aboutsummaryrefslogtreecommitdiff
path: root/lib/CodeGen
AgeCommit message (Collapse)Author
2008-09-17Improve x86 ABI compatibility.Daniel Dunbar
- Enables use of ABIArgInfo::Expand when needed. This greatly improves our x86 ABI compatibility. - As the infrastructure for target specific ABI handling isn't built yet, this change means ABI compatibility on other platforms is once again broken in a different way than before. - Upcoming: Figure out how to refactor ABI handling into targets. More documentation. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56293 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-17Print current function on verification failures.Daniel Dunbar
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56291 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-17Drop CodeGenTypes parameter to classifyArgumentType.Daniel Dunbar
Ensure that ABIArgInfo::Default implementation for aggregrate types is sensible (StructRet on return and ByVal on arguments). No functionality change. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56289 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-17Add support for ABIArgInfo::ExpandDaniel Dunbar
- No functionality change. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56269 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-11Make sure to store the exception in the catch parameter.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56102 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-11Make sure to emit the catch parameter as well as the catch body.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56101 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-11Fix stupid mistake I made in the exception handling codeAnders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56100 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-11Use ABIArgInfo for decisions about function arguments (not just returnDaniel Dunbar
value). - Added ABIArgInfo::ByVal (mostly supported) and ABIArgInfo::Expand (asserted out). - Added classifyArgumentType which currently just uses ABIArgInfo::Default or ByVal. This nearly matches old behavior, but we now set ByVal in a few situations we may have left it off before (on complex, for example). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56086 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-11Bug fix, apply default argument promotion in message sends for whichDaniel Dunbar
no method declaration was found. - This was allowing arrays to pass "by value" among other things. Add assert in CodeGen that arguments cannot have array type. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56080 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-10Implement ABIArgType::Coerce support.Daniel Dunbar
- As a test, enable basic usage for some common x86-32 cases. This increases our x86-32 compliance (on other targets our compliance will just be broken in a different way). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56051 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-10Add CodeGenTypes::GetFunctionType overload for getting the effectiveDaniel Dunbar
type of a call. Change NeXT runtime to use this instead of trying to bitcasting internally (which doesn't respect the ABI). Fix subtle bug, use of ConvertTypeRecursive instead of ConvertType is bad inside GetFunctionType. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56050 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-10Move FunctionType conversion into CGCall.cpp:Daniel Dunbar
- Added CodeGenTypes::GetFunctionType, taking a CGFunctionInfo. - Updated Obj-C runtimes to use this instead of rolling the llvm::FunctionType by hand. - Killed CodeGenTypes::{ConvertReturnType, DecodeArgumentTypes}. Add ABIArgInfo class to encapsulate ABI decision of how to lower types to LLVM. - Will move to target sometime soon. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56047 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-10Add ABIArgInfo class to encapsulate ABI decision of how to lower typesDaniel Dunbar
to LLVM. - Will move to target sometime soon. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56046 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-10Implement CodeGen support for the 'CXXConditionDeclExpr' expression node, ↵Argyrios Kyrtzidis
which represents a 'condition' declaration, e.g: "if (int x=0) {...}". git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56045 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-10Tweak CGCall functions again:Daniel Dunbar
- Realized these functions will eventually need access to more data, moved to CodeGenModule. Eventually they should probably live together in some other helper class. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56039 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-10Tweak CGCall functions:Daniel Dunbar
- Move actual param attr list creation to CodeGenFunction::ConstructParamAttrList. - Make ReturnTypeUsesSret static. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56038 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Add CodeGenFunction::ReturnTypeUsesSretDaniel Dunbar
- Hook so NeXT runtime doesn't depend on ABI. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56034 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Move ABI specific code for functions / calls to CGCall.cpp:Daniel Dunbar
- Factor out EmitFunction{Pro,Epi}log git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56031 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Factor CodeGenFunction::StartFunction out of GenerateCode andDaniel Dunbar
StartObjCMethod. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56030 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Use a unified return block.Daniel Dunbar
- For the time being this means our emitted code is somewhat worse, especially for aggregates. This will be fixed. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56013 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Move EmitAggregate{Copy,Clear} into CodeGenFunction.Daniel Dunbar
