| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | Fix ocaml bindings; add "available_externally" linkage type. | Nick Lewycky | 2009-04-13 | 2 | -0/+2 | |
| | | | | | llvm-svn: 68945 | |||||
| * | Fix warning in .ll parser, detect and reject available_externally on function | Nick Lewycky | 2009-04-13 | 1 | -0/+1 | |
| | | | | | | | declarations. llvm-svn: 68944 | |||||
| * | add AvailableExternally linkage to C bindings. | Chris Lattner | 2009-04-13 | 1 | -0/+1 | |
| | | | | | llvm-svn: 68942 | |||||
| * | Add a new "available_externally" linkage type. This is intended | Chris Lattner | 2009-04-13 | 19 | -29/+120 | |
| | | | | | | | | | to support C99 inline, GNU extern inline, etc. Related bugzilla's include PR3517, PR3100, & PR2933. Nothing uses this yet, but it appears to work. llvm-svn: 68940 | |||||
| * | eliminate unneeded parens. | Chris Lattner | 2009-04-13 | 1 | -2/+2 | |
| | | | | | llvm-svn: 68939 | |||||
| * | Regenerate. | Nick Lewycky | 2009-04-13 | 2 | -44/+64 | |
| | | | | | llvm-svn: 68938 | |||||
| * | Link against libffi if available, fall back to "no external calls from | Nick Lewycky | 2009-04-13 | 3 | -17/+13 | |
| | | | | | | | interpreter mode" when it's not. llvm-svn: 68937 | |||||
| * | Add paragraph tags. | Nick Lewycky | 2009-04-13 | 1 | -2/+2 | |
| | | | | | llvm-svn: 68934 | |||||
| * | fix PR3965:SIGINT handler not restored after calling ParseAST(), | Chris Lattner | 2009-04-12 | 1 | -1/+2 | |
| | | | | | | | patch by Alexei Svitkine! llvm-svn: 68929 | |||||
| * | In X86DAGToDAGISel::MatchWrapper, if base or index are set, avoid matching | Rafael Espindola | 2009-04-12 | 3 | -7/+25 | |
| | | | | | | | only if symbolic addresses are RIP relatives. llvm-svn: 68924 | |||||
| * | Use a hashtable for TargetRegisterClass::contains. | Owen Anderson | 2009-04-12 | 1 | -4/+7 | |
| | | | | | llvm-svn: 68922 | |||||
| * | refactor some code into X86DAGToDAGISel::MatchWrapper | Rafael Espindola | 2009-04-12 | 1 | -44/+51 | |
| | | | | | llvm-svn: 68915 | |||||
| * | Add Kaleidoscope to examples/ so that it gets built regularly and we'll notice | Nick Lewycky | 2009-04-12 | 5 | -1/+1159 | |
| | | | | | | | | if an API change causes it to be out of date. The code is copied out of LangImpl7.html. llvm-svn: 68912 | |||||
| * | Fix compile error. Pointed out by mait on #llvm IRC! | Nick Lewycky | 2009-04-12 | 2 | -2/+2 | |
| | | | | | llvm-svn: 68910 | |||||
| * | "There was a typo in my previous patch which leads to miscompilation of | Chris Lattner | 2009-04-12 | 1 | -1/+1 | |
| | | | | | | | | | | | | | | | strncat :( strncat(foo, "bar", 99) would be optimized to memcpy(foo+strlen(foo), "bar", 100, 1) instead of memcpy(foo+strlen(foo), "bar", 4, 1)" Patch by Benjamin Kramer! llvm-svn: 68905 | |||||
| * | Add tests for the parts of X86-64 TLS that are already implemented. | Rafael Espindola | 2009-04-12 | 4 | -0/+12 | |
| | | | | | llvm-svn: 68901 | |||||
| * | fix a cross-block fastisel crash handling overflow intrinsics. | Chris Lattner | 2009-04-12 | 2 | -5/+37 | |
| | | | | | | | See comment for details. This fixes rdar://6772169 llvm-svn: 68890 | |||||
| * | make UpdateValueMap handle the possiblity that we could be | Chris Lattner | 2009-04-12 | 1 | -1/+1 | |
| | | | | | | | copying into the right register, avoiding a copy. llvm-svn: 68889 | |||||
