e3b3464d4e
Fix a major regression with value-initialization of class types with
...
trivial default constructors. We're weren't zero-initializing them,
which manifested as <rdar://problem/8320532> (a regression in the GCC
test suite) and is likely to have caused significant other breakage.
llvm-svn: 111650
2010-08-20 16:57:37 +00:00
b02244969d
Translate NEON vmovl intrinsics to zero/sign-extend operations.
...
llvm-svn: 111612
2010-08-20 03:36:08 +00:00
f834bcd659
Mangle explicit template arguments in dependent or overloaded names.
...
llvm-svn: 111591
2010-08-20 00:17:19 +00:00
6bcb07ad71
Add some enum goodness as requested by Chris. Now instead of storing the
...
active C++ ABI as a raw string, we store it as an enum. This should improve
performance somewhat.
And yes, this time, I started from a clean build directory, and
all the tests passed. :)
llvm-svn: 111507
2010-08-19 02:18:14 +00:00
dcfba7b35b
Generate Attr subclasses with TableGen.
...
Now all classes derived from Attr are generated from TableGen.
Additionally, Attr* is no longer its own linked list; SmallVectors or
Attr* are used. The accompanying LLVM commit contains the updates to
TableGen necessary for this.
Some other notes about newly-generated attribute classes:
- The constructor arguments are a SourceLocation and a Context&,
followed by the attributes arguments in the order that they were
defined in Attr.td
- Every argument in Attr.td has an appropriate accessor named getFoo,
and there are sometimes a few extra ones (such as to get the length
of a variadic argument).
Additionally, specific_attr_iterator has been introduced, which will
iterate over an AttrVec, but only over attributes of a certain type. It
can be accessed through either Decl::specific_attr_begin/end or
the global functions of the same name.
llvm-svn: 111455
2010-08-18 23:23:40 +00:00
78fbb61ed7
Contextual arity is a feature of mangling expressions; kill off
...
mangleCallExpression. Also, operator names with unknown arity should
be mangled as binary operators; this is actually covered by an oddly-
positioned sentence in the ABI document. Fixes PR7891.
llvm-svn: 111395
2010-08-18 19:18:59 +00:00
396639db95
zap dead code.
...
llvm-svn: 111365
2010-08-18 16:09:06 +00:00
aa94cc3fcd
fix typo
...
llvm-svn: 111334
2010-08-18 00:13:52 +00:00
b1995dffaf
Fix PR7889 by generalizing some over specialized code. There is no
...
reason that this should be limited to simple lvalues.
llvm-svn: 111331
2010-08-18 00:08:27 +00:00
00d5141aef
fix PR7892, a crash on valid in c++ codegen of __PRETTY_FUNCTION__
...
llvm-svn: 111330
2010-08-17 23:58:10 +00:00
56df97761d
StringRef'ication of lots stuff, patch by Peter Davies!
...
llvm-svn: 111314
2010-08-17 22:39:59 +00:00
6d86202fa7
Whoops. Don't fall through into the overload case when mangling a
...
dependent call expression.
llvm-svn: 111300
2010-08-17 21:51:21 +00:00
35b077e674
Convert all uses of StringLiteral::getStrData() to StringLiteral::getString()
...
and remove getStrData(). Patch by Peter Davies (with some tweaks).
llvm-svn: 111229
2010-08-17 12:54:38 +00:00
53c59df2f7
Implement support for member pointers under the Microsoft C++ ABI in the
...
AST library.
This also adds infrastructure for supporting multiple C++ ABIs in the AST.
llvm-svn: 111117
2010-08-16 03:33:14 +00:00
1cec2cc798
Remove dead code, caught by unused function warnings.
...
llvm-svn: 111091
2010-08-15 01:15:58 +00:00
1bb53d9cd6
Explicitly handle every case in the switch in RTTIBuilder::BuildTypeInfo.
...
llvm-svn: 111085
2010-08-15 00:24:31 +00:00
32427785c0
More cleanup enabling.
...
llvm-svn: 111070
2010-08-14 07:46:19 +00:00
612942d65f
Sketch out a framework for delaying the activation of a cleanup.
...
Not yet complete or used.
llvm-svn: 111044
2010-08-13 21:20:51 +00:00
a2622dd266
Zap unused UnaryOperator::OffsetOf.
...
llvm-svn: 110996
2010-08-13 01:36:11 +00:00
f7f020bb2a
Make use of __func__ in a block actually refer to
...
block's helper function. Fixes radar 7860965.
llvm-svn: 110988
2010-08-13 00:19:55 +00:00
b3732bb3b7
Just disable the hidden-visibility optimization for now by hiding it behind
...
a -cc1 option. The Darwin linker complains about mixed visibility when linking
gcc-built objects with clang-built objects, and the optimization isn't really
that valuable. Platforms with less ornery linkers can feel free to enable this.
llvm-svn: 110979
2010-08-12 23:36:15 +00:00
398dc37d1f
Implement RTTI generation for Objective C types. Fixes PR7864.
...
llvm-svn: 110900
2010-08-12 02:17:33 +00:00
caa23f0708
Emit debug info for static const class member.
...
llvm-svn: 110885
2010-08-12 00:02:44 +00:00
5996699834
Revise r110163: don't mark weak functions nounwind, because the optimizer
...
treats that as a contract to be fulfilled by any replacements.
llvm-svn: 110864
2010-08-11 22:38:33 +00:00
d76c1dbe39
Emit a stop point for delegate constructor call. This gives user a chance to step into constructor body.
...
llvm-svn: 110853
2010-08-11 21:04:37 +00:00
ffe4630f3d
When re-raising an exception after a cleanup, we need to call _Unwind_Resume_or_Rethrow
...
instead of _Unwind_Resume. With SJLJ exceptions, this is spelled
"_Unwind_SjLj_Resume_or_Rethrow", not "_Unwind_SjLj_Resume", which has
significantly different semantics.
We should actually never be generating a call to _Unwind_SjLj_Resume directly;
even if we were generating true cleanups (which we aren't because of the
horrible hack), we should be calling __cxa_end_cleanup() on ARM. I
haven't implemented this because there's little point as long as the HH is
present.
I believe this fixes <rdar://problem/8281377>.
llvm-svn: 110851
2010-08-11 20:59:53 +00:00
961fc05418
Work in progress for PR7864. Someone more familiar with ObjC++ needs to fill
...
in the code after the "FIXME: Needs to be written".
llvm-svn: 110849
2010-08-11 20:41:51 +00:00
d539d098a4
IRGen support for functions returning objc object
...
types. Fixes PR7865.
llvm-svn: 110832
2010-08-11 17:37:35 +00:00
cebe0ca75e
Fix a bug in @finally emission in both the fragile and non-fragile EH schemes
...
where we weren't accounting for the possibility that a @finally block might
have internal cleanups and therefore might write to the cleanup destination slot.
Fixes <rdar://problem/8293901>.
llvm-svn: 110760
2010-08-11 00:16:14 +00:00
8166571a25
Simplify
...
llvm-svn: 110716
2010-08-10 20:16:57 +00:00
a5c7d706b8
Slightly revised handling of mult-alt constraints, to avoid an assert, until we have the full fix.
...
llvm-svn: 110706
2010-08-10 19:20:14 +00:00
62eb9a8e84
When initializing a static local, pop the guard-abort EH cleanup after
...
the variable is fully initialized.
llvm-svn: 110704
2010-08-10 18:51:44 +00:00
76e3b53541
Do not use DIGlobalVariable to emit debugging information for enums.
...
llvm-svn: 110697
2010-08-10 18:27:15 +00:00
dc866e119c
Simplify code and add comments, in code that generate debug info for constant integer globals, based on Chris's feedback.
...
llvm-svn: 110694
2010-08-10 17:53:33 +00:00
e03edfd3e7
Even if a constant's evaluated value is used, emit debug info for the constant variable.
...
llvm-svn: 110660
2010-08-10 07:24:25 +00:00
13257c5527
Support ObjC implementation decls for PCH.
...
Strictly speaking, implementations don't go in headers but there's no law against it.
llvm-svn: 110567
2010-08-09 10:54:20 +00:00
6ff7161d51
Thread local variables aren't considered common linkage.
...
llvm-svn: 110530
2010-08-08 01:37:14 +00:00
8b2d2fe234
Allow reference binding of a reference of Objective-C object type to
...
an lvalue of another, compatible Objective-C object type (e.g., a
subclass). Introduce a new initialization sequence step kind to
describe this binding, along with a new cast kind. Fixes PR7741.
llvm-svn: 110513
2010-08-07 11:51:51 +00:00
7cd1d97e33
Inline a special case of EmitAggregateCopy into EmitNullInitialization
...
to avoid the awesome-but-wrong-in-this-case assertion in the canon EAC.
Fixes PR7834.
Also fix a subtle address-space bug in the memset path.
llvm-svn: 110511
2010-08-07 08:21:30 +00:00
cf14216509
Store inheritance paths after CastExprs instead of inside them.
...
This takes some trickery since CastExpr has subclasses (and indeed,
is abstract).
Also, smoosh the CastKind into the bitfield from Expr.
Drops two words of storage from Expr in the common case of expressions
which don't need inheritance paths. Avoids a separate allocation and
another word of overhead in cases needing inheritance paths. Also has
the advantage of not leaking memory, since destructors for AST nodes are
never run.
llvm-svn: 110507
2010-08-07 06:22:56 +00:00
8139c98cf9
Correct -ftrapv to trap on errors, instead of calling the
...
__overflow_handler entrypoint that David Chisnall made up.
Calling __overflow_handler is not part of the contract of
-ftrapv provided by GCC, and should never have been checked
in in the first place.
According to:
http://permalink.gmane.org/gmane.comp.compilers.clang.devel/8699
David is using this for some of arbitrary precision integer stuff
or something, which is not an appropriate thing to implement on
this.
llvm-svn: 110490
2010-08-07 00:20:46 +00:00
165301d38d
Initialize variable to work around warning; unfortunately, there isn't any
...
way to tell gcc "really, values outside the enum aren't valid".
llvm-svn: 110450
2010-08-06 16:37:05 +00:00
f95e358004
Finishing up block variable layout API by supporting
...
union type variables and their nesting inside other
aggregate types.
llvm-svn: 110448
2010-08-06 16:28:55 +00:00
ad5dd42817
vdup_lane was missing
...
<rdar://problem/8278732>
llvm-svn: 110420
2010-08-06 01:24:57 +00:00
e83d2b7649
Attempt to fix uninitialized value warning reported on cfe-commits.
...
llvm-svn: 110418
2010-08-06 01:17:25 +00:00
491462425a
Don't crash when mangling empty anonymous unions. We never actually *need*
...
these, but it's convenient to mangle them when deferring them (in the 99.99%
case where it's not an anonymous union, of course).
llvm-svn: 110381
2010-08-05 22:02:13 +00:00
903aba39ee
Add support for block imported struct variable layout info.
...
(objc gc and blocks in NeXt runtime).
llvm-svn: 110377
2010-08-05 21:00:25 +00:00
5513fce96b
It turns out that linkers (at least, the Darwin linker) don't necessarily
...
do the right thing with mixed-visibility symbols, so disable the visibility
optimization where that's possible, i.e. with template classes (since it's
possible that an arbitrary template might be subject to an explicit
instantiation elsewhere). 447.dealII actually does this.
I've put the code under an option that's currently not hooked up to anything.
llvm-svn: 110374
2010-08-05 20:39:18 +00:00
95bde46bbb
Argument evaluation order is not guaranteed. Split these out to force an order.
...
llvm-svn: 110354
2010-08-05 18:11:10 +00:00
a9731a4179
Fix a major bug with -ftrapv and ++/--. Patch by David Keaton!
...
llvm-svn: 110347
2010-08-05 17:39:44 +00:00
d652ac82b0
For now skip over aggregate non-byref block variables.
...
(objc gc specific).
llvm-svn: 110340
2010-08-05 16:13:18 +00:00
89f3721dee
Trying to unbreak buildbot.
...
llvm-svn: 110339
2010-08-05 15:52:12 +00:00
d7c7232a1d
Add IRGen support for non-constant OffsetOfExpr.
