Now compiler defines 5 sets of constants to represent rounding mode.
These are:
1. `llvm::APFloatBase::roundingMode`. It specifies all 5 rounding modes
defined by IEEE-754 and is used in `APFloat` implementation.
2. `clang::LangOptions::FPRoundingModeKind`. It specifies 4 of 5 IEEE-754
rounding modes and a special value for dynamic rounding mode. It is used
in clang frontend.
3. `llvm::fp::RoundingMode`. Defines the same values as
`clang::LangOptions::FPRoundingModeKind` but in different order. It is
used to specify rounding mode in in IR and functions that operate IR.
4. Rounding mode representation used by `FLT_ROUNDS` (C11, 5.2.4.2.2p7).
Besides constants for rounding mode it also uses a special value to
indicate error. It is convenient to use in intrinsic functions, as it
represents platform-independent representation for rounding mode. In this
role it is used in some pending patches.
5. Values like `FE_DOWNWARD` and other, which specify rounding mode in
library calls `fesetround` and `fegetround`. Often they represent bits
of some control register, so they are target-dependent. The same names
(not values) and a special name `FE_DYNAMIC` are used in
`#pragma STDC FENV_ROUND`.
The first 4 sets of constants are target independent and could have the
same numerical representation. It would simplify conversion between the
representations. Also now `clang::LangOptions::FPRoundingModeKind` and
`llvm::fp::RoundingMode` do not contain the value for IEEE-754 rounding
direction `roundTiesToAway`, although it is supported natively on
some targets.
This change defines all the rounding mode type via one `llvm::RoundingMode`,
which also contains rounding mode for IEEE rounding direction `roundTiesToAway`.
Differential Revision: https://reviews.llvm.org/D77379
Instead, represent the mask as out-of-line data in the instruction. This
should be more efficient in the places that currently use
getShuffleVector(), and paves the way for further changes to add new
shuffles for scalable vectors.
This doesn't change the syntax in textual IR. And I don't currently plan
to change the bitcode encoding in this patch, although we'll probably
need to do something once we extend shufflevector for scalable types.
I expect that once this is finished, we can then replace the raw "mask"
with something more appropriate for scalable vectors. Not sure exactly
what this looks like at the moment, but there are a few different ways
we could handle it. Maybe we could try to describe specific shuffles.
Or maybe we could define it in terms of a function to convert a fixed-length
array into an appropriate scalable vector, using a "step", or something
like that.
Differential Revision: https://reviews.llvm.org/D72467
This is the Waymarking algorithm implemented as an independent utility.
The utility is operating on a range of sequential elements.
First we "tag" the elements, by calling `fillWaymarks`.
Then we can "follow" the tags from every element inside the tagged
range, and reach the "head" (the first element), by calling
`followWaymarks`.
Differential Revision: https://reviews.llvm.org/D74415
Also add a test case to wasm-ld that asserts without this change.
Internally wasm-ld builds a StringMap of exported functions and it seems
like allowing empty string in the set is preferable to adding checks.
This assert looks like it was most likely just a historical accident.
It started life here purely to support InputLanguagesSet:
eeac27e38c5c567d63bbfa5410620d955696491b
Then got extracted here:
e57a4033385c5976cbb17af1e962b1224a61183b
Then got moved to AST here
5c48bae209bcbd261886f63abac695b1e30544e6
With the `InLang` paramater name still intact which suggested is
InputLanguagesSet origins.
Differential Revision: https://reviews.llvm.org/D74589
This is safe if the iterator type is a pointer and the comparator is
stateless. The enable_if pattern I'm adding here only uses
array_pod_sort for the default comparator (std::less).
Using array_pod_sort has a potential performance impact, but I didn't
notice anything when testing clang. Sorting doesn't seem to be on the
hot path anywhere in LLVM.
Shrinks Release+Asserts clang by 73k.
Without this some instances of copy construction would use the
converting constructor & lead to the destination function_ref referring
to the source function_ref instead of the underlying functor.
Discovered in feedback from 857bf5da35af8e1f9425e1865dab5f5fce5e38f2
Thanks to Johannes Doerfert, Arthur O'Dwyer, and Richard Smith for the
discussion and debugging.
This can avoid all sorts of mistakes with implicit conversion
(indirectly) to int, etc. I'm quite surprise there aren't any things to
fixup with this - but I guess most uses of function_ref aren't
optional/nullable.
