- Mar 21, 2011
-
-
Oscar Fuentes authored
Tested that MinGW/MSYS builds fine without that. llvm-svn: 128033
-
Jim Grosbach authored
Move the dynamic linking functionality of the llvm-rtdyld program into an ExecutionEngine support library. Update llvm-rtdyld to just load an object file into memory, use the library to process it, then run the _main() function, if one is found. llvm-svn: 128031
-
Devang Patel authored
llvm-svn: 128030
-
Nick Lewycky authored
llvm-svn: 128028
-
Devang Patel authored
llvm-svn: 128027
-
Bill Wendling authored
llvm-svn: 128020
-
Devang Patel authored
llvm-svn: 128017
-
Ted Kremenek authored
Provide a means for CrashRecovery clients to determine if code is currently running while crash recovery cleanups are being processed. llvm-svn: 128008
-
Ted Kremenek authored
llvm-svn: 128007
-
Eric Christopher authored
llvm-svn: 128005
-
Eric Christopher authored
llvm-svn: 128004
-
Jay Foad authored
llvm-svn: 128002
-
Anders Carlsson authored
llvm-svn: 127997
-
Bill Wendling authored
code below it. Even though it looks very similar, it will match more precisely and geneate better functions in the long run. llvm-svn: 127991
-
Bill Wendling authored
This is a helper class that will make it easier to say which InstAliases can be printed and which cannot (because of ambiguity). llvm-svn: 127990
-
Bill Wendling authored
* Move the code that emits the reg in reg class matching into its own function. llvm-svn: 127988
-
Bill Wendling authored
the alias of an InstAlias instead of the thing being aliased. Because we need to know the features that are valid for an InstAlias. This is part of a work-in-progress. llvm-svn: 127986
-
Anders Carlsson authored
As suggested by Nick Lewycky, ignore debugging intrinsics when trying to decide whether a destructor is empty or not. llvm-svn: 127985
-
Nick Lewycky authored
llvm-svn: 127984
-
Eli Friedman authored
llvm-svn: 127982
-
Evan Cheng authored
Re-apply r127953 with fixes: eliminate empty return block if it has no predecessors; update dominator tree if cfg is modified. llvm-svn: 127981
-
- Mar 20, 2011
-
-
Anders Carlsson authored
llvm-svn: 127976
-
Anders Carlsson authored
llvm-svn: 127975
-
Anders Carlsson authored
llvm-svn: 127974
-
Jakob Stoklund Olesen authored
llvm-svn: 127973
-
Rafael Espindola authored
gun as does. This makes it a lot easier to compare the output of both as the addresses are now a lot closer. llvm-svn: 127972
-
Anders Carlsson authored
Add an optimization to GlobalOpt that eliminates calls to __cxa_atexit, if the function passed is empty. llvm-svn: 127970
-
Benjamin Kramer authored
- glibc falls back to fork+exec if a file actions object is present. - On BSDs this saves a malloc. llvm-svn: 127969
-
Argyrios Kyrtzidis authored
If a class inherits from RefCountedBaseVPTR allow all its subclasses to be used with IntrusiveRefCntPtr. llvm-svn: 127966
-
Jakob Stoklund Olesen authored
This can happen when multiple sibling registers are spilled after live range splitting. llvm-svn: 127965
-
Jakob Stoklund Olesen authored
llvm-svn: 127964
-
Daniel Dunbar authored
llvm-svn: 127962
-
Jakob Stoklund Olesen authored
llvm-svn: 127960
-
Jakob Stoklund Olesen authored
llvm-svn: 127959
-
- Mar 19, 2011
-
-
Oscar Fuentes authored
Some of those POSIX <-> Windows command line conversions ended on failure. llvm-svn: 127958
-
Oscar Fuentes authored
llvm-svn: 127957
-
Oscar Fuentes authored
llvm-svn: 127956
-
Daniel Dunbar authored
to canonicalize IR", it broke a lot of things. llvm-svn: 127954
-
Evan Cheng authored
to have single return block (at least getting there) for optimizations. This is general goodness but it would prevent some tailcall optimizations. One specific case is code like this: int f1(void); int f2(void); int f3(void); int f4(void); int f5(void); int f6(void); int foo(int x) { switch(x) { case 1: return f1(); case 2: return f2(); case 3: return f3(); case 4: return f4(); case 5: return f5(); case 6: return f6(); } } => LBB0_2: ## %sw.bb callq _f1 popq %rbp ret LBB0_3: ## %sw.bb1 callq _f2 popq %rbp ret LBB0_4: ## %sw.bb3 callq _f3 popq %rbp ret This patch teaches codegenprep to duplicate returns when the return value is a phi and where the phi operands are produced by tail calls followed by an unconditional branch: sw.bb7: ; preds = %entry %call8 = tail call i32 @f5() nounwind br label %return sw.bb9: ; preds = %entry %call10 = tail call i32 @f6() nounwind br label %return return: %retval.0 = phi i32 [ %call10, %sw.bb9 ], [ %call8, %sw.bb7 ], ... [ 0, %entry ] ret i32 %retval.0 This allows codegen to generate better code like this: LBB0_2: ## %sw.bb jmp _f1 ## TAILCALL LBB0_3: ## %sw.bb1 jmp _f2 ## TAILCALL LBB0_4: ## %sw.bb3 jmp _f3 ## TAILCALL rdar://9147433 llvm-svn: 127953
-
Evan Cheng authored
llvm-svn: 127952
-