- Jul 14, 2009
-
-
Devang Patel authored
llvm-svn: 75574
-
Evan Cheng authored
Fix pr4544. When remating, make sure the destination register fits the instruction definition. It may be mismatched due to sub-register coalescing. No test case yet because the code doesn't trigger until 75408 is re-applied. llvm-svn: 75572
-
Dale Johannesen authored
xor R, R is a common and valid idiom for zeroing a register, for example. llvm-svn: 75571
-
Dan Gohman authored
llvm-svn: 75568
-
Dan Gohman authored
llvm-svn: 75567
-
Bob Wilson authored
llvm-svn: 75566
-
Bob Wilson authored
llvm-svn: 75565
-
Chris Lattner authored
llvm-svn: 75564
-
Bob Wilson authored
llvm-svn: 75562
-
Chris Lattner authored
descriptive. Thange them to keep track of the ID of a global that is assigned, not the first mangled name returned for it. Without doing this, we are required to always use the same suffix for a global that gets mangled. This means that we can mangle the same global once with $stub and another time with $non_lazy_ptr or whatever. llvm-svn: 75561
-
Owen Anderson authored
llvm-svn: 75559
-
Chris Lattner authored
local symbols and we haven't had type planes since llvm 1.9. llvm-svn: 75558
-
Chris Lattner authored
used with globals. llvm-svn: 75557
-
Chris Lattner authored
llvm-svn: 75556
-
Chris Lattner authored
llvm-svn: 75554
-
Evan Cheng authored
llvm-svn: 75553
-
Chris Lattner authored
if present. llvm-svn: 75547
-
Owen Anderson authored
llvm-svn: 75546
-
Kevin Enderby authored
llvm-svn: 75545
-
Dan Gohman authored
doesn't print a newline. llvm-svn: 75543
-
Chris Lattner authored
1) unique globals with the existing "Count" local in Mangler, not with atomic nonsense. Using atomics will give us nondeterminstic output from the compiler when using multiple threads, which is bad. 2) Do not mangle an unknown global name with a type suffix. We don't need this anymore now that llvm ir doesn't have type planes. llvm-svn: 75541
-
Devang Patel authored
llvm-svn: 75540
-
Eli Friedman authored
llvm-svn: 75539
-
Owen Anderson authored
llvm-svn: 75538
-
Bruno Cardoso Lopes authored
llvm-svn: 75537
-
Chris Lattner authored
llvm-svn: 75532
-
Eli Friedman authored
(I think it's reasonably clear that we want to have a canonical form for constructs like this; if anyone thinks that a select is not the best canonical form, please tell me.) llvm-svn: 75531
-
Dan Gohman authored
llvm-svn: 75530
-
Dan Gohman authored
llvm-svn: 75529
-
Owen Anderson authored
llvm-svn: 75528
-
Chris Lattner authored
printSymbolOperand instead. llvm-svn: 75526
-
Dan Gohman authored
llvm-svn: 75525
-
Dan Gohman authored
llvm-svn: 75524
-
Dan Gohman authored
and related functions out of LoopBase and into Loop, since they are specific to BasicBlock-based loops. This also allows the code to be moved out-of-line. llvm-svn: 75523
-
- Jul 13, 2009
-
-
Chris Lattner authored
llvm-svn: 75520
-
Dan Gohman authored
using the Curiously Recurring Template Pattern with LoopBase. This will help further refactoring, and future functionality for Loop. Also, Headers can now foward-declare Loop, instead of pulling in LoopInfo.h or doing tricks. llvm-svn: 75519
-
Chris Lattner authored
only remaining modifier is "subreg". llvm-svn: 75516
-
Devang Patel authored
llvm-svn: 75515
-
Eli Friedman authored
the changes are allowed by not calling this function for bitcasts. The Instruction::AShr case is dead because SimplifyDemandedInstructionBits handles that case. llvm-svn: 75514
-
David Goodwin authored
Fix FP elimination code to work for Thumb-2 addrmode AddrModeT2_so. This fixes SingleSource/Benchmarks/Stanford/Queens (among others). llvm-svn: 75513
-