Commit r355068 "Fix IR/Analysis layering issue with OptBisect" uses the
template

   return Gate.isEnabled() && !Gate.shouldRunPass(this, getDescription(...));

for all pass kinds. For the RegionPass, it left out the not operator,
causing region passes to be skipped as soon as a pass gate is used.

llvm-svn: 355733

llvm-svn: 355732

When matching half of the build_vector to a load, there could still be
a hidden dependency on the other half of the build_vector the pattern
wouldn't detect. If there was an additional chain dependency on the
other value, a cycle could be introduced.

I don't think a tablegen pattern is capable of matching the necessary
conditions, so move this into PreprocessISelDAG. Check isPredecessorOf
for the other value to avoid a cycle. This has a warning that it's
expensive, so this should probably be moved into an MI pass eventually
that will have more freedom to reorder instructions to help match
this. That is currently complicated by the lack of a computeKnownBits
type mechanism for the selected function.

llvm-svn: 355731

After recent changes, Host is now dependency-free.

llvm-svn: 355730

Removed not required service variable for the debug info.

llvm-svn: 355729

This avoids breaking possible value dependencies when sorting loads by
offset.

AMDGPU has some load instructions that write into the high or low bits
of the destination register, and have a tied input for the other input
bits. These can easily have the same base pointer, but be a swizzle so
the high address load needs to come first. This was inserting glue
forcing the opposite ordering, producing a cycle the InstrEmitter
would assert on. It may be potentially expensive to look for the
dependency between the other loads, so just skip any where this could
happen.

Fixes bug 40936 by reverting r351379, which added a hacky attempt to
fix this by adding chains in this case, which I think was just working
around broken glue before the InstrEmitter. The core of the patch is
re-implementing the fix for that problem.

llvm-svn: 355728

llvm-svn: 355727

llvm-svn: 355726

This broke the windows bots.

This reverts commit 28302c66.

llvm-svn: 355725

Also fix a few cases that weren't testing what they were supposed to.

llvm-svn: 355724

This is only called in contexts that are verifying the chain itself,
and the query itself is only asking about the address.

llvm-svn: 355723

This was checking the wrong operands for the base register and the
offsets. The indexes are shifted by the number of output registers
from the machine instruction definition, and the chain is moved to the
end.

llvm-svn: 355722

There is nontrivial bug caused in lld that I need to further
investigate. Meanwhile, I'll revert this.

This reverts commit 8297e93480c636dc90fd14653c5a66406193363f.

llvm-svn: 355721

Summary:
Added a new category of checkers for non-determinism. Added a checker for non-determinism
caused due to sorting containers with pointer-like elements.

Reviewers: NoQ, george.karpenkov, whisperity, Szelethus

Reviewed By: NoQ, Szelethus

Subscribers: Charusso, baloghadamsoftware, jdoerfert, donat.nagy, dkrupp, martong, dblaikie, MTC, Szelethus, mgorny, xazax.hun, szepet, rnkovacs, a.sidorin, mikhail.ramalho, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D50488

llvm-svn: 355720

Summary:
If the LLVM module shows that it has debug info, but the file is
actually empty and the real debug info is not emitted, the ptxas tool
emits error 'Debug information not found in presence of .target debug'.
We need at leas one empty debug section to silence this message. Section
`.debug_loc` is not emitted for PTX and we can emit empty `.debug_loc`
section if `debug` option was emitted.

Reviewers: tra

Subscribers: jholewinski, aprantl, llvm-commits

Differential Revision: https://reviews.llvm.org/D57250

llvm-svn: 355719

r355348 uses builtins without proper guards, breaking the test on
various platforms.

llvm-svn: 355718

On Linux (and other non-Darwin platforms) we need to initialize
interceptors. Since tsan_libdispatch.cc is compiled optionally, add a
weak default implementation of `InitializeLibdispatchInterceptors`.

Reviewed By: dvyukov

Differential Revision: https://reviews.llvm.org/D59113

llvm-svn: 355717

Summary: This pattern is sometime created after legalization.

Reviewers: efriedma, spatel, RKSimon, zvi, bkramer

Subscribers: llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D58874

llvm-svn: 355716

Patch by Enna1!

Differential Revision: https://reviews.llvm.org/D58756

llvm-svn: 355715

many valnos.

Recently we found compile time out problem in several cases when
SpeculativeLoadHardening was enabled. The significant compile time was spent
in register coalescing pass, where register coalescer tried to join many other
live intervals with some very large live intervals with many valnos.

Specifically, every time JoinVals::mapValues is called, computeAssignment will
be called by getNumValNums() times of the target live interval. If the large
live interval has N valnos and has N copies associated with it, trying to
coalescing those copies will at least cost N^2 complexity.

The patch adds some limit to the effort trying to join those very large live
intervals with others. By default, for live interval with > 100 valnos, and
when it has been coalesced with other live interval by more than 100 times,
we will stop coalescing for the live interval anymore. That put a compile
time cap for the N^2 algorithm and effectively solves the compile time
problem we saw.

