Legalize and emit code for converting (Un)Signed Word to quad-precision via:

xscvsdqp
xscvudqp

Differential Revision: https://reviews.llvm.org/D45389

llvm-svn: 330273

Summary:
Patch adds initial emission of the debug info for NVPTX target.
Currently, only .file and .loc directives are emitted, everything else is
commented out to not break the compilation of Cuda.

Reviewers: echristo, jlebar, tra, jholewinski

Subscribers: mgorny, aprantl, JDevlieghere, llvm-commits

Differential Revision: https://reviews.llvm.org/D41827

llvm-svn: 330271

a zero register.

Previously I tried this and saw LLVM unable to transform this to fold
with memory operands such as spill slot rematerialization. However, it
clearly works as shown in this patch. We turn these into `cmpb $0,
<mem>` when useful for folding a memory operand without issue. This form
has no disadvantage compared to `testb $-1, <mem>`. So overall, this is
likely no worse and may be slightly smaller in some cases due to the
`testb %reg, %reg` form.

Differential Revision: https://reviews.llvm.org/D45475

llvm-svn: 330269

Path.inc/widenPath tries to decode the path using both UTF-8 and the default Windows code page.
This is no longer necessary with the new InitLLVM method which ensures that the command line
arguemnts are already UTF-8 on Windows.
 

llvm-svn: 330266

across basic blocks in the limited cases where it is very straight
forward to do so.

This will also be useful for other places where we do some limited
EFLAGS propagation across CFG edges and need to handle copy rewrites
afterward. I think this is rapidly approaching the maximum we can and
should be doing here. Everything else begins to require either heroic
analysis to prove how to do PHI insertion manually, or somehow managing
arbitrary PHI-ing of EFLAGS with general PHI insertion. Neither of these
seem at all promising so if those cases come up, we'll almost certainly
need to rewrite the parts of LLVM that produce those patterns.

We do now require dominator trees in order to reliably diagnose patterns
that would require PHI nodes. This is a bit unfortunate but it seems
better than the completely mysterious crash we would get otherwise.

Differential Revision: https://reviews.llvm.org/D45673

llvm-svn: 330264

Patch by Dávid Bolvanský!

Differential Revision: https://reviews.llvm.org/D45413

llvm-svn: 330259

Summary:
A change to use divergence analysis in the AMDGPU backend was getting formal
arguments incorrect (not tagged as divergent) unless they were VGPR0, VGPR1 or
VGPR2

For graphics shaders it is possible to have more than these passed in as VGPR

Modified the checking code to check for any VGPR registers passed in as formal
arguments.

Also, some intrinsics that are sources of divergence may have been lowered
during instruction selection and are missed on subsequent calls to
isSDNodeSourceOfDivergence - added the relevant AMDGPUISD checks as well.

Finally, the FunctionLoweringInfo tracks virtual registers that are live across
basic block boundaries. This is used to check for divergence of CopyFromRegister
registers using the DivergenceAnalysis analysis. For multiple blocks the lazily
evaluated inverted map VirtReg2Value was not cleared when the ValueMap map was.

Subscribers: arsenm, kzhuravl, wdng, nhaehnle, yaxunl, tpr, t-tye, llvm-commits

Differential Revision: https://reviews.llvm.org/D45372

Change-Id: I112f3bd6dfe0f62e63ce9b43b893982778e4bee3
llvm-svn: 330257

After investigation discussed in D45439, it would seem that the nsw
flag restriction is unnecessary in most cases. So the IsInductionVar
lambda has been removed, the functionality extracted, and now only
require nsw when using eq/ne predicates.

Differential Revision: https://reviews.llvm.org/D45617

llvm-svn: 330256

Summary:
Due to some android peculiarities, in some build configurations
(statically linked executables targeting older releases) we could detect
the presence of these functions (because they are present in libc.a,
where check_library_exists searches), but then fail to build because the
headers did not include the definition.

This attempts to remedy that by upgrading the check_library_exists to
check_symbol_exists, which will check that the function is declared too.

I am hoping that a more thorough check will make the messy #ifdef we
have accumulated in the code obsolete, so I optimistically try to remove
them.

Reviewers: zturner, kparzysz, danalbert

Subscribers: srhines, mgorny, krytarowski, llvm-commits

Differential Revision: https://reviews.llvm.org/D45359

llvm-svn: 330251

If a predicate does not become known after peeling, peeling is unlikely
to be beneficial.