- No functionality change. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@56010 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Check in half-assed implementation of @try/@catch.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55994 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Simple @throw support.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55991 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Add types and functions related to exceptions.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55984 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Move handling of @try and @throw to the runtime class.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55983 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-09Change CodeGen to emit calls using (RValue,Type) list:Daniel Dunbar
- Add CodeGenFunction::EmitAnyExprToTemp o Like EmitAnyExpr, but emits aggregates to a temporary location if none is available. Seems like this should be simpler (even aside from using first class aggregates). - Killed CodeGenFunction::EmitCallArg (just append the pair) - Conversion of RValues to actual call arguments is now isolated in CodeGenFunction::EmitCall. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55970 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-08Fix a number of issues w.r.t. emission of global for functions andDaniel Dunbar
aliases. - Attributes specific to a definition are only set when the definition is seen. - Alias generation is delayed until the end of the module; necessary since the alias may reference forward. - Fixes: PR2743, <rdr://6140807&6094512> - Improves: <rdr://6095112> (added XFAIL) Also, print module on verification failures. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55966 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-08Refactor parameter attribute handling:Daniel Dunbar
- Add CGCall.h for dealing with ABI issues related to calls. - Add CGFunctionInfo and CGCallInfo for capturing ABI relevant information about functions and calls. - Isolate LLVM parameter attribute handling inside CGCall.cpp git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55963 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-06Key LLVM types for TagDecl's off of the clang Type, since there is nowDaniel Dunbar
a many-to-one relationship between TagDecl's and types. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55870 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-05Change struct forward declarations and definitions to use unique ↵Ted Kremenek
RecordDecls, as opposed to creating a single RecordDecl and reusing it. This change effects both RecordDecls and CXXRecordDecls, but does not effect EnumDecls (yet). The motivation of this patch is as follows: - Capture more source information, necessary for refactoring/rewriting clients. - Pave the way to resolve ownership issues with RecordDecls with the forthcoming addition of DeclGroups. Current caveats: - Until DeclGroups are in place, we will leak RecordDecls not explicitly referenced by the AST. For example: typedef struct { ... } x; The RecordDecl for the struct will be leaked because the TypedefDecl doesn't refer to it. This will be solved with DeclGroups. - This patch also (temporarily) breaks CodeGen. More below. High-level changes: - As before, TagType still refers to a TagDecl, but it doesn't own it. When a struct/union/class is first referenced, a RecordType and RecordDecl are created for it, and the RecordType refers to that RecordDecl. Later, if a new RecordDecl is created, the pointer to a RecordDecl in RecordType is updated to point to the RecordDecl that defines the struct/union/class. - TagDecl and RecordDecl now how a method 'getDefinition()' to return the TagDecl*/RecordDecl* that refers to the TagDecl* that defines a particular enum/struct/class/union. This is useful from going from a RecordDecl* that defines a forward declaration to the RecordDecl* that provides the actual definition. Note that this also works for EnumDecls, except that in this case there is no distinction between forward declarations and definitions (yet). - Clients should no longer assume that 'isDefinition()' returns true from a RecordDecl if the corresponding struct/union/class has been defined. isDefinition() only returns true if a particular RecordDecl is the defining Decl. Use 'getDefinition()' instead to determine if a struct has been defined. - The main changes to Sema happen in ActOnTag. To make the changes more incremental, I split off the processing of enums and structs et al into two code paths. Enums use the original code path (which is in ActOnTag) and structs use the ActOnTagStruct. Eventually the two code paths will be merged, but the idea was to preserve the original logic both for comparison and not to change the logic for both enums and structs all at once. - There is NO CHAINING of RecordDecls for the same RecordType. All RecordDecls that correspond to the same type simply have a pointer to that type. If we need to figure out what are all the RecordDecls for a given type we can build a backmap. - The diff in CXXRecordDecl.[cpp,h] is actually very small; it just mimics the changes to RecordDecl. For some reason 'svn' marks the entire file as changed. Why is CodeGen broken: - Codegen assumes that there is an equivalence between RecordDecl* and RecordType*. This was true before because we only created one RecordDecl* for a given RecordType*, but it is no longer true. I believe this shouldn't be too hard to change, but the patch was big enough as it is. I have tested this patch on both the clang test suite, and by running the static analyzer over Postgresql and a large Apple-internal project (mix of Objective-C and C). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55839 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-05Remove "NextDecl" from RecordDecl. This change touches many files that ↵Ted Kremenek
