The types of SEH aren't x86(-32) vs x64 but rather stack-based exception chaining
vs table-based exception handling. x86-32 is the only arch for which Windows
uses the former. 32-bit ARM would use what is called Win64SEH today, which
is a bit confusing so instead let's just rename it to be a bit more clear.
Reviewed By: compnerd, rnk
Differential Revision: https://reviews.llvm.org/D90117
We can sharpen the range of a AddRec if we know that it does not
self-wrap and know the symbolic iteration count in the loop. If we can
evaluate the value of AddRec on the last iteration and prove that at least
one its intermediate value lies between start and end, then no-wrap flag
allows us to conclude that all of them also lie between start and end. So
the estimate of range can be improved to union of ranges of start and end.
Switched off by default, can be turned on by flag.
Differential Revision: https://reviews.llvm.org/D89381
Reviewed By: lebedev.ri, nikic
CUDA buildbots complained about a redefinition when I landed D89971.
This is odd and I fail to understand where in the CUDA headers the other
definition is supposed to be. For now, given that CUDA doesn't need the
overload (AFAIKT), we simply restrict it to the OpenMP mode.
This patch removes extraneous calls to setEdgeProbability introduced
in c91487769d80487eba1712a7a172a1c8977a9b4f.
The follow-up patch, a7b662d0f4098371b96ce4446fb0eba79b0b649f, has
since fixed BranchProbabilityInfo::eraseBlock, so we don't need to
worry about getting stale values from getEdgeProbability.
Also, since getEdgeProbability(BB, BB->getSingleSuccessor()) returns
edge probability 1/1 by default for BB with exactly one successor
edge, we don't need to explicitly call setEdgeProbability.
This patch introduces almost no functional change, but we do end up
reducing debug messages from setEdgeProbability.
Differential Revision: https://reviews.llvm.org/D90284
This reverts commit 5b3bf8b453b8cc00efd5269009a1e63c4442a30e.
This caused a regression in the ASan buildbot. See comments at
https://reviews.llvm.org/D89817 for more information.
For any newly added parse function, clang-tidy complains. New parse
functions are implicitly defined by a macro "Parse##CLASS(N, IsDistinct)".
Now this macro and exising function definitions are corrected (lower case
first character). Some other variable/function names are also corrected
to comply LLVM coding style.
Reviewed By: djtodoro
Differential Revision: https://reviews.llvm.org/D90243
SIPreAllocateWWMRegs was being inserted after RegisterCoalescer
but this pass does not exist during FastAlloc so pre-allocation
pass was never being run.
Insert pre-allocation after TwoAddressInstructionPass instead.
Reviewed By: rampitec
Differential Revision: https://reviews.llvm.org/D90236
Before we used to only mark unreachable static functions as dead if all
uses were known dead. Now we optimistically assume uses to be dead until
proven otherwise.
If we are looking at a call site argument it might be a load or call
which is in a different context than the call site argument. We cannot
simply use the call site argument range for the call or load.
Bug reported and reduced by Whitney Tsang <whitneyt@ca.ibm.com>.
In the AANoAlias logic we determine if a pointer may have been captured
before a call. We need to look at other uses in the call not uses of the
call.
The new code is not perfect as it does not allow trivial cases where the
call has multiple arguments but it is at least not unsound and a TODO
was added.
In the OpenACC specification, there are two different self clause. One for the
update directive with a var-list argument. This clause is a synonym of the host clause.
The second self clause is present for most of the compute construct and takes an optional
condition. To solve this ambiguity, the self clause for the update directive is directly
translated to a host clause during the parsing. The self clause in AccClause refers always
to the compute construct clause.
Reviewed By: kiranktp
Differential Revision: https://reviews.llvm.org/D90185
Summary:
Propagate driver commandline remarks options to linker when LTO is enabled.
This gives novice user a convenient way to collect and filter remarks throughout
a typical toolchain invocation with sample profile and LTO using single switch
from the clang driver.
