The use of 'nocapture' for parameters and return values is incorrect for C++ types, as they can actually capture a pointer into its own value (e.g. std::string in libstdc++)
rdar://115062687
The move checker was converting some kinds of copies
into their `explicit_copy_*` versions, despite the
type of the copy being a copyable type. This was
causing random crashes in some narrow circumstances.
resolves rdar://106669967
rdar://105837040
* WIP: Store layout string in type metadata
* WIP: More cases working
* WIP: Layout strings almost working
* Add layout string pointer to struct metadata
* Fetch bytecode layout strings from metadata in runtime
* More efficient bytecode layout
* Add support for interpreted generics in layout strings
* Layout string instantiation, take and more
* Remove duplicate information from layout strings
* Include size of previous object in next objects offset to reduce number of increments at runtime
* Add support for existentials
* Build type layout strings with StructBuilder to support target sizes and metadata pointers
* Add support for resilient types
* Properly cache layout strings in compiler
* Generic resilient types working
* Non-generic resilient types working
* Instantiate resilient type in layout when possible
* Fix a few issues around alignment and signing
* Disable generics, fix static alignment
* Fix MultiPayloadEnum size when no extra tag is necessary
* Fixes after rebase
* Cleanup
* Fix most tests
* Fix objcImplementattion and non-Darwin builds
* Fix BytecodeLayouts on non-Darwin
* Fix Linux build
* Fix sizes in linux tests
* Sign layout string pointers
* Use nullptr instead of debug value
This patch replace all in-memory objects of DebugValueAddrInst with
DebugValueInst + op_deref, and duplicates logics that handles
DebugValueAddrInst with the latter. All related check in the tests
have been updated as well.
Note that this patch neither remove the DebugValueAddrInst class nor
remove `debug_value_addr` syntax in the test inputs.
This bleeds into the implementation where "guaranteed" is used
everywhere to talk about optimization of guaranteed values. We need to
use mandatory to indicate we're talking about the pass pipeline.
The pass is already not being run during normal compilation scenarios today
since it bails on OSSA except in certain bit-rot situations where a test wasn't
updated and so was inadvertently invoking the pass. I discovered these while
originally just trying to eliminate the pass from the diagnostic pipeline. The
reason why I am doing this in one larger change is that I found there were a
bunch of sil tests inadvertently relying on guaranteed arc opts to eliminate
copy traffic. So, if I just removed this and did this in two steps, I would
basically be unoptimizing then re-optimizing the tests.
Some notes:
1. The new guaranteed arc opts is based off of SemanticARCOpts and runs only on
ossa. Specifically, in this new pass, we just perform simple
canonicalizations that do not involve any significant analysis. Some
examples: a copy_value all of whose uses are destroys. This will do what the
original pass did and more without more compile time. I did a conservative
first approximation, but we can probably tune this a bit.
2. the reason why I am doing this now is that I was trying to eliminate the
enable-ownership-stripping-after-serialization flag and discovered that the
test opaque_value_mandatory implicitly depends on this since sil-opt by
default was the only place left in the compiler with that option set to false
by default. So I am eliminating that dependency before I land the larger
change.
Failing tests that do not test mandatory combine are updated to skip
the mandatory combine pass. Othere tests are updated to use otherwise
removed values.
The uniquing key for these was just the number of witness tables,
but the function itself referenced the specific existential type
it was instantiated with.
Everything still worked because getOrCreateHelperFunction() would
bitcast an existing function to the correct type, and in practice
the layout of an existential type only depends on the number of
witness tables.
However on master-next, other changes were made that stripped
off the bitcasts. This would result in assertions or LLVM
verifier failures when multiple existential types were used in
a single translation unit.
Fixes <rdar://problem/54780404>.
The uniquing key for these was just the number of witness tables,
but the function itself referenced the specific existential type
it was instantiated with.
Everything still worked because getOrCreateHelperFunction() would
bitcast an existing function to the correct type, and in practice
the layout of an existential type only depends on the number of
witness tables.
However on master-next, other changes were made that stripped
off the bitcasts. This would result in assertions or LLVM
verifier failures when multiple existential types were used in
a single translation unit.
Fixes <rdar://problem/54780404>.
This is a large patch; I couldn't split it up further while still
keeping things working. There are four things being changed at
once here:
- Places that call SILType::isAddressOnly()/isLoadable() now call
the SILFunction overload and not the SILModule one.
- SILFunction's overloads of getTypeLowering() and getLoweredType()
now pass the function's resilience expansion down, instead of
hardcoding ResilienceExpansion::Minimal.
- Various other places with '// FIXME: Expansion' now use a better
resilience expansion.
- A few tests were updated to reflect SILGen's improved code
generation, and some new tests are added to cover more code paths
that previously were uncovered and only manifested themselves as
standard library build failures while I was working on this change.
