- Mar 13, 2009
-
-
Bill Wendling authored
llvm-svn: 66866
-
Dale Johannesen authored
codegen (speculative execution). llvm-svn: 66859
-
Dale Johannesen authored
right; did the wrong thing when there are exactly 11 non-debug instructions, followed by debug info. Remove a FIXME since it's apparently been fixed along the way. llvm-svn: 66840
-
- Mar 12, 2009
-
-
Duncan Sands authored
in the Ada testcase. Reverting this only covers up the real problem, which is a nasty conceptual difficulty in the phi elimination pass: when eliminating phi nodes in landing pads, the register copies need to come before the invoke, not at the end of the basic block which is too late... See PR3784. llvm-svn: 66826
-
Dale Johannesen authored
sorting of ConstantInt's; unreinvent wheel. llvm-svn: 66824
-
Dale Johannesen authored
llvm-svn: 66800
-
Dale Johannesen authored
llvm-svn: 66751
-
Dale Johannesen authored
from a switch table. Multiple table entries that branch to the same place were being sorted by the pointer value of the ConstantInt*; changed to sort by the actual value of the ConstantInt. llvm-svn: 66749
-
- Mar 11, 2009
-
-
Dale Johannesen authored
allocations. Apparently the assumption is there is an instruction (terminator?) following the allocation so I am allowing the same assumption. llvm-svn: 66716
-
Anton Korobeynikov authored
llvm-svn: 66699
-
Anton Korobeynikov authored
llvm-svn: 66692
-
Anton Korobeynikov authored
This fixes fallout from recent PIC/delibtoolize changes and unbreaks build on cygming. llvm-svn: 66686
-
- Mar 10, 2009
-
-
Dale Johannesen authored
whether a callee to be inlined is a leaf. llvm-svn: 66588
-
Dale Johannesen authored
another instcombine pass if we weren't going to make one without debug info. llvm-svn: 66576
-
Devang Patel authored
llvm-svn: 66554
-
John Criswell authored
Also fixed a punctuation error in the header comment. This fixes PR3775. llvm-svn: 66542
-
Evan Cheng authored
If a function is marked alwaysinline, it must be inlined (possibly for correctness). Do so even if the callee has dynamic alloca and the caller doesn't. llvm-svn: 66539
-
Devang Patel authored
llvm-svn: 66490
-
- Mar 09, 2009
-
-
Dan Gohman authored
if needed. This simplifies the code a little, and is needed for an upcoming refactoring. llvm-svn: 66479
-
Devang Patel authored
llvm-svn: 66471
-
Dan Gohman authored
where memory access types are needed. llvm-svn: 66470
-
Dan Gohman authored
llvm-svn: 66469
-
Dan Gohman authored
llvm-svn: 66467
-
Dan Gohman authored
Use VoidTy instead, to be properly conservative. llvm-svn: 66463
-
Dan Gohman authored
of an instruction into a helper function. llvm-svn: 66460
-
Devang Patel authored
llvm-svn: 66454
-
Dan Gohman authored
have to be done twice. llvm-svn: 66449
-
Dan Gohman authored
llvm-svn: 66446
-
Dan Gohman authored
before it does any processing. llvm-svn: 66443
-
Duncan Sands authored
be needed now that these intrinsics are marked as not accessing memory. llvm-svn: 66420
-
Chris Lattner authored
hopefully no functionality change. llvm-svn: 66398
-
- Mar 08, 2009
-
-
Nick Lewycky authored
llvm-svn: 66384
-
Nick Lewycky authored
llvm-svn: 66381
-
Nick Lewycky authored
llvm-svn: 66369
-
Chris Lattner authored
a vector type instead of into an integer type. llvm-svn: 66368
-
Chris Lattner authored
memcpy/memmove'd into or out of. This fixes a serious perf issue that Nate ran into. llvm-svn: 66366
-
Chris Lattner authored
instead of a Constant*, which is what the clients of it really want. llvm-svn: 66364
-
Chris Lattner authored
llvm-svn: 66362
-
Chris Lattner authored
parent between MemCpyInst and MemMoveInst, simplify some code to use it. llvm-svn: 66361
-
Chris Lattner authored
"boolify" pointers, generating really awful code because getting the pointer value requires a load itself. Before: _foo: movb $1, _X.b ret _get: xorl %ecx, %ecx movb _X.b, %al testb %al, %al movl $_Y, %eax cmove %ecx, %eax ret With the xform disabled: _foo: movl $_Y, _X ret _get: movl _X, %eax ret llvm-svn: 66351
-