A typical use of this option from clang command-line:
* Using -Rpass* options to print remarks to screen:
clang -fuse-ld=lld -flto=thin -fprofile-sample-use=foo_sample.txt
-Rpass=inline -Rpass-missed=inline -Rpass-analysis=inline
-fdiagnostics-show-hotness -fdiagnostics-hotness-threshold=100 -o foo foo.cpp
Remarks will be dumped to screen from both pre-lto and lto
compilation.
* Using serialized remarks options
clang -fuse-ld=lld -flto=thin -fprofile-sample-use=foo_sample.txt
-fsave-optimization-record
-fdiagnostics-show-hotness -fdiagnostics-hotness-threshold=100 -o foo foo.cpp
This will produce multiple yaml files containing optimization remarks:
1. foo.opt.yaml : remarks from pre-lto
2. foo.opt.ld.yaml.thin.1.yaml: remark during lto
Differential Revision: https://reviews.llvm.org/D85810
Since Wasm comdat sections work similarly to ELF, we can use that mechanism
to eliminate duplicate dwarf type information in the same way.
Differential Revision: https://reviews.llvm.org/D88603
If `null_pointer_is_valid` is present, `dereferenceable` does not imply
`nonnull`, make it clear.
Came up in D17993.
Reviewed By: aqjune
Differential Revision: https://reviews.llvm.org/D89417
Reported by Colleen Bertoni <bertoni@anl.gov> after running the OvO test
suite: https://github.com/TApplencourt/OvO/
The template overload is still hidden behind an ifdef for OpenMP. In the
future we probably want to remove the ifdef but that requires further
testing.
Reviewed By: JonChesterfield, tra
Differential Revision: https://reviews.llvm.org/D89971
At this point, these methods are just carbon copies of OpBuilder::create and aren't necessary given that PatternRewriter inherits from OpBuilder.
Differential Revision: https://reviews.llvm.org/D90087
An InterfaceMap is generated for every single operation type, and is responsible for a large amount of the code size from MLIR given that its internals highly utilize templates. This revision refactors the internal implementation to use bare malloc/free for interface instances as opposed to static variables and moves as much code out of templates as possible. This led to a decrease of over >1mb (~12% of total MLIR related code size) for a downstream MLIR library with a large amount of operations.
Differential Revision: https://reviews.llvm.org/D90086
When compiling for code size, the use of a vtable causes a destructor(and constructor in certain cases) to be generated for the class. Interface models don't need a complex constructor or a destructor, so this can lead to many megabytes of code size increase(even in opt). This revision switches to a simpler struct of function pointers approach that accomplishes the same API requirements as before. This change requires no updates to user code, or any other code aside from the generator, as the user facing API is still exactly the same.
Differential Revision: https://reviews.llvm.org/D90085
All InterfaceMethods will have a corresponding entry in the interface model, and by extension have an implementation generated for every operation type. This can result in large binary size increases when a large amount of operations use an interface, such as the side effect interface.
Differential Revision: https://reviews.llvm.org/D90084
This patch adds support for fusing linalg.indexed_generic op with
linalg.tensor_reshape op by expansion, i.e.
- linalg.indexed_generic op -> linalg.tensor_reshape op when the
latter is expanding.
- linalg.tensor_reshape op -> linalg.indexed_generic op when the
former is folding.
Differential Revision: https://reviews.llvm.org/D90082
This patch ensures that BranchProbabilityInfo::eraseBlock(BB) deletes
all entries in Probs associated with with BB.
Without this patch, stale entries for BB may remain in Probs after
eraseBlock(BB), leading to a situation where a newly created basic
block has an edge probability associated with it even before the pass
responsible for creating the basic block adds any edge probability to
it.
Consider the current implementation of eraseBlock(BB):
for (const_succ_iterator I = succ_begin(BB), E = succ_end(BB); I != E; ++I) {
auto MapI = Probs.find(std::make_pair(BB, I.getSuccessorIndex()));
if (MapI != Probs.end())
Probs.erase(MapI);
}
Notice that it uses succ_begin(BB) and succ_end(BB), which are based
on BB->getTerminator(). This means that if the terminator changes
between calls to setEdgeProbability and eraseBlock, then we may not
examine all pairs associated with BB.
