aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen/CodeGenFunction.h
AgeCommit message (Collapse)Author
2010-07-13Teach IR generation how to lazily emit cleanups. This has a lot of advantages,John McCall
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@108270 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-08Reinstate the fix for PR7556. A silly use of isTrivial() wasDouglas Gregor
suppressing copies of objects with trivial copy constructors. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@107857 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-07Revert r107828 and r107827, the fix for PR7556, which seems to beDouglas Gregor
breaking bootstrap on Linux. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@107837 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-07Rename CXXZeroInitValueExpr to CXXScalarValueInitExpr, to reflect itsDouglas Gregor
newly-narrowed scope. No functionality change. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@107828 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-07Teach function-try-blocks on constructors and destructors to implicitlyJohn McCall
rethrow. Fixes rdar://problem/7696603 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@107757 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-06Provide a hook for the benefit of clients using clang IR gen as a subroutine:John McCall
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. :) git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@107739 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-06Validated by nightly-test runs on x86 and x86-64 darwin, including afterJohn McCall
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@107631 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-05in the "coerce" case, the ABI handling code ends up making theChris Lattner
alloca for an argument. Make sure the argument gets the proper decl alignment, which may be different than the type alignment. This fixes PR7567 git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@107627 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-27finally get around to doing a significant cleanup to irgen:Chris Lattner
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106977 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-27Change IR generation for return (in the simple case) to avoid doing sillyChris Lattner
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. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106970 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-26Change EmitReferenceBindingToExpr to take a decl instead of a boolean.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106949 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-25IRGen for trivial initialization of dynamiccaly allocatedFariborz Jahanian
array of other done c++ objects. Fixes PR7490. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106869 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-24Patch to correctly mangle block helper functionsFariborz Jahanian
when block literal is declared inside a ctor/dtor. Fixes radr 8096995. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106700 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-22implement support for -finstrument-functions, patch by NelsonChris Lattner
Elhage! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106507 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-19Fix PR7097, a bad interaction between -fno-use-cxa-atexit andChris Lattner
-mconstructor-aliases by using a WeakVH instead of a raw pointer. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106384 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-17Objective-c++ IRGen. Support for @selector expression asFariborz Jahanian
an lvalue. Fixes PR7390. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@106235 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-14Add some missing shiftsNate Begeman
Fix multiplies by scalar Add SemaChecking code for all immediates Add SemaChecking-gen support to arm_neon.td git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@105930 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-11Most of remaining builtins, 2 generics, vld, and rounding shfits remain.Nate Begeman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@105848 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-10support _lane ops, and multiplies by scalar.Nate Begeman
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@105770 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-08Rename __tcf_ to __cxx_global_array_dtor. Remove the ↵Anders Carlsson
UniqueAggreDestructorCount from CodeGenFunction and let LLVM handle uniquing the internal functions instead. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@105648 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-08Simplify GenerateCXXAggrDestructorHelper.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@105646 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-08Fix NEON intrinsic argument passing, support vext. Most now successfully ↵Nate Begeman
make it through codegen to the .s file git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@105599 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-26Correctly pass aggregates by reference when emitting thunks.John McCall
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@104778 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-22Re-land the fix for PR7139.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@104446 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-22Push a return-value slot throughout ObjC message-send codegen. Will beJohn McCall
critical for ObjC++ correctness; hard to test independently of various required Sema changes, though. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@104422 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-21Unbreak self-host.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@104390 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-21Rename CodeGenFunction::EmitMemSetToZero to EmitNullInitialization. Handle ↵Anders Carlsson
setting null data member pointers correctly. Fixes PR7139. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@104387 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-21Allocate space in a block record for implicit references to the Objective CJohn McCall
'self' variable arising from uses of the 'super' keyword. Also reorganize some code so that BlockInfo (now CGBlockInfo) can be opaque outside of CGBlocks.cpp. Fixes rdar://problem/8010633. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@104312 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-21When emitting an lvalue for an anonymous struct or union member duringJohn McCall
class initialization, drill down through an arbitrary number of anonymous records. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@104310 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-20Support implicitly closing on 'this' in a block. Fixed PR7165.John McCall
(the codegen works here, too, but that's annoying to test without execution) git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@104202 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-17Ensure that destructors are called for NRVO'd objects when theDouglas Gregor
function does not return. Thanks to Eli for pointing out this corner case. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@103941 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-16When initializing thread-safe statics, put the call toDouglas Gregor
