.
Patch by Lorenzo Chelini <l.chelini@icloud.com>
Differential Revision: https://reviews.llvm.org/D59263
llvm-svn: 356965
Reid Kleckner [Mon, 25 Mar 2019 23:20:18 +0000 (23:20 +0000)]
[MS] Add frontend support for __declspec(allocator)
The intention is to add metadata to direct call sites of functions
marked with __declspec(allocator), which will ultimately result in some
S_HEAPALLOCSITE debug info records when emitting codeview.
This is a piece of PR38491
llvm-svn: 356964
Eli Friedman [Mon, 25 Mar 2019 22:42:30 +0000 (22:42 +0000)]
[ARM] Add missing memory operands to a bunch of instructions.
This should hopefully lead to minor improvements in code generation, and
more accurate spill/reload comments in assembly.
Also fix isLoadFromStackSlotPostFE/isStoreToStackSlotPostFE so they
don't lead to misleading assembly comments for merged memory operands;
this is technically orthogonal, but in practice the relevant memory
operand lists don't show up without this change.
Differential Revision: https://reviews.llvm.org/D59713
llvm-svn: 356963
Jonathan Peyton [Mon, 25 Mar 2019 22:36:31 +0000 (22:36 +0000)]
[OpenMP] Add LLVM license header to file
This file was missing the LLVM license header
llvm-svn: 356962
Jonathan Peyton [Mon, 25 Mar 2019 22:31:00 +0000 (22:31 +0000)]
[OpenMP] Add Intel 19.0 to list of compilers in kmp_version.cpp
llvm-svn: 356961
Michal Gorny [Mon, 25 Mar 2019 22:29:29 +0000 (22:29 +0000)]
[lldb] [Process/NetBSD] Use nullptr to fix template arg deduction
llvm-svn: 356960
Sanjay Patel [Mon, 25 Mar 2019 22:08:45 +0000 (22:08 +0000)]
[x86] add tests for vector cmps; NFC
llvm-svn: 356959
Matt Arsenault [Mon, 25 Mar 2019 21:41:40 +0000 (21:41 +0000)]
Revert "AMDGPU: Scavenge register instead of findUnusedReg"
This reverts r356149.
This is crashing on rocBLAS.
llvm-svn: 356958
Bruno Ricci [Mon, 25 Mar 2019 21:37:10 +0000 (21:37 +0000)]
[Sema] Don't check for array bounds when the types in the base expression are dependent
Bail-out of CheckArrayAccess when the types of the base expression before
and after eventual casts are dependent. We will get another chance to check
for array bounds during instantiation. Fixes PR41087.
Differential Revision: https://reviews.llvm.org/D59776
Reviewed By: efriedma
llvm-svn: 356957
Matt Arsenault [Mon, 25 Mar 2019 21:28:53 +0000 (21:28 +0000)]
AMDGPU: Remove unnecessary check for isFullCopy
Subregister indexes are not used for physical register operands, so
isFullCopy is implied by the physical register check.
llvm-svn: 356956
Matt Arsenault [Mon, 25 Mar 2019 21:28:51 +0000 (21:28 +0000)]
AMDGPU: Make collapse-endcf test more useful
Without a VALU instruction in the return block, these were mostly
testing the path to delete exec mask code before s_endpgm rather than
the end cf handling.
llvm-svn: 356955
Eli Friedman [Mon, 25 Mar 2019 21:25:28 +0000 (21:25 +0000)]
[AArch64] Prefer "mov" over "orr" to materialize constants.
This is generally more readable due to the way the assembler aliases
work.
(This causes a lot of test changes, but it's not really as scary as it
looks at first glance; it's just mechanically changing a bunch of checks
for orr to check for mov instead.)
Differential Revision: https://reviews.llvm.org/D59720
llvm-svn: 356954
Sam Clegg [Mon, 25 Mar 2019 21:14:26 +0000 (21:14 +0000)]
[WebAssembly] Don't use default GetLinkerPath
We can't (don't want to) honor the same set of "-fuse-ld" flags with
WebAssembly since the ELF linkers (ld.lld, ld.gnu, etc) don't work with
wasm object files.
