Previously in PredMemOpts, we would insert any extracts at the load site, i.e.:
store %foo to %0
...
%1 = struct_element_addr %0, $Type, $Type.field
%2 = load %1
...
apply %use(%2)
would transform to:
store %foo to %0
...
%2 = struct_extract %foo
apply %use(%2)
This framework will not work with Ownership enabled since the value stored is
considered consumed by the store. This change fixes the issue by moving such
non-destructive extracts to occur while %foo is considered live, i.e. before the
store:
%2 = struct_extract %foo
store %foo to %0
...
apply %use(%2)
This means that we have to store insertion points for each store that provides
us with available values and insert the extracts at those points. This creates
some complications in the case where we have multiple stores since we need to
deal with phi nodes. Rather than dealing with it by hand, we just insert the
extracts at each point and then use the SSA updater to insert the relevant phi
nodes.
rdar://31521023
This is just convenient if you want to have a pass work transparently with both
ownership and non-ownership SIL.
Chopped off from the larger pred-memopt update commit to ease review.
rdar://31521023
- The overload of operator^ on FormalLinkage, while cute, was only used
in this one place, and does not behave like an XOR.
- The structural type walk was totally unnecessary in the first place,
because we were only ever calling getTypeLinkage() with builtin types
and nominal types.
- Furthermore, the structural type walk was doing the wrong thing with
nested nominal types, because the linkage of a nested type A.B
should not be the intersection of the linkage of A and A.B. If A is an
imported type and A.B is defined in an extension of A, we would give
the metadata of A.B shared linkage, which is wrong.
@_silgen_name and @_cdecl functions are assumed to be referenced from
C code. Public and internal functions marked as such must not be deleted
by the optimizer, and their C symbols must be public or hidden respectively.
rdar://33924873, SR-6209
lldb is not aware of using directives currently so if you try to invoke the
given methods you will get an ambiguous name lookup error from lldb. These are
important methods for debugging in lldb, so instead in this commit we just
redeclare them and delegate to the appropriate parent class.
TypeConverter::getConstantInfo was changed to return its found records by reference, but by reference into the inline storage of a DenseMap, which is a bad idea if someone else makes a getConstantInfo call that causes the cache to rehash before we're done with the reference from a previous call. Change it so that the cached SILConstantInfos get allocated out of the SILModule's arena, and so that we store the pointers in the hashtable, so that the references remain stable. Should fix rdar://problem/35132592.
This brings the capability from clang to save remarks in an external YAML files.
YAML files can be viewed with tools like the opt-viewer.
Saving the remarks is activated with the new option -save-optimization-record.
Similarly to -emit-tbd, I've only added support for single-compile mode for now.
In this case the default filename is determined by
getOutputFilenameFromPathArgOrAsTopLevel, i.e. unless explicitly specified
with -save-optimization-record-path, the file is placed in the directory of the
main output file as <modulename>.opt.yaml.
I am doing this to ensure that parts of the optimizer that have not yet been
updated for such operations do not see said instructions. This will ensure no
surprise perf regressions from this work.
rdar://31521023
This allows reporting successful and unsuccessful optimizations similar to
clang/llvm.
This first patch adds support for the
options -Rpass=<pass-name-regex> -Rpass-missed=<pass-name-regex>. These allow
reporting successful/unsuccessful optimization on the compiler output for passes
specified by the regex. I've also added one missed and one passed remark type
to the inliner to test the infrastructure.
Clang also has the option of collecting these records in an external YAML data
file. This will be added in a later patch.
A few notes:
* The goal is to use this facility for both user-lever "performance" warnings
and expert-level performance analysis. There will probably be a flag in the
future differentiating the verbosity.
* The intent is match clang/llvm as much as it makes sense. On the other hand I
did make some changes. Unlike in llvm, the emitter is not a pass which
simplifies things. Also the remark class hierarchy is greatly simplified since
we don't derive from DiagnosticInfo. We also don't derive from Diagnostic to
support the streaming API for arbitrary named-value pairs.
* Currently function names are printed mangled which should be fixed.