...
llvm-svn: 110326
2010-08-05 09:58:49 +00:00
570024a8d9
Implement #pragma GCC visibility.
...
llvm-svn: 110315
2010-08-05 06:57:20 +00:00
1f78a9ad56
Clean up of my last patch.
...
llvm-svn: 110290
2010-08-05 00:19:48 +00:00
9659f6b540
Block variable layout bitmap API generation.
...
llvm-svn: 110287
2010-08-04 23:55:24 +00:00
c8bd9c277b
Extend the visibility-hidden optimization to linkonce_odr thunks for
...
functions with in-line definitions, since such thunks will be emitted at any
use of the function.
Completes the feature work for rdar://problem/7523229.
llvm-svn: 110285
2010-08-04 23:46:35 +00:00
afa3c0a875
More objc block variable layout info. work.
...
llvm-svn: 110239
2010-08-04 18:44:59 +00:00
c05349e53a
Some early work for providing block layout info.
...
for objective-c/c++ blocks (NeXt runtime).
llvm-svn: 110213
2010-08-04 16:57:49 +00:00
e16adc2b1e
Emit standard-library RTTI with external linkage, not weak_odr.
...
Apply hidden visibility to most RTTI; libstdc++ does not rely on exact
pointer equality for the type info (just the type info names). Apply
the same optimization to RTTI that we do to vtables.
Fixes PR5962.
llvm-svn: 110192
2010-08-04 08:34:44 +00:00
8a705c15d9
Extend the hidden-visibility vtables optimization to template classes that
...
haven't been explicitly instantiated.
llvm-svn: 110189
2010-08-04 06:38:15 +00:00
2dd7d44135
Some more correctness fixes and code-size optimizations for fragile-ABI
...
ObjC exceptions:
- don't enter a try for the catch blocks unless there's a finally
- put the setjmp buffer in the locals set for liveness reasons
- dump the sync object into an alloca in the locals set for liveness reasons
Some of this can go away if the backend starts to properly calculate liveness
in the presence of setjmp (which would also be a *much* stabler solution).
llvm-svn: 110188
2010-08-04 05:59:32 +00:00
8601a75118
Do a very simple pass over every function we emit to infer whether we can
...
mark it nounwind based on whether it contains any non-nounwind calls.
<rdar://problem/8087431>
llvm-svn: 110163
2010-08-03 22:46:07 +00:00
f568b074db
Add support for VFP status & control operations for ARM.
...
llvm-svn: 110153
2010-08-03 21:32:34 +00:00
5696d6d316
Emit weak vtables of non-template classes with hidden visibility.
...
llvm-svn: 110107
2010-08-03 07:24:12 +00:00
8b0f4ff317
Further adjustments to -Wglobal-constructors; works for references and direct
...
initializations now.
llvm-svn: 110063
2010-08-02 21:13:48 +00:00
42227edc79
Fix fragile-ABI ObjC exceptions in the presence of optimization with
...
the magic of inline assembly. Essentially we use read and write hazards
on the set of local variables to force flushing locals to memory
immediately before any protected calls and to inhibit optimizing locals
across the setjmp->catch edge. Fixes rdar://problem/8160285
llvm-svn: 109960
2010-07-31 23:20:56 +00:00
45d494785d
Emit global destructors even if the destroyed object has no initializers or has
...
an initializer requiring temporary object disposal.
Fixes rdar:://problem/8246444.
llvm-svn: 109849
2010-07-30 04:56:58 +00:00
8a2f3c778e
fix PR5179 and correctly fix PR5831 to not miscompile.
...
The X86-64 ABI code didn't handle the case when a struct
would get classified and turn up as "NoClass INTEGER" for
example. This is perfectly possible when the first slot
is all padding (e.g. due to empty base classes). In this
situation, the first 8-byte doesn't take a register at all,
only the second 8-byte does.
This fixes this by enhancing the x86-64 abi stuff to allow
and handle this case, reverts the broken fix for PR5831,
and enhances the target independent stuff to be able to
handle an argument value in registers being accessed at an
offset from the memory value.
This is the last x86-64 calling convention related miscompile
that I'm aware of.
llvm-svn: 109848
2010-07-30 04:02:24 +00:00
cda4307308
Revert "Re-apply fixed version of 108749, correctly conditionalizing the new
...
sections on", this change uncovered a possible linker bug which resulted in the
wrong messages getting dispatched. Backing this out while we investigate...
llvm-svn: 109817
2010-07-29 22:57:21 +00:00
1194bd2bd8
Wire up sema checking for __builtin_arm_usat and __builtin_arm_ssat immediates.
...
llvm-svn: 109814
2010-07-29 22:48:34 +00:00
1f3a063f00
move the last hunk of getCoerceResult into the place
...
that needs it and remove getCoerceResult.
llvm-svn: 109807
2010-07-29 21:42:50 +00:00
60fbd7744f
now that direct and coerce are merged, getCoerceResult gets simpler.
...
llvm-svn: 109805
2010-07-29 21:29:53 +00:00
09794695ef
now that GetSSETypeAtOffset handles passing SSE class values as
...
float, the special case hack in getCoerceResult can go away.
llvm-svn: 109804
2010-07-29 21:22:50 +00:00
c904933aac
Change the name to something less terrible; suggestion by Doug. No functionality change.
...
llvm-svn: 109797
2010-07-29 20:08:05 +00:00
e556a71859
Implement the clang-side of detection for when to pass as
...
<2 x float> instead of double. This works but can't be turned
on until I teach codegen to pass <2 x float> as one XMM register
instead of two.
llvm-svn: 109790
2010-07-29 18:39:32 +00:00
50a357e962
Look at me, I can count!
...
llvm-svn: 109786
2010-07-29 18:19:50 +00:00
c81af03fb3
Merge PCHWriterDecl.cpp's isRequiredDecl and CodeGenModule::MayDeferGeneration into a new function,
...
DeclIsRequiredFunctionOrFileScopedVar.
This is essentially a CodeGen predicate that is also needed by the PCH mechanism to determine whether a decl
needs to be deserialized during PCH loading for codegen purposes.
Since this logic is shared by CodeGen and the PCH mechanism, move it to the ASTContext,
thus CodeGenModule's GetLinkageForFunction/GetLinkageForVariable and the GVALinkage enum is moved out of CodeGen.
This fixes current (and avoids future) codegen-from-PCH bugs.
llvm-svn: 109784
2010-07-29 18:15:58 +00:00
7f4b81af7a
fix rdar://8251384, another case where we could access beyond the
...
end of a struct. This improves the case when the struct being passed
contains 3 floats, either due to a struct or array of 3 things. Before
we'd generate this IR for the testcase:
define float @bar(double %X.coerce0, double %X.coerce1) nounwind {
entry:
%X = alloca %struct.foof, align 8 ; <%struct.foof*> [#uses=2]
%0 = bitcast %struct.foof* %X to %1* ; <%1*> [#uses=2]
%1 = getelementptr %1* %0, i32 0, i32 0 ; <double*> [#uses=1]
store double %X.coerce0, double* %1
%2 = getelementptr %1* %0, i32 0, i32 1 ; <double*> [#uses=1]
store double %X.coerce1, double* %2
%tmp = getelementptr inbounds %struct.foof* %X, i32 0, i32 2 ; <float*> [#uses=1]
%tmp1 = load float* %tmp ; <float> [#uses=1]
ret float %tmp1
}
which compiled (with optimization) to:
_bar: ## @bar
## BB#0: ## %entry
movd %xmm1, %rax
movd %eax, %xmm0
ret
Now we produce:
define float @bar(double %X.coerce0, float %X.coerce1) nounwind {
entry:
%X = alloca %struct.foof, align 8 ; <%struct.foof*> [#uses=2]
%0 = bitcast %struct.foof* %X to %0* ; <%0*> [#uses=2]
%1 = getelementptr %0* %0, i32 0, i32 0 ; <double*> [#uses=1]
store double %X.coerce0, double* %1
%2 = getelementptr %0* %0, i32 0, i32 1 ; <float*> [#uses=1]
store float %X.coerce1, float* %2
%tmp = getelementptr inbounds %struct.foof* %X, i32 0, i32 2 ; <float*> [#uses=1]
%tmp1 = load float* %tmp ; <float> [#uses=1]
ret float %tmp1
}
and:
_bar: ## @bar
## BB#0: ## %entry
movaps %xmm1, %xmm0
ret
llvm-svn: 109776
2010-07-29 18:13:09 +00:00
c95a398947
start setting up infrastructure for passing multi-floats
...
as <2 x float> instead of as double. The backend isn't ready
yet, but infrastructure in the frontend can come up.
llvm-svn: 109768
2010-07-29 17:49:08 +00:00
1c56d9ab56
rename Get8ByteTypeAtOffset -> GetINTEGERTypeAtOffset to
...
make it clear that this function should only return a type
that the codegen will classify the same as an INTEGER type.
llvm-svn: 109763
2010-07-29 17:40:35 +00:00
3f76342cfc
handle a case where we could access off the end of a function
...
that Eli pointed out, rdar://8249586
llvm-svn: 109762
2010-07-29 17:34:39 +00:00
cd84084f02
fix PR7742 / rdar://8250764, a miscompilation of struct
...
return where the struct has a base but no fields. This
was because the x86-64 abi logic was checking the wrong
predicate in one place.
This was introduced in r91874, which was a fix for PR5831,
which lacked a CHECK line, so I verified and added it.
llvm-svn: 109759
2010-07-29 17:04:54 +00:00
98076a25ce
This is a little bit far, but optimize cases like:
...
struct a {
struct c {
double x;
int y;
} x[1];
};
void foo(struct a A) {
}
into:
define void @foo(double %A.coerce0, i32 %A.coerce1) nounwind {
entry:
%A = alloca %struct.a, align 8 ; <%struct.a*> [#uses=1]
%0 = bitcast %struct.a* %A to %struct.c* ; <%struct.c*> [#uses=2]
%1 = getelementptr %struct.c* %0, i32 0, i32 0 ; <double*> [#uses=1]
store double %A.coerce0, double* %1
%2 = getelementptr %struct.c* %0, i32 0, i32 1 ; <i32*> [#uses=1]
store i32 %A.coerce1, i32* %2
instead of:
define void @foo(double %A.coerce0, i64 %A.coerce1) nounwind {
entry:
%A = alloca %struct.a, align 8 ; <%struct.a*> [#uses=1]
%0 = bitcast %struct.a* %A to %0* ; <%0*> [#uses=2]
%1 = getelementptr %0* %0, i32 0, i32 0 ; <double*> [#uses=1]
store double %A.coerce0, double* %1
%2 = getelementptr %0* %0, i32 0, i32 1 ; <i64*> [#uses=1]
store i64 %A.coerce1, i64* %2
I only do this now because I never want to look at this code again :)
llvm-svn: 109738
2010-07-29 07:43:55 +00:00
c8b7b53a1e
implement a todo: pass a eight-byte that consists of a
...
small integer + padding as that small integer. On code
like:
struct c { double x; int y; };
void bar(struct c C) { }
This means that we compile to:
define void @bar(double %C.coerce0, i32 %C.coerce1) nounwind {
entry:
%C = alloca %struct.c, align 8 ; <%struct.c*> [#uses=2]
%0 = getelementptr %struct.c* %C, i32 0, i32 0 ; <double*> [#uses=1]
store double %C.coerce0, double* %0
%1 = getelementptr %struct.c* %C, i32 0, i32 1 ; <i32*> [#uses=1]
store i32 %C.coerce1, i32* %1
instead of:
define void @bar(double %C.coerce0, i64 %C.coerce1) nounwind {
entry:
%C = alloca %struct.c, align 8 ; <%struct.c*> [#uses=3]
%0 = bitcast %struct.c* %C to %0* ; <%0*> [#uses=2]
%1 = getelementptr %0* %0, i32 0, i32 0 ; <double*> [#uses=1]
store double %C.coerce0, double* %1
%2 = getelementptr %0* %0, i32 0, i32 1 ; <i64*> [#uses=1]
store i64 %C.coerce1, i64* %2
which gives SRoA heartburn.
This implements rdar://5711709, a nice low number :)
llvm-svn: 109737
2010-07-29 07:30:00 +00:00
2cdfda44a1
fix a builder, why didn't clang++ catch this?