Instead we implement our own linker finding logic, similar or other
non-ELF platforms like MSVC.
We've had a few issues with CLANG_DEFAULT_LINKER overriding the
WebAssembly linker which doesn't make sense since there is no generic
linker that can handle WebAssembly today.
Differential Revision: https://reviews.llvm.org/D59743
llvm-svn: 356953
Matt Arsenault [Mon, 25 Mar 2019 21:10:12 +0000 (21:10 +0000)]
AMDGPU: Set hasSideEffects 0 on _term instructions
These were defaulting to true, but they are just wrappers around bit
operations. This avoids regressions in the exec mask optimization
passes in a future commit.
llvm-svn: 356952
Ali Tamur [Mon, 25 Mar 2019 21:09:07 +0000 (21:09 +0000)]
Revert "[llvm] Prevent duplicate files in debug line header in dwarf 5."
This reverts commit
312ab05887d0e2caa29aaf843cefe39379a98d36.
My commit broke the build; I will revert and find out what happened.
llvm-svn: 356951
Jonas Devlieghere [Mon, 25 Mar 2019 21:07:53 +0000 (21:07 +0000)]
[PythonTestSuite] Fix usage of InitializePrivate in PythonTestSuite
llvm-svn: 356950
Ali Tamur [Mon, 25 Mar 2019 21:06:02 +0000 (21:06 +0000)]
Revert "[lld] Prevent duplicate files in debug line header in dwarf 5."
This reverts commit
5aa7478a55fa86dae45913b82bd05ed32d9c0973.
Some things are broken; I will fix and try to commit again.
llvm-svn: 356949
Robert Widmann [Mon, 25 Mar 2019 20:58:58 +0000 (20:58 +0000)]
[LLVM-C] Add binding to look up intrinsic by name
Summary: Add a binding to Function::lookupIntrinsicID so clients don't have to go searching the ID table themselves.
Reviewers: whitequark, deadalnix
Reviewed By: whitequark
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59697
llvm-svn: 356948
Konstantin Zhuravlyov [Mon, 25 Mar 2019 20:54:00 +0000 (20:54 +0000)]
AMDGPU: Add support for cross address space synchronization scopes (clang)
Differential Revision: https://reviews.llvm.org/D59494
llvm-svn: 356947
Konstantin Zhuravlyov [Mon, 25 Mar 2019 20:50:21 +0000 (20:50 +0000)]
AMDGPU: Add support for cross address space synchronization scopes
Differential Revision: https://reviews.llvm.org/D59517
llvm-svn: 356946
Louis Dionne [Mon, 25 Mar 2019 20:44:07 +0000 (20:44 +0000)]
[pstl][NFC] Remove some warnings
llvm-svn: 356945
Jonas Devlieghere [Mon, 25 Mar 2019 20:39:08 +0000 (20:39 +0000)]
[ScriptInterpreterPython] Make SWIG callback private (NFC)
With the initialization taking place inside the Python script
interpreter, these function no longer need to be public. The exception
is the g_swig_init_callback which is used from the RAII object.
llvm-svn: 356944
Ali Tamur [Mon, 25 Mar 2019 20:18:27 +0000 (20:18 +0000)]
[lld] Prevent duplicate files in debug line header in dwarf 5.
Summary:
Changes to a couple of tests that will start failing after https://reviews.llvm.org/D59515 is submitted.
Reviewers: echristo, ruiu, espindola
Reviewed By: echristo, ruiu
Subscribers: emaste, arichardson, MaskRay, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59671
llvm-svn: 356943
Jonas Devlieghere [Mon, 25 Mar 2019 20:14:31 +0000 (20:14 +0000)]
[ScriptInterpreterPython] Move SWIG initialization into the Python plugin (NFC)
Abstract initialization of the Python SWIG support in the Python plugin.
llvm-svn: 356942
Ali Tamur [Mon, 25 Mar 2019 20:08:00 +0000 (20:08 +0000)]
[llvm] Prevent duplicate files in debug line header in dwarf 5.
