- 614cdcd Add support for building the compactiontable for bytecode files. This shrinks by Chris Lattner · 22 years ago
- af894e9 Add support for representing the "compaction table" by Chris Lattner · 22 years ago
- 80b9734 Bytecode format for LLVM 1.2 no longer explicitly encodes zeros in primitive by Chris Lattner · 22 years ago
- 0baa0af If these blocks are empty, there is no reason to even emit the bytecode blocks. by Chris Lattner · 22 years ago
- d5c59d5 Cleanups & efficiency improvements by Chris Lattner · 22 years ago
- 8c202cd The bcwriter does not want ConstantPointerRef's to be indexed, and the asmwriter never did! by Chris Lattner · 22 years ago
- 7be08bf ConstantPointerRef's are no longer emitted. This saves 20028 bytes in the by Chris Lattner · 22 years ago
- 9336199 Allow bytecode files to refer directly to global values as constants, instead by Chris Lattner · 22 years ago
- ff47e7d Fix more breakage with string change. by Chris Lattner · 22 years ago
- f4c0e7f Fix PR73: bytecode format inconsistent by Chris Lattner · 22 years ago
- 44d0eeb Fix PR73 by Chris Lattner · 22 years ago
- 570fb1c Fix brokenness in my last checking by Chris Lattner · 22 years ago
- 7969dc2 Change all of the bytecode reader primitives to throw exceptions instead of by Chris Lattner · 22 years ago
- c8434e3 Remove the optimization that depends on ENDIAN_LITTLE: it's not worth it. by Chris Lattner · 22 years ago
- 7851e1b Improve comments, add support for remembering the constants strings that by Chris Lattner · 22 years ago
- 83bb3d2 The new bytecode format supports emitting strings a special case. This is by Chris Lattner · 22 years ago
- 9e893e8 Version 1.2 now supports encoding strings as a special case, to avoid having by Chris Lattner · 22 years ago
- dcea630 When emitting bytecode, handle strings specially. Do not add the characters by Chris Lattner · 22 years ago
- 4e86836 Like output_data, it's obvious that input_data was only used with 1 character by Chris Lattner · 22 years ago
- 0a3a27b It is obvious that this has never been used for outputing more than a single by Chris Lattner · 22 years ago
- 44e3803 Be const correct by Chris Lattner · 22 years ago
- d6942d7 Ok, I can't handle it. This is a temporary checkin of a ton of statistics that by Chris Lattner · 22 years ago
- c08912f Remove support for the pre-1.0 bytecode version #1. This will become by Chris Lattner · 22 years ago
- 277bafb Toggle sense of flag by Chris Lattner · 22 years ago
- 8ce7501 The only clients of the slot calculator are now the asmwriter and bcwriter. by Chris Lattner · 22 years ago
- b16689b Don't use ConstantExpr::getShift anymore by Chris Lattner · 22 years ago
- 92fa44c Urg, remove testing code. by Chris Lattner · 22 years ago
- 9380297 Fix a regression that I introduced yesterday. :( by Chris Lattner · 22 years ago
- a14b0d4 Check in two changes: by Chris Lattner · 22 years ago
- 737d3cd Do not bother to emit a BytecodeBlock for an empty symbol table. This commonly by Chris Lattner · 22 years ago
- 32abce6 Err, we don't need Config/*.h files for things that are standard C++ by Chris Lattner · 22 years ago
- 1499934 minor comment tweaks by Chris Lattner · 22 years ago
- 3446ae8 * finegrainify namespacification of ArchiveReader.cpp by Chris Lattner · 22 years ago
- 4443019 Refactor writer stuff out of include/llvm/Bytecode/Primitives.h. This is by Chris Lattner · 22 years ago
- 44f549b Finegrainify namespacification by Chris Lattner · 22 years ago
- 258b537 Inching our way towards fixing PR82 by Chris Lattner · 22 years ago
- f63aeec remove some useless #includes by Chris Lattner · 22 years ago
- b70abe1 Use new getFileSize function instead of sys/stat.h directly. by Chris Lattner · 22 years ago
- 2d6481c Factor FDHandle out of the bytecode reader into the FileUtilities.h support by Chris Lattner · 22 years ago
- ddceeb7 minor cleanups by Chris Lattner · 22 years ago
- 9059158 Reverted back to revision 1.11. The previous fix doesn't really fix by John Criswell · 22 years ago
- 1118d0f Fix PR#193. by John Criswell · 22 years ago
- 6ca5b8f Make getObjectType() smarter about ranlibbed Mac OS X archives. by Brian Gaeke · 22 years ago
- 27b40bc Throw better error messages, by calling strerror(errno) when we by Brian Gaeke · 22 years ago
- 3c09636 In getObjectType(), doxygenify comment, and take a pointer to the by Brian Gaeke · 22 years ago
- 8d1dbd2 Emit & read more compressed bytecode by not emitting a bytecodeblock for by Chris Lattner · 22 years ago
- 31f8499 Finegrainify namespacification by Chris Lattner · 22 years ago
- 29b789b * Finegrainify namespacification by Chris Lattner · 22 years ago
- 8eb4993 Minor efficiency improvements, finegrainify namespacification by Chris Lattner · 22 years ago
- c2f4b11 In the future, these slabs can be null... by Chris Lattner · 22 years ago
- ebfb9f4 Minor efficiency improvements all over. Finegrainify Namespacification by Chris Lattner · 22 years ago
