Changes

[lldb] More memory allocation test fixes

XFAIL nodefaultlib.cpp on darwin - the test does not pass there

XFAIL TestGdbRemoteMemoryAllocation on windows - memory is allocated
with incorrect permissions

[MBP] Add whole chain to BlockFilterSet instead of individual BB

Currently we add individual BB to BlockFilterSet if its frequency satisfies

LoopFreq / Freq <= LoopToColdBlockRatio

LoopFreq is edge frequency from outside to loop header.
LoopToColdBlockRatio is a command line parameter.

It doesn't make sense since we always layout whole chain, not individual BBs.

It may also cause a tricky problem. Sometimes it is possible that the LoopFreq
of an inner loop is smaller than LoopFreq of outer loop. So a BB can be in
BlockFilterSet of inner loop, but not in BlockFilterSet of outer loop,
like .cold in the test case. So it is added to the chain of inner loop. When
work on the outer loop, .cold is not added to BlockFilterSet, so the edge to
successor .problem is not counted in UnscheduledPredecessors of .problem chain.
But other blocks in the inner loop are added BlockFilterSet, so the whole inner
loop chain can be layout, and markChainSuccessors is called to decrease
UnscheduledPredecessors of following chains. markChainSuccessors calls
markBlockSuccessors for every BB, even it is not in BlockFilterSet, like .cold,
so .problem chain's UnscheduledPredecessors is decreased, but this edge was not
counted on in fillWorkLists, so .problem chain's UnscheduledPredecessors
becomes 0 when it still has an unscheduled predecessor .pred! And it causes
problems in following various successor BB selection algorithms.

Differential Revision: https://reviews.llvm.org/D89088

[llvm] Set the default for -bbsections-cold-text-prefix to .text.split.

After using this for a while, we find that it is generally useful to
have it set to .text.split. by default, removing the need for an
additional -mllvm option.

Differential Revision: https://reviews.llvm.org/D88997

[clang] Add -fc++-abi= flag for specifying which C++ ABI to use

This implements the flag proposed in RFC http://lists.llvm.org/pipermail/cfe-dev/2020-August/066437.html.

The goal is to add a way to override the default target C++ ABI through
a compiler flag. This makes it easier to test and transition between different
C++ ABIs through compile flags rather than build flags.

In this patch:
- Store `-fc++-abi=` in a LangOpt. This isn't stored in a
  CodeGenOpt because there are instances outside of codegen where Clang
  needs to know what the ABI is (particularly through
  ASTContext::createCXXABI), and we should be able to override the
  target default if the flag is provided at that point.
- Expose the existing ABIs in TargetCXXABI as values that can be passed
  through this flag.
  - Create a .def file for these ABIs to make it easier to check flag
    values.
  - Add an error for diagnosing bad ABI flag values.

Differential Revision: https://reviews.llvm.org/D85802

[mlir] Linalg refactor for using "bufferize" terminology.

Part of the refactor discussed in:
https://llvm.discourse.group/t/what-is-the-strategy-for-tensor-memref-conversion-bufferization/1938/17

Differential Revision: https://reviews.llvm.org/D89261

[mlir] Rename ShapeTypeConversion to ShapeBufferize

Once we have tensor_to_memref ops suitable for type materializations,
this pass can be split into a generic type conversion pattern.

Part of the refactor discussed in:
https://llvm.discourse.group/t/what-is-the-strategy-for-tensor-memref-conversion-bufferization/1938/17

Differential Revision: https://reviews.llvm.org/D89258

[mlir] Refactor code out of BufferPlacement.cpp

Now BufferPlacement.cpp doesn't depend on Bufferize.h.

Part of the refactor discussed in:
https://llvm.discourse.group/t/what-is-the-strategy-for-tensor-memref-conversion-bufferization/1938/17

Differential Revision: https://reviews.llvm.org/D89268

[mlir][bufferize] Rename BufferAssignment* to Bufferize*

Part of the refactor discussed in:
https://llvm.discourse.group/t/what-is-the-strategy-for-tensor-memref-conversion-bufferization/1938/17

Differential Revision: https://reviews.llvm.org/D89271

[mlir] Fix some style comments from D89268

That change was a pure move, so split out the stylistic changes into
this patch.

Differential Revision: https://reviews.llvm.org/D89272

[llvm] Update default cutoff threshold for machine function splitter.

Based on internal testing at Google we found that setting the profile
summary cutoff threshold to 999950 yields the best results in terms of
itlb and icache metrics (as observed on Intel CPUs).

*default* = Split out code if no profile count available for block
*size-%*  = The fraction of bytes split out of .text and .text.hot
*itlb*    = Misses per kilo instructions (MPKI) for itlb
*icache*  = Misses per kilo instructions (MPKI) for L1 icache

Search1

| cutoff  | size-%  | itlb      | icache  |
|---------|---------|-----------|---------|
| default | 42.5861 | 0.0822151 | 2.46363 |
|  999999 | 44.9350 | 0.0767194 | 2.44416 |
|  999950 | 50.0660 |  0.075744 |  2.4091 |
|  999500 | 56.9158 |  0.082564 |  2.4188 |
|  995000 | 63.8625 | 0.0814927 | 2.42832 |
|  990000 | 71.7314 |  0.106906 | 2.57785 |

Search2

| cutoff  | size-% | itlb     | icache  |
|---------|--------|----------|---------|
| default | 2.8845 | 0.626712 | 4.73245 |
|  999999 | 3.3291 | 0.602309 | 4.70045 |
|  999950 | 3.8577 | 0.587842 | 4.71632 |
|  999500 | 4.4170 |  0.63577 | 4.68351 |
|  995000 | 5.1020 | 0.657969 | 4.82272 |
|  990000 | 5.7153 | 0.719122 | 5.39496 |

Differential Revision: https://reviews.llvm.org/D89085

clang/Basic: Replace ContentCache::getBuffer with Optional semantics

Remove `ContentCache::getBuffer`, which always returned a
dereferenceable `MemoryBuffer*` and had a `bool*Invalid` out parameter,
and replace it with:

- `ContentCache::getBufferOrNone`, which returns
  `Optional<MemoryBufferRef>`. This is the new API that consumers should
  use. Later it could be renamed to `getBuffer`, but intentionally using
  a different name to root out any unexpected callers.
- `ContentCache::getBufferPointer`, which returns `MemoryBuffer*` with
  "optional" semantics. This is `private` to avoid growing callers and
  `SourceManager` has temporarily been made a `friend` to access it.
  Later paches will update the transitive callers to not need a raw
  pointer, and eventually this will be deleted.

No functionality change intended here.

Differential Revision: https://reviews.llvm.org/D89348

Make header self-contained. NFC.

[mlir][Linalg] Rethink fusion of linalg ops with reshape ops.

The current fusion on tensors fuses reshape ops with generic ops by
linearizing the indexing maps of the fused tensor in the generic
op. This has some limitations
- It only works for static shapes
- The resulting indexing map has a linearization that would be
  potentially prevent fusion later on (for ex. tile + fuse).

Instead, try to fuse the reshape consumer (producer) with generic op
producer (consumer) by expanding the dimensionality of the generic op
when the reshape is expanding (folding).  This approach conflicts with
the linearization approach. The expansion method is used instead of
the linearization method.

Further refactoring that changes the fusion on tensors to be a
collection of patterns.

Differential Revision: https://reviews.llvm.org/D89002