This hack seems to only have been necessary because of the
constructor bug noted in 33125cffd.
Once again, it's hard to prove NFC, but that's the hope...
This should be NFC unless some target was expecting that
some form of cttz/ctlz/memcpy is free in terms of size/latency
but not free in throughput cost.
This should be close to NFC (no-functional-change), but I
can't completely rule out that some call on some target
travels down a different path. There's an especially large
amount of code spaghetti in this part of the cost model.
The goal is to clean up the intrinsic cost handling so
we can canonicalize to the new min/max intrinsics without
causing regressions.
When we see this:
```
%and = G_AND %x, %y
%xor = G_XOR %and, %y
```
Produce this:
```
%not = G_XOR %x, -1
%new_and = G_AND %not, %y
```
as long as we are guaranteed to eliminate the original G_AND.
Also matches all commuted forms. E.g.
```
%and = G_AND %y, %x
%xor = G_XOR %y, %and
```
will be matched as well.
Differential Revision: https://reviews.llvm.org/D88104
bug 45566 shows the process of building coroutine frame won't consider
that the lifetimes of different local variables are not overlapped,
which means the compiler could generates smaller frame.
This patch calculate the lifetime range of each alloca by StackLifetime
class. Then the patch build non-overlapped sets for allocas whose
lifetime ranges are not overlapped. We use the largest type in a
non-overlapped set as the field type in the frame. In insertSpills
process, if we find the type of field is not the same with the alloca,
we cast the pointer to the field type to the pointer to the alloca type.
Since the lifetime range of alloca in one non-overlapped set is not
overlapped with each other, it should be ok to reuse the storage space
in the frame.
Test plan: check-llvm, check-clang, cppcoro, folly
Reviewers: junparser, lxfind, modocache
Differential Revision: https://reviews.llvm.org/D87596
After some recent upstream discussion we decided that it was best
to avoid having the / operator for both ElementCount and TypeSize,
since this could give the impression that these classes can be used
in the same way as basic integer integer types. However, division
for scalable types is a bit odd because we are only dividing the
minimum quantity by a value, as opposed to something like:
(MinSize * Vscale) / SomeValue
This is why when performing division it's important the caller
first establishes whether the operation makes sense, perhaps by
calling isKnownMultipleOf() prior to division. The caller must now
explictly call divideCoefficientBy() on the class to perform the
operation.
Differential Revision: https://reviews.llvm.org/D87700
Add a flag to getPredicateAt() that allows making use of the block
value. This allows us to take into account range information from
the current block, rather than only information that is threaded
over edges, making the icmp simplification in CVP a lot more
powerful.
I'm not changing getPredicateAt() to use the block value
unconditionally to avoid any impact on the JumpThreading pass,
which is somewhat picky about LVI query order.
Most test changes here are just icmps that now get dropped (while
previously only a result used in a return was replaced). The three
tests in icmp.ll show some representative improvements. Some of
the folds this enables have been covered by IPSCCP in the meantime,
but LVI can reason about some cases which are hard to support in
IPSCCP, such as in test_br_cmp_with_offset.
The compile-time time cost of doing this is fairly minimal, with
a ~0.05% CTMark regression for ReleaseThinLTO:
https://llvm-compile-time-tracker.com/compare.php?from=709d03f8af4da4204849a70f01798e7cebba2e32&to=6236fd503761f43c99f4537121e057a01056f185&stat=instructions
This is because the block values will typically already be queried
and cached by other CVP optimizations anyway.
Differential Revision: https://reviews.llvm.org/D69686
If -enable-constraint-elimination is specified, add it to the -O2/-O3 pipeline.
(-O1 uses a separate function now.)
Reviewed By: fhahn, aeubanks
Differential Revision: https://reviews.llvm.org/D88365
Require CxtI in getConstant() and getConstantRange() APIs.
Accordingly drop the BB parameter, as it is implied by
CxtI->getParent().
This makes sure we don't forget to pass the context instruction,
and makes the API contract clearer (also clean up the comments to
that effect -- the value holds at the context instruction, not
the end of the block).
It is not a good idea to expose raw constants in the LLVM C API. Replace this with an explicit getter.
Differential Revision: https://reviews.llvm.org/D88367
This is like FastMathFlagGuard in IR. Since we use SDAG instance to get
values, it's with SelectionDAG. By creating a FlagInserter in current
scope, all values created by getNode will get the flags if no Flags
argument provided.
In this patch, I applied it to floating point operations folding part in
DAG combiner, and removed Flags passing to getNode to show its effect.