where RecordDecl or CXXRecordDecl was constructed, always with an argument of 'NULL' for the previous declaration. The motivation behind this change is that chaining the RecordDecls is simply unnecessary. Once we create multiple RecordDecls for the same struct/union/class, clients that care about all the declarations of the same struct can build a back map by seeing which Decls refer to the same RecordType. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55821 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-05Set sext/zext on function result.Daniel Dunbar
- <rdar://problem/6156739> git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55815 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-04Set function attributes (sext, zext, etc.) on Objective-C methods.Daniel Dunbar
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55812 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-04Fix infinite loop in for ... in code generation.Daniel Dunbar
- Patch via Thomas Clement, thanks! git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55804 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-04NeXT: Emit lazy reference to Protocol class for forward protocolDaniel Dunbar
references (to match gcc). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55760 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-04Avoid superfluous errors regarding variable-length arrays (casts).Daniel Dunbar
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55759 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-04Implement codegen of aggregates as lvalues in binary expressions,Daniel Dunbar
e.g. "(a = b).somefield". git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55758 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-03Add __builtin_object_size support.Daniel Dunbar
- Currently CodeGen always returns a conservative value for this (-1 or 0 depending on the context). git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55735 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-03Fix ObjCPropertRefExpr to be able to encode all the information forDaniel Dunbar
uses which refer to methods not properties. - Not yet wired in Sema. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55681 91177308-0d34-0410-b5e6-96231b3b80d8
2008-09-01codegen constant data as such. add QualType::isConstant()Nuno Lopes
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55603 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-31Handle mutation while enumerating correctly. Fix some bugs.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55583 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-31Initial support for CodeGen of for ... in statements.Anders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55580 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-30Stub out CodeGenFunction::EmitObjCForCollectionStmt.Anders Carlsson
Add CodeGenFunction::EmitMemSetToZero and make AggExprEmitter::EmitAggregateClear use it. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55573 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-30Simplify some calls to Builder.CreateCallAnders Carlsson
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55567 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-30Add Objective-C property setter support.Daniel Dunbar
- Change Obj-C runtime message API, drop the ObjCMessageExpr arg in favor of just result type and selector. Necessary so it can be reused in situations where we don't want to cons up an ObjCMessageExpr. - Update aggregate binary assignment to know about special property ref lvalues. - Add CodeGenFunction::EmitCallArg overload which takes an already emitted rvalue. Add CodeGenFunction::StoreComplexIntoAddr. Disabled logic in Sema for parsing Objective-C dot-syntax that accesses methods. This code does not search in the correct order and the AST node has no way of properly representing its results. Updated StmtDumper to print a bit more information about ObjCPropertyRefExprs. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55561 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-30Refactor handling of calls:Daniel Dunbar
- Added CodeGenFunction::EmitCall which just takes the callee, return type, and a list of (Value*,QualType) pairs. - Added CodeGenFunction::EmitCallArg which handles emitting code for a call argument and turning it into an appropriate (Value*,QualType) pair. - Changed Objective-C runtime interface so that the actual emission of arguments for message sends is (once again) done in the code to emit a message send. No intended functionality change, this is prep work for better ABI support and for Objective-C property setter support. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55560 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-29Downgrade a number of FIXME asserts to ErrorUnsupported.Daniel Dunbar
- Notably VLAs git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55544 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-29Add special "property reference" CodeGen::LValue type for emittingDaniel Dunbar
Objective-C property references. - This handles property references "more correctly" but setters still don't work. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55534 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-28Fix double-free error with sizeof applied to VLA types.Daniel Dunbar
- PR2727. Also, fix warning in CodeGenTypes for new BlockPointer type. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55479 91177308-0d34-0410-b5e6-96231b3b80d8
2008-08-28NeXT: Emit mostly-correct property type encoding.Daniel Dunbar
- Added ASTContext::getObjCEncodingForPropertyDecl. git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@55461 91177308-0d34-0410-b5e6-96231b3b80d8
generated by cgit v1.2.3-18-g5258 (git 2.32.0) at 2025-05-12 10:08:17 +0000