...
llvm-svn: 109735
2010-07-29 06:44:09 +00:00
fe34c1d53e
Kill off the 'coerce' ABI passing form. Now 'direct' and 'extend' always
...
have a "coerce to" type which often matches the default lowering of Clang
type to LLVM IR type, but the coerce case can be handled by making them
not be the same.
This simplifies things and fixes issues where X86-64 abi lowering would
return coerce after making preferred types exactly match up. This caused
us to compile:
typedef float v4f32 __attribute__((__vector_size__(16)));
v4f32 foo(v4f32 X) {
return X+X;
}
into this code at -O0:
define <4 x float> @foo(<4 x float> %X.coerce) nounwind {
entry:
%retval = alloca <4 x float>, align 16 ; <<4 x float>*> [#uses=2]
%coerce = alloca <4 x float>, align 16 ; <<4 x float>*> [#uses=2]
%X.addr = alloca <4 x float>, align 16 ; <<4 x float>*> [#uses=3]
store <4 x float> %X.coerce, <4 x float>* %coerce
%X = load <4 x float>* %coerce ; <<4 x float>> [#uses=1]
store <4 x float> %X, <4 x float>* %X.addr
%tmp = load <4 x float>* %X.addr ; <<4 x float>> [#uses=1]
%tmp1 = load <4 x float>* %X.addr ; <<4 x float>> [#uses=1]
%add = fadd <4 x float> %tmp, %tmp1 ; <<4 x float>> [#uses=1]
store <4 x float> %add, <4 x float>* %retval
%0 = load <4 x float>* %retval ; <<4 x float>> [#uses=1]
ret <4 x float> %0
}
Now we get:
define <4 x float> @foo(<4 x float> %X) nounwind {
entry:
%X.addr = alloca <4 x float>, align 16 ; <<4 x float>*> [#uses=3]
store <4 x float> %X, <4 x float>* %X.addr
%tmp = load <4 x float>* %X.addr ; <<4 x float>> [#uses=1]
%tmp1 = load <4 x float>* %X.addr ; <<4 x float>> [#uses=1]
%add = fadd <4 x float> %tmp, %tmp1 ; <<4 x float>> [#uses=1]
ret <4 x float> %add
}
This implements rdar://8248065
llvm-svn: 109733
2010-07-29 06:26:06 +00:00
9fa15c3608
ignore structs that wrap vectors in IR, the abstraction shouldn't add penalty.
...
Before we'd compile the example into something like:
%coerce.dive2 = getelementptr %struct.v4f32wrapper* %retval, i32 0, i32 0 ; <<4 x float>*> [#uses=1]
%1 = bitcast <4 x float>* %coerce.dive2 to <2 x double>* ; <<2 x double>*> [#uses=1]
%2 = load <2 x double>* %1, align 1 ; <<2 x double>> [#uses=1]
ret <2 x double> %2
Now we produce:
%coerce.dive2 = getelementptr %struct.v4f32wrapper* %retval, i32 0, i32 0 ; <<4 x float>*> [#uses=1]
%0 = load <4 x float>* %coerce.dive2, align 1 ; <<4 x float>> [#uses=1]
ret <4 x float> %0
llvm-svn: 109732
2010-07-29 05:02:29 +00:00
4200fe4e50
move the 'pretty 16-byte vector' inferring code up to be shared
...
with return values, improving stuff that returns __m128 etc.
llvm-svn: 109731
2010-07-29 04:56:46 +00:00
ce1bd754d8
simplify code by eliminating a premature optimization.
...
llvm-svn: 109730
2010-07-29 04:51:12 +00:00
3a44c7e55d
now that we have CGT around, we can start using preferred types
...
for return values too. Instead of compiling something like:
struct foo {
int *X;
float *Y;
};
struct foo test(struct foo *P) { return *P; }
to:
%1 = type { i64, i64 }
define %1 @test(%struct.foo* %P) nounwind {
entry:
%retval = alloca %struct.foo, align 8 ; <%struct.foo*> [#uses=2]
%P.addr = alloca %struct.foo*, align 8 ; <%struct.foo**> [#uses=2]
store %struct.foo* %P, %struct.foo** %P.addr
%tmp = load %struct.foo** %P.addr ; <%struct.foo*> [#uses=1]
%tmp1 = bitcast %struct.foo* %retval to i8* ; <i8*> [#uses=1]
%tmp2 = bitcast %struct.foo* %tmp to i8* ; <i8*> [#uses=1]
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* %tmp2, i64 16, i32 8, i1 false)
%0 = bitcast %struct.foo* %retval to %1* ; <%1*> [#uses=1]
%1 = load %1* %0, align 1 ; <%1> [#uses=1]
ret %1 %1
}
We now get the result more type safe, with:
define %struct.foo @test(%struct.foo* %P) nounwind {
entry:
%retval = alloca %struct.foo, align 8 ; <%struct.foo*> [#uses=2]
%P.addr = alloca %struct.foo*, align 8 ; <%struct.foo**> [#uses=2]
store %struct.foo* %P, %struct.foo** %P.addr
%tmp = load %struct.foo** %P.addr ; <%struct.foo*> [#uses=1]
%tmp1 = bitcast %struct.foo* %retval to i8* ; <i8*> [#uses=1]
%tmp2 = bitcast %struct.foo* %tmp to i8* ; <i8*> [#uses=1]
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* %tmp2, i64 16, i32 8, i1 false)
%0 = load %struct.foo* %retval ; <%struct.foo> [#uses=1]
ret %struct.foo %0
}
That memcpy is completely terrible, but I don't know how to fix it.
llvm-svn: 109729
2010-07-29 04:46:19 +00:00
029c0f1681
sink preferred type stuff lower. It's possible that this might
...
improve codegen for vaarg or something, because its codepath is
getting preferred types now.
llvm-svn: 109728
2010-07-29 04:41:05 +00:00
22326a10a7
dissolve some more complexity: make the x86-64 abi lowering code
...
compute its own preferred types instead of having CGT compute
them then pass them (circuituously) down into ABIInfo.
llvm-svn: 109726
2010-07-29 02:31:05 +00:00
c11301c76e
simplify Get8ByteTypeAtOffset by making it a member of X86_64ABIInfo
...
llvm-svn: 109724
2010-07-29 02:20:19 +00:00
458b2aaee0
now that ABIInfo depends on CGT, it has trivial access to such
...
things as TargetData, ASTContext, LLVMContext etc. Stop passing
them through so many APIs.
llvm-svn: 109723
2010-07-29 02:16:43 +00:00
2b03797222
cave in to reality and make ABIInfo depend on CodeGenTypes.
...
This will simplify a bunch of code, coming up next.
llvm-svn: 109722
2010-07-29 02:01:43 +00:00
f4ba08aeaf
pass argument vectors in a type that corresponds to the user type if
...
possible. This improves the example to pass <4 x float> instead of
<2 x double> but we still get awful code, and still don't get the
return value right.
llvm-svn: 109700
2010-07-28 23:47:21 +00:00
4b8585ef6a
tidy up
...
llvm-svn: 109699
2010-07-28 23:46:15 +00:00
c7f16ab3e3
Override selected builtin names (e.g. "long int" instead of "long") to match names used by gcc in debug info. This makes gdb testsuite happy.
...
llvm-svn: 109694
2010-07-28 23:23:29 +00:00
fa560fedb7
fix some break statements to be formatted more consistently,
...
remove some now-dead code.
llvm-svn: 109690
2010-07-28 23:12:33 +00:00
31faff5d58
use Get8ByteTypeAtOffset for the return value path as well so we
...
don't get errors similar to PR7714 on the return path.
llvm-svn: 109689
2010-07-28 23:06:14 +00:00
b22f1c8bf7
refactor the autosizing code, eliminating duplication
...
and making Get8ByteTypeAtOffset always succeed and documented.
llvm-svn: 109685
2010-07-28 22:44:07 +00:00
4c1e484f39
fix PR7714 by not referencing off the end of a struct when passed by value in
...
x86-64 abi. This also improves codegen as well. Some refactoring is needed of
this code.
llvm-svn: 109681
2010-07-28 22:15:08 +00:00
d5010898ab
Fix flags in global block descriptor when
...
block returns structs. Fies radar 8241648.
Executable test added to llvm test suite.
llvm-svn: 109620
2010-07-28 19:07:18 +00:00
ff941a666a
some cleanups and get alignments correct for various coerce cases.
...
llvm-svn: 109607
2010-07-28 18:24:28 +00:00
79ac9ed7ac
we are not supposed to create an improper callsite using a CallInstr; leave a fixme mentioning the simplification when CallSite can clone itself
...
llvm-svn: 109575
2010-07-28 09:19:33 +00:00
a652fab052
construct debug info for "id" by hand.
...
Tested by mi1-var-obj.exp in gdb testsuite.
llvm-svn: 109571
2010-07-28 01:33:15 +00:00
ba80390307
When creating a jump destination, its scope should be the scope of the
...
enclosing normal cleanup, not the top of the EH stack. I'm *really*
surprised this hasn't been causing more problems.
Fixes rdar://problem/8231514.
llvm-svn: 109569
2010-07-28 01:07:35 +00:00
c0279a9826
Revert r109546, it broke linux build.
...
llvm-svn: 109550
2010-07-27 22:37:14 +00:00
4fac280618
Merge PCHWriterDecl.cpp's isRequiredDecl and CodeGenModule::MayDeferGeneration into a new function,
...
DeclIsRequiredFunctionOrFileScopedVar.
This function is part of the public CodeGen interface since it's essentially a CodeGen predicate that is also
needed by the PCH mechanism to determine whether a decl needs to be deserialized during PCH loading for codegen purposes.
This fixes current (and avoids future) codegen-from-PCH bugs.
llvm-svn: 109546
2010-07-27 22:01:17 +00:00
4f6e73b168
Always use current working directory for DW_AT_comp_dir.
...
llvm-svn: 109535
2010-07-27 20:49:59 +00:00
6014edd117
Reapply 109303.
...
llvm-svn: 109507
2010-07-27 15:17:16 +00:00
0ebca28f1d
2nd argument of __builtin_expect must be evaluated
...
if it hs side-effect to matchgcc's behaviour.
Addresses radar 8172109.
llvm-svn: 109467
2010-07-26 23:11:03 +00:00
5cbe152ffc
Test for the presence of EH branch-throughs instead of normal branch-throughs.
...
I knew this code duplication would bite me.
llvm-svn: 109463
2010-07-26 22:44:58 +00:00
6005082354
Revert 109303.
...
llvm-svn: 109426
2010-07-26 18:49:27 +00:00
b412e174db
Remove the vast majority of the Destroy methods from the AST library,
...
since we aren't going to be calling them ever.
llvm-svn: 109377
2010-07-25 18:17:45 +00:00
268b576b72
Mangle enum constant expressions. Fixes rdar://problem/8204122
...
llvm-svn: 109315
2010-07-24 01:17:35 +00:00
268ad093a7
Untangle filename/dirname confusion. Store constructed strings on the side. Avoid use of Path.makeAbsolute().
...
DW_TAG_compile_unit uses two attributes DW_AT_name and DW_AT_comp_dir. Their expected values are:
$ clang foo.c -g
DW_AT_name - foo.c
DW_AT_comp_dir - `pwd`
$ clang one/two/foo.c -g
DW_AT_name - one/two/foo.c
DW_AT_comp_dir - `pwd`
$ clang /tmp/one/foo.c -g
DW_AT_name - /tmp/one/foo.c
DW_AT_comp_dir - empty
llvm-svn: 109303
2010-07-24 00:59:16 +00:00
2ca705eb13
Support catching Objective C pointers in C++ under the non-fragile NeXT runtime.
...
Diagnose attempts to do this under the GNU or fragile NeXT runtimes.
llvm-svn: 109298
2010-07-24 00:37:23 +00:00
72cdffa401
Return type of a setter call caused by
...
use of property-dot syntax using 'super' as receiver
is 'void'. This fixes a bug in generating correct
API for setter call. Fixes radar 8203426.
llvm-svn: 109297
2010-07-24 00:34:08 +00:00
cb9fe9ec16
Revert r109263.