| * | optimize FastISel::UpdateValueMap to avoid duplicate map lookups, | Chris Lattner | 2009-04-12 | 2 | -8/+13 | |
| | | | | | | | and make it return the assigned register. llvm-svn: 68888 | |||||
| * | simplify code by using IntrinsicInst. | Chris Lattner | 2009-04-12 | 1 | -69/+54 | |
| | | | | | llvm-svn: 68887 | |||||
| * | Add new TargetInstrDesc::hasImplicitUseOfPhysReg and | Chris Lattner | 2009-04-12 | 2 | -15/+21 | |
| | | | | | | | | hasImplicitDefOfPhysReg methods. Use them to remove a look in X86 fast isel. llvm-svn: 68886 | |||||
| * | add some optimizations for strncpy/strncat and factor some | Chris Lattner | 2009-04-12 | 3 | -22/+166 | |
| | | | | | | | code. Patch by Benjamin Kramer! llvm-svn: 68885 | |||||
| * | Revert r68847. It breaks the build on non-Darwin targets, with this message | Dan Gohman | 2009-04-11 | 7 | -178/+7 | |
| | | | | | | | | from the assembler: Error: unknown pseudo-op: `.debug_inlined' llvm-svn: 68863 | |||||
| * | Keep track of inlined functions and their locations. This information is ↵ | Devang Patel | 2009-04-11 | 7 | -7/+178 | |
| | | | | | | | | | collected when nested llvm.dbg.func.start intrinsics are seen. (Right now, inliner removes nested llvm.dbg.func.start intrinisics during inlining.) Create debug_inlined dwarf section using these information. This info is used by gdb, at least on Darwin, to enable better experience debugging inlined functions. See DwarfWriter.cpp for more information on structure of debug_inlined section. llvm-svn: 68847 | |||||
| * | move a target-specific test into its directory so it isn't run if you | Chris Lattner | 2009-04-10 | 1 | -0/+0 | |
| | | | | | | | don't configure the ARM target in. llvm-svn: 68843 | |||||
| * | Fixed spelling. | Misha Brukman | 2009-04-10 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68821 | |||||
| * | DebugLabelFolder ruthlessly deletes redundant labels. However, sometimes the ↵ | Devang Patel | 2009-04-10 | 2 | -1/+18 | |
| | | | | | | | redundant labels is referenced by debug info somewhere else. This patch provies a way so that dwarf writer can mark labels as used. llvm-svn: 68813 | |||||
| * | Clean up a bunch of whitespace issues and fix a comment typo. | Bob Wilson | 2009-04-10 | 1 | -72/+74 | |
| | | | | | | | No functional changes. llvm-svn: 68808 | |||||
| * | Don't install the libLLVMHello.dylib example. | Bill Wendling | 2009-04-10 | 2 | -1/+3 | |
| | | | | | llvm-svn: 68807 | |||||
| * | Add compatibility_version and current_version flags when creating dylibs. | Bill Wendling | 2009-04-10 | 1 | -0/+1 | |
| | | | | | llvm-svn: 68801 | |||||
| * | Don't run "dsymutils" on .a files. | Bill Wendling | 2009-04-10 | 1 | -2/+3 | |
| | | | | | llvm-svn: 68795 | |||||
| * | fix two problems with machine sinking: | Chris Lattner | 2009-04-10 | 2 | -12/+42 | |
| | | | | | | | | | | | | 1. Sinking would crash when the first instruction of a block was sunk due to iterator problems. 2. Instructions could be sunk to their current block, causing an infinite loop. This fixes PR3968 llvm-svn: 68787 | |||||
| * | Now that register classes have names, include the name in debug output. | Dan Gohman | 2009-04-10 | 1 | -2/+4 | |
| | | | | | llvm-svn: 68786 | |||||
