This new OSSA invariant simplifies many optimizations because they don't have to take care of the corner case of incomplete lifetimes in dead-end blocks.
The implementation basically consists of these changes:
* add the lifetime completion utility
* add a flag in SILFunction which tells optimization that they need to run the lifetime completion utility
* let all optimizations complete lifetimes if necessary
* enable the ownership verifier to check complete lifetimes
This pass has been disabled since a very long time (because it's terrible for code size).
It does not work for OSSA. Therefore it cannot be enabled anymore (as is) once we have OSSA throughout the pipeline.
So it's time to completely remove it.
Replace the one-off compiler flag for Library Evolution with an
optional language feature. This makes the
`hasFeature(LibraryEvolution)` check work in an `#if`, and is
otherwise just cleanup.
Tracked by rdar://161125572.
This does not enable it by default. Use either of the flags:
```
-enable-copy-propagation
-enable-copy-propagation=always
```
to enable it in -Onone. The previous frontend flag
`-enable-copy-propagation=true` has been renamed to
`-enable-copy-propagation=optimizing`, which is currently default.
rdar://107610971
If these programs crash, we want them to print the Swift bug report
message, not the default LLVM one, which leads to
https://github.com/llvm/llvm-project/issues.
While here, hoist the setting of the bug report message to the
START_PROGRAM macro so that we don't forget to set it in the future.
Since LayoutPrespecialization has been enabled by default in all compiler
invocations for quite some time, it doesn't make sense for it to be treated as
experimental feature. Make it a baseline feature and remove all the
checks for it from the compiler.
The Runtime module requires C++ interop enabled, because that's
the only way to robustly fix things so that it is able to declare various
types and constants without clashing with system headers.
To make this work, we also need to add a missing invocation to sil-opt to
make it pick the correct C++ runtime library.
rdar://143050566
When Swift passes search paths to clang, it does so directly into the HeaderSearch. That means that those paths get ordered inconsistently compared to the equivalent clang flag, and causes inconsistencies when building clang modules with clang and with Swift. Instead of touching the HeaderSearch directly, pass Swift search paths as driver flags, just do them after the -Xcc ones.
Swift doesn't have a way to pass a search path to clang as -isystem, only as -I which usually isn't the right flag. Add an -Isystem Swift flag so that those paths can be passed to clang as -isystem.
rdar://93951328
The old analysis pass doesn't take into account profile data, nor does
it consider post-dominance. It primarily dealt with _fastPath/_slowPath.
A block that is dominated by a cold block is itself cold. That's true
whether it's forwards or backwards dominance.
We can also consider a call to any `Never` returning function as a
cold-exit, though the block(s) leading up to that call may be executed
frequently because of concurrency. For now, I'm ignoring the concurrency
case and assuming it's cold. To make use of this "no return" prediction,
use the `-enable-noreturn-prediction` flag, which is currently off by
default.
Although I don't plan to bring over new assertions wholesale
into the current qualification branch, it's entirely possible
that various minor changes in main will use the new assertions;
having this basic support in the release branch will simplify that.
(This is why I'm adding the includes as a separate pass from
rewriting the individual assertions)
TLDR: This makes it so that we always can parse sending/transferring but changes
the semantic language effects to be keyed on RegionBasedIsolation instead.
----
The key thing that makes this all work is that I changed all of the "special"
semantic changes originally triggered on *ArgsAndResults to now be triggered
based on RegionBasedIsolation being enabled. This makes a lot of sense since we
want these semantic changes specifically to be combined with the checkers that
RegionBasedIsolation turns on. As a result, even though this causes these two
features to always be enabled, we just parse it but we do not use it for
anything semantically.
rdar://128961672
A few things:
1. Internally except for in the parser and the clang importer, we only represent
'sending'. This means that it will be easy to remove 'transferring' once enough
time has passed.
2. I included a warning that suggested to the user to change 'transferring' ->
'sending'.
3. I duplicated the parsing diagnostics for 'sending' so both will still get
different sets of diagnostics for parsing issues... but anywhere below parsing,
I have just changed 'transferring' to 'sending' since transferring isn't
represented at those lower levels.