...
llvm-svn: 109284
2010-07-23 23:04:28 +00:00
ad5d61e227
Revise cleanup IR generation to fix a major bug with cleanups (PR7686)
...
as well as some significant asymptotic inefficiencies with threading
multiple jumps through deep cleanups.
llvm-svn: 109274
2010-07-23 21:56:41 +00:00
28f167699a
There is no need to use separate dir name for AT_comp_dir attribute. Using absolute path for filename allows clients to query complete file location info from gdb breakpoints. Save constructed full file name.
...
llvm-svn: 109263
2010-07-23 20:38:37 +00:00
5cc2c8b9c3
Vectors are not integer types, so the type system should not classify
...
them as such. Type::is(Signed|Unsigned|)IntegerType() now return false
for vector types, and new functions
has(Signed|Unsigned|)IntegerRepresentation() cover integer types and
vector-of-integer types. This fixes a bunch of latent bugs.
Patch from Anton Yartsev!
llvm-svn: 109229
2010-07-23 15:58:24 +00:00
0884a60eb5
Keep track of artificial scopes introduced by line directives. For example,
...
#line 41 "bar.c"
dummy (1, i);
#line 24 "bar.h"
i = f2 (i);
#line 44 "bar.c"
This is tested by step-line.exp in gdb testsuite.
llvm-svn: 109189
2010-07-22 22:29:16 +00:00
775f9f9919
Turn off EH cleanups for __block variables; they caused some internal buildbot
...
failures. There's a radar tracking this.
llvm-svn: 109170
2010-07-22 21:25:44 +00:00
6e7e8cc19d
atch for implementation of objective-c's -Wselector
...
warning flag in clang. Little more to do
for a PCH issue. Radar 6507158.
llvm-svn: 109129
2010-07-22 18:24:20 +00:00
222f4be834
ObjCId is special "struct objc_object". Make this explicit in debug info.
...
This is tested by objc-rbreak.exp in gdb testsuite.
llvm-svn: 109050
2010-07-21 22:41:25 +00:00
65497583b5
Fix regression caused by r108911.
...
Do not override known debug loc with unknown debug loc.
This is tested by sections.exp in gdb testsuite.
llvm-svn: 109022
2010-07-21 18:08:50 +00:00
4715d161c0
Don't crash when sending a message inside a block with the non-fragile ABI (GNU runtime).
...
llvm-svn: 109012
2010-07-21 15:28:28 +00:00
7fa204edc7
Mark the load after calling objc_msg_lookup_sender() so that it doesn't get optimised away (GNU runtime).
...
llvm-svn: 109010
2010-07-21 12:55:25 +00:00
cda666ccd8
Rename LazyCleanup -> Cleanup. No functionality change for these last three
...
commits.
llvm-svn: 109000
2010-07-21 07:22:38 +00:00
20141f2d8c
Rip out EHCleanupScope.
...
llvm-svn: 108999
2010-07-21 07:11:21 +00:00
7535ee0352
Kill the CleanupBlock API.
...
llvm-svn: 108998
2010-07-21 07:04:01 +00:00
65bea08879
Switch the fragile-ABI @finally/@synchronized cleanup over to using a lazy
...
cleanup.
llvm-svn: 108997
2010-07-21 06:59:36 +00:00
0d42a41a63
Code simplification.
...
llvm-svn: 108996
2010-07-21 06:45:54 +00:00
fb442ae7b9
Switch the main possibly-conditional temporary cleanup over to being lazy.
...
llvm-svn: 108995
2010-07-21 06:44:28 +00:00
8680f87d99
Switch the destructor for a temporary arising from a reference binding over to
...
using a lazy cleanup.
llvm-svn: 108994
2010-07-21 06:29:51 +00:00
da650e091f
Switch the __cxa_guard_abort cleanup to being a lazy cleanup.
...
llvm-svn: 108993
2010-07-21 06:20:50 +00:00
a464ff9d15
Switch some random local-decl cleanups over to using lazy cleanups. Turn on
...
the block-release unwind cleanup: we're never going to test it if we don't turn
it on.
llvm-svn: 108992
2010-07-21 06:13:08 +00:00
906da4bb52
Switch finally cleanups over to being lazy cleanups. We get basically nothing
...
from the laziness features here except better block ordering, but it removes yet
another CleanupBlock use.
llvm-svn: 108990
2010-07-21 05:47:49 +00:00
f99a631e4e
Implement proper base/member destructor EH chaining.
...
llvm-svn: 108989
2010-07-21 05:30:47 +00:00
53cad2ef04
Change PushDestructorCleanup to use lazy cleanups.
...
llvm-svn: 108979
2010-07-21 01:41:18 +00:00
1d9875654f
Convert the EH cleanups for base and member destructors in a constructor into
...
lazy cleanups.
llvm-svn: 108978
2010-07-21 01:23:41 +00:00
05fc5be32f
Implement zero-initialization for array new when there is an
...
initializer of (). Make sure to use a simple memset() when we can, or
fall back to generating a loop when a simple memset will not
suffice. Fixes <rdar://problem/8212208>, a regression due to my work
in r107857.
llvm-svn: 108977
2010-07-21 01:10:17 +00:00
1e67040fb7
Convert the end-catch call for finally blocks to a lazy cleanup. This kills off
...
the last of the shared-code cleanups.
llvm-svn: 108975
2010-07-21 00:52:03 +00:00
c20acd31f7
Convert the ObjC @synchronized cleanups to laziness. This is not actually
...
a big deal, except that I want to eliminate the shared-code EH cleanups
in preparation for a significant algorithmic fix.
llvm-svn: 108973
2010-07-21 00:41:47 +00:00
8f51041184
Add a little helper method which will be useful soon.
...
llvm-svn: 108972
2010-07-21 00:40:03 +00:00
0043977f09
Remove unintended code that was checked in as part of r108916.
...
llvm-svn: 108951
2010-07-20 22:32:37 +00:00
11a42a4655
Remove unused argument.
...
llvm-svn: 108946
2010-07-20 22:20:10 +00:00
5add20cefa
Fix the IR generation for catching pointers by references.
...
Fixes <rdar://problem/8212123>.
llvm-svn: 108944
2010-07-20 22:17:55 +00:00
32ac583d91
in 'new int[4]', constant fold the 4*4=16 instead of
...
doing an overflow check.
llvm-svn: 108943
2010-07-20 21:55:52 +00:00
f2f3870189
Follow the implementation approach suggested by PR6687,
...
which generates more efficient and more obviously conformant
code. We now test for overflow of the multiply then force
the result to -1 if so. On X86, this generates nice code
like this:
__Z4testl: ## @_Z4testl
## BB#0: ## %entry
subl $12, %esp
movl $4, %eax
mull 16(%esp)
testl %edx, %edx
movl $-1, %ecx
cmovel %eax, %ecx
movl %ecx, (%esp)
call __Znam
addl $12, %esp
ret
llvm-svn: 108927
2010-07-20 21:07:09 +00:00
217af240b5
Adopt objc_assign_threadlocal() for __thread variables of GC types.
...
Implements radar 8203301.
llvm-svn: 108917
2010-07-20 20:30:03 +00:00
6c01820900
Print template argument names for template class.
...
llvm-svn: 108916
2010-07-20 20:24:18 +00:00
26008e07de
implement rdar://5739832 - operator new should check for overflow in multiply,
...
causing clang to compile this code into something that correctly throws a
length error, fixing a potential integer overflow security attack:
void *test(long N) {
return new int[N];
}
int main() {
test(1L << 62);
}
We do this even when exceptions are disabled, because it is better for the
code to abort than for the attack to succeed.
This is heavily based on a patch that Fariborz wrote.
llvm-svn: 108915
2010-07-20 20:19:24 +00:00
481e40c681
Use getDebugLoc and setDebugLoc instead of getDbgMetadata and setDbgMetadata,
...
avoiding MDNode overhead.
llvm-svn: 108911
2010-07-20 20:13:52 +00:00
b7a95cf165
delete a loop that just generates dead code. In an example
...
like this:
void *test(long N) {
return new int[N][42][42];
}
the loop generates two dead mul instructions:
%tmp = load i64* %N.addr ; <i64> [#uses=2]
%0 = mul i64 %tmp, 7056 ; <i64> [#uses=1]
%1 = mul i64 %tmp, 42 ; <i64> [#uses=1]
%2 = mul i64 %1, 42 ; <i64> [#uses=0]
%call = call noalias i8* @_Znam(i64 %0) ; <i8*> [#uses=1]
The scale of these multiplies is already handled by the typesize stuff.
llvm-svn: 108884
2010-07-20 18:49:33 +00:00
cb46bdc34f
remove the special case for constant array sizes from
...
EmitCXXNewAllocSize. This code uses IRBuilder, which does
constant folding already.
llvm-svn: 108882
2010-07-20 18:45:57 +00:00
71d963e954
Re-apply fixed version of 108749, correctly conditionalizing the new sections on
...
ObjC ABI version 2 this time.
llvm-svn: 108847
2010-07-20 16:20:26 +00:00
c57d34bc35
Update ImplicitCastExpr to be able to represent an XValue.
...
llvm-svn: 108807
2010-07-20 04:20:21 +00:00
7d7bc561bf
Correct line info for declarations/definitions. Radar 8063111.
...
llvm-svn: 108785
2010-07-19 23:56:31 +00:00
50c63168ed
Temporarily revert. Some odd internal breakage is likely related.
...
llvm-svn: 108764
2010-07-19 22:43:34 +00:00
f699594de8
Put ObjC method names, method types and class names in separate string literal
...
sections. rdar://8207705
llvm-svn: 108749
2010-07-19 20:54:43 +00:00
f8340deacd
Fix mangling for static member variables of classes inside an extern "C"
...
linkage specification. Not sure if this is the ideal fix, but I'm reasonably
sure it's correct vs. gcc.
llvm-svn: 108656
2010-07-18 20:49:59 +00:00
bc8cab16c5
Improve the representation of the atomic builtins in a few ways. First, we make
...
their call expressions synthetically have the "deduced" types based on their
first argument. We only insert conversions in the AST for arguments whose
values require conversion to match the value type expected. This keeps PR7600
closed by maintaining the return type, but avoids assertions due to unexpected
implicit casts making the type unsigned (test case added from Daniel).
The magic is moved into the codegen for the atomic builtin which inserts the
casts as needed at the IR level to raise the type to an integer suitable for
the LLVM intrinsic. This shouldn't cause any real change in functionality, but
now we can make the builtin be more truly polymorphic.
llvm-svn: 108638
2010-07-18 07:23:17 +00:00
ad4f38b1a9
BUILD_ARCHIVE is the default for libraries, no need to set it.
...
llvm-svn: 108633
2010-07-18 00:14:47 +00:00
3ee1022f37
Fix crash initializing a bit-field with a non-constant in a place where we
...
try to evaluate the initializer as a constant.
llvm-svn: 108632
2010-07-17 23:55:01 +00:00
36ea3723c0
The GNU-runtime ObjC personality function doesn't let us rethrow with URR for
...
multiple reasons. Rethrow with _objc_exception_throw instead. Fixes PR7656.
llvm-svn: 108595
2010-07-17 00:43:08 +00:00
eca55afea3
Fix for PR3800: make sure not to evaluate the expression for a read-write
...
asm operand twice.
llvm-svn: 108489
2010-07-16 00:55:21 +00:00
3348e2d175
IRgen: Support user defined attributes on block runtime functions.
...
- This issue here is that /usr/include/Blocks.h wants to define some of the
block runtime globals as weak, depending on the target. This doesn't work in
Clang because we aren't using the AST decl for these globals.
- The fix is a pretty gross hack which just watches all the decls for the
specific blocks globals we need to know about; if we see one we use it,
otherwise we use the hand coded type.
In time, I would like to clean this up by changing IRgen to ask Sema/AST for
the decl, which would then be lazily loaded from the builtin table if
necessary. This could be used in a whole host of places in IRgen and would
get rid of a lot of grotty hand coding of LLVM IR; however, we need some
extra Sema support for this as well as support for builtin global variables.
llvm-svn: 108482
2010-07-16 00:00:19 +00:00
900546d2e3
IRgen: Move blocks runtime interfaces to CodeGenModule.