Summary:
Motivation: In previous dwarf versions, file name indexes started from 1, and
the primary source file was not explicit. Dwarf 5 standard (6.2.4) prescribes
the primary source file to be explicitly given an entry with an index number 0.
The current implementation honors the specification by just duplicating the
main source file, once with index number 0, and later maybe with another
index number. While this is compliant with the letter of the standard, the
duplication causes problems for consumers of this information such as lldb.
(Some files are duplicated, where only some of them have a line table although
all refer to the same file)
With this change, dwarf 5 debug line section files always start from 0, and
the zeroth entry is not duplicated whenever possible. This requires different
handling of dwarf 4 and dwarf 5 during generation (e.g. when a function returns
an index zero for a file name, it signals an error in dwarf 4, but not in dwarf 5)
However, I think the minor complication is worth it, because it enables all
consumers (lldb, gdb, dwarfdump, objdump, and so on) to treat all files in the
file name list homogenously.
Reviewers: dblaikie, probinson, aprantl, espindola
Reviewed By: probinson
Subscribers: emaste, jvesely, nhaehnle, aprantl, javed.absar, arichardson, hiraditya, MaskRay, rupprecht, jdoerfert, llvm-commits
Tags: #llvm, #debug-info
Differential Revision: https://reviews.llvm.org/D59515
llvm-svn: 356941
JF Bastien [Mon, 25 Mar 2019 20:06:32 +0000 (20:06 +0000)]
Thread Safety: also look at ObjC methods
Summary:
SExprBuilder::translateDeclRefExpr was only looking at FunctionDecl and not also looking at ObjCMethodDecl. It should consider both because the attributes can be used on Objective-C as well.
<rdar://problem/
48941331>
Reviewers: dexonsmith, erik.pilkington
Subscribers: jkorous, jdoerfert, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D59523
llvm-svn: 356940
Simon Pilgrim [Mon, 25 Mar 2019 20:05:27 +0000 (20:05 +0000)]
[SLPVectorizer] Merge reorderAltShuffleOperands into reorderInputsAccordingToOpcode
As discussed on D59738, this generalizes reorderInputsAccordingToOpcode to handle multiple + non-commutative instructions so we can get rid of reorderAltShuffleOperands and make use of the extra canonicalizations that reorderInputsAccordingToOpcode brings.
Differential Revision: https://reviews.llvm.org/D59784
llvm-svn: 356939
Simon Pilgrim [Mon, 25 Mar 2019 18:51:57 +0000 (18:51 +0000)]
[SelectionDAG] Add icmp UNDEF handling to SelectionDAG::FoldSetCC
First half of PR40800, this patch adds DAG undef handling to icmp instructions to match the behaviour in llvm::ConstantFoldCompareInstruction and SimplifyICmpInst, this permits constant folding of vector comparisons where some elements had been reduced to UNDEF (by SimplifyDemandedVectorElts etc.).
This involved a lot of tweaking to reduced tests as bugpoint loves to reduce icmp arguments to undef........
Differential Revision: https://reviews.llvm.org/D59363
llvm-svn: 356938
Teresa Johnson [Mon, 25 Mar 2019 18:38:48 +0000 (18:38 +0000)]
[CGP] Build the DominatorTree lazily
Summary:
In r355512 CGP was changed to build the DominatorTree only once per
function traversal, to avoid repeatedly building it each time it was
accessed. This solved one compile time issue but introduced another. In
the second case, we now were building the DT unnecessarily many times
when we performed many function traversals (i.e. more than once per
function when running CGP because of changes made each time).
Change to saving the DT in the CodeGenPrepare object, and building it
lazily when needed. It is reset whenever we need to rebuild it.
The case that exposed the issue there are 617 functions, and we walk
them (i.e. execute the "while (MadeChange)" loop in runOnFunction) a
total of 12083 times (so previously we were building the DT 12083
times). With this patch we only build the DT 844 times (average of 1.37
times per function). We dropped the total time to compile this file from
538.11s without this patch to 339.63s with it.
