This patch optionally replaces the CRT allocator (i.e., malloc and free) with rpmalloc (mixed public domain licence/MIT licence) or snmalloc (MIT licence) or mimalloc (MIT licence). Please note that the source code for these allocators must be available outside of LLVM's tree.
To enable, use `cmake ... -DLLVM_INTEGRATED_CRT_ALLOC=D:/git/rpmalloc -DLLVM_USE_CRT_RELEASE=MT` where `D:/git/rpmalloc` has already been git clone'd from `https://github.com/mjansson/rpmalloc`. The same applies to snmalloc and mimalloc.
When enabled, the allocator will be embeded (statically linked) into the LLVM tools & libraries. This currently only works with the static CRT (/MT), although using the dynamic CRT (/MD) could potentially work as well in the future.
When enabled, this changes the memory stack from:
new/delete -> MS VC++ CRT malloc/free -> HeapAlloc -> VirtualAlloc
to:
new/delete -> {rpmalloc|snmalloc|mimalloc} -> VirtualAlloc
The goal of this patch is to bypass the application's global heap - which is thread-safe thus inducing locking - and instead take advantage of a modern lock-free, thread cache, allocator. On a 6-core Xeon Skylake we observe a 2.5x decrease in execution time when linking a large scale application with LLD and ThinLTO (12 min 20 sec -> 5 min 34 sec), when all hardware threads are being used (using LLD's flag /opt:lldltojobs=all). On a dual 36-core Xeon Skylake with all hardware threads used, we observe a 24x decrease in execution time (1 h 2 min -> 2 min 38 sec) when linking a large application with LLD and ThinLTO. Clang build times also see a decrease in the range 5-10% depending on the configuration.
Differential Revision: https://reviews.llvm.org/D71786
When dealing with dialects that will results in function calls to
external libraries, it is important to be able to handle maps as some
dialects may require mapped data. Before this patch, the detection of
whether normalization can apply or not, operations are compared to an
explicit list of operations (`alloc`, `dealloc`, `return`) or to the
presence of specific operation interfaces (`AffineReadOpInterface`,
`AffineWriteOpInterface`, `AffineDMAStartOp`, or `AffineDMAWaitOp`).
This patch add a trait, `MemRefsNormalizable` to determine if an
operation can have its `memrefs` normalized.
This trait can be used in turn by dialects to assert that such
operations are compatible with normalization of `memrefs` with
nontrivial memory layout specification. An example is given in the
literal tests.
Differential Revision: https://reviews.llvm.org/D86236
Currently we bail out early for strlen calls with a GEP operand, if none
of the GEP specific optimizations fire. But there could be later
optimizations that still apply, which we currently miss out on.
An example is that we do not apply the following optimization
strlen(x) == 0 --> *x == 0
Unless I am missing something, there seems to be no reason for bailing
out early there.
Fixes PR47149.
Reviewed By: lebedev.ri, xbolva00
Differential Revision: https://reviews.llvm.org/D85886
The function was returning an incorrect (empty) value on the first
invocation. Given that this only affected the first invocation, this
bug/typo went mostly unaffected. DW_AT_const_value were particularly
badly affected by this as the GetByteSize call is
SymbolFileDWARF::ParseVariableDIE is likely to be the first call of this
function, and its effects cannot be undone by retrying.
Depends on D86348.
Differential Revision: https://reviews.llvm.org/D86436
This applies the same fix that D84748 did for macro definitions.
Appropriate include path is now automatically set for all libraries
which link against gtest targets, which avoids the need to set
include_directories in various parts of the project.
Differential Revision: https://reviews.llvm.org/D86616
Motivating use case is ".cu.cc" extensions used in some bazel projects.
Alternative is to work around this with IncludeIsMainRegex in styles.
I proposed this approach because it seems like a better default.
Differential Revision: https://reviews.llvm.org/D86597
Class-level static constexpr variables can have both DW_AT_const_value
(in the "declaration") and a DW_AT_location (in the "definition")
attributes. Our code was trying to handle this, but it was brittle and
hard to follow (and broken) because it was processing the attributes in
the order in which they were found.