Summary:
This patch introduces command-line support for the Armv8.6-a architecture and assembly support for BFloat16. Details can be found
https://community.arm.com/developer/ip-products/processors/b/processors-ip-blog/posts/arm-architecture-developments-armv8-6-a
in addition to the GCC patch for the 8..6-a CLI:
https://gcc.gnu.org/legacy-ml/gcc-patches/2019-11/msg02647.html
In detail this patch
- march options for armv8.6-a
- BFloat16 assembly
This is part of a patch series, starting with command-line and Bfloat16
assembly support. The subsequent patches will upstream intrinsics
support for BFloat16, followed by Matrix Multiplication and the
remaining Virtualization features of the armv8.6-a architecture.
Based on work by:
- labrinea
- MarkMurrayARM
- Luke Cheeseman
- Javed Asbar
- Mikhail Maltsev
- Luke Geeson
Reviewers: SjoerdMeijer, craig.topper, rjmccall, jfb, LukeGeeson
Reviewed By: SjoerdMeijer
Subscribers: stuij, kristof.beyls, hiraditya, dexonsmith, danielkiss, cfe-commits, llvm-commits
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D76062
This is the Waymarking algorithm implemented as an independent utility.
The utility is operating on a range of sequential elements.
First we "tag" the elements, by calling `fillWaymarks`.
Then we can "follow" the tags from every element inside the tagged
range, and reach the "head" (the first element), by calling
`followWaymarks`.
Differential Revision: https://reviews.llvm.org/D74415
advanceToLowerBound moves an iterator to the first bit set at, or after,
the given index. This can be faster than doing IntervalMap::find.
rdar://60046261
Differential Revision: https://reviews.llvm.org/D76466
Avoid making a heap allocation when constructing a CoalescingBitVector.
This reduces time spent in LiveDebugValues when compiling sqlite3 by
700ms (0.5% of the total User Time).
rdar://60046261
Differential Revision: https://reviews.llvm.org/D76465
Summary:
These implement the usual IEEE-style floating point comparison
semantics, e.g. +0.0 == -0.0 and all operators except != return false
if either argument is NaN.
Subscribers: arsenm, jvesely, nhaehnle, hiraditya, dexonsmith, kerbowa, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D75237
Summary:
We already have overloaded binary arithemetic operators so you can write
A+B etc. This patch lets you write -A instead of neg(A).
Subscribers: hiraditya, dexonsmith, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D75236
DenseMap requires two sentinel values for keys: empty and tombstone
values. To avoid undefined behavior, LLVM aligns the two sentinel
pointers to alignof(T). This requires T to be complete, which is
needlessly restrictive.
Instead, assume that DenseMap pointer keys have a maximum alignment of
4096, and use the same sentinel values for all pointer keys. The new
sentinels are:
empty: static_cast<uintptr_t>(-1) << 12
tombstone: static_cast<uintptr_t>(-2) << 12
These correspond to the addresses of -4096 and -8192. Hopefully, such a
key is never inserted into a DenseMap.
I encountered this while looking at making clang's SourceManager not
require FileManager.h, but it has several maps keyed on classes defined
in FileManager.h. FileManager depends on various LLVM FS headers, which
cumulatively take ~200ms to parse, and are generally not needed.
Reviewed By: hans
Differential Revision: https://reviews.llvm.org/D75301
Add CoalescingBitVector to ADT. This is part 1 of a 3-part series to
address a compile-time explosion issue in LiveDebugValues.
---
CoalescingBitVector is a bitvector that, under the hood, relies on an
IntervalMap to coalesce elements into intervals.
CoalescingBitVector efficiently represents sets which predominantly
contain contiguous ranges (e.g. the VarLocSets in LiveDebugValues,
which are very long sequences that look like {1, 2, 3, ...}). OTOH,
CoalescingBitVector isn't good at representing sets with lots of gaps
between elements. The first N coalesced intervals of set bits are stored
in-place (in the initial heap allocation).
Compared to SparseBitVector, CoalescingBitVector offers more predictable
performance for non-sequential find() operations. This provides a
crucial speedup in LiveDebugValues.
Differential Revision: https://reviews.llvm.org/D74984
With this --shuffle-sections=seed produces the same result in every
host.
Reviewed By: grimar, MaskRay
Differential Revision: https://reviews.llvm.org/D74971
This patch adds DenseMapInfo<> support for BitVector and SmallBitVector.