There is still an issue as CGP is taking much longer than all other
passes even with this patch, and before a recent compiler release cut at
r355392 the total time to this compile was only 97 sec with a huge
reduction in CGP time. I suspect that one of the other recent changes to
CGP led to iterating each function many more times on average, but I
need to do some more investigation.
Reviewers: spatel
Subscribers: jdoerfert, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59696
llvm-svn: 356937
Dimitry Andric [Mon, 25 Mar 2019 18:37:49 +0000 (18:37 +0000)]
Fix interoperability test compilation on FreeBSD
Summary:
While building the 8.0 releases on FreeBSD, I encountered the following
error in the regression tests, where ompt/misc/interoperability.cpp
failed to compile, with:
```
projects/openmp/runtime/test/ompt/misc/interoperability.cpp:7:10: fatal error: 'alloca.h' file not found
#include <alloca.h>
^~~~~~~~~~
```
Like on NetBSD, alloca(3) is defined in <stdlib.h> instead.
Reviewers: emaste, jlpeyton, krytarowski, mgorny, protze.joachim
Reviewed By: jlpeyton
Subscribers: jdoerfert, openmp-commits
Tags: #openmp
Differential Revision: https://reviews.llvm.org/D59736
llvm-svn: 356936
Louis Dionne [Mon, 25 Mar 2019 18:37:28 +0000 (18:37 +0000)]
[pstl] Gate usage of TBB in test on whether TBB is used as a backend
__PSTL_USE_PAR_POLICIES is the wrong macro to use there, since the PSTL
could conceivably be using parallel policies but not TBB as a backend.
llvm-svn: 356935
Dimitry Andric [Mon, 25 Mar 2019 18:37:14 +0000 (18:37 +0000)]
Fix gettid warnings on FreeBSD
Summary:
[Split off from D59451 to get this fix in separately]
While building the 8.0 releases on FreeBSD, I encountered the following
warnings in openmp quite a few times:
```
In file included from projects/openmp/runtime/src/kmp_settings.cpp:27:
projects/openmp/runtime/src/kmp_wrapper_getpid.h:35:2: warning: #warning is a language extension [-Wpedantic]
#warning No gettid found, use getpid instead
^
projects/openmp/runtime/src/kmp_wrapper_getpid.h:35:2: warning: No gettid found, use getpid instead [-W#warnings]
2 warnings generated.
```
I added a gettid wrapper that uses FreeBSD's pthread_getthreadid_np(3)
function for this.
Reviewers: emaste, jlpeyton, krytarowski, mgorny, protze.joachim
Reviewed By: jlpeyton
Subscribers: jfb, jdoerfert, openmp-commits
Tags: #openmp
Differential Revision: https://reviews.llvm.org/D59735
llvm-svn: 356934
Jonathan Peyton [Mon, 25 Mar 2019 18:17:55 +0000 (18:17 +0000)]
[OpenMP] Fix pause check with version info
Add 5.0 guard to pause code for now.
Patch by Terry Wilmarth
Differential Revision: https://reviews.llvm.org/D59428
llvm-svn: 356933
Thomas Lively [Mon, 25 Mar 2019 18:10:26 +0000 (18:10 +0000)]
Revert "[WebAssembly] Add linker options to control feature checking"
This reverts commit
5991328c96b6146999cfa74ede42901f8c01f2fa.
llvm-svn: 356932
Sylvestre Ledru [Mon, 25 Mar 2019 18:06:20 +0000 (18:06 +0000)]
clang -dumpversion returns 4.2.1 for legacy reason, update it
Summary:
It has been introduced in 2011 for gcc compat:
https://github.com/llvm-mirror/clang/commit/
95a907fc0f2f717dca38a379c1d2353bfea06d4f
it is probably time to remove it to remove the confusion.
reported: https://bugs.llvm.org/show_bug.cgi?id=38836
https://bugs.launchpad.net/ubuntu/+source/llvm-defaults/+bug/
1810860
Reviewers: ddunbar, rnk
Reviewed By: rnk
Subscribers: jdoerfert, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D56803
llvm-svn: 356931
Sanjay Patel [Mon, 25 Mar 2019 17:53:56 +0000 (17:53 +0000)]
[x86] add another vector zext test; NFC
Goes with the proposal in D59777
llvm-svn: 356930
Mikhail R. Gadelha [Mon, 25 Mar 2019 17:47:45 +0000 (17:47 +0000)]
Moved everything SMT-related to LLVM and updated the cmake scripts.
