- 9a5a3f2 Rename BaseLLVMType to NonVirtualBaseLLVMType. by Anders Carlsson · 14 years ago
- f4a15b0 Add getCGRecordLayout helper function. No functionality change. by Anders Carlsson · 14 years ago
- cb0511c Remove debugging printf. Fix linux build. by Nick Lewycky · 14 years ago
- 3d155e6 Introduce the concept of a non-virtual base type to CGRecordLayoutBuilder as a first step towards fixing PR6995. by Anders Carlsson · 14 years ago
- a14f597 Rename getBaseClassOffset to getBaseClassOffsetInBits and introduce a getBaseClassOffset which returns the offset in CharUnits. Do the same thing for getVBaseClassOffset. by Anders Carlsson · 14 years ago
- e7a80bd IRgen: Move CGBitFieldInfo strategy computation helpers to static member by Daniel Dunbar · 14 years ago
- f16aa10 Go back to asking CodeGenTypes whether a type is zero-initializable. by John McCall · 14 years ago
- 8a2c92c AST: Rename PragmaPackAttr to MaxFieldAlignmentAttr, which is more accurate. by Daniel Dunbar · 14 years ago
- a83fb4b Correctly initialize bases with member pointers. This should fix PR6441 but that test case is a bit weird and I'd like to investigate further before closing that bug. by Anders Carlsson · 14 years ago
- c6772ce Keep track of the LLVM field numbers for non-virtual bases. by Anders Carlsson · 14 years ago
- 15ddfdc Start laying out bases as individual fields. We still use ugly i8 arrays but this is a step in the right direction. by Anders Carlsson · 14 years ago
- 68e3013 Use a more appropriate LLVM type for the vtable pointer. by Anders Carlsson · 14 years ago
- 52968a1 IRgen: Fix another case where we generated an invalid access component when we by Daniel Dunbar · 14 years ago
- 4651efb IRgen: Fix case where we might generate an access component with width == 0, if by Daniel Dunbar · 14 years ago
- 89da874 IRgen: Set alignment correctly on bit-field accesses. by Daniel Dunbar · 14 years ago
- e1467a4 IRgen: Rewrite bit-field access policy to not access data beyond the bounds of the structure, which we also now verify as part of the post-layout consistency checks. by Daniel Dunbar · 14 years ago
- ad75953 IRgen: Fix CGRecordLayout::print to print the bit-field infos in a consistent order. by Daniel Dunbar · 14 years ago
- 3b2ae7a IRgen: Add checking that the LLVM and AST record layout offsets agree (for by Daniel Dunbar · 14 years ago
- 8ab78a7 IRgen: Always use i8 arrays to access union bit-fields. This is ugly, but by Daniel Dunbar · 14 years ago
- 8d8ab74 AST: Dump ASTRecordLayout objects when they are created with -fdump-record-layouts. by Daniel Dunbar · 14 years ago
- 2e7b7c2 IRgen: Kill unused function and move the type match assert to after record dumping. by Daniel Dunbar · 14 years ago
- 6ba3815 Simplify wide bit-field layout in CGRecordLayoutBuilder, and also fix a bug where assigning to a bit-field member would overwrite other parts of the struct. by Anders Carlsson · 14 years ago
- d62328e Unnamed bit-fields in a union should be laid out with a type that doesn't affect alignment. by Anders Carlsson · 14 years ago
- 8666446 Factor union field layout code out into a separate function. No functionality change. by Anders Carlsson · 14 years ago
- 046c294 Vtable -> VTable renames across the board. by Anders Carlsson · 14 years ago
- 1c7658f Make CGRecordLayoutBuilder deal with wide bit-fields. Will land tests shortly (Daniel, please review). by Anders Carlsson · 14 years ago
- 2df2569 IRgen: Change CGBitFieldInfo to take the AccessInfo as constructor arguments, it is now an immutable object. by Daniel Dunbar · 14 years ago
- 7fb6195 IRgen: Eliminate now unused fields from CGBitFieldInfo. by Daniel Dunbar · 14 years ago
- ab970f9 IRgen: Enhance CGBitFieldInfo with enough information to fully describe the "policy" with which a bit-field should be accessed. by Daniel Dunbar · 14 years ago
- 9b28daf IRgen: Factor out ComputeBitFieldInfo. by Daniel Dunbar · 14 years ago
- 93c6296 IRgen: Add CGRecordLayout::dump, and dump (irgen) record layouts as part of -fdump-record-layouts. by Daniel Dunbar · 14 years ago
- 7f28964 IRgen: Move the bit-field access type into CGBitFieldInfo, and change bit-field LValues to just store the base address of object containing the bit-field. by Daniel Dunbar · 14 years ago
- efbf487 IRgen: Move BitFieldIsSigned bit into CGBitFieldInfo. by Daniel Dunbar · 14 years ago
- c7a984a Simplify. by Daniel Dunbar · 14 years ago
- 2eec0b2 IRgen: Lift BitFieldInfo to CGBitFieldInfo at namespace level. by Daniel Dunbar · 14 years ago
- 198bcb4 IRGen: Move the auxiliary data structures tracking AST -> LLVM mappings out of CodeGenTypes, to per-record CGRecordLayout structures. by Daniel Dunbar · 14 years ago