- deab9a7 Fine grainify namespacification, #include file that defines the interface! by Chris Lattner · 22 years ago
- 1c3673b Minor speedup to do less linear time searches of information we already have. by Chris Lattner · 22 years ago
- 02071d0 Remove usage of ConstantPointer by Chris Lattner · 22 years ago
- 2c61d7b Implement skeletal support for __.SYMDEF (ranlib) sections in archives. by Brian Gaeke · 22 years ago
- db9546e I can't believe the incompetence of the people I have to deal with around here. by Chris Lattner · 22 years ago
- 8dc6ba9 Fix PR 110: the constant folding code assumes that if something is not a by Chris Lattner · 22 years ago
- d0fde30 Put all LLVM code into the llvm namespace, as per bug 109. by Brian Gaeke · 22 years ago
- 095be96 no need for endl by Chris Lattner · 22 years ago
- 29e3b2b Add support for constant expr shifts by Chris Lattner · 22 years ago
- 088eb45 Use already existing classes for post order iterator. by Alkis Evlogimenos · 22 years ago
- 74fa84f Output types in reverse postorder. This will allow the ByteCode/Reader by Alkis Evlogimenos · 22 years ago
- 60048b8 Put DEBUG_OUTPUT at the top along with TRACE_LEVEL. Also fix the code by Alkis Evlogimenos · 22 years ago
- bc839b5 Explicitly cast type, so we aren't passing output_vbr a size_t. by Brian Gaeke · 22 years ago
- 5d1bdcd Add more debugging info to help tracing the SlotCalculator. by Alkis Evlogimenos · 22 years ago
- 2765c41 Make sure to add a type for va_next operands, which may not other wise by Chris Lattner · 22 years ago
- 389dbfb Fix problem which effected linking apache by Chris Lattner · 22 years ago
- 856ba76 Added LLVM copyright header. by John Criswell · 22 years ago
- e488e93 Added LLVM copyright notice to Makefiles. by John Criswell · 22 years ago
- 6fbcc26 Added LLVM copyright header (for lack of a better term). by John Criswell · 22 years ago
- b576c94 Added LLVM project notice to the top of every C++ source file. by John Criswell · 22 years ago
- ce439b5 "Function"[0] == 'F', not 'M' by Chris Lattner · 22 years ago
- 3b237fc Change the Opcode enum for PHI nodes from "Instruction::PHINode" to "Instruction::PHI" to be more consistent with the other instructions. by Chris Lattner · 22 years ago
- b91d971 Fix warning by Chris Lattner · 22 years ago
- 22482a1 add support for new linkage types by Chris Lattner · 22 years ago
- eff112c Updated to emit the final 1.0 bytecode format. This supports weak linkage, by Chris Lattner · 22 years ago
- cb7e2e2 * New revised variable argument handling support by Chris Lattner · 22 years ago
- 6059638 Eliminate some extraneous code in SlotCalculator::insertVal(). by Alkis Evlogimenos · 22 years ago
- 9fcdadb Add separator between different types of readers. by Misha Brukman · 22 years ago
- 6b25242 Add support for 'weak' linkage. by Chris Lattner · 22 years ago
- cb18d18 Stop using "reverse depth first" order by Chris Lattner · 22 years ago
- 7dc3a2e Minor cleanups by Chris Lattner · 22 years ago
- f0d9273 Avoid calling getTypeSlot more by Chris Lattner · 22 years ago
- d9113c7 This is a header file, thus it needs a -*-C++-*- by Chris Lattner · 22 years ago
- cf3056d Regularize header file comments by Chris Lattner · 22 years ago
- 39cacce Ok, the "fix" for this is to do a real associative container. Symbol tables by Chris Lattner · 22 years ago
- 97330cf Another 10% performance improvement by not using replaceAllUsesWith by Chris Lattner · 22 years ago
- 4c52392 Reserve space for PHI nodes when we read them in. This provides a VERY by Chris Lattner · 22 years ago
- 35d2ca6 Use the version of getValue that takes the type plane instead of the type by Chris Lattner · 22 years ago
- 6fcf503 Pass a vector around to reduce dynamic allocation by Chris Lattner · 22 years ago
- c9456ca Change getConstantValue to throw an exception on error, not return null by Chris Lattner · 22 years ago
- 9073613 Remove potentially N^2 algorithm from symbol table reader. No speedup by Chris Lattner · 22 years ago
- 927b185 Major refactoring of the bytecode reader. This includes the following by Chris Lattner · 22 years ago
- 3483f54 Significantly clean up parsing of instructions. This exceptionizes and by Chris Lattner · 22 years ago
- bf43ac6 Eliminate the instruction placeholder. Simplify a bunch of code. by Chris Lattner · 22 years ago
- 8eb10ce Eliminate the old LateResolveValues data structure, replacing it with a by Chris Lattner · 22 years ago
- 13eb871 Remove a dead method by Chris Lattner · 22 years ago
- 4ee8ef2 This patch substantially simplifies and cleans up handling of basic blocks by Chris Lattner · 22 years ago
- 6e44802 Inline the postResolveValues method. It was poorly named anyway by Chris Lattner · 22 years ago
- 36392bc Various cleanups and simplifications. This speeds up reading a bytecode file by Chris Lattner · 22 years ago