aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen
AgeCommit message (Collapse)Author
2010-08-05Fix a major bug with -ftrapv and ++/--. Patch by David Keaton!John McCall
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110347 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-05For now skip over aggregate non-byref block variables.Fariborz Jahanian
(objc gc specific). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110340 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-05Trying to unbreak buildbot.Fariborz Jahanian
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110339 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-05Add IRGen support for non-constant OffsetOfExpr.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110326 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-05Implement #pragma GCC visibility.Eli Friedman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110315 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-05Clean up of my last patch.Fariborz Jahanian
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110290 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04Block variable layout bitmap API generation.Fariborz Jahanian
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110287 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04Extend the visibility-hidden optimization to linkonce_odr thunks forJohn McCall
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110285 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04More objc block variable layout info. work.Fariborz Jahanian
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110239 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04Some early work for providing block layout info. Fariborz Jahanian
for objective-c/c++ blocks (NeXt runtime). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110213 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04Emit standard-library RTTI with external linkage, not weak_odr.John McCall
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110192 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04Extend the hidden-visibility vtables optimization to template classes thatJohn McCall
haven't been explicitly instantiated. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110189 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-04Some more correctness fixes and code-size optimizations for fragile-ABIJohn McCall
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). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110188 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-03Do a very simple pass over every function we emit to infer whether we canJohn McCall
mark it nounwind based on whether it contains any non-nounwind calls. <rdar://problem/8087431> git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110163 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-03Add support for VFP status & control operations for ARM.Nate Begeman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110153 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-03Emit weak vtables of non-template classes with hidden visibility.John McCall
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110107 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-02Further adjustments to -Wglobal-constructors; works for references and directJohn McCall
initializations now. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@110063 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-31Fix fragile-ABI ObjC exceptions in the presence of optimization withJohn McCall
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 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109960 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-30Emit global destructors even if the destroyed object has no initializers or hasJohn McCall
an initializer requiring temporary object disposal. Fixes rdar:://problem/8246444. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109849 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-30fix PR5179 and correctly fix PR5831 to not miscompile.Chris Lattner
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109848 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29Revert "Re-apply fixed version of 108749, correctly conditionalizing the newDaniel Dunbar
sections on", this change uncovered a possible linker bug which resulted in the wrong messages getting dispatched. Backing this out while we investigate... git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109817 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29Wire up sema checking for __builtin_arm_usat and __builtin_arm_ssat immediates.Nate Begeman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109814 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29move the last hunk of getCoerceResult into the placeChris Lattner
that needs it and remove getCoerceResult. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109807 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29now that direct and coerce are merged, getCoerceResult gets simpler.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109805 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29now that GetSSETypeAtOffset handles passing SSE class values asChris Lattner
float, the special case hack in getCoerceResult can go away. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109804 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29Change the name to something less terrible; suggestion by Doug. No ↵Argyrios Kyrtzidis
functionality change. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109797 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29Implement the clang-side of detection for when to pass asChris Lattner
<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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109790 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29Look at me, I can count!Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109786 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29Merge PCHWriterDecl.cpp's isRequiredDecl and ↵Argyrios Kyrtzidis
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109784 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29fix rdar://8251384, another case where we could access beyond theChris Lattner
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 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109776 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29start setting up infrastructure for passing multi-floatsChris Lattner
as <2 x float> instead of as double. The backend isn't ready yet, but infrastructure in the frontend can come up. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109768 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29rename Get8ByteTypeAtOffset -> GetINTEGERTypeAtOffset toChris Lattner
make it clear that this function should only return a type that the codegen will classify the same as an INTEGER type. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109763 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29handle a case where we could access off the end of a functionChris Lattner
that Eli pointed out, rdar://8249586 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109762 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29fix PR7742 / rdar://8250764, a miscompilation of structChris Lattner
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109759 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29This is a little bit far, but optimize cases like:Chris Lattner
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 :) git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109738 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29implement a todo: pass a eight-byte that consists of aChris Lattner
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 :) git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109737 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29fix a builder, why didn't clang++ catch this?Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109735 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29Kill off the 'coerce' ABI passing form. Now 'direct' and 'extend' alwaysChris Lattner
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 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109733 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29ignore structs that wrap vectors in IR, the abstraction shouldn't add penalty.Chris Lattner
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 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109732 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29move the 'pretty 16-byte vector' inferring code up to be sharedChris Lattner
with return values, improving stuff that returns __m128 etc. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109731 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29simplify code by eliminating a premature optimization.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109730 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29now that we have CGT around, we can start using preferred typesChris Lattner
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109729 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29sink preferred type stuff lower. It's possible that this mightChris Lattner
improve codegen for vaarg or something, because its codepath is getting preferred types now. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109728 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29dissolve some more complexity: make the x86-64 abi lowering codeChris Lattner
compute its own preferred types instead of having CGT compute them then pass them (circuituously) down into ABIInfo. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109726 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29simplify Get8ByteTypeAtOffset by making it a member of X86_64ABIInfoChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109724 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29now that ABIInfo depends on CGT, it has trivial access to suchChris Lattner
things as TargetData, ASTContext, LLVMContext etc. Stop passing them through so many APIs. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109723 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-29cave in to reality and make ABIInfo depend on CodeGenTypes.Chris Lattner
This will simplify a bunch of code, coming up next. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109722 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-28pass argument vectors in a type that corresponds to the user type ifChris Lattner
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109700 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-28tidy upChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109699 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-28Override selected builtin names (e.g. "long int" instead of "long") to match ↵Devang Patel
names used by gcc in debug info. This makes gdb testsuite happy. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@109694 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-07-19 12:00:41 +0000