- Nov 03, 2010
-
-
Chris Lattner authored
ins/outs list that isn't specified by their asmstring. Previously the asmmatcher would just force a 0 register into it, which clearly isn't right. Mark a bunch of ARM instructions that use this as isCodeGenOnly. Some of them are clearly pseudo instructions (like t2TBB) others use a weird hasExtraSrcRegAllocReq thing that will either need to be removed or the asmmatcher will need to be taught about it (someday). llvm-svn: 118119
-
Jakob Stoklund Olesen authored
clang are using. llvm-svn: 118118
-
Jakob Stoklund Olesen authored
threshold given to createFunctionInliningPass(). Both opt -O3 and clang would silently ignore the -inline-threshold option. llvm-svn: 118117
-
Dan Gohman authored
limits on their own. llvm-svn: 118113
-
Chris Lattner authored
that have complicated tying going on. llvm-svn: 118112
-
Dan Gohman authored
llvm-svn: 118110
-
John Thompson authored
llvm-svn: 118107
-
- Nov 02, 2010
-
-
Dan Gohman authored
llvm-svn: 118106
-
Dan Gohman authored
llvm-svn: 118105
-
Chris Lattner authored
filling them in one at a time. Previously this iterated over the asmoperands, which left the problem of "holes". The new approach simplifies things. llvm-svn: 118104
-
Bill Wendling authored
llvm-svn: 118103
-
Dan Gohman authored
llvm-svn: 118102
-
Bill Wendling authored
llvm-svn: 118099
-
Bill Wendling authored
is handled with the MC encoder. llvm-svn: 118098
-
Owen Anderson authored
Since these fields are not exactly equivalent to the encoded field, rename them to something with semantic meaning. llvm-svn: 118097
-
Dan Gohman authored
llvm-svn: 118096
-
Bill Wendling authored
with immediates up to 16-bits in size. The same logic is applied to other LDR encodings, e.g. VLDR, but which use a different immediate bit width (8-bits in VLDR's case). Removing the "12" allows it to be more generic. llvm-svn: 118094
-
Owen Anderson authored
llvm-svn: 118093
-
Chris Lattner authored
llvm-svn: 118092
-
Mikhail Glushenkov authored
Additionally, move the implementation of appendSuffix to Path.cpp: it is platform-independent. llvm-svn: 118089
-
Mikhail Glushenkov authored
llvm-svn: 118088
-
Owen Anderson authored
llvm-svn: 118087
-
Dan Gohman authored
llvm-svn: 118086
-
Owen Anderson authored
llvm-svn: 118084
-
Chris Lattner authored
in the generated .inc files. llvm-svn: 118083
-
Owen Anderson authored
llvm-svn: 118082
-
Rafael Espindola authored
// FIXME: We should compute this sooner, we don't want to recurse here, and // we would like to be more functional. In MCAssembler::ComputeFragmentSize. llvm-svn: 118080
-
Oscar Fuentes authored
llvm-svn: 118072
-
Bob Wilson authored
llvm-svn: 118069
-
Owen Anderson authored
llvm-svn: 118068
-
Owen Anderson authored
llvm-svn: 118067
-
Dan Gohman authored
llvm-svn: 118058
-
Dan Gohman authored
llvm-svn: 118057
-
Mikhail Glushenkov authored
llvm-svn: 118056
-
Owen Anderson authored
llvm-svn: 118055
-
Devang Patel authored
llvm-svn: 118054
-
Owen Anderson authored
llvm-svn: 118053
-
Mikhail Glushenkov authored
This allows using GetDLLSuffix() with appendSuffix(). llvm-svn: 118051
-
Mikhail Glushenkov authored
llvm-svn: 118050
-
Mikhail Glushenkov authored
This makes the behaviour of FindExecutable more consistent across platforms, but I'm not very happy with the name... llvm-svn: 118049
-