Refactor the code to make the intent clearer -- DW_AT_location trumps
DW_AT_const_value, and fix the bug which meant that we were not
displaying these variables properly (the culprit was the delayed parsing
of the const_value attribute due to a need to fetch the variable type.
Differential Revision: https://reviews.llvm.org/D86615
It isn't very wise to pass an assembly file to the compiler and tell it to compile as a C file and hope that the compiler recognizes it as assembly instead.
Simply don't mark the file as C and CMake will recognize the rest.
This was attempted earlier in https://reviews.llvm.org/D85706, but reverted due to architecture issues on Apple.
Subsequent digging revealed a similar change was done earlier for libunwind in https://reviews.llvm.org/rGb780df052dd2b246a760d00e00f7de9ebdab9d09.
Afterwards workarounds were added for MinGW and Apple:
* https://reviews.llvm.org/rGb780df052dd2b246a760d00e00f7de9ebdab9d09
* https://reviews.llvm.org/rGd4ded05ba851304b26a437896bc3962ef56f62cb
The workarounds in libunwind and compiler-rt are unified and comments added pointing to each other.
The workaround is updated to only be used for MinGW for CMake versions before 3.17, which fixed the issue (https://gitlab.kitware.com/cmake/cmake/-/merge_requests/4287).
Additionally fixed Clang not being passed as the assembly compiler for compiler-rt runtime build.
Example error:
[525/634] Building C object lib/tsan/CMakeFiles/clang_rt.tsan-aarch64.dir/rtl/tsan_rtl_aarch64.S.o
FAILED: lib/tsan/CMakeFiles/clang_rt.tsan-aarch64.dir/rtl/tsan_rtl_aarch64.S.o
/opt/tooling/drive/host/bin/clang --target=aarch64-linux-gnu -I/opt/tooling/drive/llvm/compiler-rt/lib/tsan/.. -isystem /opt/tooling/drive/toolchain/opt/drive/toolchain/include -x c -Wall -Wno-unused-parameter -fno-lto -fPIC -fno-builtin -fno-exceptions -fomit-frame-pointer -funwind-tables -fno-stack-protector -fno-sanitize=safe-stack -fvisibility=hidden -fno-lto -O3 -gline-tables-only -Wno-gnu -Wno-variadic-macros -Wno-c99-extensions -Wno-non-virtual-dtor -fPIE -fno-rtti -Wframe-larger-than=530 -Wglobal-constructors --sysroot=. -MD -MT lib/tsan/CMakeFiles/clang_rt.tsan-aarch64.dir/rtl/tsan_rtl_aarch64.S.o -MF lib/tsan/CMakeFiles/clang_rt.tsan-aarch64.dir/rtl/tsan_rtl_aarch64.S.o.d -o lib/tsan/CMakeFiles/clang_rt.tsan-aarch64.dir/rtl/tsan_rtl_aarch64.S.o -c /opt/tooling/drive/llvm/compiler-rt/lib/tsan/rtl/tsan_rtl_aarch64.S
/opt/tooling/drive/llvm/compiler-rt/lib/tsan/rtl/tsan_rtl_aarch64.S:29:1: error: expected identifier or '('
.section .text
^
1 error generated.
Differential Revision: https://reviews.llvm.org/D86308
This is the first of a set of DAGCombiner changes enabling strictfp
optimizations. I want to test to waters with this to make sure changes
like these are acceptable for the strictfp case- this particular change
should preserve exception ordering and result precision perfectly, and
many other possible changes appear to be able to as well.
Copied from regular fadd combines but modified to preserve ordering via
the chain, this change allows strict_fadd x, (fneg y) to become
struct_fsub x, y and strict_fadd (fneg x), y to become strict_fsub y, x.
Differential Revision: https://reviews.llvm.org/D85548
Binutils generated sections seem to be padded to a multiple of 16 bytes,
but the aux section definition contains the original, unpadded section
length.
The size check used for IMAGE_COMDAT_SELECT_SAME_SIZE previously
only checked the size of the section itself. When checking the
currently processed object file against the previously chosen
comdat section, we easily have access to the aux section definition
of the currently processed section, but we have to iterate over the
symbols of the previously selected object file to find the section
definition of the previously picked section. (We don't want to
inflate SectionChunk to carry more data, for something that is only
needed in corner cases.) Only do this when the mingw flag is set.