Differential Revision: https://reviews.llvm.org/D54978
llvm-svn: 356929
Krasimir Georgiev [Mon, 25 Mar 2019 17:29:16 +0000 (17:29 +0000)]
[clang-format] Refine structured binding detection
Summary:
Revision r356575 had the unfortunate consequence that now clang-format never
detects an ObjC call expression after `&&`.
This patch tries harder to distinguish between C++17 structured bindings and
ObjC call expressions and adds a few regression tests.
Reviewers: klimek
Reviewed By: klimek
Subscribers: cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D59774
llvm-svn: 356928
Jonas Devlieghere [Mon, 25 Mar 2019 17:27:14 +0000 (17:27 +0000)]
[Args] Handle backticks to prevent crash.
Currently LLDB crashes when autocompleting a command that ends with a
backtick because the quote character wasn't handled. This fixes that and
adds a unit test for this function.
Differential revision: https://reviews.llvm.org/D59779
llvm-svn: 356927
Matt Arsenault [Mon, 25 Mar 2019 17:15:44 +0000 (17:15 +0000)]
MISched: Don't schedule regions with 0 instructions
I think this is correct, but may not necessarily be the correct fix
for the assertion I'm really trying to solve. If a scheduling region
was found that only has dbg_value instructions, the RegPressure
tracker would end up in an inconsistent state because it would skip
over any debug instructions and point to an instruction outside of the
scheduling region. It may still be possible for this to happen if
there are some real schedulable instructions between dbg_values, but I
haven't managed to break this.
The testcase is extremely sensitive and I'm not sure how to make it
more resistent to future scheduler changes that would avoid stressing
this situation.
llvm-svn: 356926
Bruno Ricci [Mon, 25 Mar 2019 17:08:51 +0000 (17:08 +0000)]
[Sema][NFCI] Don't allocate storage for the various CorrectionCandidateCallback unless we are going to do some typo correction
The various CorrectionCandidateCallbacks are currently heap-allocated
unconditionally. This was needed because of delayed typo correction.
However these allocations represent currently 15.4% of all allocations
(number of allocations) when parsing all of Boost (!), mostly because
of ParseCastExpression, ParseStatementOrDeclarationAfterAttrtibutes
and isCXXDeclarationSpecifier. Note that all of these callback objects
are small. Let's not do this.
Instead initially allocate the callback on the stack, and only do a
heap allocation if we are going to do some typo correction. Do this by:
1. Adding a clone function to each callback, which will do a polymorphic
clone of the callback. This clone function is required to be implemented
by every callback (of which there is a fair amount). Make sure this is
the case by making it pure virtual.
2. Use this clone function when we are going to try to correct a typo.
This additionally cut the time of -fsyntax-only on all of Boost by 0.5%
(not that much, but still something). No functional changes intended.
Differential Revision: https://reviews.llvm.org/D58827
Reviewed By: rnk
llvm-svn: 356925
Tom Stellard [Mon, 25 Mar 2019 17:01:29 +0000 (17:01 +0000)]
merge-request.sh: Update 8.0 metabug for 8.0.1
llvm-svn: 356924
Nico Weber [Mon, 25 Mar 2019 16:50:24 +0000 (16:50 +0000)]
Attempt to fix sphinx bot after r356897
llvm-svn: 356923
Matt Arsenault [Mon, 25 Mar 2019 16:47:42 +0000 (16:47 +0000)]
AMDGPU: Preserve LiveIntervals in WQM
This seems to already be done, but wasn't marked.
llvm-svn: 356922
Louis Dionne [Mon, 25 Mar 2019 16:44:08 +0000 (16:44 +0000)]
[pstl] Make sure unit tests exit with a non-zero error code when they fail
The previous test system would only print errors to stderr, however CMake
(and lit) detect failure via the program returning a non-zero error code.
