This reverts commits r335794 and r335797. Breaks ThinLTO+FDO selfhost.

llvm-svn: 335851

Armv6 introduced instructions to perform 32-bit SIMD operations. The purpose of
this pass is to do some straightforward IR pattern matching to create ACLE DSP
intrinsics, which map on these 32-bit SIMD operations.

Currently, only the SMLAD instruction gets recognised. This instruction
performs two multiplications with 16-bit operands, and stores the result in an
accumulator. We will follow this up with patches to recognise SMLAD in more
cases, and also to generate other DSP instructions (like e.g. SADD16).

Patch by: Sam Parker and Sjoerd Meijer

Differential Revision: https://reviews.llvm.org/D48128

llvm-svn: 335850

It is https://github.com/llvm-mirror/lld/blob/master/ELF/EhFrame.cpp#L144,
and we had no test case to cover it.

llvm-svn: 335849

llvm-svn: 335848

It's https://github.com/llvm-mirror/lld/blob/master/ELF/EhFrame.cpp#L147 line,
which was untested before this patch.

llvm-svn: 335847

llvm-svn: 335846

llvm-svn: 335845

This reverts commit r335839, because it breaks the MSVC build.

llvm-svn: 335844

The :option: syntax was generating Sphinx build warnings; switched to double backticks to silence the warning; NFC.

llvm-svn: 335843

Correct the code highlighting marker to be Objective-C rather than C++ which fixes a Sphinx build warning; NFC.

llvm-svn: 335842

llvm-svn: 335841

This test case adds test for the line that was uncovered previously:
https://github.com/llvm-mirror/lld/blob/master/ELF/EhFrame.cpp#L175

it errors out, but we now can test that we have expected flow,
which is different for CIE of version 1 and CIE of version 3.

llvm-svn: 335840

Summary:
This allows the implicit ArrayRef conversions to kick in when e.g.
comparing ArrayRef to a SmallVector.

Reviewers: zturner, dblaikie

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D48632

llvm-svn: 335839

It was https://github.com/llvm-mirror/lld/blob/master/ELF/EhFrame.cpp#L195
and we did not cover it with test.

Patch fixes it.

llvm-svn: 335838

Rather than calling std::find in a loop, just sort the vector and remove
duplicate entries at the end of the function.

Also, move the debug print at the end of the function, and query the
MCRegisterInfo to print register names rather than physreg IDs.

No functional change intended.

llvm-svn: 335837

By recomputing CompilerInvocation instead of copying it.
The problem was caused by the fact that copies of CompilerInvocation
store references to the shared state (DiagnosticOptions) when copied,
causing data races when two different copies are destroyed from
different threads.

llvm-svn: 335836

Add handling of the begin() funcion of containers to the iterator checkers,
together with the pre- and postfix ++ and -- operators of the iterators. This
makes possible the checking of iterators dereferenced ahead of the begin of the
container.

Differential Revision: https://reviews.llvm.org/D32642

llvm-svn: 335835

Follow-up commit for r335757 to address some inconsistencies.

llvm-svn: 335834

It is the following line:
https://github.com/llvm-mirror/lld/blob/master/ELF/EhFrame.cpp#L120

We did not cover it yet.

llvm-svn: 335833

Summary: Just a silly one-character correction.

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D48709

llvm-svn: 335832

llvm-svn: 335831

We have too many mechanisms for tracking the various offsets
used for kernel arguments, so remove one. There's still a lot of
confusion with these because there are two different "implicit"
argument areas located at the beginning and end of the kernarg
segment.

Additionally, the offset was determined based on the memory
size of the split element types. This would break in a future
commit where v3i32 is decomposed into separate i32 pieces.

llvm-svn: 335830

In principle nothing should stop these from working, but
work is necessary to create an ABI for dealing with the stack
related registers.

llvm-svn: 335829

Not sure how this wasn't noticed before.

llvm-svn: 335828

Just fix the crash for now by not doing the optimization since
figuring out how to properly convert the bits for an arbitrary
struct is a pain.

Also fix a crash when there is only an empty struct argument.

llvm-svn: 335827

llvm-svn: 335826

It was uncovered by our test cases.

llvm-svn: 335824

These are all benign races and only visible in !NDEBUG. tsan complains
about it, but a simple atomic bool is sufficient to make it happy.

llvm-svn: 335823

Summary:
This patch fixes a problem with retrieving a function symbol by an
address in a nested block. In the current implementation of
ResolveSymbolContext function it retrieves a symbol with
PDB_SymType::None and then checks if found symbol's tag equals to
PDB_SymType::Function. So, if nested block's symbol was found,
ResolveSymbolContext does not resolve a function.

Reviewers: asmith, labath, zturner

Reviewed By: asmith, labath

Differential Revision: https://reviews.llvm.org/D47939
Patch by Aleksandr Urakov <aleksandr.urakov@jetbrains.com>

llvm-svn: 335822

We could get away with it for constant folded cases, but not for rL335719.

Thanks to Krzysztof Parzyszek for noticing.

llvm-svn: 335821

SCCP does not change the CFG, so we can mark it as preserved.

Reviewers: dberlin, efriedma, davide

Reviewed By: davide

Differential Revision: https://reviews.llvm.org/D47149

llvm-svn: 335820

It was uncovered by our test cases.

llvm-svn: 335819

We has precompiled object with unsupported FDE version (=2).
It is possible to use llvm-mc instead for this test.

Patch do this change.

llvm-svn: 335818

Noticed in D45806 review.

llvm-svn: 335817

If a trunc has a user in a block which is not reachable from entry,
we can safely perform trunc elimination as if this user didn't exist.

llvm-svn: 335816

Summary:
This enables the X86-specific X86FloatingPointStackifierPass, and allow
llvm-exegesis to generate and measure X87 latency/uops for some FP ops.

Reviewers: gchatelet

Subscribers: tschuett, llvm-commits

Differential Revision: https://reviews.llvm.org/D48592

llvm-svn: 335815

If range [m .. n] is stored for symbolic expression A - B, then we can deduce the range for B - A which is [-n .. -m]. This is only true for signed types, unless the range is [0 .. 0].

Differential Revision: https://reviews.llvm.org/D35110

llvm-svn: 335814

Suggested-by: Michael Kruse
llvm-svn: 335813

Previously we were using default logic when building Linux runtimes
in Fuchsia toolchain, but that leads to various issues due to how
the CMake logic in compiler-rt for determining the platform support
is implemented. With this change, we will use explicit target for
every provided Linux sysroot.

Differential Revision: https://reviews.llvm.org/D48563

llvm-svn: 335812

Remove unused ByteStreamer argument from function emitDebugLocValue.

Patch by Nikola Prica.

Differential Revision: https://reviews.llvm.org/D48590

llvm-svn: 335811