...
llvm-svn: 108481
2010-07-16 00:00:15 +00:00
70013b640f
When deferring the emission of declarations with initializers in C++, remember
...
the order they appeared in the translation unit. If they get emitted, put them
in their proper order. Fixes rdar://problem/7458115
llvm-svn: 108477
2010-07-15 23:40:35 +00:00
8fd6499a97
Set "optimization is ON" and supply other optional parameters. This helps codegenerator preserve info in case the symbol is deleted.
...
llvm-svn: 108471
2010-07-15 23:09:46 +00:00
8997690ff1
Don't suppress the emission of available_externally functions marked
...
with always_inline attribute. Thanks to Howard for the tip.
llvm-svn: 108469
2010-07-15 22:58:18 +00:00
b3026df9fa
Mark implementation generated methods as artificial.
...
Tested by namespace.exp and virtfunc.exp from gdb testsuite.
llvm-svn: 108468
2010-07-15 22:57:00 +00:00
ec624b2d19
After some discussion, the ABI list settled on a much more sensible mangling
...
for string literals.
llvm-svn: 108464
2010-07-15 21:53:03 +00:00
cdb466e58f
Reinstate the scalar-cast-to-const-reference improvements, this time
...
with the proper spelling of "non-class prvalue". Silly me, I think
class rvalues were xvalues rather than prvalues!
Hah hah hah.
llvm-svn: 108443
2010-07-15 18:58:16 +00:00
31e225cc90
Revert r108431 and r108433 (the cast-to-const-reference fixes), which
...
broke nightlytest.
llvm-svn: 108439
2010-07-15 18:41:54 +00:00
99c372134d
Revert 108220 and subsequent patch.
...
This is not required (I am not 100% sure why) but method.exp from gdb testsuite flagged regression due to this patch.
llvm-svn: 108434
2010-07-15 18:16:09 +00:00
6744cbda03
Spell isPRValue() properly.
...
llvm-svn: 108433
2010-07-15 18:12:28 +00:00
1bae6a1be9
Teach CodeGenFunction::EmitCastLValue() to handle casts to an lvalue
...
that involve binding a reference to a pure rvalue temporary (e.g., not
a class temporary), by creating a new temporary and copying the result
there. Fixes PR6024.
llvm-svn: 108431
2010-07-15 18:04:13 +00:00
cee039093a
CK_BitCast is not an lvalue bitcast any longer
...
llvm-svn: 108428
2010-07-15 16:39:55 +00:00
1d14dd1dc6
CodeGen: Tweak ABI handling for Minix, patch by Kees van Reeuwijk!
...
llvm-svn: 108423
2010-07-15 15:02:28 +00:00
6f2e839693
CodeGen/ObjC/NeXT: Fix Obj-C message send to match llvm-gcc when choosing
...
whether to use objc_msgSend_fpret; the choice is target dependent, not Obj-C ABI
dependent.
- <rdar://problem/8139758> arm objc _objc_msgSend_fpret bug
llvm-svn: 108379
2010-07-14 23:39:36 +00:00
c357f4121e
Add lvalue-bitcast support for complex numbers.
...
llvm-svn: 108363
2010-07-14 21:35:45 +00:00
4f3b5f302c
Fix the mangling of template template arguments, which do not always
...
follow <name>; instead they follow <type>, which has <name> as a subset.
Fixes PR7446.
llvm-svn: 108326
2010-07-14 06:43:17 +00:00
afe20780ee
Remove a few mangling FIXMEs:
...
- TSTs whose template is a template template parameter already work
- we don't provide an imaginary type, so we can't mangle one
- we don't need a generic FIXME for vendor type qualifiers
llvm-svn: 108317
2010-07-14 04:38:21 +00:00
05ddbb80d2
Implement the standard mangling for array-subscript expressions, and implement
...
the current proposals from David Vandervoorde for new, delete, throw, typeid,
imaginary literals, string literals, and null literals.
llvm-svn: 108315
2010-07-14 04:20:34 +00:00
11e577b99a
Work around an obnoxious GCC warning by changing semantics in a hopefully-
...
harmless way.
llvm-svn: 108295
2010-07-13 23:19:49 +00:00
51954276cc
Introduce a new cast kind for an "lvalue bitcast", which handles
...
reinterpret_casts (possibly indirectly via C-style/functional casts)
on values, e.g.,
int i;
reinterpret_cast<short&>(i);
The IR generated for this is essentially the same as for
*reinterpret_cast<short*>(&i).
Fixes PR6437, PR7593, and PR7344.
llvm-svn: 108294
2010-07-13 23:17:26 +00:00
5c0e69ef0e
Switch the __cxa_rethrow cleanup to be lazy.
...
llvm-svn: 108288
2010-07-13 22:24:23 +00:00
5c08ab956b
Allow for the possibility that __cxa_end_catch might throw for a catch-all block
...
or a catch of a record type by value or reference. Also convert this to a
lazy cleanup.
llvm-svn: 108287
2010-07-13 22:12:14 +00:00
bb0260139a
Switch the __cxa_free_exception cleanup to be lazy.
...
llvm-svn: 108276
2010-07-13 21:17:51 +00:00
2b7fc3828e
Teach IR generation how to lazily emit cleanups. This has a lot of advantages,
...
mostly in avoiding unnecessary work at compile time but also in producing more
sensible block orderings.
Move the destructor cleanups for local variables over to use lazy cleanups.
Eventually all cleanups will do this; for now we have some awkward code
duplication.
Tell IR generation just to never produce landing pads in -fno-exceptions.
This is a much more comprehensive solution to a problem which previously was
half-solved by checks in most cleanup-generation spots.
llvm-svn: 108270
2010-07-13 20:32:21 +00:00
a8a089bfd5
Whenever we're creating an expression that is typically an rvalue
...
(e.g., a call, cast, etc.), immediately adjust the expression's type
to strip cv-qualifiers off of all non-class types (in C++) or all
types (in C). This effectively extends my previous fix for PR7463,
which was restricted to calls, to other kinds of expressions within
similar characteristics. I've audited every use of
getNonReferenceType() in the code base, switching to the newly-renamed
getNonLValueExprType() where necessary.
Big thanks to Eli for pointing out just how incomplete my original fix
for PR7463 actually was. We've been handling cv-qualifiers on rvalues
wrong for a very, very long time. Fixes PR7463.
llvm-svn: 108253
2010-07-13 18:40:04 +00:00
01c8c100c2
Add volatile qualifiers for "this".
...
llvm-svn: 108245
2010-07-13 16:23:13 +00:00
a700f68828
Reinstate the optimization suppressing available_externally functions
...
at -O0. The only change from the previous patch is that we don't try
to generate virtual method thunks for an available_externally
function.
llvm-svn: 108230
2010-07-13 06:02:28 +00:00
0a34e31d81
const qualify debug info for "this" for const methods.
...
llvm-svn: 108220
2010-07-13 00:24:30 +00:00
0aabb1200f
While collecting members for a class, always create delcaration entry for methods. Debug info for method definition will be generated while generating code for method body.
...
Tested by classes.exp in gdb testsuite.
llvm-svn: 108205
2010-07-12 22:54:41 +00:00
553f3a9b30
Speculatively revert r108156; it appears to be breaking self-host.
...
llvm-svn: 108194
2010-07-12 21:08:32 +00:00
dbb2806a7b
Do not generate LLVM IR for available_externally function bodies at
...
-O0, since we won't be using the definitions for anything anyway. For
lib/System/Path.o when built in Debug+Asserts mode, this leads to a 4%
improvement in compile time (and suppresses 440 function bodies).
<rdar://problem/7987644>
llvm-svn: 108156
2010-07-12 17:24:55 +00:00
8f3f76f991
Handle forward declarations properly in debug info.
...
Patch by Alexander Kabaev.
PR 7595.
llvm-svn: 107900
2010-07-08 19:56:29 +00:00
be349def4b
Mark calls to 'throw()' functions as nounwind, and mark the functions nounwind
...
as well.
llvm-svn: 107858
2010-07-08 06:48:12 +00:00
747eb7840a
Reinstate the fix for PR7556. A silly use of isTrivial() was
...
suppressing copies of objects with trivial copy constructors.
llvm-svn: 107857
2010-07-08 06:14:04 +00:00
cb7696cf35
fix the clang side of PR7437: EmitAggregateCopy
...
was not producing a memcpy with the right address
spaces because of two places in it doing casts of
the arguments to i8, one of which that didn't
preserve the address space.
There is also an optimizer bug here.
llvm-svn: 107842
2010-07-08 00:07:45 +00:00
e182370eda
Revert r107828 and r107827, the fix for PR7556, which seems to be
...
breaking bootstrap on Linux.
llvm-svn: 107837
2010-07-07 23:37:33 +00:00
6df2b8c3ac
Rename CXXZeroInitValueExpr to CXXScalarValueInitExpr, to reflect its
...
newly-narrowed scope. No functionality change.
llvm-svn: 107828
2010-07-07 22:43:56 +00:00
442612c285
Do not use CXXZeroValueInitExpr for class types. Instead, use
...
CXXConstructExpr/CXXTemporaryObjectExpr/CXXNewExpr as
appropriate. Fixes PR7556, and provides a slide codegen improvement
when copy-initializing a POD class type from a value-initialized
temporary. Previously, we weren't eliding the copy.
llvm-svn: 107827
2010-07-07 22:35:13 +00:00
3e2ee147d0
add driver support for minix, patch by Kees van Reeuwijk
...
from PR7583
llvm-svn: 107788
2010-07-07 16:01:42 +00:00
54fcbc7345
getBody() -> hasBody()
...
llvm-svn: 107773
2010-07-07 12:24:18 +00:00
36ea322579
Introduce Decl::hasBody() and FunctionDecl::hasBody() and use them instead of getBody() when we are just checking the existence of a body, to avoid de-serialization of the body from PCH.
...
Makes de-serialization of the function body even more "lazier".
llvm-svn: 107768
2010-07-07 11:31:19 +00:00
b609d3f5f9
Teach function-try-blocks on constructors and destructors to implicitly
...
rethrow. Fixes rdar://problem/7696603
llvm-svn: 107757
2010-07-07 06:56:46 +00:00
09ae03299a
Provide a hook for the benefit of clients using clang IR gen as a subroutine:
...
emit metadata associating allocas and global values with a Decl*. This feature
is controlled by an option that (intentionally) cannot be enabled on the command
line.
To use this feature, simply set
CodeGenOptions.EmitDeclMetadata = true;
and then interpret the completely underspecified metadata. :)
llvm-svn: 107739
2010-07-06 23:57:41 +00:00
189223e222
Make CGBuilderTy a typedef again; its functionality has been rolled back
...
into IRBuilder.
llvm-svn: 107687
2010-07-06 18:43:48 +00:00
2d605ac1f5
When destroying a cleanup, kill any references to instructions in the entry
...
block before deleting it. Fixes PR7575.
This really just a short-term fix before implementing lazy cleanups.
llvm-svn: 107676
2010-07-06 17:35:03 +00:00
466e221037
When creating functions to run global initializers and destructors, mark them
...
as nounwind in -fno-exceptions. Fixes rdar://problem/8090834.
llvm-svn: 107639
2010-07-06 04:38:10 +00:00
bd30929e4d
Validated by nightly-test runs on x86 and x86-64 darwin, including after
...
self-host. Hopefully these results hold up on different platforms.
I tried to keep the GNU ObjC runtime happy, but it's hard for me to test.
Reimplement how clang generates IR for exceptions. Instead of creating new
invoke destinations which sequentially chain to the previous destination,
push a more semantic representation of *why* we need the cleanup/catch/filter
behavior, then collect that information into a single landing pad upon request.
Also reorganizes how normal cleanups (i.e. cleanups triggered by non-exceptional
control flow) are generated, since it's actually fairly closely tied in with
the former. Remove the need to track which cleanup scope a block is associated
with.
Document a lot of previously poorly-understood (by me, at least) behavior.
