- Aug 15, 2003
-
-
Brian Gaeke authored
llvm-svn: 7906
-
Brian Gaeke authored
Many changes suggested by Chris. It's okay, I'll recover from the emotional damage...maybe someday. :-) Collapse ParseCStringVector into ParseStringVector. Comment it. Make it take a const input. Use std::string::npos instead of -1 (what a mouthful!) Make ParseEnvironmentOptions take const inputs. Check its args at the very beginning. Strdup all the contents of newArgv and free them all at the end. include/Support/CommandLine.h: Constify progName and envVar arguments to ParseEnvironmentOptions(). llvm-svn: 7905
-
Misha Brukman authored
rounding up to a page, we need to request (num instrs * 4) bytes. However, I think that 64 bytes is overkill for the stub function. llvm-svn: 7888
-
Misha Brukman authored
llvm-svn: 7884
-
Chris Lattner authored
llvm-svn: 7880
-
Chris Lattner authored
llvm-svn: 7878
-
Chris Lattner authored
llvm-svn: 7877
-
Chris Lattner authored
llvm-svn: 7875
-
Chris Lattner authored
llvm-svn: 7873
-
Chris Lattner authored
llvm-svn: 7872
-
Chris Lattner authored
llvm-svn: 7870
-
Misha Brukman authored
registers are global, are NOT windowed, and hence can be clobbered! * Removed unused register shorthand notations * Fixed and cleaned up comments in inline assembly llvm-svn: 7853
-
Brian Gaeke authored
llvm-svn: 7851
-
- Aug 14, 2003
-
-
Vikram S. Adve authored
another delayed instr. would cause the later sanity-check (assertion) in PhyRegAlloc.cpp to fail, even though there is really no error. llvm-svn: 7848
-
Chris Lattner authored
implement more constant expressions so that 176.gcc compiles with the CBE llvm-svn: 7847
-
Chris Lattner authored
speeds up program execution by 15% pretty consistently for large programs llvm-svn: 7845
-
Misha Brukman authored
to mark TableGen description files with "C++ mode". llvm-svn: 7841
-
Chris Lattner authored
llvm-svn: 7840
-
Brian Gaeke authored
llvm-svn: 7839
-
Brian Gaeke authored
llvm-svn: 7838
-
Brian Gaeke authored
addPassesToEmitMachineCode now both take a FunctionPassManager. llvm-svn: 7837
-
Brian Gaeke authored
Get rid of RegisterLLC, which can't handle FunctionPasses anyway. llvm-svn: 7836
-
Brian Gaeke authored
addPassesToEmitMachineCode now both take a FunctionPassManager. Factory methods for function passes now return type FunctionPass *. llvm-svn: 7835
-
Chris Lattner authored
This also enables -time-passes for FunctionPassManagers, which allows it to be used for the JIT llvm-svn: 7834
-
Chris Lattner authored
llvm-svn: 7832
-
- Aug 13, 2003
-
-
Chris Lattner authored
llvm-svn: 7831
-
Chris Lattner authored
(A <setcc1> B) logicalop (A <setcc2> B) -> (A <setcc3> B) or true or false Where setcc[123] is one of the 6 setcc instructions, and logicalop is one of: And, Or, Xor llvm-svn: 7828
-
Brian Gaeke authored
Function at a time: smacked Who you gonna call!!! :-) llvm-svn: 7826
-
Chris Lattner authored
llvm-svn: 7825
-
Brian Gaeke authored
llvm-svn: 7823
-
Brian Gaeke authored
llvm-svn: 7822
-
Brian Gaeke authored
FunctionPassManager, to support function-at-a-time compilation and emission of code. llvm-svn: 7821
-
Brian Gaeke authored
Use a FunctionPassManager instead of a PassManager. llvm-svn: 7820
-
Brian Gaeke authored
Run passes on single function (hey, just-in-time compilation!) instead of the entire module that contains it. llvm-svn: 7819
-
Brian Gaeke authored
on Function * when passing it to getPointerToFunction(). llvm-svn: 7818
-
Brian Gaeke authored
function-at-a-time compilation and emission of code. Separate addPassesToEmitAssembly from addPassesToJITCompile, because the latter requires you to use FunctionPasses, and the former might diverge anyway. llvm-svn: 7817
-
Brian Gaeke authored
Correct one of the functions' comments. llvm-svn: 7816
-
Brian Gaeke authored
Revert (to v1.55) one of the hunks of Chris's change that messed up the %-register workaround. llvm-svn: 7815
-
Chris Lattner authored
llvm-svn: 7813
-
Chris Lattner authored
llvm-svn: 7812
-