- 490fc90 CGRecordLayoutBuilder: Switch unions to use same mechanism for tracking field and bit-field info as structs. by Daniel Dunbar · 14 years ago
- 270e203 IRGen: Hide CGRecordLayoutBuilder class, because I can. by Daniel Dunbar · 14 years ago
- 2924ade IRgen: Move CGRecordLayout to its own happy little file. by Daniel Dunbar · 14 years ago
- 2c12d03 Improve handling of emitting 'null' pointers to data members. by Anders Carlsson · 15 years ago
- 21fd7d7 Fix an incorrect union layout assert. Fixes PR6164. by Anders Carlsson · 15 years ago
- 4b3e5be Baby steps towards fixing PR5589. If a class needs a vtable pointer, add one. by Anders Carlsson · 15 years ago
- c245682 No need to add tail padding if the resulting LLVM struct type will have the same size as the final record size. by Anders Carlsson · 15 years ago
- 1a5e0d7 Have ASTRecordLayout keep track of the key function, in preparation of fixing a synthetic ctor/dtor bug. by Anders Carlsson · 15 years ago
- a84b404 Fixup key function calculations. by Mike Stump · 15 years ago
- 88bd009 This doesn't work yet. by Mike Stump · 15 years ago
- e504170 Prevent a code gen. crash on empty unions - pr5408. by Fariborz Jahanian · 15 years ago
- 5017a15 Store the key function of a record decl inside CGRecordLayout. by Anders Carlsson · 15 years ago
- 1eb4433 Remove tabs, and whitespace cleanups. by Mike Stump · 15 years ago
- 3662000 If the alignment of the chosen field in a union is greater than the alignment of the union, we need to use a packed LLVM struct. Fixes <rdar://problem/7184250>. by Anders Carlsson · 15 years ago
- d0eb3b9 Packed unions should be packed. Fixes an assert Daniel reported. by Anders Carlsson · 15 years ago
- fc3eaa4 More work towards zero-initializing structs that contain member pointers in constant expressions. by Anders Carlsson · 15 years ago
- 341bdf8 Remove the PaddingFields member from CGRecordLayout, it wasn't used anyway. by Anders Carlsson · 15 years ago
- 0032b27 Update for LLVM API change. by Owen Anderson · 15 years ago
- a5dd722 Take #pragma pack into account when laying out structs. Fixes rdar://problem/7095436. by Anders Carlsson · 15 years ago
- a860e75 Introduce a new PragmaPack attribute, and use it for #pragma pack. The PackedAttr now only represents __attribute__((packed)). by Anders Carlsson · 15 years ago
- df31e09 Assert that the LLVM type has the same size as the RecordDecl size. by Anders Carlsson · 15 years ago
- 47a434f Update for LLVM API change. by Owen Anderson · 15 years ago
- de9f2c9 Simplify alignment handling in the record builder. by Anders Carlsson · 15 years ago
- 8a8a629 No longer need to keep base class offsets in the offset by Fariborz Jahanian · 15 years ago
- c2cc1d5 More CGRecordLayoutBuilder cleanup. by Anders Carlsson · 15 years ago
- cad8665 Some minor changes toward support of data by Fariborz Jahanian · 15 years ago
- abdad35 Enable the new struct type builder now that the constant struct builder works. (The old code will still be there until we know that everything works well. by Anders Carlsson · 15 years ago
- c1efe36 Fix a tail padding bug in the record layout builder code. The bug was found by an existing test. by Anders Carlsson · 15 years ago
- 4c98efd Check in a half finished new constant struct builder (Obviously not used yet). by Anders Carlsson · 15 years ago
- 728d7cd Get rid of the size parameter to AppendField. No functionality change. by Anders Carlsson · 15 years ago
- 94ae95f Don't just store the field/bit field info one field, do it for all fields in the union. by Anders Carlsson · 15 years ago
- 177d4d8 Fix another thinko. by Anders Carlsson · 15 years ago
- faaec22 Correct a thinko in bitfield layout code. Fixes PR4611. by Anders Carlsson · 15 years ago
- 4b5584b We don't need to keep track of the packed alignment, just whether the struct is packed or not. Fixes PR4610. by Anders Carlsson · 15 years ago
- 8330cee Move the LLVM field number for bit fields into the BitFieldInfo structure, since it's meaning is completely different than for non-bit fields. by Anders Carlsson · 15 years ago
- cfc6758 Set field info for unions. by Anders Carlsson · 15 years ago
- 2cc8f17 Handle zero width bit fields in unions correctly (by ignoring them). by Anders Carlsson · 15 years ago
- 5a6e398 Implement union layout support. by Anders Carlsson · 15 years ago
- 45372a6 Check in CGRecordLayoutBuilder which is a reimplementation of the record layout code. (Yay, no more packed structs unless absolutely necessary). We currently don't use the layouts being built but that will change when the new code is mature enough :) by Anders Carlsson · 15 years ago