| * | fix a broken assertion in ilist_iterator, which caused it to crash in the | Chris Lattner | 2009-04-10 | 1 | -1/+1 | |
| | | | | | | | | bad case instead of actually assert. Bug pointed out by Jakob Stoklund Olesen! llvm-svn: 68785 | |||||
| * | fix typo, patch by Shlomi Fish! | Chris Lattner | 2009-04-10 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68784 | |||||
| * | Added code to handle spilling and reloading of FSRs. | Sanjiv Gupta | 2009-04-10 | 4 | -14/+61 | |
| | | | | | llvm-svn: 68783 | |||||
| * | Don't fold a load if the other operand is a TLS address. | Rafael Espindola | 2009-04-10 | 6 | -14/+49 | |
| | | | | | | | | | | | | | | | With this we generate movl %gs:0, %eax leal i@NTPOFF(%eax), %eax instead of movl $i@NTPOFF, %eax addl %gs:0, %eax llvm-svn: 68778 | |||||
| * | Cleanup. Remove redundant llvm:: , we don't need them since we're inside | Nick Lewycky | 2009-04-10 | 1 | -18/+18 | |
| | | | | | | | namespace llvm already. llvm-svn: 68774 | |||||
| * | Add a new Type::getPointerTo method, which is shorthand for | Chris Lattner | 2009-04-10 | 2 | -0/+8 | |
| | | | | | | | llvm::PointerType::get(). Patch by Anders Johnsen! llvm-svn: 68772 | |||||
| * | Add utility function to IRBuilder that takes the difference between two | Nick Lewycky | 2009-04-10 | 1 | -0/+14 | |
| | | | | | | | | pointers, taking into account the size of the pointed-to object. Patch by Jeffrey Yasskin! llvm-svn: 68768 | |||||
| * | Regenerate. | Nick Lewycky | 2009-04-10 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68767 | |||||
| * | a few fixes to "addrspace(256) is reference offset of GS segment register". | Chris Lattner | 2009-04-10 | 1 | -17/+60 | |
| | | | | | | | It turns out that there are still several problems with this, will file a bugzilla. llvm-svn: 68749 | |||||
| * | Pass in the std::string parameter instead of returning it by value. | Bill Wendling | 2009-04-10 | 4 | -28/+34 | |
| | | | | | llvm-svn: 68747 | |||||
| * | Constify getter methods. | Bill Wendling | 2009-04-10 | 3 | -3/+4 | |
| | | | | | llvm-svn: 68745 | |||||
| * | Remove the obsolete SelectionDAG::getNodeValueTypes and simplify | Dan Gohman | 2009-04-09 | 7 | -95/+60 | |
| | | | | | | | code that uses it by using SelectionDAG::getVTList instead. llvm-svn: 68744 | |||||
| * | Change the NumVTs field from short to int, since there's no | Dan Gohman | 2009-04-09 | 1 | -1/+1 | |
| | | | | | | | real need for it to be short. llvm-svn: 68742 | |||||
| * | StringMap<DIE*>::iterator::first() returns a pointer to the first character of | Bill Wendling | 2009-04-09 | 3 | -5/+10 | |
| | | | | | | | | | the key. This will cause it to create a new std::string, which isn't wanted. Instead, pass back the "const char*". Modify the EmitString() method to take a "const char*". llvm-svn: 68741 | |||||
| * | Silence unused variable warning. | Devang Patel | 2009-04-09 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68735 | |||||
| * | ignore register zero in isRegTiedToUseOperand, following the example of | Chris Lattner | 2009-04-09 | 1 | -1/+1 | |
| | | | | | | | isRegTiedToDefOperand. Thanks to Bob for pointing this out! llvm-svn: 68734 | |||||
| * | Give register alias checking the hash table treatment too. | Owen Anderson | 2009-04-09 | 3 | -5/+97 | |
| | | | | | llvm-svn: 68730 | |||||