So all the tests would always pretend they passed.
llvm-svn: 356921
Evandro Menezes [Mon, 25 Mar 2019 16:38:48 +0000 (16:38 +0000)]
[clang] Remove cmake warning message (NFC)
Recognize an empty string for CLANG_DEFAULT_UNWINDLIB as a valid option.
llvm-svn: 356920
James Henderson [Mon, 25 Mar 2019 16:36:26 +0000 (16:36 +0000)]
[llvm-objcopy]Preserve data in segments not covered by sections
llvm-objcopy previously knew nothing about data in segments that wasn't
covered by section headers, meaning that it wrote zeroes instead of what
was there. As it is possible for this data to be useful to the loader,
this patch causes llvm-objcopy to start preserving this data. Data in
sections that are explicitly removed continues to be written as zeroes.
This fixes https://bugs.llvm.org/show_bug.cgi?id=41005.
Reviewed by: jakehehrlich, rupprecht
Differential Revision: https://reviews.llvm.org/D59483
llvm-svn: 356919
Marshall Clow [Mon, 25 Mar 2019 16:35:59 +0000 (16:35 +0000)]
Cleaup of requirements for optional. While researching LWG3196, I realized that optional did not reject 'const in_place_t' like it should. Added a test as well, and a check for arrays (which were already disallowed, but now we get a better error message). Should not affect anyone's code.
llvm-svn: 356918
Louis Dionne [Mon, 25 Mar 2019 16:20:49 +0000 (16:20 +0000)]
[PSTL][NFC] Add missing #include
llvm-svn: 356917
Ilya Biryukov [Mon, 25 Mar 2019 16:18:56 +0000 (16:18 +0000)]
[clangd] Add .cu files to VSCode extension
Summary:
clangd should be able to handle those with a proper compilation
database. However, users using 'nvcc' might start seeing spurious errors
in '.cu' files after this change.
My plan is to land and release this, but be ready to revert in
case of negative user feedback.
Reviewers: hokein
Reviewed By: hokein
Subscribers: ioeric, MaskRay, jkorous, arphaman, kadircet, jdoerfert, cfe-commits
Tags: #clang
Differential Revision: https://reviews.llvm.org/D59759
llvm-svn: 356916
Simon Pilgrim [Mon, 25 Mar 2019 16:14:21 +0000 (16:14 +0000)]
[SLPVectorizer] Update file missed in rL356913
Differential Revision: https://reviews.llvm.org/D59738
llvm-svn: 356915
Sanjay Patel [Mon, 25 Mar 2019 15:54:34 +0000 (15:54 +0000)]
[x86] add tests for vector zext; NFC
The AVX1 lowering is poor.
llvm-svn: 356914
Simon Pilgrim [Mon, 25 Mar 2019 15:53:55 +0000 (15:53 +0000)]
[SLPVectorizer] reorderInputsAccordingToOpcode - remove non-Instruction canonicalization
Remove attempts to commute non-Instructions to the LHS - the codegen changes appear to rely on chance more than anything else and also have a tendency to fight existing instcombine canonicalization which moves constants to the RHS of commutable binary ops.
This is prep work towards:
(a) reusing reorderInputsAccordingToOpcode for alt-shuffles and removing the similar reorderAltShuffleOperands
(b) improving reordering to optimized cases with commutable and non-commutable instructions to still find splat/consecutive ops.
Differential Revision: https://reviews.llvm.org/D59738
llvm-svn: 356913
Haojian Wu [Mon, 25 Mar 2019 15:46:07 +0000 (15:46 +0000)]
Revert "[clang-format] Keep protobuf "package" statement on one line"
This reverts commit r356835. This patch causes a regression, see the
test below:
verifyFormat("// Detached comment\n\n"
"// Leading comment\n"
"syntax = \"proto2\"; // trailing comment\n\n"
"// in foo.bar package\n"
"package foo.bar; // foo.bar package\n");
llvm-svn: 356912