Differential revision: https://reviews.llvm.org/D59143

llvm-svn: 355714

llvm-svn: 355713

llvm-svn: 355712

Re-ran lldb-inst on the API folder to insert missing LLDB_RECORD_DUMMY
macros.

llvm-svn: 355711

Extend lldb-instr to insert LLDB_RECORD_DUMMY macros for currently
unsupported signatures (void and function pointers).

llvm-svn: 355710

Add a macro that doesn't actually record anything but still toggles the
API boundary. Removing just the register macros for lldb::thread_t
wasn't sufficient on NetBSD because the serialization logic needed the
underlying type to be complete.

This macro should be used by functions that are currently unsupported,
as they might trip the API boundary logic. This should be easy using the
lldb-instr tool.

llvm-svn: 355709

I changed the variable to an unsigned to get rid of a signed and
unsigned compare without realizing the value could be negative. This
fixes the assert instead.

llvm-svn: 355708

The indexed variant of vfmal.f16 and vfmsl.f16
instructions use the uppser bits of the indexed
operand to store the index (1 bit for the double
variant, 2 bits for the quad).

This limits the usable registers to d0 - d7 or
s0 - s15. This patch enforces this limitation.

Differential Revision: https://reviews.llvm.org/D59021

llvm-svn: 355707

The last round of logging taught us that when the test fails, lldb
is indeed aware of the thread it's failing to associate to a given
queue. Add more logging to try to figure out why the thread and the
queue do not appear related to the Queue APIs.

llvm-svn: 355706

In the commited testfile, macro expansion (the one implemented for the plist
output) runs into an infinite recursion. The issue originates from the algorithm
being faulty, as in

#define value REC_MACRO_FUNC(value)

the "value" is being (or at least attempted) expanded from the same macro.

The solved this issue by gathering already visited macros in a set, which does
resolve the crash, but will result in an incorrect macro expansion, that would
preferably be fixed down the line.

Patch by Tibor Brunner!

Differential Revision: https://reviews.llvm.org/D57891

llvm-svn: 355705

Asserting on invalid input isn't very nice, hence the patch to emit an error
instead.

This is the first of many patches to overhaul the way we handle checker options.

Differential Revision: https://reviews.llvm.org/D57850

llvm-svn: 355704

In D55734, we implemented a far more general way of describing taint propagation
rules for functions, like being able to specify an unlimited amount of
source and destination parameters. Previously, we didn't have a particularly
elegant way of expressing the propagation rules for functions that always return
(either through an out-param or return value) a tainted value. In this patch,
we model these functions similarly to other ones, by assigning them a
TaintPropagationRule that describes that they "create a tainted value out of
nothing".

The socket C function is somewhat special, because for certain parameters (for
example, if we supply localhost as parameter), none of the out-params should
be tainted. For this, we added a general solution of being able to specify
custom taint propagation rules through function pointers.

Patch by Gábor Borsik!

Differential Revision: https://reviews.llvm.org/D59055

llvm-svn: 355703

This function is always used in a context where its result was also
negated, which made for confusing naming and code.

llvm-svn: 355702

r355343 was landed and was reverted in r355363 due to build breakage.
This patch adds Linux/Windows support on top of r355343.

In this patch, Darwin should be working with testing case. Linux should be working,
I will enable the testing case in a follwup diff. Windows/Other should be building.
Correct implementation for Other platforms will be added.

Thanks David for reviewing the original diff, helping me with issues on Linux, and
giving suggestions for adding support for Other platforms.

llvm-svn: 355701

The type of wide string literals varies depending on the target.

llvm-svn: 355700

llvm-svn: 355699

r355322 fixed this, however is being reverted due to concerns with
enabling it in other modes.

Change-Id: I6a939b7469b8fa196d5871a627eb2330dbd30f29
llvm-svn: 355698

This reverts commit 24400dafe16716f28cd0e7e5fa6e004c0e50686a.

llvm-svn: 355697

llvm-readelf prints relocation addends as:

  <symbol value>[+-]<absolute addend>

where [+-] is determined from whether addend is less than zero or not.
However, it does not print the +/- if there is no symbol, which meant
that negative addends became their positive value with no indication
that this had happened. This patch stops the absolute conversion when
addends are negative and there is no associated symbol.

Reviewed by: Higuoxing, mattd, MaskRay

Differential Revision: https://reviews.llvm.org/D59095

llvm-svn: 355696

llvm-svn: 355695

From the Python subprocess docs:

   If shell is True, it is recommended to pass args as a string rather than as
   a sequence.

   [...]

   If args is a sequence, the first item specifies the command string, and any
   additional items will be treated as additional arguments to the shell itself.

Prior to this change, the `--version` would be passed to the shell, not to
a potential gn binary on $PATH, and running `gn` without any arguments makes
it exit with an exit code != 0, so the script would think that there wasn't
a working gn binary on $PATH.

Fix this by following the documentation's recommendation of using a string
now that we pass shell=True. I tested this on macOS and Windows, each with
the three cases of

- no gn on PATH (should run gn downloaded by get.py if present,
  else suggest running get.py)
- broken gn wrapper on PATH (should behave like the previous item)
- working gn on PATH (should use gn on PATH)

llvm-svn: 355694