The new framework implements the Horrible Hack (tm), which requires every
landing pad to have a catch-all so that inlining will work. Clang no longer
requires the Horrible Hack just to make exceptions flow correctly within
a function, however. The HH is an unfortunate requirement of LLVM's EH IR.
llvm-svn: 107631
2010-07-06 01:34:17 +00:00
ceddafb846
Generate fewer first class aggregate values for other
...
coerce cases (e.g. {double,int}) which avoids fastisel
bailing out at -O0.
llvm-svn: 107628
2010-07-05 20:41:41 +00:00
c401de9998
in the "coerce" case, the ABI handling code ends up making the
...
alloca for an argument. Make sure the argument gets the proper
decl alignment, which may be different than the type alignment.
This fixes PR7567
llvm-svn: 107627
2010-07-05 20:21:00 +00:00
53b479ff6a
fix PR7564 a cast where the bitfield struct init code
...
wasn't handling array padding elements right.
llvm-svn: 107621
2010-07-05 18:03:30 +00:00
9a3459f1a6
AppendBitField really can never fail, so return its return value.
...
Everyone knows that no bugs are ever possible with bitfields.
llvm-svn: 107620
2010-07-05 17:04:23 +00:00
3babfba597
Mangle Objective-C pointers and block pointers in the Microsoft C++ Mangler.
...
ObjC pointers were easy enough (as far as the ABI is concerned, they're
just pointers to structs), but I had to invent a new mangling for block
pointers. This is particularly worrying with the Microsoft ABI, because
it is a vendor-specific ABI; extending it could come back to bite us
later when MS extends it on their own (and you know they will).
llvm-svn: 107572
2010-07-03 16:56:59 +00:00
1180f8e90f
Provide convenience routines to save and restore the current insertion
...
point.
llvm-svn: 107570
2010-07-03 09:25:20 +00:00
3b10dd1bec
Fix mangling of array dimensions in the Microsoft C++ Mangler.
...
llvm-svn: 107568
2010-07-03 08:15:16 +00:00
d49950a317
Mangle member pointer types in the Microsoft C++ Mangler.
...
llvm-svn: 107567
2010-07-03 08:01:32 +00:00
0029a2a957
Fix mangling of function pointers in the Microsoft C++ Mangler.
...
llvm-svn: 107564
2010-07-03 05:53:41 +00:00
77552766d9
Fix mangling of array parameters for functions in the Microsoft C++ Mangler.
...
Only actual functions get mangled correctly; I don't know how to fix it for
function pointers yet. Thanks to John McCall for the hint.
Also, mangle anonymous tag types. I don't have a suitable testcase yet; I have
a feeling that that's going to need support for static locals, and I haven't
figured out exactly how MSVC's scheme for mangling those works.
llvm-svn: 107561
2010-07-03 02:41:45 +00:00
bac7490ffc
Remove unnecessary ASTContext parameter from
...
CXXRecordDecl::getDestructor(); no functionality change.
llvm-svn: 107394
2010-07-01 14:13:13 +00:00
0e7929f30c
fix rdar://8147692 - yet another crash due to my abi work.
...
llvm-svn: 107387
2010-07-01 06:20:47 +00:00
bb7ac52e02
Driver/IRgen: Add support for -momit-leaf-frame-pointer.
...
llvm-svn: 107367
2010-07-01 01:31:45 +00:00
3de064dfdf
Revert "IRgen: Make sure any prolog instructions get debug info.", the lexical
...
scope hasn't been set up yet so this isn't valid. It was just a cleanup to the
IR, so I'm going to ignore it for now.
llvm-svn: 107356
2010-07-01 00:16:31 +00:00
6696e22cc9
IRgen: Fix debug info regression in r106970; when we eliminate the return value
...
store make sure to move the debug metadata from the store (which is actual
'return' statement location) to the return instruction (which otherwise would
have the function end location as its debug info).
- Tested by gdb test suite.
llvm-svn: 107322
2010-06-30 21:27:58 +00:00
d7ac9a375a
IRgen: Make sure any prolog instructions get debug info.
...
llvm-svn: 107320
2010-06-30 21:27:56 +00:00
5c740f1523
Reapply:
...
r107173, "fix PR7519: after thrashing around and remembering how all this stuff"
r107216, "fix PR7523, which was caused by the ABI code calling ConvertType instead"
This includes a fix to make ConvertTypeForMem handle the "recursive" case, and call
it as such when lowering function types which have an indirect result.
llvm-svn: 107310
2010-06-30 19:14:05 +00:00
02995320e9
Use isFunctionOrMethod for vars declared locallly
...
in method/blocks to decide not to mangle them.
llvm-svn: 107309
2010-06-30 18:57:21 +00:00
0f19afb9d4
extern variable declared locally to objective-c++ method
...
should not be mangled either. Fixes radar 8016412.
llvm-svn: 107303
2010-06-30 18:27:47 +00:00
749b8ed5af
reduce nesting.
...
llvm-svn: 107292
2010-06-30 16:58:07 +00:00
2a47730767
Mangle arrays in the Microsoft C++ Mangler. It's not quite finished (it
...
doesn't mangle array parameters right), but I think that should be fixed
in Sema (Doug, John, what do you think?).
Also, stub out the remaining mangleType() routines.
llvm-svn: 107264
2010-06-30 08:09:57 +00:00
e422266926
Revert r107173, "fix PR7519: after thrashing around and remembering how all this stuff", it broke bootstrap.
...
llvm-svn: 107232
2010-06-30 00:22:35 +00:00
8386469d7d
Revert r107216, "fix PR7523, which was caused by the ABI code calling ConvertType instead", it is part of a boostrap breaking sequence.
...
llvm-svn: 107231
2010-06-30 00:22:30 +00:00
8c94ffe776
IRgen: Assignment to Objective-C properties shouldn't reload the value, for
...
complex values either. Previously we did this properly for regular assignment,
but not for compound assignment.
- Also, tidy up assignment code a bit to look more like the scalar path.
llvm-svn: 107217
2010-06-29 22:44:21 +00:00
466b1419c6
fix PR7523, which was caused by the ABI code calling ConvertType instead
...
of ConvertTypeRecursive when it needed to in a few cases, causing pointer
types to get resolved at the wrong time.
llvm-svn: 107216
2010-06-29 22:39:04 +00:00
c85ea8e175
IRgen: Assignment to Objective-C properties shouldn't reload the value (which
...
would trigger an extra method call).
- While in the area, I also changed Clang to not emit an unnecessary load from
'x' in cases like 'y = (x = 1)'.
llvm-svn: 107210
2010-06-29 22:00:45 +00:00
cccaad9584
change ABIArgInfo to hold its llvm type with PATypeHolder so that
...
it doesn't dangle as types get refined. This fixes Shootout-C++/lists1
and probably also PR7522.
llvm-svn: 107196
2010-06-29 19:21:36 +00:00
34d6281ae5
relax the CGFunctionInfo::CGFunctionInfo ctor to allow any sequence
...
of CanQualTypes to be passed in.
llvm-svn: 107176
2010-06-29 18:13:52 +00:00
ab1e65e2ea
fix PR7519: after thrashing around and remembering how all this stuff
...
works, the fix is quite simple: just make sure to call ConvertTypeRecursive
when the function type being lowered is in the midst of ConvertType.
llvm-svn: 107173
2010-06-29 17:56:33 +00:00
e70a007b36
minor cleanups.
...
llvm-svn: 107150
2010-06-29 16:40:28 +00:00
22a931e3bb
Change X86_64ABIInfo to have ASTContext and TargetData ivars to
...
avoid passing ASTContext down through all the methods it has.
When classifying an argument, or argument piece, as INTEGER, check
to see if we have a pointer at exactly the same offset in the
preferred type. If so, use that pointer type instead of i64. This
allows us to compile A function taking a stringref into something
like this:
define i8* @foo(i64 %D.coerce0, i8* %D.coerce1) nounwind ssp {
entry:
%D = alloca %struct.DeclGroup, align 8 ; <%struct.DeclGroup*> [#uses=4]
%0 = getelementptr %struct.DeclGroup* %D, i32 0, i32 0 ; <i64*> [#uses=1]
store i64 %D.coerce0, i64* %0
%1 = getelementptr %struct.DeclGroup* %D, i32 0, i32 1 ; <i8**> [#uses=1]
store i8* %D.coerce1, i8** %1
%tmp = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i64*> [#uses=1]
%tmp1 = load i64* %tmp ; <i64> [#uses=1]
%tmp2 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 1 ; <i8**> [#uses=1]
%tmp3 = load i8** %tmp2 ; <i8*> [#uses=1]
%add.ptr = getelementptr inbounds i8* %tmp3, i64 %tmp1 ; <i8*> [#uses=1]
ret i8* %add.ptr
}
instead of this:
define i8* @foo(i64 %D.coerce0, i64 %D.coerce1) nounwind ssp {
entry:
%D = alloca %struct.DeclGroup, align 8 ; <%struct.DeclGroup*> [#uses=3]
%0 = insertvalue %0 undef, i64 %D.coerce0, 0 ; <%0> [#uses=1]
%1 = insertvalue %0 %0, i64 %D.coerce1, 1 ; <%0> [#uses=1]
%2 = bitcast %struct.DeclGroup* %D to %0* ; <%0*> [#uses=1]
store %0 %1, %0* %2, align 1
%tmp = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i64*> [#uses=1]
%tmp1 = load i64* %tmp ; <i64> [#uses=1]
%tmp2 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 1 ; <i8**> [#uses=1]
%tmp3 = load i8** %tmp2 ; <i8*> [#uses=1]
%add.ptr = getelementptr inbounds i8* %tmp3, i64 %tmp1 ; <i8*> [#uses=1]
ret i8* %add.ptr
}
This implements rdar://7375902 - [codegen quality] clang x86-64 ABI lowering code punishing StringRef
llvm-svn: 107123
2010-06-29 06:01:59 +00:00
399d22ac1b
plumb preferred types down into X86_64ABIInfo::classifyArgumentType,
...
no functionality change.
llvm-svn: 107115
2010-06-29 01:14:09 +00:00
1d7c9f7f4b
Pass the LLVM IR version of argument types down into computeInfo.
...
This is somewhat annoying to do this at this level, but it avoids
having ABIInfo know depend on CodeGenTypes for a hint.
Nothing is using this yet, so no functionality change.
llvm-svn: 107111
2010-06-29 01:08:48 +00:00
9e748e9d6e
add IR names to coerced arguments.
...
llvm-svn: 107105
2010-06-29 00:14:52 +00:00
15ec361bd6
make the argument passing stuff in the FCA case smarter still, by
...
avoiding making the FCA at all when the types exactly line up. For
example, before we made:
%struct.DeclGroup = type { i64, i64 }
define i64 @_Z3foo9DeclGroup(i64, i64) nounwind {
entry:
%D = alloca %struct.DeclGroup, align 8 ; <%struct.DeclGroup*> [#uses=3]
%2 = insertvalue %struct.DeclGroup undef, i64 %0, 0 ; <%struct.DeclGroup> [#uses=1]
%3 = insertvalue %struct.DeclGroup %2, i64 %1, 1 ; <%struct.DeclGroup> [#uses=1]
store %struct.DeclGroup %3, %struct.DeclGroup* %D
%tmp = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i64*> [#uses=1]
%tmp1 = load i64* %tmp ; <i64> [#uses=1]
%tmp2 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 1 ; <i64*> [#uses=1]
%tmp3 = load i64* %tmp2 ; <i64> [#uses=1]
%add = add nsw i64 %tmp1, %tmp3 ; <i64> [#uses=1]
ret i64 %add
}
... which has the pointless insertvalue, which fastisel hates, now we
make:
%struct.DeclGroup = type { i64, i64 }
define i64 @_Z3foo9DeclGroup(i64, i64) nounwind {
entry:
%D = alloca %struct.DeclGroup, align 8 ; <%struct.DeclGroup*> [#uses=4]
%2 = getelementptr %struct.DeclGroup* %D, i32 0, i32 0 ; <i64*> [#uses=1]
store i64 %0, i64* %2
%3 = getelementptr %struct.DeclGroup* %D, i32 0, i32 1 ; <i64*> [#uses=1]
store i64 %1, i64* %3
%tmp = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i64*> [#uses=1]
%tmp1 = load i64* %tmp ; <i64> [#uses=1]
%tmp2 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 1 ; <i64*> [#uses=1]
%tmp3 = load i64* %tmp2 ; <i64> [#uses=1]
%add = add nsw i64 %tmp1, %tmp3 ; <i64> [#uses=1]
ret i64 %add
}
This only kicks in when x86-64 abi lowering decides it likes us.
llvm-svn: 107104
2010-06-29 00:06:42 +00:00
3dd716c3c3
Change CGCall to handle the "coerce" case where the coerce-to type
...
is a FCA to pass each of the elements as individual scalars. This
produces code fast isel is less likely to reject and is easier on
the optimizers.
