* [Coverage] Only instrument ClosureExprs once
ClosureExprs should only be visited for profiling purposes once, when
the SILFunction definition for the closure is being emitted.
This fixes an issue where the coverage tooling can't figure out how to
attribute the code coverage data of a closure to the right function.
rdar://39200851
* [Coverage] Assert that we don't emit duplicate coverage mappings
While generating SIL for a function with a default argument, we don't
need to emit two identical coverage mappings for the function body. The
same goes for functions which reference foreign functions.
This PR introduces an assertion which should catch similar problems in
the future.
rdar://39297172
* [Coverage] Only instrument nested functions once
Coverage counters for a nested function can be assigned once (in its
parent's scope), and then again (in its own scope). Nested functions
should only be visited for coverage mapping purposes once.
This is related to r://39200851, which is the same bug but for closures.
* [Coverage] Remove special handling of autoclosures
Treating AutoClosureExprs the same as ClosureExprs allows for some nice
code simplifications, and should be more robust.
* [Coverage] Only instrument curried instance methods once
Another fix related to r://39297172, in which we avoid instrumenting the
function associated with a curried thunk more than once.
This patch moves the ownership of profiling state from SILGenProfiling
to SILFunction, where it always belonged. Similarly, it moves ownership
of the profile reader from SILGenModule to SILModule.
The refactor sets us up to fix a few outstanding code coverage bugs and
does away with sad hacks like ProfilerRAII. It also allows us to locally
guarantee that a profile counter increment actually corresponds to the
SILFunction at hand.
That local guarantee causes a bugfix to accidentally fall out of this
refactor: we now set up the profiling state for delayed functions
correctly. Previously, we would set up a ProfilerRAII for the delayed
function, but its counter increment would never be emitted :(. This fix
constitutes the only functional change in this patch -- the rest is NFC.
As a follow-up, I plan on removing some dead code in the profiling
logic and fixing a few naming inconsistencies. I've left that for later
to keep this patch simple.
This patch moves the ownership of profiling state from SILGenProfiling
to SILFunction, where it always belonged. Similarly, it moves ownership
of the profile reader from SILGenModule to SILModule.
The refactor sets us up to fix a few outstanding code coverage bugs and
does away with sad hacks like ProfilerRAII. It also allows us to locally
guarantee that a profile counter increment actually corresponds to the
SILFunction at hand.
That local guarantee causes a bugfix to accidentally fall out of this
refactor: we now set up the profiling state for delayed functions
correctly. Previously, we would set up a ProfilerRAII for the delayed
function, but its counter increment would never be emitted :(. This fix
constitutes the only functional change in this patch -- the rest is NFC.
As a follow-up, I plan on removing some dead code in the profiling
logic and fixing a few naming inconsistencies. I've left that for later
to keep this patch simple.
Changes:
* Terminate all namespaces with the correct closing comment.
* Make sure argument names in comments match the corresponding parameter name.
* Remove redundant get() calls on smart pointers.
* Prefer using "override" or "final" instead of "virtual". Remove "virtual" where appropriate.
Fix a crash in emitBuiltinCall() which occurs because we drop function
linkage information when creating SILCoverageMaps.
This re-applies 45c7e4e86 with the MachO-specific checks in the test
case removed.
If multiple swift files are compiled together, then guessing as to the
file when we emit IR obviously doesn't work. Find the filename when we
generate a function's coverage map and propagate it through SIL.
Swift SVN r25436
Keeping a reference to the function here is dangerous. We only
actually care about the name, so save ourselves a copy of that
instead.
This fixes a crash that seems to happen only when the coverage data is
very large.
Swift SVN r25433
