There was one case which was not handled and that's inlining of a compiler intrinsic.
In this case there is a different location type on the function call.
rdar://problem/49651421
The assert currently assumes that the size of SILLocation is 3 pointers big, or 24-bytes as it is on 64-bit platforms. The catch is that this math works because it’s having to make a rough assumption of the size of unsigned and pointer in relation to each other. 4 unsigned + 1 pointer = 3 pointer in this example.
This breaks down on 32-bit where 4 unsigned + 1 pointer = 5 unsigned = 20 bytes. It technically doesn’t violate the spirit of the assert, but the way the assert is written simply doesn’t work for 32-bit.
It can be written as `sizeof(SILLocation) <= 6*sizeof(unsigned)` without an ifdef, but it means that if someone was to make it even smaller at some point, the assert wouldn’t fail, signaling to the developer that they should update the assert with the new, smaller, value. But it would be cleaner. The way this change is written, if the size changes in either direction, the developer is signaled that they should either update the assert, or fix their regression.
Since SIL instructions get used a /lot/, this may result in a
compile-time speedup and/or lowered peak memory usage.
(It's not a lot in practice, though.)
Add an abstraction to retrieve a name suitable for display for a given
source location for use with diagnostics, debug locations, and magic
file literals.
Before this change, stepping through the code
1 foo(x,
2 f(a)
3 f(b)
4 )
would visit the code in the order 2, 3, 4, with the function call
being on line 4. After this patch the order is 2, 3, 1 with the
function call being on line 1. This is both closer to what clang
generates for simialar C code and more useful to the programmer since
it is easier to understand which function is being called in a nested
expression.
rdar://problem/35430708
A lot of files transitively include Expr.h, because it was
included from SILInstruction.h, SILLocation.h and SILDeclRef.h.
However in reality most of these files don't do anything
with Exprs, especially not anything in IRGen or the SILOptimizer.
Now we're down to 171 files in the frontend which depend on
Expr.h, which is still a lot but much better than before.
In all cases the DeclCtx field was supposed to be initialized from the
SILLocation of the function, so we can save one pointer per
SILFunction.
There is one test case change where a different (more precise)
diagnostic is being generated after this change.
This ireapplies commit 255c52de9f.
Original commit message:
Serialize debug scope and location info in the SIL assembler language.
At the moment it is only possible to test the effects that SIL
optimization passes have on debug information by observing the
effects of a full .swift -> LLVM IR compilation. This change enable us
to write targeted testcases for single SIL optimization passes.
The new syntax is as follows:
sil-scope-ref ::= 'scope' [0-9]+
sil-scope ::= 'sil_scope' [0-9]+ '{'
sil-loc
'parent' scope-parent
('inlined_at' sil-scope-ref )?
'}'
scope-parent ::= sil-function-name ':' sil-type
scope-parent ::= sil-scope-ref
sil-loc ::= 'loc' string-literal ':' [0-9]+ ':' [0-9]+
Each instruction may have a debug location and a SIL scope reference
at the end. Debug locations consist of a filename, a line number, and
a column number. If the debug location is omitted, it defaults to the
location in the SIL source file. SIL scopes describe the position
inside the lexical scope structure that the Swift expression a SIL
instruction was generated from had originally. SIL scopes also hold
inlining information.
<rdar://problem/22706994>
At the moment it is only possible to test the effects that SIL
optimization passes have on debug information by observing the
effects of a full .swift -> LLVM IR compilation. This change enable us
to write targeted testcases for single SIL optimization passes.
The new syntax is as follows:
sil-scope-ref ::= 'scope' [0-9]+
sil-scope ::= 'sil_scope' [0-9]+ '{'
sil-loc
'parent' scope-parent
('inlined_at' sil-scope-ref )?
'}'
scope-parent ::= sil-function-name ':' sil-type
scope-parent ::= sil-scope-ref
sil-loc ::= 'loc' string-literal ':' [0-9]+ ':' [0-9]+
Each instruction may have a debug location and a SIL scope reference
at the end. Debug locations consist of a filename, a line number, and
a column number. If the debug location is omitted, it defaults to the
location in the SIL source file. SIL scopes describe the position
inside the lexical scope structure that the Swift expression a SIL
instruction was generated from had originally. SIL scopes also hold
inlining information.