This is essentially a long-belated follow-up to Arnold's #12606.
The key observation here is that the enum-tag-single-payload witnesses
are strictly more powerful than the XI witnesses: you can simulate
the XI witnesses by using an extra case count that's <= the XI count.
Of course the result is less efficient than the XI witnesses, but
that's less important than overall code size, and we can work on
fast-paths for that.
The extra inhabitant count is stored in a 32-bit field (always present)
following the ValueWitnessFlags, which now occupy a fixed 32 bits.
This inflates non-XI VWTs on 32-bit targets by a word, but the net effect
on XI VWTs is to shrink them by two words, which is likely to be the
more important change. Also, being able to access the XI count directly
should be a nice win.
Previously, when a tuple type had non-fixed layout, we would compute
a layout by building the metadata for that tuple type and then
extracting the layout from the VWT. This can be quite expensive
because it involves constructing the exact metadata for types like
arrays and functions despite those types being fixed-layout across
all instantiations. It also tends to cause unnecessary recursive-type
issues, especially with enums where tuples are currently used to model
cases with mutliple payloads. Since we just need a layout, computing
it directly from element layouts instead of constructing metadata for
the formal type lets us take advantage of all the other fast paths for
layout construction, e.g. for fixed types and single-field aggregates.
This is a good improvement overall, but it also serves to alleviate
some of the problems of rdar://40810002 / SR-7876 in a way that
might be suitable for integration to 4.2.
The central thrust of this patch is to get these metadata initializations
off of `swift_once` and onto the metadata-request system where we can
properly detect and resolve dependencies. We do this by first introducing
runtime support for resolving metadata requests for "in-place"
initializations (committed previously) and then teaching IRGen to actually
generate code to use them (this patch).
A non-trivial amount of this patch is just renaming and refactoring some of
existing infrastructure that was being used for in-place initializations to
try to avoid unnecessary confusion.
The remaining cases that are still using `swift_once` resolution of
metadata initialization are:
- non-generic classes that can't statically fill their superclass or
have resilient internal layout
- foreign type metadata
Classes require more work because I'd like to switch at least the
resilient-superclass case over to using a pattern much more like what
we do with generic class instantiation. That is, I'd like in-place
initialization to be reserved for classes that actually don't need
relocation.
Foreign metadata should also be updated to the request/dependency scheme
before we declare ABI stability. I'm not sure why foreign metadata
would ever require a type to be resolved, but let's assume it's possible.
Fixes part of SR-7876.
It hass been a longstanding principle in LLVM that the presence of
debug info shall not affect code generation. This patch brings the
Swift frontend closer to this ideal:
- unconditionally emit shadow copies
- unconditionally bind type metadata
The extra allocas, bitcasts, geps, and stores being emitted get
optimized away when compiling at anything but -Onone. There are few
use-cases for compiling at -Onone without -g, so this shouldn't affect
performance for any real-world use-cases.
And update the existential container's initializeWithTake implementation
in the runtime. After only allowing bitwise takable values in the
inline buffer we can use memcpy to move existential container values.
rdar://31414907
SR-343
The previous change to this test for master-next (adbc4103d5) made
it work only for 64-bit targets. This changes it back to match the
IR for both 32-bit and 64-bit targets. rdar://problem/39116363
This adds the dllstorage annotations on the tests. This first pass gets
most of the IRGen tests passing on Windows (though has dependencies on
other changes). However, this allows for the changes to be merged more
easily as we cannot regress other platforms here.
I am going to leave in the infrastructure around this just in case. But there is
no reason to keep this in the tests themselves. I can always just revert this
and I don't think merge conflicts are likely due to previous work I did around
the tooling for this.
This includes global generic and non-generic global access
functions, protocol associated type access functions,
swift_getGenericMetadata, and generic type completion functions.
The main part of this change is that the functions now need to take
a MetadataRequest and return a MetadataResponse, which is capable
of expressing that the request can fail. The state of the returned
metadata is reported as an second, independent return value; this
allows the caller to easily check the possibility of failure without
having to mask it out from the returned metadata pointer, as well
as allowing it to be easily ignored.
Also, change metadata access functions to use swiftcc to ensure that
this return value is indeed returned in two separate registers.
Also, change protocol associated conformance access functions to use
swiftcc. This isn't really related, but for some reason it snuck in.
Since it's clearly the right thing to do, and since I really didn't
want to retroactively tease that back out from all the rest of the
test changes, I've left it in.
Also, change generic metadata access functions to either pass all
the generic arguments directly or pass them all indirectly. I don't
know how we ended up with the hybrid approach. I needed to change all
the code-generation and calls here anyway in order to pass the request
parameter, and I figured I might as well change the ABI to something
sensible.