For example, before we would compile:
struct DeclGroup { long NumDecls; char * Y; };
char * foo(DeclGroup D) {
return D.NumDecls+D.Y;
}
to:
%struct.DeclGroup = type { i64, i64 }
define i64 @_Z3foo9DeclGroup(%struct.DeclGroup) nounwind {
entry:
%D = alloca %struct.DeclGroup, align 8 ; <%struct.DeclGroup*> [#uses=3]
store %struct.DeclGroup %0, %struct.DeclGroup* %D, align 1
%tmp = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i64*> [#uses=1]
%tmp1 = load i64* %tmp ; <i64> [#uses=1]
%tmp2 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 1 ; <i64*> [#uses=1]
%tmp3 = load i64* %tmp2 ; <i64> [#uses=1]
%add = add nsw i64 %tmp1, %tmp3 ; <i64> [#uses=1]
ret i64 %add
}
Now we get:
%0 = type { i64, i64 }
%struct.DeclGroup = type { i64, i8* }
define i8* @_Z3foo9DeclGroup(i64, i64) nounwind {
entry:
%D = alloca %struct.DeclGroup, align 8 ; <%struct.DeclGroup*> [#uses=3]
%2 = insertvalue %0 undef, i64 %0, 0 ; <%0> [#uses=1]
%3 = insertvalue %0 %2, i64 %1, 1 ; <%0> [#uses=1]
%4 = bitcast %struct.DeclGroup* %D to %0* ; <%0*> [#uses=1]
store %0 %3, %0* %4, align 1
%tmp = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i64*> [#uses=1]
%tmp1 = load i64* %tmp ; <i64> [#uses=1]
%tmp2 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 1 ; <i8**> [#uses=1]
%tmp3 = load i8** %tmp2 ; <i8*> [#uses=1]
%add.ptr = getelementptr inbounds i8* %tmp3, i64 %tmp1 ; <i8*> [#uses=1]
ret i8* %add.ptr
}
Elimination of the FCA inside the function is still-to-come.
llvm-svn: 107099
2010-06-28 23:44:11 +00:00
d200eda487
make the trivial forms of CreateCoerced{Load|Store} trivial.
...
llvm-svn: 107091
2010-06-28 22:51:39 +00:00
93af332819
pass/return structs of char and short as i8/i16 to avoid
...
aweful through-memory coersion, just like we do for i32 now.
llvm-svn: 107078
2010-06-28 21:59:07 +00:00
d776fb150e
more tidying up.
...
llvm-svn: 107076
2010-06-28 21:43:59 +00:00
0cf2419cd7
random acts of tidying.
...
llvm-svn: 107050
2010-06-28 20:05:43 +00:00
a7d81ab7f3
X86-64:
...
pass/return structs of float/int as float/i32 instead of double/i64
to make the code generated for ABI cleaner. Passing in the low part
of a double is the same as passing in a float.
For example, we now compile:
struct DeclGroup { float NumDecls; };
float foo(DeclGroup D);
void bar(DeclGroup *D) {
foo(*D);
}
into:
%struct.DeclGroup = type { float }
define void @_Z3barP9DeclGroup(%struct.DeclGroup* %D) nounwind {
entry:
%D.addr = alloca %struct.DeclGroup*, align 8 ; <%struct.DeclGroup**> [#uses=2]
%agg.tmp = alloca %struct.DeclGroup, align 4 ; <%struct.DeclGroup*> [#uses=2]
store %struct.DeclGroup* %D, %struct.DeclGroup** %D.addr
%tmp = load %struct.DeclGroup** %D.addr ; <%struct.DeclGroup*> [#uses=1]
%tmp1 = bitcast %struct.DeclGroup* %agg.tmp to i8* ; <i8*> [#uses=1]
%tmp2 = bitcast %struct.DeclGroup* %tmp to i8* ; <i8*> [#uses=1]
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* %tmp2, i64 4, i32 4, i1 false)
%coerce.dive = getelementptr %struct.DeclGroup* %agg.tmp, i32 0, i32 0 ; <float*> [#uses=1]
%0 = load float* %coerce.dive, align 1 ; <float> [#uses=1]
%call = call float @_Z3foo9DeclGroup(float %0) ; <float> [#uses=0]
ret void
}
instead of:
%struct.DeclGroup = type { float }
define void @_Z3barP9DeclGroup(%struct.DeclGroup* %D) nounwind {
entry:
%D.addr = alloca %struct.DeclGroup*, align 8 ; <%struct.DeclGroup**> [#uses=2]
%agg.tmp = alloca %struct.DeclGroup, align 4 ; <%struct.DeclGroup*> [#uses=2]
%tmp3 = alloca double ; <double*> [#uses=2]
store %struct.DeclGroup* %D, %struct.DeclGroup** %D.addr
%tmp = load %struct.DeclGroup** %D.addr ; <%struct.DeclGroup*> [#uses=1]
%tmp1 = bitcast %struct.DeclGroup* %agg.tmp to i8* ; <i8*> [#uses=1]
%tmp2 = bitcast %struct.DeclGroup* %tmp to i8* ; <i8*> [#uses=1]
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* %tmp2, i64 4, i32 4, i1 false)
%coerce.dive = getelementptr %struct.DeclGroup* %agg.tmp, i32 0, i32 0 ; <float*> [#uses=1]
%0 = bitcast double* %tmp3 to float* ; <float*> [#uses=1]
%1 = load float* %coerce.dive ; <float> [#uses=1]
store float %1, float* %0, align 1
%2 = load double* %tmp3 ; <double> [#uses=1]
%call = call float @_Z3foo9DeclGroup(double %2) ; <float> [#uses=0]
ret void
}
which is this machine code (at -O0):
__Z3barP9DeclGroup:
subq $24, %rsp
movq %rdi, 16(%rsp)
movq 16(%rsp), %rdi
leaq 8(%rsp), %rax
movl (%rdi), %ecx
movl %ecx, (%rax)
movss 8(%rsp), %xmm0
callq __Z3foo9DeclGroup
addq $24, %rsp
ret
vs this:
__Z3barP9DeclGroup:
subq $24, %rsp
movq %rdi, 16(%rsp)
movq 16(%rsp), %rdi
leaq 8(%rsp), %rax
movl (%rdi), %ecx
movl %ecx, (%rax)
movss 8(%rsp), %xmm0
movss %xmm0, (%rsp)
movsd (%rsp), %xmm0
callq __Z3foo9DeclGroup
addq $24, %rsp
ret
At -O3, it is the difference between this now:
__Z3barP9DeclGroup:
movss (%rdi), %xmm0
jmp __Z3foo9DeclGroup # TAILCALL
vs this before:
__Z3barP9DeclGroup:
movl (%rdi), %eax
movd %rax, %xmm0
jmp __Z3foo9DeclGroup # TAILCALL
llvm-svn: 107048
2010-06-28 19:56:59 +00:00
c42461e145
Minor refactorin of my last patch (radar 7860965 related).
...
llvm-svn: 107047
2010-06-28 19:42:10 +00:00
36ad0e99d5
Have __func__ and siblings point to block's implementation function
...
name. Fixes radar 7860965.
llvm-svn: 107044
2010-06-28 18:58:34 +00:00
c1028f689e
Fix UnitTests/2004-02-02-NegativeZero.c, which regressed when
...
I broke negate of FP values.
llvm-svn: 107019
2010-06-28 17:12:37 +00:00
3f48c603fb
Correctly destroy reference temporaries with global storage. Remove ErrorUnsupported call when binding a global reference to a non-lvalue. Fixes PR7326.
...
llvm-svn: 106983
2010-06-27 17:52:15 +00:00
18c205ecdf
Add a CreateReferenceTemporary that will do the right thing for variables with global storage.
...
llvm-svn: 106982
2010-06-27 17:23:46 +00:00
2969c8c69d
Simplify CodeGenFunction::EmitReferenceBindingToExpr as a first step towards fixing PR7326.
...
llvm-svn: 106981
2010-06-27 16:56:04 +00:00
ca68d357d4
Reduce indentation.
...
llvm-svn: 106980
2010-06-27 15:24:55 +00:00
818efb64a3
misc tidying
...
llvm-svn: 106978
2010-06-27 07:40:06 +00:00
5e016ae983
finally get around to doing a significant cleanup to irgen:
...
have CGF create and make accessible standard int32,int64 and
intptr types. This fixes a ton of 80 column violations
introduced by LLVMContextification and cleans up stuff a lot.
llvm-svn: 106977
2010-06-27 07:15:29 +00:00
e000907e13
tidy up OrderGlobalInits
...
llvm-svn: 106976
2010-06-27 06:32:58 +00:00
055097f024
If coercing something from int or pointer type to int or pointer type
...
(potentially after unwrapping it from a struct) do it without going through
memory. We now compile:
struct DeclGroup {
unsigned NumDecls;
};
int foo(DeclGroup D) {
return D.NumDecls;
}
into:
%struct.DeclGroup = type { i32 }
define i32 @_Z3foo9DeclGroup(i64) nounwind ssp noredzone {
entry:
%D = alloca %struct.DeclGroup, align 4 ; <%struct.DeclGroup*> [#uses=2]
%coerce.dive = getelementptr %struct.DeclGroup* %D, i32 0, i32 0 ; <i32*> [#uses=1]
%coerce.val.ii = trunc i64 %0 to i32 ; <i32> [#uses=1]
store i32 %coerce.val.ii, i32* %coerce.dive
%tmp = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i32*> [#uses=1]
%tmp1 = load i32* %tmp ; <i32> [#uses=1]
ret i32 %tmp1
}
instead of:
%struct.DeclGroup = type { i32 }
define i32 @_Z3foo9DeclGroup(i64) nounwind ssp noredzone {
entry:
%D = alloca %struct.DeclGroup, align 4 ; <%struct.DeclGroup*> [#uses=2]
%tmp = alloca i64 ; <i64*> [#uses=2]
%coerce.dive = getelementptr %struct.DeclGroup* %D, i32 0, i32 0 ; <i32*> [#uses=1]
store i64 %0, i64* %tmp
%1 = bitcast i64* %tmp to i32* ; <i32*> [#uses=1]
%2 = load i32* %1, align 1 ; <i32> [#uses=1]
store i32 %2, i32* %coerce.dive
%tmp1 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i32*> [#uses=1]
%tmp2 = load i32* %tmp1 ; <i32> [#uses=1]
ret i32 %tmp2
}
... which is quite a bit less terrifying.
llvm-svn: 106975
2010-06-27 06:26:04 +00:00
895c52ba8b
Same patch as the previous on the store side. Before we compiled this:
...
struct DeclGroup {
unsigned NumDecls;
};
int foo(DeclGroup D) {
return D.NumDecls;
}
to:
%struct.DeclGroup = type { i32 }
define i32 @_Z3foo9DeclGroup(i64) nounwind ssp noredzone {
entry:
%D = alloca %struct.DeclGroup, align 4 ; <%struct.DeclGroup*> [#uses=2]
%tmp = alloca i64 ; <i64*> [#uses=2]
store i64 %0, i64* %tmp
%1 = bitcast i64* %tmp to %struct.DeclGroup* ; <%struct.DeclGroup*> [#uses=1]
%2 = load %struct.DeclGroup* %1, align 1 ; <%struct.DeclGroup> [#uses=1]
store %struct.DeclGroup %2, %struct.DeclGroup* %D
%tmp1 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i32*> [#uses=1]
%tmp2 = load i32* %tmp1 ; <i32> [#uses=1]
ret i32 %tmp2
}
which caused fast isel bailouts due to the FCA load/store of %2. Now
we generate this just blissful code:
%struct.DeclGroup = type { i32 }
define i32 @_Z3foo9DeclGroup(i64) nounwind ssp noredzone {
entry:
%D = alloca %struct.DeclGroup, align 4 ; <%struct.DeclGroup*> [#uses=2]
%tmp = alloca i64 ; <i64*> [#uses=2]
%coerce.dive = getelementptr %struct.DeclGroup* %D, i32 0, i32 0 ; <i32*> [#uses=1]
store i64 %0, i64* %tmp
%1 = bitcast i64* %tmp to i32* ; <i32*> [#uses=1]
%2 = load i32* %1, align 1 ; <i32> [#uses=1]
store i32 %2, i32* %coerce.dive
%tmp1 = getelementptr inbounds %struct.DeclGroup* %D, i32 0, i32 0 ; <i32*> [#uses=1]
%tmp2 = load i32* %tmp1 ; <i32> [#uses=1]
ret i32 %tmp2
}
This avoids fastisel bailing out and is groundwork for future patch.