This is part of https://reviews.llvm.org/D71775, where a BitVector is used as a thread affinity mask.
This patch is replacements missed in my last change doing this across LLVM.
No functional change, although I think there was a missing typename
in struct conjunction that is now fixed.
Summary:
As far as I can tell, the SFINAE was broken; there is no such thing as
std::is_trivially_constructible<T>::type.
Subscribers: dexonsmith, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74380
In a previous patch I changed `std::decay<T>::type` to `std::decay<T>`
rather than `std::decay_t<T>`. This seems to have broken the build
*only for clang-cl*. I don't know why.
Submitting with post-commit review because this is an obvious fix for a
build breakage and we've verified that it fixes the breakage.
Summary:
Simplifies the C++11-style "-> decltype(...)" return-type deduction.
Note that you have to be careful about whether the function return type
is `auto` or `decltype(auto)`. The difference is that bare `auto`
strips const and reference, just like lambda return type deduction. In
some cases that's what we want (or more likely, we know that the return
type is a value type), but whenever we're wrapping a templated function
which might return a reference, we need to be sure that the return type
is decltype(auto).
No functional change.
Subscribers: dexonsmith, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74383
As detailed on PR43943, we're seeing static analyzer use after move warnings in the iplist_impl move constructor/operator as they call std::move to both the TraitsT and IntrusiveListT base classes.
As suggested by @dexonsmith this patch casts the moved value to the base classes to silence the warnings.
Differential Revision: https://reviews.llvm.org/D74062
AMDGPU and x86 at least both have separate controls for whether
denormal results are flushed on output, and for whether denormals are
implicitly treated as 0 as an input. The current DAGCombiner use only
really cares about the input treatment of denormals.
The fix in b3d7d1061dc375bb5ea725e6597382fcd37f41d6 compiled nicely,
but didn't link because at least the VS 2017 version I use doesn't
have the builtin yet. Instead, make use of the builtin with MSVC
conditional on VS 2019 or later.
Otherwise Visual Studio 2017 will complain about
llvm::StringRef::strlen not being constexpr:
StringRef.h(80): error C3615: constexpr function 'llvm::StringRef::strLen' cannot result in a constant expression
StringRef.h(84): note: failure was caused by call of undefined function or one not declared 'constexpr'
This is the first of multiple parts to make OpenMP context/trait
handling reusable and generic. This patch was originally part of D71830
but with the unit tests it can be tested independently.
This patch implements an almost complete handling of OpenMP
contexts/traits such that we can reuse most of the logic in Flang
through the OMPContext.{h,cpp} in llvm/Frontend/OpenMP.
All but construct SIMD specifiers, e.g., inbranch, and the device ISA
selector are define in llvm/lib/Frontend/OpenMP/OMPKinds.def. From
these definitions we generate the enum classes TraitSet,
TraitSelector, and TraitProperty as well as conversion and helper
functions in llvm/lib/Frontend/OpenMP/OMPContext.{h,cpp}.
The OpenMP context is now an explicit object (see `struct OMPContext`).
This is in anticipation of construct traits that need to be tracked. The
OpenMP context, as well as the VariantMatchInfo, are basically made up
of a set of active or respectively required traits, e.g., 'host', and an
ordered container of constructs which allows duplication. Matching and
scoring is kept as generic as possible to allow easy extension in the
future.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D71847
Summary:
The return type of 'PointerUnion::is' has been 'int' since it was first
added in March 2009, in SVN r67987, or
https://github.com/llvm/llvm-project/commit/a9c6de15fb3.
The only other change to this member function was a clang-format applied
in December 2015, in SVN r256513, or
https://github.com/llvm/llvm-project/commit/548a49aacc0.
However, since the return value is the result of a `==` comparison, an
implicit cast must be made converting the boolean result to an `int`.
Change the return type to `bool` to remove the need for such a cast.
Test Plan:
I ran llvm-project `check-all` under ASAN, no failures were reported
(other than obviously unrelated tests that were already failing in
ASAN buildbots).
Reviewers: gribozavr, gribozavr2, rsmith, bkramer, dblaikie
Subscribers: dexonsmith, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D73836
Saves 2427 unneeded includes of TypeSize.h, which instantiates
std::tie<uint64_t, bool>, which instantiates std::tuple<uint64_t, bool>,
which is slow.
I'll remove the tie in a follow-up, since it's just for operator==.