This is exactly what happens in MaybeMergeBasicBlockIntoOnlyPred,
which merges basic blocks A into B if A is the sole predecessor of B,
and B is the sole successor of A. It replaces the terminator of A
with UnreachableInst before (indirectly) calling eraseBlock(A).
The patch fixes the problem by keeping track of all edge probablities
entered with setEdgeProbability in a map from BasicBlock* to a
successor index.
Differential Revision: https://reviews.llvm.org/D90272
This patch teaches the jump threading pass to set edge probabilities
whenever the pass creates new basic blocks.
Without this patch, the compiler sometimes produces non-deterministic
results. The non-determinism comes from the jump threading pass using
stale edge probabilities in BranchProbabilityInfo. Specifically, when
the jump threading pass creates a new basic block, we don't initialize
its outgoing edge probability.
Edge probabilities are maintained in:
DenseMap<Edge, BranchProbability> Probs;
in class BranchProbabilityInfo, where Edge is an ordered pair of
BasicBlock * and a successor index declared as:
using Edge = std::pair<const BasicBlock *, unsigned>;
Probs maps edges to their corresponding probabilities.
Now, we rarely remove entries from this map, so if we happen to
allocate a new basic block at the same address as a previously deleted
basic block with an edge probability assigned, the newly created basic
block appears to have an edge probability, albeit a stale one.
This patch fixes the problem by explicitly setting edge probabilities
whenever the jump threading pass creates new basic blocks.
Differential Revision: https://reviews.llvm.org/D90106
This field to represents the amount of static data needed by
an dynamic library or executable it should not include things
like heap or stack areas, which in the case of `-pie` are
not determined until runtime (e.g. __stack_pointer is imported).
Differential Revision: https://reviews.llvm.org/D90261
This was originally part of:
f2c25c70791d
but that was reverted because there was an underlying bug in
processing the vector type of these intrinsics. That was
fixed with:
74ffc823ed21
This is similar in spirit to 01ea93d85d6e (memcpy) except that
here the underlying caller assumptions were created for vectorizer
use (throughput) rather than other passes.
That meant targets could have an enormous throughput cost with no
corresponding size, latency, or blended cost increase.
Paraphrasing from the previous commits:
This may not make sense for some callers, but at least now the
costs will be consistently wrong instead of mysteriously wrong.
Targets should provide better overrides if the current modeling
is not accurate.
On windows, wchar_t is 16 bit, while we might be widening chars to
char32_t.
This cast had been present since the initial commit, and removing it
doesn't seem to make any tests fail.
Differential Revision: https://reviews.llvm.org/D90228
When converting a BUILD_VECTOR or VECTOR_SHUFFLE to a splatting load
as of 1461fb6e783cb946b061f66689b419f74f7fad63, we inaccurately check
for a single user of the load and neglect to update the users of the
output chain of the original load. As a result, we can emit a new
load when the original load is kept and the new load can be reordered
after a dependent store. This patch fixes those two issues.
Fixes https://bugs.llvm.org/show_bug.cgi?id=47891
Summary:
This patch adds basic support for priting the source location and names for the
mapped variables. This patch does not support names for custom mappers. This is
based on D89802. The names information currently will be printed out only in
debug mode or using env LIBOMPTARGET_INFO during execution. But the information
is added when availible to the Device and Private data structures. To get the
information out the code must be built with debug symbols on using -g or
-Rpass=openmp-opt
Reviewers: jdoerfert
Differential Revision: https://reviews.llvm.org/D90172
On windows, the previous path replacement using forward slashes wouldn't
work, and so we'd end up including the same file again. We would do that
until we'd hit the recursion limit of the Python interpreter.
Instead, use `os.path` to properly replace without assuming a specific
path separator.