<rdar://problem/22706994>
remove the mixed concept that was SILFileLocation.
Also add support for a third type of underlying storage that will be used
for deserialized debug lcoations from textual SIL.
NFC
<rdar://problem/22706994>
with user code and that the boilerplate is counted towards the prologue.
<rdar://problem/18563763> Setting a breakpoint on "main" in a Swift program doesn't stop at user code
Swift SVN r22611
Eliminate the intermediate top_level_code function. Now that SIL is expressive enough to express a "main" function, there's no reason for it, and this eliminates a bunch of mystery code in IRGen to thunk from main to top_level_code by reaching for hardcoded symbol names. Demystify the special code for setting up C_ARGC and C_ARGV by having SILGen look for a transparent "_didEnterMain" hook in the stdlib and emit a call to it.
Swift SVN r22525
Given a destructor that is not implicit, one will have a decl that is not
autogenerated implying that this assert will always be hit.
I am going to submit an additional patch after this that changes
InlineLocation/MandatoryInlinedLocation/CleanupLocation that ensures that
the special flags on SILLocation are propagated appropriately (which they are
not right now).
Swift SVN r11595
The empty tuple used for return value when none is supplied does not really correspond to user code. (We don’t have epilog, so mark it as pert of cleanup.)
The destroy_addr instructions are doing the cleanup.
Swift SVN r9765
This will allow us to differentiate from regular inlined code (as in optimization inlining). Some passes(AllocBocToStack) and debug info would be allowed to be more aggressive/preserve less debug info with mandatory inlined code.
Swift SVN r9339
Also, change the location of "looping back to head" branch to point to the closing brace of the body, which is a more natural source location than the beginning of the enclosing loop statement.
Swift SVN r7989
Auto generated location should not be a separate location kind since
we might have different kinds of auto generated locations (ex: cleanup,
return). Even though the kind info might not be necessary for diagnostics or
debug info, it allows us to better use type system, for example, only pass
CleanupLocation where expected.
Swift SVN r7816
I've decided to keep only the location of the scope AST node that corresponds to the cleanup. (Currently, there is no user that needs the originator expression, which caused the cleanup. So keeping things simple.)
Added the cleanup location to the Scope and JumpDest classes, which gets assigned on construction of those. The Scope's and JumpDest locations are used when we emit the cleanup instructions.
We now give better location info for 2 existing tests for definitive initialization.
(+ Rather sparse testing of all this.)
Swift SVN r7764
The test displays "<invalid loc>" even though the SIL location is
non-null because the corresponding AST node (Expr) does not have a location.
Swift SVN r7066
Added a -v(verbose) option to swift that will trigger verbose printing in SIL
Printer. SIL Printer will print the location info only in the verbose mode.
Here is the example of the format - only the line and colon are displayed for
brevity:
%24 = apply %13(%22) : $[cc(method), thin] ((), [byref] Bool) -> Builtin.Int1 // user: %26 line:46:10
(This will be used to test the validity of SILLocation info.)
Swift SVN r6991
This doubles SILLocation to being two words, since SourceLocations
can't be dumped into a PointerUnion, but it is due for a redesign someday
anyway, so I'm not going to worry about it.
Swift SVN r6989
As per Chris's suggestion (review of r6152), further refactored SILLocation not to derive from PointerUnion3 but to include it as a member.
In addition, added some template magic to make sure we don't have to chain dyn_casts, which I suspect will be/is happening a lot with SILLocation:
Ex:
- if (auto E = Func.dyn_cast<Expr*>()) {
- if (const FuncExpr *FE = dyn_cast<FuncExpr>(E))
- return SILLocation(FE->getBody())
+ if (const FuncExpr *FE = Func.getAs<FuncExpr>())
+ return SILLocation(FE->getBody());
Swift SVN r6283
