Adds a basic logging class to Dexter that uses the existing PrettyOutput
class for printing and supports 3 levels of verbosity (note, warning,
error). Intended to consolidate the logging logic for Dexter into one
place, removing the need for conditional log statements and making it
easier for us later if we wish to use a more complete logging class.
Reviewed By: Orlando
Differential Revision: https://reviews.llvm.org/D144983
For the Playstation platform, mimick MSVC in propagating dllexport/dllimport attributes
to an instantiated template base class.
Reviewed By: hans
Differential Revision: https://reviews.llvm.org/D145970
There are no 32-bit targets that have LZCNT but not CMOV, and this allows us to test the straight line i64 pattern - otherwise we're doing the same branchy code as the 32-bit BSR test
By the psABI, any test case where a FPR would be used for a float, it
should also be used if you replaced that float with a _Float16. This
doesn't hold true in current Clang for the special cases in the FP
calling convention involving structs. This patch doesn't attempt to fix
that, simply to add coverage. D145074 contains the fix.
Differential Revision: https://reviews.llvm.org/D145070
This commits adds support for folding subview into GPU subgroup
MMA load/store ops.
Reviewed By: ThomasRaoux
Differential Revision: https://reviews.llvm.org/D146150
This patch is extracted from D96035. It adds AddressesMap map interface
to the DWARFLinkerParallel library. This interface mostly match with the
paired interface from the DWARFLinker library, except that it does
not depend on DIEInfo class.
Reviewed By: JDevlieghere
Differential Revision: https://reviews.llvm.org/D140788
This is an implementation detail of the flattening scheme, so hide it in the implementation thereof. This does require one caller to go through the appropriate utility, but doing that makes the code cleaner anyways.
Adds the ability to load a plugin to control the inline order.
This allows developing and distributing inlining heuristics
outside of tree. And together with the inline advisor plugins
allows for fine grained control of the inliner.
The PluginInlineOrderAnalysis class serves as the entry point
for dynamic advisors. Plugins must register instances of this
class to provide their own InlineOrder.
Reviewed By: kazu
Differential Revision: https://reviews.llvm.org/D140637
Adds an option to Dexter that passes command line arguments to the
debugged process, following (and in addition to) any arguments given by
the DexCommandLine command.
Differential Revision: https://reviews.llvm.org/D144979
Currently we don't emit any CFI instructions for the SCS register when
enabling SCS on RISCV. This causes problems when unwinding, since the
SCS register isn't being handled properly.
Reviewed By: mcgrathr
Differential Revision: https://reviews.llvm.org/D145205
After D134050, it makes sense to combine the RV64 ABI tests into a
single file in order to make it more maintainable (i.e. not having to
split tests based on the combinations of ABIs they're expected to
impact). This patch deletes duplicated tests but doesn't do much further
reorganisation beyond that.
I imagine the logical ordering of tests in the file and comments could
be further improved in the future. My personal feeling is that it's
probably not worth investing the time to try to get this "perfect", and
to instead settle for this incremental step forward. But if there's
reviewer interest in attempting to further iterate, I'm happy to do so.
Differential Revision: https://reviews.llvm.org/D140400
Having the header granularized makes it possible to remove the
dependency on this header in <format>. This <format> header gets
included in more headers due to more usage of std::formatter in the
library. This should reduce the number of transitive includes.
Note formatting the new headers will be done in a followup patch.
Reviewed By: #libc, ldionne
Differential Revision: https://reviews.llvm.org/D145590
Catch invalid element type in fir.box in the verifier so
it does not propagate later in lowering.
Reviewed By: PeteSteinfeld
Differential Revision: https://reviews.llvm.org/D146078
Adapat the fix made in D146079 to just avoid the type
to be wrapped with an extra fir.box or fir.class. The potential
load is delegated to the code that is after.
Reviewed By: PeteSteinfeld
Differential Revision: https://reviews.llvm.org/D146120
This makes it so that calling `llvm::is_contained` no longer degrades
performance over member contains, even though both have almost identical
names. This would be the case in most set/map classes that can check for
an element being present in O(1) or O(log n) time vs. linear scan with
`std::find`. For C++17 maps/sets without `.contains`, use `.find` when available,
falling back to a linear scan with `std::find`.
I also considered detecting member contains and triggering a
`static_assert` instead, but decided against it because it's just as easy
to do the right thing and call `.contains`. This would also make some code fail
only when compiled in the C++20 mode when more container types come with
`.contains` member functions.
This was actually already the case with `CommandLine.h` calling `is_contained`
on `SmallPtrSet` and in a recent BOLT patch.
Reviewed By: kazu, dblaikie, MaskRay
Differential Revision: https://reviews.llvm.org/D146061
This patch implements an initial step towards refactoring our ABI tests
(moving them to update_cc_test_checks.py). Future patches combine them.
Differential Revision: https://reviews.llvm.org/D134050
This allows instrumentation to inspect cached analyses to verify them.
The CGSCC PassInstrumentation previously ran `runAfterPass()` on the original SCC, but really it should be running on UpdatedC when relevant since that's the relevant SCC after the pass.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D146096
When fsource or tsource is not polymorphic, the result is not
polymorphic. Rebox the polymoprhic arguement so the dynamic
type of the result is correct.
Reviewed By: PeteSteinfeld
Differential Revision: https://reviews.llvm.org/D146133
This patch adjusts the memory instrumentation to account for scalable vector types in allocas. Note that we don't allow scalable vector globals, so we don't need to update that codepath.
A couple points.
First, this simply disables the optimization for scalable allocas. We can revisit this in the future, but it requires a bit of plumbing to get scalable object sizes through the visitor to be useful.
Second, I am simply disabling stack poisoning for scalable vector allocas. This is mostly for staging the change as I can't write a working test for memory instrumentation without doing so. I don't think it's unreasonable to do on it's own basis as without the bailout, we crash the compiler.
Differential Revision: https://reviews.llvm.org/D145259
Replace references to enumerate results with either result_pairs
(reference wrapper type) or structured bindings. I did not use
structured bindings everywhere as it wasn't clear to me it would
improve readability.
This is in preparation to the switch to zip semantics which won't
support non-const lvalue reference to elements:
https://reviews.llvm.org/D144503.
I chose to use values instead of const lvalue-refs because MLIR is
biased towards avoiding `const` local variables. This won't degrade
performance because currently `result_pair` is cheap to copy (size_t
+ iterator), and in the future, the enumerator iterator dereference
will return temporaries anyway.
Reviewed By: dblaikie
Differential Revision: https://reviews.llvm.org/D146006
Implements parts of P1614R2: `operator<=>` for `map` and `multimap`
Reviewed By: #libc, philnik
Spies: philnik, libcxx-commits, yaxunl
Differential Revision: https://reviews.llvm.org/D145976
For D141247 - if that pattern was used by GISel it could cause constant bus limitation failures.
Just use inline immediates instead of S_MOV to avoid the issue.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D146131