Reviewers: mcrosier, efriedma, mkazantsev, junbuml

Reviewed By: mkazantsev

Differential Revision: https://reviews.llvm.org/D44983

llvm-svn: 330250

Summary:
Previously we crashed for the combination of the two features because we
tried to reference the dwo CU from the main object file. The fix
consists of two items:
- reference the skeleton CU from the name index (the consumer is
  expected to use the skeleton CU to find the real data).
- use the main object file string pool for the strings in the index

Reviewers: JDevlieghere, aprantl, dblaikie

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D45566

llvm-svn: 330249

Summary:
When sinking an instruction in InstCombine we now also sink
the DbgInfoIntrinsics that are using the sunken value.

Example)

When sinking the load in this input

bb.X:
  %0 = load i64, i64* %start, align 4, !dbg !31
  tail call void @llvm.dbg.value(metadata i64 %0, ...)
  br i1 %cond, label %for.end, label %for.body.lr.ph
for.body.lr.ph:
  br label %for.body

we now also move the dbg.value, like this

bb.X:
  br i1 %cond, label %for.end, label %for.body.lr.ph
for.body.lr.ph:
  %0 = load i64, i64* %start, align 4, !dbg !31
  tail call void @llvm.dbg.value(metadata i64 %0, ...)
  br label %for.body

In the past we haven't moved the dbg.value so we got

bb.X:
  tail call void @llvm.dbg.value(metadata i64 %0, ...)
  br i1 %cond, label %for.end, label %for.body.lr.ph
for.body.lr.ph:
  %0 = load i64, i64* %start, align 4, !dbg !31
  br label %for.body


So in the past we got a debug-use before the def of %0.
And that dbg.value was also on the path jumping to %for.end, for
which %0 never was defined.

CodeGenPrepare normally comes to rescue later (when not moving
the dbg.value), since it moves dbg.value instrinsics quite
brutally, without really analysing if it is correct to move
the intrinsic (see PR31878).
So at the moment this patch isn't expected to have much impact,
besides that it is moving the dbg.value already in opt, making
the IR look more sane directly.

This can be seen as a preparation to (hopefully) make it possible
to turn off CodeGenPrepare::placeDbgValues later as a solution
to PR31878.

I also adjusted test/DebugInfo/X86/sdagsplit-1.ll to make the
IR in the test case up-to-date with this behavior in InstCombine.

Reviewers: rnk, vsk, aprantl

Reviewed By: vsk, aprantl

Subscribers: mattd, JDevlieghere, llvm-commits

Tags: #debug-info

Differential Revision: https://reviews.llvm.org/D45425

llvm-svn: 330243

llvm-svn: 330241

llvm-svn: 330239

Summary: Previously if a modifer was placed on a non-GPR register class we would hit an assert or crash.

Reviewers: echristo

Reviewed By: echristo

Subscribers: eraman, llvm-commits

Differential Revision: https://reviews.llvm.org/D45751

llvm-svn: 330238

The bitcast may be interfering with other combines or vectorization 
as shown in PR16739:
https://bugs.llvm.org/show_bug.cgi?id=16739

Most pointer-related optimizations are probably able to look through 
this bitcast, but removing the bitcast shrinks the IR, so it's at
least a size savings.

Differential Revision: https://reviews.llvm.org/D44833

llvm-svn: 330237

Summary:
Statistic and ManagedStatic both use mutexes. There was a lock order
inversion where, during initialization, Statistic's mutex would be
held while taking ManagedStatic's, and in llvm_shutdown,
ManagedStatic's mutex would be held while taking Statistic's
mutex. This change causes Statistic's initialization code to avoid
holding its mutex while calling ManagedStatic's methods, avoiding the
inversion.

Reviewers: dsanders, rtereshin

Reviewed By: dsanders

Subscribers: hiraditya, llvm-commits

Differential Revision: https://reviews.llvm.org/D45398

llvm-svn: 330236

Literal encoding needs op_sel_hi to select low 16 bit in this case.

Differential Revision: https://reviews.llvm.org/D45745

llvm-svn: 330230

Track the debug locations of the incoming values to newly-created phis,
and apply merged debug locations to the phis.

A merged location will be on line 0, but will have the correct scope
set. This improves crash reporting when an inlined instruction with a
merged location triggers a machine exception. A debugger will be able to
narrow down the crash to the correct inlined scope, instead of simply
pointing to the outer scope of the caller.

