- May 10, 2010
-
-
Evan Cheng authored
llvm-svn: 103436
-
Evan Cheng authored
llvm-svn: 103435
-
Evan Cheng authored
Be careful with operand promotion. For a binary operation, the source operands may be the same. PR7018. rdar://7939869. llvm-svn: 103419
-
Evan Cheng authored
llvm-svn: 103410
-
Bob Wilson authored
SSAUpdater for the value from the first def may see uses of undefined values, because the later defs will not have been updated yet. llvm-svn: 103407
-
Duncan Sands authored
Based on a patch by Javier Martinez. llvm-svn: 103391
-
- May 08, 2010
-
-
Devang Patel authored
llvm-svn: 103327
-
Devang Patel authored
llvm-svn: 103325
-
Dan Gohman authored
and not the real problem itself, by dropping debug info for i128 values. rdar://7958162. llvm-svn: 103310
-
Devang Patel authored
llvm-svn: 103305
-
- May 07, 2010
-
-
Chris Lattner authored
llvm-svn: 103304
-
Devang Patel authored
llvm-svn: 103303
-
Dale Johannesen authored
getConstantFP to accept the two supported long double target types. This was not the original intent, but there are other places that assume this works and it's easy enough to do. llvm-svn: 103299
-
Devang Patel authored
llvm-svn: 103295
-
Devang Patel authored
llvm-svn: 103272
-
Chris Lattner authored
patch by Peter Housel! llvm-svn: 103267
-
Nick Lewycky authored
llvm-svn: 103233
-
Dan Gohman authored
lowered copies. llvm-svn: 103228
-
Dan Gohman authored
llvm-svn: 103227
-
Dan Gohman authored
increase in the debug line info section, and it's causing regressions in a gdb testsuite. llvm-svn: 103226
-
- May 06, 2010
-
-
Dan Gohman authored
doesn't have to guess. llvm-svn: 103194
-
Evan Cheng authored
llvm-svn: 103193
-
Evan Cheng authored
llvm-svn: 103185
-
Evan Cheng authored
with the fix in 103157. %reg1039:1<def> = VMOVS %S1<kill>, pred:14, pred:%reg0 is not coalescable since none of the super-registers of S1 are in reg1039's register class: DPR_VFP2. But it is still a legal copy instruction so it should not assert. llvm-svn: 103170
-
Dan Gohman authored
llvm-svn: 103145
-
Chris Lattner authored
Users can write broken code that emits the same label twice with asm renaming, detect this and emit a fatal backend error instead of aborting. llvm-svn: 103140
-
Dan Gohman authored
llvm-svn: 103139
-
Dan Gohman authored
support a new bottom-up mode. llvm-svn: 103138
-
Dan Gohman authored
of just letting them inherit the debug locations of adjacent instructions. Debug info should aim to be either accurate or absent. llvm-svn: 103135
-
Jakob Stoklund Olesen authored
llvm-svn: 103133
-
Devang Patel authored
llvm-svn: 103126
-
- May 05, 2010
-
-
Evan Cheng authored
llvm-svn: 103109
-
Evan Cheng authored
Teach liveintervalanalysis about virtual registers which are defined by reg_sequence instructions that are formed by registers defined by distinct instructions. e.g. 80 %reg1041:6<def> = VSHRNv4i16 %reg1034<kill>, 12, pred:14, pred:%reg0 . . . 120 %reg1041:5<def> = VSHRNv4i16 %reg1039<kill>, 12, pred:14, pred:%reg0 llvm-svn: 103102
-
Bob Wilson authored
MachineSSAUpdater to avoid duplicating all the code. llvm-svn: 103060
-
- May 04, 2010
-
-
Evan Cheng authored
Teach PHI elimination to remove REG_SEQUENCE instructions and update references of the source operands with references of the destination with subreg indices. e.g. %reg1029<def>, %reg1030<def> = VLD1q16 %reg1024<kill>, ... %reg1031<def> = REG_SEQUENCE %reg1029<kill>, 5, %reg1030<kill>, 6 => %reg1031:5<def>, %reg1031:6<def> = VLD1q16 %reg1024<kill>, ... PHI elimination now does more than phi elimination. It is really a de-SSA pass. llvm-svn: 103039
-
Evan Cheng authored
llvm-svn: 103013
-
Devang Patel authored
llvm-svn: 103001
-
Evan Cheng authored
Instruction selection optimizations may have moved the def of a function argument out of the entry block. rdar://7937489 llvm-svn: 102993
-
Evan Cheng authored
llvm-svn: 102984
-
Dan Gohman authored
llvm-svn: 102981
-