This fixes statically linking clang-built C++ object files against
libstdc++ built with GCC, if the object files contain e.g. typeinfo.
Differential Revision: https://reviews.llvm.org/D86659
This matches lld-link's own default.
Add a new command line option --no-dynamicbase for disabling it.
(Unfortunately, GNU ld doesn't yet have a matching --no-dynamicbase
option, as that's the default there.)
Differential Revision: https://reviews.llvm.org/D86654
The existing implementations are almost identical except for width of the
integer type.
Factor them out to int_mulo_impl.inc for better maintainability.
This patch is almost identical to D86277.
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D86289
The existing implementations are almost identical except for width of the
integer type.
Factor them out to int_mulv_impl.inc for better maintainability.
Reviewed By: MaskRay
Differential Revision: https://reviews.llvm.org/D86277
This reverts commit b9d977b0ca.
This cutoff is no longer required. The commit 34ffa7fc501 (D86153) introduces a
performance improvement which was tested against the motivating case for this
patch.
Discussed in differential revision: https://reviews.llvm.org/D86153
Almost NFC (see end).
The backwards scan in validThroughout significantly contributed to compile time
for a pathological case, causing the 'X86 Assembly Printer' pass to account for
roughly 70% of the run time. This patch guards the loop against running
unnecessarily, bringing the pass contribution down to 4%.
Almost NFC: There is a hack in validThroughout which promotes single constant
value DBG_VALUEs in the prologue to be live throughout the function. We're more
likely to hit this code path with this patch applied. Similarly to the parent
patches there is a small coverage change reported in the order of 10s of bytes.
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D86153
With the changes introduced in D86151 we can now check for single locations
which span multiple blocks for inlined scopes and blocks.
D86151 introduced the InstructionOrdering parameter, replacing a scan through
MBB instructions. The functionality to compare instruction positions across
blocks was add there, and this patch just removes the exit checks that were
previously (but no longer) required.
CTMark shows a geomean binary size reduction of 2.2% for RelWithDebInfo builds.
llvm-locstats (using D85636) shows a very small variable location coverage
change in 5 of 10 binaries, but just like in D86151 it is only in the order of
10s of bytes.
Reviewed By: djtodoro
Differential Revision: https://reviews.llvm.org/D86152
With this patch we're now accounting for two more cases which should be
considered 'valid throughout': First, where RangeEnd is ScopeEnd. Second, where
RangeEnd comes before ScopeEnd when including meta instructions, but are both
preceded by the same non-meta instruction.
CTMark shows a geomean binary size reduction of 1.5% for RelWithDebInfo builds.
`llvm-locstats` (using D85636) shows a very small variable location coverage
change in 2 of 10 binaries, but it is in the order of 10s of bytes which lines
up with my expectations.
I've added a test which checks both of these new cases. The first check in the
test isn't strictly necessary for this patch. But I'm not sure that it is
explicitly tested anywhere else, and is useful for the final patch in the
series.
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D86151
Group the map and methods used to query instruction ordering for trimVarLocs
(D82129) into a class. This will make it easier to reuse the functionality
upcoming patches.
Reviewed By: aprantl
Differential Revision: https://reviews.llvm.org/D86150
This patch implements the semantics for the 'arm_sve_vector_bits' type
attribute, defined by the Arm C Language Extensions (ACLE) for SVE [1].
The purpose of this attribute is to define vector-length-specific (VLS)
versions of existing vector-length-agnostic (VLA) types.
The semantics were already implemented by D83551, although the
implementation approach has since changed to represent VLSTs as
VectorType in the AST and fixed-length vectors in the IR everywhere
except in function args/returns. This is described in the prototype
patch D85128 demonstrating the new approach.
The semantic changes added in D83551 are changed since the
AttributedType is replaced by VectorType in the AST. Minimal changes
were necessary in the previous patch as the canonical type for both VLA
and VLS was the same (i.e. sizeless), except in constructs such as
globals and structs where sizeless types are unsupported. This patch
reverts the changes that permitted VLS types that were represented as
sizeless types in such circumstances, and adds support for implicit
casting between VLA <-> VLS types as described in section 3.7.3.2 of the
ACLE.