Taken together with a change allows generating merged line-0 locations
for  instructions which aren't calls, this results in a 0.5% increase in
the uncompressed size of the .debug_line section of a stage2+Release
build of clang (-O3 -g).

rdar://33858697

Differential Revision: https://reviews.llvm.org/D45397

llvm-svn: 330227

llvm-svn: 330226

The implementation follows the MIPS backend and expands the
pseudo instruction directly during asm parsing. As the result, only
real MC instructions are emitted to the MCStreamer. Additionally,
PseudoLI instructions are emitted during codegen. The actual
expansion to real instructions is performed during MI to MC lowering
and is similar to the expansion performed by the GNU Assembler.

Differential Revision: https://reviews.llvm.org/D41949
Patch by Mario Werner.

llvm-svn: 330224

When we skip bitcasts while looking for GEP in LoadSoreVectorizer
we should also verify that the type is sized otherwise we assert

Differential Revision: https://reviews.llvm.org/D45709

llvm-svn: 330221

Summary:
Add an LLVM intrinsic for type discriminated event logging with XRay.
Similar to the existing intrinsic for custom events, but also accepts
a type tag argument to allow plugins to be aware of different types
and semantically interpret logged events they know about without
choking on those they don't.

Relies on a symbol defined in compiler-rt patch D43668. I may wait
to submit before I can see demo everything working together including
a still to come clang patch.

Reviewers: dberris, pelikan, eizan, rSerge, timshen

Subscribers: llvm-commits

Differential Revision: https://reviews.llvm.org/D45633

llvm-svn: 330219

Summary:
It was not easy to provide a test case for D45648 (rL330079) because the bug
didn't manifest itself in the set of currently valid IRs. Added an assertion to
check this faster, thanks to @dblaikie's suggestion.

Reviewers: dblaikie

Subscribers: jfb, dschuff, sbc100, jgravelle-google, llvm-commits, dblaikie

Differential Revision: https://reviews.llvm.org/D45711

llvm-svn: 330217

GetArgumentVector (or GetCommandLineArguments) is very Windows-specific.
I think it doesn't make much sense to provide that function from sys::Process.

I also made a change so that the function takes a BumpPtrAllocator
instead of a SpecificBumpPtrAllocator. The latter is the class to call
dtors, but since char * is trivially destructible, we should use the
former class.

Differential Revision: https://reviews.llvm.org/D45641

llvm-svn: 330216

Fixes PR36564.

llvm-svn: 330215

The DBI stream contains a list of module descriptors.  At the
beginning of each descriptor is a structure representing the first
section contribution in the output file for that module.  LLD
currently doesn't fill out this structure at all, but link.exe
does.  So as a precursor to emitting this data in LLD, we first
need a way to dump it so that it can be checked.

This patch adds support for the dumping, and verifies via a test
that LLD emits bogus information.

llvm-svn: 330208

llvm-svn: 330204

llvm-svn: 330203

Stack addressing needs addressing modes that provide an offset field
immediately following the frame index. An initializer from a non-stack
addressing could force the stack address to use a form that does not
provide an offset field.

llvm-svn: 330191

LLVMDump* functions are available in Release builds too.

Patch by Brenton Bostick.

Differential Revision: https://reviews.llvm.org/D44600

llvm-svn: 330189

Make the shrink wrapping pass pay attention to uses/defs of the stack pointer.

Differential revision: https://reviews.llvm.org/D45524

llvm-svn: 330183

This reverts r330175. There are still stage3/stage4 miscompares.

llvm-svn: 330180

Split VCMP/VMAX/VMIN instructions off to WriteFCmp and VCOMIS instructions off to WriteFCom instead of assuming they match WriteFAdd

Differential Revision: https://reviews.llvm.org/D45656

llvm-svn: 330179

llvm-svn: 330178

llvm-svn: 330176

One more, hopefully the last, bug is fixed: when forming UsesToRewrite
we should ignore phi operands coming from edges that we want to delete.

This reverts r329910.

llvm-svn: 330175

Older compiler issued '#' instead of ';'

llvm-svn: 330173

Differential Revision: https://reviews.llvm.org/D45714

llvm-svn: 330172

Apparently, DebugInfoFinder::processCompileUnit doesn't process all
of the possible kinds of DIImportedEntit'ies, e.g. DIGlobalVariable's.

Previously introduced `llvm_unreachable` is therefore incorrect.
Removing it here.

llvm-svn: 330167