maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
85,913 Commits 23 Branches 0 Tags 2.2 GiB
Commit Graph

3736 Commits

Author SHA1 Message Date
Chris Lattner 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
Chris Lattner 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
Chris Lattner 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
Chris Lattner 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
Chris Lattner 726b3d09cd reduce indentation 
llvm-svn: 106967
 2010-06-26 23:13:19 +00:00
Chris Lattner 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
Chris Lattner 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
Chris Lattner 431bef4409 fix inc/dec to honor -fwrapv and -ftrapv, implementing PR7426. 
llvm-svn: 106962
 2010-06-26 22:18:28 +00:00
Chris Lattner 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
Chris Lattner fa20e95043 use more efficient type comparison predicates. 
llvm-svn: 106958
 2010-06-26 21:52:32 +00:00
Chris Lattner 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
Chris Lattner 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
Chris Lattner 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
Anders Carlsson 04775f8413 Change EmitReferenceBindingToExpr to take a decl instead of a boolean. 
llvm-svn: 106949
 2010-06-26 16:35:32 +00:00
Anders Carlsson 709ef8e46c Add function for mangling reference temporaries. 
llvm-svn: 106948
 2010-06-26 16:09:40 +00:00
Charles Davis 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
Fariborz Jahanian b66b08ef01 Minor change to my last patch to fix PR7490. 
llvm-svn: 106875
 2010-06-25 20:01:13 +00:00
Fariborz Jahanian 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
Fariborz Jahanian 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
Chris Lattner 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
Chris Lattner b48a2d5d2f polish the -finstrument-functions implementation, patch by Nelson Elhage! 
llvm-svn: 106618
 2010-06-23 05:21:28 +00:00
Douglas Gregor 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
Fariborz Jahanian 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
Anders Carlsson ea836bc41c Switch over to the new caching version of getMangledName. 
llvm-svn: 106549
 2010-06-22 16:16:50 +00:00
Anders Carlsson 2e2f4d2e6d Add a new variant of getMangledName that caches the mangling for decls. 
llvm-svn: 106547
 2010-06-22 16:05:32 +00:00
Anders Carlsson 16d9f1788e Add GlobalDecl::getCanonicalDecl. 
llvm-svn: 106546
 2010-06-22 16:00:14 +00:00
Fariborz Jahanian 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
Chris Lattner 3c77a355e0 implement support for -finstrument-functions, patch by Nelson 
Elhage!

llvm-svn: 106507
 2010-06-22 00:03:40 +00:00
Fariborz Jahanian 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
Fariborz Jahanian 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
Fariborz Jahanian 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
Anders Carlsson 280e61f148 Fix an Obj-C++ miscompile when calling an Obj-C method that returns a C++ reference. 
llvm-svn: 106477
 2010-06-21 20:59:55 +00:00
Fariborz Jahanian 090e4e5773 Use more efficient API of SmallVector/array_pod_sort 
for sorting (radar 8076356).

llvm-svn: 106453
 2010-06-21 19:49:38 +00:00
Fariborz Jahanian 9f2a4eed0d IRGen for implementation of init-priority attribute. 
Test case will be checked in llvm test suite.
(finishes off radar 8076356).

llvm-svn: 106441
 2010-06-21 18:45:05 +00:00
Douglas Gregor 5dd34744c5 Instantiations subject to an explicit template instantiation 
declaration have default visibility even under
-fvisibility=hidden. Fixes <rdar://problem/8109763>.

llvm-svn: 106440
 2010-06-21 18:41:26 +00:00
Nate Begeman ed48c857dc Implement remaining codegen for NEON, all operations should now work. 
llvm-svn: 106407
 2010-06-20 23:05:28 +00:00
Anton Korobeynikov cc50b7d7d5 More AltiVec support. 
Patch by Anton Yartsev!

llvm-svn: 106387
 2010-06-19 09:47:18 +00:00
Eli Friedman c96b2496fc Fix for PR7415: refactor CodeGenModule::MayDeferGeneration and make it less 
conservative for static variables in templated classes.

llvm-svn: 106385
 2010-06-19 06:24:06 +00:00
Chris Lattner 87233f785b Fix PR7097, a bad interaction between -fno-use-cxa-atexit and 
-mconstructor-aliases by using a WeakVH instead of a raw pointer.

llvm-svn: 106384
 2010-06-19 05:52:45 +00:00
Charles Davis 108f5a2748 Mangle tag types (unions, structs, classes, enums) in the Microsoft C++ Mangler. 
Also, test that static members with default visibility in a struct have the
right mangling.

llvm-svn: 106276
 2010-06-18 07:51:00 +00:00
Fariborz Jahanian 64cda8b5f3 objective-C++ IRGen: property reference as an 
lvalue when performing a derived-to-base conversion.
Fixes radar 7501812. Added an executable test to
llvm-test suite.

llvm-svn: 106247
 2010-06-17 23:00:29 +00:00
Fariborz Jahanian 9240f3dac7 Objective-c++ IRGen. Support for @selector expression as 
an lvalue. Fixes PR7390.

llvm-svn: 106235
 2010-06-17 19:56:20 +00:00
Charles Davis 8c02c13e19 Mangle operator names in the Microsoft C++ Mangler. 
llvm-svn: 106211
 2010-06-17 06:47:31 +00:00
Nate Begeman dbafec1f3e Remove last of the bool shifts for MS VC++, patch by dimitry andric 
llvm-svn: 106206
 2010-06-17 02:26:59 +00:00
Alexis Hunt c675ec09f0 Update CMake build for new attribute changes. 
llvm-svn: 106188
 2010-06-17 00:37:02 +00:00
Rafael Espindola 23a8a06554 Change the test for which ABI/CC to use on ARM to be base on the environment 
(the last argument of the triple).

llvm-svn: 106131
 2010-06-16 19:01:17 +00:00
Fariborz Jahanian 4a30307840 Fixed conflict between objc_memmove_collectable builtin 
decl. and one ddefined in darwin header file.

llvm-svn: 106107
 2010-06-16 16:22:04 +00:00
Rafael Espindola a92c442437 Don't set the calling convention for ARM if it is already the default. 
llvm-svn: 106106
 2010-06-16 16:13:39 +00:00
Charles Davis 89338af1ff Start mangling function types in the Microsoft C++ Mangler. 
llvm-svn: 106081
 2010-06-16 05:33:16 +00:00
Douglas Gregor 6972a62c8f Give Type::isIntegralType() an ASTContext parameter, so that it 
provides C "integer type" semantics in C and C++ "integral type"
semantics in C++. 

Note that I still need to update isIntegerType (and possibly other
predicates) using the same approach I've taken for
isIntegralType(). The two should have the same meaning, but currently
don't (!).

llvm-svn: 106074
 2010-06-16 00:35:25 +00:00