This reduces bailouts on CGStmt.ll to 911 from 935.
llvm-svn: 106974
2010-06-27 06:04:18 +00:00
1cd6698a7c
improve CreateCoercedLoad a bit to generate slightly less awful
...
IR when handling X86-64 by-value struct stuff. For example, we
use to compile this:
struct DeclGroup {
unsigned NumDecls;
};
int foo(DeclGroup D);
void bar(DeclGroup *D) {
foo(*D);
}
into:
define void @_Z3barP9DeclGroup(%struct.DeclGroup* %D) ssp nounwind {
entry:
%D.addr = alloca %struct.DeclGroup*, align 8 ; <%struct.DeclGroup**> [#uses=2]
%agg.tmp = alloca %struct.DeclGroup, align 4 ; <%struct.DeclGroup*> [#uses=2]
%tmp3 = alloca i64 ; <i64*> [#uses=2]
store %struct.DeclGroup* %D, %struct.DeclGroup** %D.addr
%tmp = load %struct.DeclGroup** %D.addr ; <%struct.DeclGroup*> [#uses=1]
%tmp1 = bitcast %struct.DeclGroup* %agg.tmp to i8* ; <i8*> [#uses=1]
%tmp2 = bitcast %struct.DeclGroup* %tmp to i8* ; <i8*> [#uses=1]
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* %tmp2, i64 4, i32 4, i1 false)
%0 = bitcast i64* %tmp3 to %struct.DeclGroup* ; <%struct.DeclGroup*> [#uses=1]
%1 = load %struct.DeclGroup* %agg.tmp ; <%struct.DeclGroup> [#uses=1]
store %struct.DeclGroup %1, %struct.DeclGroup* %0, align 1
%2 = load i64* %tmp3 ; <i64> [#uses=1]
call void @_Z3foo9DeclGroup(i64 %2)
ret void
}
which would cause fastisel to bail out due to the first class aggregate load %1. With
this patch we now compile it into the (still awful):
define void @_Z3barP9DeclGroup(%struct.DeclGroup* %D) nounwind ssp noredzone {
entry:
%D.addr = alloca %struct.DeclGroup*, align 8 ; <%struct.DeclGroup**> [#uses=2]
%agg.tmp = alloca %struct.DeclGroup, align 4 ; <%struct.DeclGroup*> [#uses=2]
%tmp3 = alloca i64 ; <i64*> [#uses=2]
store %struct.DeclGroup* %D, %struct.DeclGroup** %D.addr
%tmp = load %struct.DeclGroup** %D.addr ; <%struct.DeclGroup*> [#uses=1]
%tmp1 = bitcast %struct.DeclGroup* %agg.tmp to i8* ; <i8*> [#uses=1]
%tmp2 = bitcast %struct.DeclGroup* %tmp to i8* ; <i8*> [#uses=1]
call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* %tmp2, i64 4, i32 4, i1 false)
%coerce.dive = getelementptr %struct.DeclGroup* %agg.tmp, i32 0, i32 0 ; <i32*> [#uses=1]
%0 = bitcast i64* %tmp3 to i32* ; <i32*> [#uses=1]
%1 = load i32* %coerce.dive ; <i32> [#uses=1]
store i32 %1, i32* %0, align 1
%2 = load i64* %tmp3 ; <i64> [#uses=1]
%call = call i32 @_Z3foo9DeclGroup(i64 %2) noredzone ; <i32> [#uses=0]
ret void
}
which doesn't bail out. On CGStmt.ll, this reduces fastisel bail outs from 958 to 935,
and is the precursor of better things to come.
llvm-svn: 106973
2010-06-27 05:56:15 +00:00
3fcc790cd8
Change IR generation for return (in the simple case) to avoid doing silly
...
load/store nonsense in the epilog. For example, for:
int foo(int X) {
int A[100];
return A[X];
}
we used to generate:
%arrayidx = getelementptr inbounds [100 x i32]* %A, i32 0, i64 %idxprom ; <i32*> [#uses=1]
%tmp1 = load i32* %arrayidx ; <i32> [#uses=1]
store i32 %tmp1, i32* %retval
%0 = load i32* %retval ; <i32> [#uses=1]
ret i32 %0
}
which codegen'd to this code:
_foo: ## @foo
## BB#0: ## %entry
subq $408, %rsp ## imm = 0x198
movl %edi, 400(%rsp)
movl 400(%rsp), %edi
movslq %edi, %rax
movl (%rsp,%rax,4), %edi
movl %edi, 404(%rsp)
movl 404(%rsp), %eax
addq $408, %rsp ## imm = 0x198
ret
Now we generate:
%arrayidx = getelementptr inbounds [100 x i32]* %A, i32 0, i64 %idxprom ; <i32*> [#uses=1]
%tmp1 = load i32* %arrayidx ; <i32> [#uses=1]
ret i32 %tmp1
}
and:
_foo: ## @foo
## BB#0: ## %entry
subq $408, %rsp ## imm = 0x198
movl %edi, 404(%rsp)
movl 404(%rsp), %edi
movslq %edi, %rax
movl (%rsp,%rax,4), %eax
addq $408, %rsp ## imm = 0x198
ret
This actually does matter, cutting out 2000 lines of IR from CGStmt.ll
for example.
Another interesting effect is that altivec.h functions which are dead
now get dce'd by the inliner. Hence all the changes to
builtins-ppc-altivec.c to ensure the calls aren't dead.
llvm-svn: 106970
2010-06-27 01:06:27 +00:00
726b3d09cd
reduce indentation
...
llvm-svn: 106967
2010-06-26 23:13:19 +00:00
6c5abe88bf
Implement rdar://7530813 - collapse multiple GEP instructions in IRgen
...
This avoids generating two gep's for common array operations. Before
we would generate something like:
%tmp = load i32* %X.addr ; <i32> [#uses=1]
%arraydecay = getelementptr inbounds [100 x i32]* %A, i32 0, i32 0 ; <i32*> [#uses=1]
%arrayidx = getelementptr inbounds i32* %arraydecay, i32 %tmp ; <i32*> [#uses=1]
%tmp1 = load i32* %arrayidx ; <i32> [#uses=1]
Now we generate:
%tmp = load i32* %X.addr ; <i32> [#uses=1]
%arrayidx = getelementptr inbounds [100 x i32]* %A, i32 0, i32 %tmp ; <i32*> [#uses=1]
%tmp1 = load i32* %arrayidx ; <i32> [#uses=1]
Less IR is better at -O0.
llvm-svn: 106966
2010-06-26 23:03:20 +00:00
57ce97151f
minor cleanup: don't emit the base of an array subscript until after
...
we're done diddling around with the index stuff. Use a cheaper type
comparison.
llvm-svn: 106963
2010-06-26 22:40:46 +00:00
431bef4409
fix inc/dec to honor -fwrapv and -ftrapv, implementing PR7426.
...
llvm-svn: 106962
2010-06-26 22:18:28 +00:00
05dc78c096
move scalar inc/dec codegen into ScalarExprEmitter instead
...
of being in CGF. No functionality change.
llvm-svn: 106961
2010-06-26 22:09:34 +00:00
fa20e95043
use more efficient type comparison predicates.
...
llvm-svn: 106958
2010-06-26 21:52:32 +00:00
0bf27620f0
Fix unary minus to trap on overflow with -ftrapv, refactoring binop
...
code so we can use it from VisitUnaryMinus.
llvm-svn: 106957
2010-06-26 21:48:21 +00:00
51924e517b
Implement support for -fwrapv, rdar://7221421
...
As part of this, pull together trapv handling into the same enum.
This also add support for NSW multiplies.
This also makes PCH disagreement on overflow behavior silent, since it
really doesn't matter except for warnings and codegen (no macros get
defined etc).
llvm-svn: 106956
2010-06-26 21:25:03 +00:00
217e056e40
implement rdar://7432000 - signed negate should codegen as NSW.
...
While I'm in there, adjust pointer to member adjustments as well.
llvm-svn: 106955
2010-06-26 20:27:24 +00:00
04775f8413
Change EmitReferenceBindingToExpr to take a decl instead of a boolean.
...
llvm-svn: 106949
2010-06-26 16:35:32 +00:00
709ef8e46c
Add function for mangling reference temporaries.
...
llvm-svn: 106948
2010-06-26 16:09:40 +00:00
f4db33cbdf
Mangle pointer and (lvalue) reference types in the Microsoft C++ Mangler.
...
Also, fix mangling of throw specs. Turns out MSVC totally ignores throw
specs when mangling names.
llvm-svn: 106937
2010-06-26 03:50:05 +00:00
b66b08ef01
Minor change to my last patch to fix PR7490.
...
llvm-svn: 106875
2010-06-25 20:01:13 +00:00
d5202e0926
IRGen for trivial initialization of dynamiccaly allocated
...
array of other done c++ objects. Fixes PR7490.
llvm-svn: 106869
2010-06-25 18:26:07 +00:00
9b5528d278
Patch to correctly mangle block helper functions
...
when block literal is declared inside a ctor/dtor.
Fixes radr 8096995.
llvm-svn: 106700
2010-06-24 00:08:06 +00:00
37141f4fb4
improve altivec vector bool/pixel support, patch by Anton Yartsev
...
with several tweaks by me.
llvm-svn: 106619
2010-06-23 06:00:24 +00:00
b48a2d5d2f
polish the -finstrument-functions implementation, patch by Nelson Elhage!
...
llvm-svn: 106618
2010-06-23 05:21:28 +00:00
49b4d73451
Type Type::isRealFloatingType() that vectors are not floating-point
...
types, updating callers of both isFloatingType() and
isRealFloatingType() accordingly. Caught at least one issue where we
allowed one to declare a vector of vectors (!), along with cleaning up
the standard-conversion logic for C++.
llvm-svn: 106595
2010-06-22 23:07:26 +00:00
7966aff294
Fixes a problem in generation of meta-data for
...
category implementation whereby property list was missing.
NeXt ObjC runtime (radar 8093297).
llvm-svn: 106550
2010-06-22 16:33:55 +00:00
ea836bc41c
Switch over to the new caching version of getMangledName.
...
llvm-svn: 106549
2010-06-22 16:16:50 +00:00
2e2f4d2e6d
Add a new variant of getMangledName that caches the mangling for decls.
...
llvm-svn: 106547
2010-06-22 16:05:32 +00:00
16d9f1788e
Add GlobalDecl::getCanonicalDecl.
...
llvm-svn: 106546
2010-06-22 16:00:14 +00:00
469b200321
Per Chris's comment, remove another static (this one
...
is a static comparator operator).
llvm-svn: 106511
2010-06-22 00:23:08 +00:00
3c77a355e0
implement support for -finstrument-functions, patch by Nelson
...
Elhage!
llvm-svn: 106507
2010-06-22 00:03:40 +00:00
c9fec975e6
Avoid using a local static for providing lexical
...
order of priorotized global object initializations.
llvm-svn: 106503
2010-06-21 23:31:29 +00:00
9adb2e646c
In fragile-abi (32bit mode abi) generate global symbol
...
objc_category_name_xxx for each category implementation.
(fixes PR7431) patch by Nico Weber.
llvm-svn: 106492
2010-06-21 22:05:18 +00:00
89bdd14f2f
In supporting init-priority, globals with the same init_priority must be
...
emitted in the order in which they are seen (still radar 8076356).
llvm-svn: 106485
2010-06-21 21:27:42 +00:00
Douglas Gregor