Other places in DAG combiner and other helper methods similar to getNode
also need this. They can be done in follow-up patches.
Reviewed By: spatel
Differential Revision: https://reviews.llvm.org/D87361
This matches the legacy PM pass by having one constructor use command
line flags, and the other use parameters to the pass.
This fixes all tests under Transforms/LowerTypeTests using NPM.
Reviewed By: ychen, pcc
Differential Revision: https://reviews.llvm.org/D87845
This patch performs a minor cleanup of the class Slice:
static methods and constructors which take a pointer but assume that
it's not null now take the argument by reference.
NFC.
Test plan: make check-all
Differential revision: https://reviews.llvm.org/D88320
This commit fixes a regression (from LLVM 10 to LLVM 11 RC3) in the LLVM
C API.
Previously, commit 1ee6ec2bf removed the mask operand from the
ShuffleVector instruction, storing the mask data separately in the
instruction instead; this reduced the number of operands of
ShuffleVector from 3 to 2. AFAICT, this change unintentionally caused
a regression in the LLVM C API. Specifically, it is no longer possible
to get the mask of a ShuffleVector instruction through the C API. This
patch introduces new functions which together allow a C API user to get
the mask of a ShuffleVector instruction, restoring the functionality
which was previously available through LLVMGetOperand().
This patch also adds tests for this change to the llvm-c-test
executable, which involved adding support for InsertElement,
ExtractElement, and ShuffleVector itself (as well as constant vectors)
to echo.cpp. Previously, vector operations weren't tested at all in
echo.ll.
I also fixed some typos in comments and help-text nearby these changes,
which I happened to spot while developing this patch. Since the typo
fixes are technically unrelated other than being in the same files, I'm
happy to take them out if you'd rather they not be included in the patch.
Differential Revision: https://reviews.llvm.org/D88190
The intrinsics don't have any pointer arguments, so "argmemonly" makes
optimizations think they don't write to memory at all.
Differential Revision: https://reviews.llvm.org/D88186
1c5a3c4d3823 updated the variables inserted by Emscripten SjLj lowering to be
thread-local, depending on the CoalesceFeaturesAndStripAtomics pass to downgrade
them to normal globals if the target features did not support TLS. However, this
had the unintended side effect of preventing all non-TLS-supporting objects from
being linked into modules with shared memory, because stripping TLS marks an
object as thread-unsafe. This patch fixes the problem by only making the SjLj
lowering variables thread-local if the target machine supports TLS so that it
never introduces new usage of TLS that will be stripped. Since SjLj lowering
works on Modules instead of Functions, this required that the
WebAssemblyTargetMachine have its feature string updated to reflect the
coalesced features collected from all the functions so that a
WebAssemblySubtarget can be created without using any particular function.
Differential Revision: https://reviews.llvm.org/D88323
Make the corresponding change that was made for byval in
b7141207a483d39b99c2b4da4eb3bb591eca9e1a. Like byval, this requires a
bulk update of the test IR tests to include the type before this can
be mandatory.
This change adds an option to basic block sections to allow cold
clusters to be assigned a custom text prefix. With a custom prefix such
as ".text.split." (D87840), lld can place them in a separate output section.
The benefits are -
* Empirically shown to improve icache and itlb metrics by 3-5%
(absolute) compared to placing split parts in .text.unlikely.
* Mitigates against poor profiles, eg samplePGO profiles used with the
machine function splitter. Optimizations such as hugepage remapping can
make different decisions at the section granularity.
* Enables section granularity hotness monitoring (checking on the
decisions made during compilation vs sample data from production).
Differential Revision: https://reviews.llvm.org/D87813
Summary:
This patch add support for printing analysis messages relating to data
globalization on the GPU. This occurs when data is shared between the
threads in a GPU context and must be pushed to global or shared memory.
Reviewers: jdoerfert
Subscribers: guansong hiraditya llvm-commits ormris sstefan1 yaxunl
Tags: #OpenMP #LLVM
Differential Revision: https://reviews.llvm.org/D88243
Introduce a helper which can be used to update the debug location of an
Instruction after the instruction is hoisted. This can be used to safely
drop a source location as recommended by the docs.
For more context, see the discussion in https://reviews.llvm.org/D60913.
Differential Revision: https://reviews.llvm.org/D85670
This change adds the support for __builtin_return_address
for ARMv8.3A Pointer Authentication.
Location of the authentication code in the pointer depends on
the system configuration, therefore a dedicated instruction is used for
effectively removing the authentication code without
authenticating the pointer.