Since the SVE builtin types for bool and uint8 are both represented as
BuiltinType::UChar in VLSTs, two new vector kinds are implemented to
distinguish predicate and data vectors.
[1] https://developer.arm.com/documentation/100987/latest
Reviewed By: aaron.ballman
Differential Revision: https://reviews.llvm.org/D85736
This patch adds code to recognize vector shuffles which can be
represented as VDUP (splat) of a vector lane with of a different
(wider) type than the original vector lane type.
For example:
shufflevector <4 x i16> %v, <4 x i16> undef, <4 x i32> <i32 0, i32 1, i32 0, i32 1>
is essentially:
shufflevector <2 x i32> %v, <2 x i32> undef, <2 x i32> <i32 0, i32 0>
Such patterns are generated by the SelectionDAG machinery in some cases
(see DAGCombiner::visitBITCAST in DAGCombiner.cpp, the "Remove double
bitcasts from shuffles" part).
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D86225
This patch adds type information for SVE ACLE vector types,
by describing them as vectors, with a lower bound of 0, and
an upper bound described by a DWARF expression using the
AArch64 Vector Granule register (VG), which contains the
runtime multiple of 64bit granules in an SVE vector.
Reviewed By: efriedma
Differential Revision: https://reviews.llvm.org/D86101
Whilst reviewing some internal testing, I noticed a couple of holes in
coverage of mergeable sections containing symbols. This patch addresses
these holes:
1) Show that mid-piece symbols have their values updated properly when
pieces are merged.
2) Show the behaviour of symbols in mergeable pieces when --gc-sections
is enabled.
Reviewed by: grimar, MaskRay
Differential Revision: https://reviews.llvm.org/D86543
Since the canonical floatig-point move is fsgnj rd, rs, rs, we should
handle this case in RISCVInstrInfo::isAsCheapAsAMove().
Reviewed By: lenary
Differential Revision: https://reviews.llvm.org/D86518
The isTriviallyRematerializable hook is only called for instructions that are
tagged as isAsCheapAsAMove. Since ADDI 0 is used for "mv" it should definitely
be marked with "isAsCheapAsAMove". This change avoids one stack spill in most of
the atomic-rmw.ll tests functions. It also avoids stack spills in two of our
out-of-tree CHERI tests.
ORI/XORI with zero may or may not be the same as a move micro-architecturally,
but since we are already doing it for register == x0, we might as well
do the same if the immediate is zero.
Reviewed By: luismarques
Differential Revision: https://reviews.llvm.org/D86480
Specify type when constructing PromotionKeys,
this fixes error:
"chosen constructor is explicit in copy-initialization"
when compiling lldb with GCC 5.4.0.
This is due to std::tuple having an explicit
default constructor, see:
http://cplusplus.github.io/LWG/lwg-defects.html#2193
Reviewed By: labath
Differential Revision: https://reviews.llvm.org/D86690
For DSE with MemorySSA it is beneficial to manually traverse the
defining access, instead of using a MemorySSA walker, so we can
better control the number of steps together with other limits and
also weed out invalid/unprofitable paths early on.
This patch requires a follow-up patch to be most effective, which I will
share soon after putting this patch up.
This temporarily XFAIL's the limit tests, because we now explore more
MemoryDefs that may not alias/clobber the killing def. This will be
improved/fixed by the follow-up patch.
This patch also renames some `Dom*` variables to `Earlier*`, because the
dominance relation is not really used/important here and potentially
confusing.
This patch allows us to aggressively cut down compile time, geomean
-O3 -0.64%, ReleaseThinLTO -1.65%, at the expense of fewer stores
removed. Subsequent patches will increase the number of removed stores
again, while keeping compile-time in check.
http://llvm-compile-time-tracker.com/compare.php?from=d8e3294118a8c5f3f97688a704d5a05b67646012&to=0a929b6978a068af8ddb02d0d4714a2843dd8ba9&stat=instructions
Reviewed By: asbirlea
Differential Revision: https://reviews.llvm.org/D86486
Alexandre Ganea