4. Since SendingArgsAndResults is always enabled when TransferringArgsAndResults
is enabled (NOTE not vis-a-versa), we know that we can always parse sending. So
we import "transferring" as "sending". This means that even if one marks a
function with "transferring", the compiler will guard it behind a
SendingArgsAndResults -D flag and in the imported header print out sending.
rdar://128216574
Specifically:
1. I added the "upcoming feature already enabled in this swift-version"
diagnostic.
2. I taught sil-opt that in swift-6 mode strict concurrency is enabled by
default.
I implemented this by adding a new option called swift-version. If not set, we
just leave LangOpts.EffectiveLanguageVersion alone and get whatever that default
behavior is. If it is set, we use the same version parsing infrastructure used
by the normal frontend. If the parse fails, we exit -1.
To write the test that I wanted to write to show this behavior, I also needed to
add support to sil-opt for verify-additional-prefixes which is just useful
goodness.
Running finishDiagProcessing and potentially silently exiting when -verify is
set is the correct behavior for failures later in sil-opt when running Sema or
when running SIL passes. This is incorrect behavior when just setting up the
compiler instance since in such a case, we want to hard exit -1 since this means
that a test has not properly setup the compiler instance itself. It is better to
hard fail and have the test maintainer just update its command line arguments as
appropriate.
I added a disable flag -disable-region-based-isolation-with-strict-concurrency
so that we do not need to update the current tests. It is only available when
asserts are enabled to ensure users cannot use it.
rdar://125918028
It's not thread safe and can cause false alarms in case multiple modules exist in different threads. E.g. when building swiftmodules from interfaces.
The leaking check is not important anymore because the builder APIs enforce that instructions are not leaking.
I.e. it's not possible to create an instruction without inserting it into a basic block. Also, it's not possible to remove an instruction from a block without deleting it.
rdar://122169263
There were a handful of different places trying to enable the
feature-flag when the stdlib has been built with the feature enabled.
This change cleans that up and unifies it in one spot for all sub-tools
like sil-opt and sil-func-extractor to pick-up.
Adding `move_value [lexical]` and `begin_borrow [lexical]` should happen
all the time at this point. Remove the ability to omit these
instructions and update the corresponding tests.
Merge `$<Feature>` and `hasFeature` implementations.
- `$<Feature>` did not support upcoming language features.
- `hasFeature` did not support promoted language features and also
didn't take into account `Options` in `Features.def`.
Remove `Options` entirely, it was always one of three cases:
- `true`
- `langOpts.hasFeature`
- `hasSwiftSwiftParser`
Since `LangOptions::hasFeature` should always be used anyway, it's no
longer necessary. `hasSwiftSwiftParser` can be special cased when adding
the default promoted language features (by removing those features).
Resolves rdar://117917456.
Some notes:
This is not emitted by SILGen. This is just intended to be used so I can write
SIL test cases for transfer non sendable. I did this by adding an
ActorIsolationCrossing field to all FullApplySites rather than adding it into
the type system on a callee. The reason that this makes sense from a modeling
perspective is that an actor isolation crossing is a caller concept since it is
describing a difference in between the caller's and callee's isolation. As a
bonus it makes this a less viral change.
For simplicity, I made it so that the isolation is represented as an optional
modifier on the instructions:
apply [callee_isolation=XXXX] [caller_isolation=XXXX]
where XXXX is a printed representation of the actor isolation.
When neither callee or caller isolation is specified then the
ApplyIsolationCrossing is std::nullopt. If only one is specified, we make the
other one ActorIsolation::Unspecified.
This required me to move ActorIsolationCrossing from AST/Expr.h ->
AST/ActorIsolation.h to work around compilation issues... Arguably that is where
it should exist anyways so it made sense.
rdar://118521597
Anthony Latsis
Erik Eckstein
Yuta Saito
Doug Gregor
Kavon Farvardin
Michael Gottesman
Allan Shortlidge
Anthony
Nate Chandler
Kuba Mracek
Andrew Trick
Meghana Gupta
Alastair Houghton
Ian Anderson
Arnold Schwaighofer
Dmitrii Galimzianov
Tim Kientzle
Holly Borla
Ben Barham
Slava Pestov
Michael Gottesman