- May 30, 2008
-
-
Owen Anderson authored
Make the renumbering correct in the face of deleted instructions that have been removed from the LiveIntervals maps. llvm-svn: 51714
-
Gabor Greif authored
llvm-svn: 51712
-
Bill Wendling authored
llvm-svn: 51709
-
- May 29, 2008
-
-
Anton Korobeynikov authored
llvm-svn: 51708
-
Dan Gohman authored
cases due to an isel deficiency already noted in lib/Target/X86/README.txt, but they can be matched in this fold-call.ll testcase, for example. This is interesting mainly because it exposes a tricky tblgen bug; tblgen was incorrectly computing the starting index for variable_ops in the case of a complex pattern. llvm-svn: 51706
-
Bill Wendling authored
function to flush a specified std::ostream. llvm-svn: 51705
-
Bill Wendling authored
llvm-svn: 51704
-
Owen Anderson authored
llvm-svn: 51701
-
Dan Gohman authored
definitions. This adds a new construct, "discard", for indicating that a named node in the input matching pattern is to be discarded, instead of corresponding to a node in the output pattern. This allows tblgen to know where the arguments for the varaible_ops are supposed to begin. This fixes "rdar://5791600", whatever that is ;-). llvm-svn: 51699
-
Dan Gohman authored
llvm-svn: 51698
-
Dan Gohman authored
llvm-svn: 51697
-
Dan Gohman authored
memmove to a more plausible value, now that it's actually being used. llvm-svn: 51696
-
Anton Korobeynikov authored
llvm-svn: 51695
-
Owen Anderson authored
llvm-svn: 51694
-
Anton Korobeynikov authored
llvm-svn: 51693
-
Anton Korobeynikov authored
llvm-svn: 51692
-
Owen Anderson authored
Renumbering needs to account for instruction slot offsets when performing lookups in the index maps. llvm-svn: 51691
-
Anton Korobeynikov authored
llvm-svn: 51688
-
Anton Korobeynikov authored
llvm-svn: 51687
-
Owen Anderson authored
llvm-svn: 51686
-
Owen Anderson authored
llvm-svn: 51685
-
Duncan Sands authored
llvm-svn: 51680
-
Owen Anderson authored
Dhrystone introduced by its removal. llvm-svn: 51669
-
Owen Anderson authored
the one case that ADCE catches that normal DCE doesn't: non-induction variable loop computations. This implementation handles this problem without using postdominators. llvm-svn: 51668
-
Evan Cheng authored
llvm-svn: 51667
-
Owen Anderson authored
llvm-svn: 51666
-
Evan Cheng authored
llvm-svn: 51665
-
Bill Wendling authored
llvm-svn: 51664
-
Gabor Greif authored
llvm-svn: 51663
-
Bill Wendling authored
like. llvm-svn: 51662
-
Gabor Greif authored
llvm-svn: 51661
-
Bill Wendling authored
instruction to execute. This can be used for transformations (like two-address conversion) to remat an instruction instead of generating a "move" instruction. The idea is to decrease the live ranges and register pressure and all that jazz. llvm-svn: 51660
-
Bill Wendling authored
that it's safe to remat an instruction. llvm-svn: 51659
-
Owen Anderson authored
llvm-svn: 51658
-
- May 28, 2008
-
-
Owen Anderson authored
Factor the numbering computation into a separate method, and add the slightest attempt at some renumbering logic, which is currently unused. llvm-svn: 51652
-
Evan Cheng authored
llvm-svn: 51648
-
Evan Cheng authored
llvm-svn: 51647
-
Nate Begeman authored
llvm-svn: 51644
-
Chris Lattner authored
their release. llvm-svn: 51638
-
Chris Lattner authored
llvm-svn: 51636
-