Reviewed By: chill
Differential Revision: https://reviews.llvm.org/D75044
This introduces an analysis pass that wraps IRSimilarityIdentifier,
and adds a printer pass to examine in what function similarities are
being found.
Test for what the printer pass can find are in
test/Analysis/IRSimilarityIdentifier.
Reviewed by: paquette, jroelofs
Differential Revision: https://reviews.llvm.org/D86973
Before this patch, the CrashRecoveryContext was returning -2 upon a signal, like ExecuteAndWait does. This didn't match the behavior on Windows, where the the exception code was returned.
We now return the signal's code, which optionally allows for re-throwing the signal later. Doing so requires all custom handlers to be removed first, through llvm::sys::unregisterHandlers() which we made a public API.
This is part of https://reviews.llvm.org/D70378
Before this patch, the CrashRecoveryContext would only catch the first abort(). Any further calls to abort() inside subsquent CrashRecoveryContexts would not be catched. This is because the Windows CRT removes the abort() handler before calling it.
This is part of https://reviews.llvm.org/D70378
For some expressions, we can use information from loop guards when
we are looking for a maximum. This patch applies information from
loop guards to the expression used to compute the maximum backedge
taken count in howFarToZero. It currently replaces an unknown
expression X with UMin(X, Y), if the loop is guarded by
X ult Y.
This patch is minimal in what conditions it applies, and there
are a few TODOs to generalize.
This partly addresses PR40961. We will also need an update to
LV to address it completely.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D67178
This patch introduces four new comparison functions:
isKnownLT, isKnownLE, isKnownGT, isKnownGE
that return true if we know at compile time that a particular
condition is met, i.e. that one size is definitely greater than
another. The existing operators <,>,<=,>= remain in the code for
now, but over time we would like to remove them and change the
code to use the isKnownXY routines instead. These functions do
not assert like the existing operators because the caller is
expected to properly deal with cases where we return false by
analysing the scalable properties. I've made more of an effort
to deal with cases where there are mixed comparisons, i.e. between
fixed width and scalable types.
I've also added some knownBitsXY routines to the EVT and MVT
classes that call the equivalent TypeSize::isKnownXY routines.
I've changed the existing bitsXY functions to call their knownBitsXY
equivalents and added asserts that the scalable properties match.
Again, over time we expect to migrate callers to use knownBitsXY
and make the code more aware of the scalable nature of the sizes.
Differential revision: https://reviews.llvm.org/D88098
This takes the mapped instructions from the IRInstructionMapper, and
passes it to the Suffix Tree to find the repeated substrings. Within
each set of repeated substrings, the IRSimilarityCandidates are compared
against one another for structure, and ensuring that the operands in the
instructions are used in the same way. Each of these structurally
similarity IRSimilarityCandidates are contained in a SimilarityGroup.
Tests checking for identifying identity of structure, different
isomorphic structure, and different
nonisomoprhic structure are found in
unittests/Analysis/IRSimilarityIdentifierTest.cpp.
Differential Revision: https://reviews.llvm.org/D86972
This patch makes the 'ExtLen' field of extended opcodes optional. We
don't need to manually calculate it in the future.
Reviewed By: jhenderson, MaskRay
Differential Revision: https://reviews.llvm.org/D88136
Since DWARFv5 places TUs in debug_info, some of DWARFContext's APIs have
become a bit erroneous, including TUs in the CU list by accident.
Correct that by providing compile_units (& dwo_compile_units) that
filter out the type units from the debug_info units.
Differential Revision: https://reviews.llvm.org/D87935
Just because sequences of instructions are similar to one another,
doesn't mean they are doing the same thing.
This introduces a structural check for the IRSimilarityCandidate that
compares two IRSimilarityCandidates against one another, and in each
instruction creates a mapping between the operands and results, or
checks that the existing mapping is valid. If this check passes, it
means we have structurally similar IRSimilarityCandidates.
Tests for whether the candidates are found in
unittests/Analysis/IRSimilarityIdentifierTest.cpp.
Recommit of: b27db2bb68163fa5bcb4a8f631a305eb5adb44e5 for Differential
URL.
Differential Revision: https://reviews.llvm.org/D86971
Just because sequences of instructions are similar to one another,
doesn't mean they are doing the same thing.
This introduces a structural check for the IRSimilarityCandidate that
compares two IRSimilarityCandidates against one another, and in each
instruction creates a mapping between the operands and results, or
checks that the existing mapping is valid. If this check passes, it
means we have structurally similar IRSimilarityCandidates.
Tests for whether the candidates are found in
unittests/Analysis/IRSimilarityIdentifierTest.cpp.