__cxa_guard_abort along the exceptional edge into (in effect) a nested "try" that rethrows after aborting. Fixes PR7144 and the remaining Boost.ProgramOptions failures, along with the regressions that r103880 caused. The crucial difference between this and r103880 is that we now follow LLVM's little dance with the llvm.eh.exception and llvm.eh.selector calls, then use _Unwind_Resume_or_Rethrow to rethrow. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@103892 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-16Revert r103880 (thread-safe static initialization w/ exceptions),Douglas Gregor
because it's causing strange linker errors. Unfixes PR7144. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@103890 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-15When initializing thread-safe statics, put the call toDouglas Gregor
__cxa_guard_abort along the exceptional edge into (in effect) a nested "try" that rethrows after aborting. Fixes PR7144 and the remaining Boost.ProgramOptions failures. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@103880 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-14Remove an unused function.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@103793 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-05Reimplement code generation for copying fields in theDouglas Gregor
implicitly-generated copy constructor. Previously, Sema would perform some checking and instantiation to determine which copy constructors, etc., would be called, then CodeGen would attempt to figure out which copy constructor to call... but would get it wrong, or poke at an uninstantiated default argument, or fail in other ways. The new scheme is similar to what we now do for the implicit copy-assignment operator, where Sema performs all of the semantic analysis and builds specific ASTs that look similar to the ASTs we'd get from explicitly writing the copy constructor, so that CodeGen need only do a direct translation. However, it's not quite that simple because one cannot explicit write elementwise copy-construction of an array. So, I've extended CXXBaseOrMemberInitializer to contain a list of indexing variables used to copy-construct the elements. For example, if we have: struct A { A(const A&); }; struct B { A array[2][3]; }; then we generate an implicit copy assignment operator for B that looks something like this: B::B(const B &other) : array[i0][i1](other.array[i0][i1]) { } CodeGen will loop over the invented variables i0 and i1 to visit all elements in the array, so that each element in the destination array will be copy-constructed from the corresponding element in the source array. Of course, if we're dealing with arrays of scalars or class types with trivial copy-assignment operators, we just generate a memcpy rather than a loop. Fixes PR6928, PR5989, and PR6887. Boost.Regex now compiles and passes all of its regression tests. Conspicuously missing from this patch is handling for the exceptional case, where we need to destruct those objects that we have constructed. I'll address that case separately. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@103079 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-03Remove OldGetAddressOfBaseClass - bye bye ambiguities.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102889 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-03More work towards getting rid of OldGetAddressOfBaseClass.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102887 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-02Add the same 'ForVirtualBase' parameter to EmitCXXDestructorCall.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102882 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-02Revert my last change and add a 'ForVirtualBase' parameter to ↵Anders Carlsson
EmitCXXConstructorCall instead. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102881 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-02Pass the construction kind down to EmitCXXConstructorCall.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102880 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-02Remove another unused function.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102871 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-02Remove an unused function.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102870 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-02As per Chris' request, return the Instruction from EmitCall and add the ↵David Chisnall
metadata in the caller. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102862 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-01Complete reimplementation of the synthesis for implicitly-defined copyDouglas Gregor
assignment operators. Previously, Sema provided type-checking and template instantiation for copy assignment operators, then CodeGen would synthesize the actual body of the copy constructor. Unfortunately, the two were not in sync, and CodeGen might pick a copy-assignment operator that is different from what Sema chose, leading to strange failures, e.g., link-time failures when CodeGen called a copy-assignment operator that was not instantiation, run-time failures when copy-assignment operators were overloaded for const/non-const references and the wrong one was picked, and run-time failures when by-value copy-assignment operators did not have their arguments properly copy-initialized. This implementation synthesizes the implicitly-defined copy assignment operator bodies in Sema, so that the resulting ASTs encode exactly what CodeGen needs to do; there is no longer any special code in CodeGen to synthesize copy-assignment operators. The synthesis of the body is relatively simple, and we generate one of three different kinds of copy statements for each base or member: - For a class subobject, call the appropriate copy-assignment operator, after overload resolution has determined what that is. - For an array of scalar types or an array of class types that have trivial copy assignment operators, construct a call to __builtin_memcpy. - For an array of class types with non-trivial copy assignment operators, synthesize a (possibly nested!) for loop whose inner statement calls the copy constructor. - For a scalar type, use built-in assignment. This patch fixes at least a few tests cases in Boost.Spirit that were failing because CodeGen picked the wrong copy-assignment operator (leading to link-time failures), and I suspect a number of undiagnosed problems will also go away with this change. Some of the diagnostics we had previously have gotten worse with this change, since we're going through generic code for our type-checking. I will improve this in a subsequent patch. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102853 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-01Simplify EmitClassAggrMemberwiseCopy.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102848 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-01Clean up EmitClassMemberwiseCopy further.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102846 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-01Get rid of a parameter from EmitClassMemberwiseCopy.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102845 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-01Tweaked EmitCall() to permit the caller to provide some metadata to attach ↵David Chisnall
to the call site. Used this in CGObjCGNU to attach metadata about message sends to permit speculative inlining. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102833 91177308-0d34-0410-b5e6-96231b3b80d8
2010-04-30Remove an unnecessary parameter from EmitClassCopyAssignment.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@102747 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-05-10 21:35:18 +0000