summaryrefslogtreecommitdiffstats
path: root/llvm/lib
Commit message (Collapse)AuthorAgeFilesLines
...
* Use MachineConstantPoolEntry getOffset() and getType() accessors.Evan Cheng2006-09-141-13/+7
| | | | llvm-svn: 30326
* Use new config.h macroChris Lattner2006-09-141-2/+2
| | | | llvm-svn: 30321
* On Mac OS/X, make Process::PreventCoreFiles disable crash reporter forChris Lattner2006-09-141-7/+29
| | | | | | | the process in addition to disabling core file emission. This speeds up bugpoint on default-configured macs by several orders of magnitude. llvm-svn: 30317
* A MachineConstantPool may have mixed Constant* and MachineConstantPoolValue* ↵Evan Cheng2006-09-141-2/+8
| | | | | | values. llvm-svn: 30316
* Avoid recursion in assignDFSNumber(). Move def from ET-Forest.hDevang Patel2006-09-141-0/+33
| | | | | | to Dominators.h llvm-svn: 30309
* add note about switch loweringChris Lattner2006-09-131-0/+29
| | | | llvm-svn: 30308
* Second half of the fix for Transforms/Inline/inline_cleanup.llChris Lattner2006-09-131-2/+28
| | | | | | | This folds unconditional branches that are often produced by code specialization. llvm-svn: 30307
* Add some more consistency checks.Nick Lewycky2006-09-131-1/+20
| | | | llvm-svn: 30305
* Fix unionSets so that it can merge correctly.Nick Lewycky2006-09-131-22/+34
| | | | llvm-svn: 30304
* Implement the first half of Transforms/Inline/inline_cleanup.llChris Lattner2006-09-131-1/+9
| | | | llvm-svn: 30303
* Skip over first operand when determining REX prefix for two-address code.Evan Cheng2006-09-131-7/+14
| | | | llvm-svn: 30300
* Erase dead instructions.Nick Lewycky2006-09-131-2/+3
| | | | llvm-svn: 30298
* Turn X < 0 -> TEST X,X jsChris Lattner2006-09-131-8/+11
| | | | llvm-svn: 30294
* The sense of this branch was inverted :(Chris Lattner2006-09-131-1/+1
| | | | llvm-svn: 30293
* Fix a ton of jit failuresChris Lattner2006-09-131-1/+1
| | | | llvm-svn: 30292
* add shifts to addressing mode 1Rafael Espindola2006-09-136-32/+81
| | | | llvm-svn: 30291
* If LSR went through a lot of trouble to put constants (e.g. the addr of a globalChris Lattner2006-09-131-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | in a specific BB, don't undo this!). This allows us to compile CodeGen/X86/loop-hoist.ll into: _foo: xorl %eax, %eax *** movl L_Arr$non_lazy_ptr, %ecx movl 4(%esp), %edx LBB1_1: #cond_true movl %eax, (%ecx,%eax,4) incl %eax cmpl %edx, %eax jne LBB1_1 #cond_true LBB1_2: #return ret instead of: _foo: xorl %eax, %eax movl 4(%esp), %ecx LBB1_1: #cond_true *** movl L_Arr$non_lazy_ptr, %edx movl %eax, (%edx,%eax,4) incl %eax cmpl %ecx, %eax jne LBB1_1 #cond_true LBB1_2: #return ret This was noticed in 464.h264ref. This doesn't usually affect PPC, but strikes X86 all the time. llvm-svn: 30290
* Fix a regression in the 32-bit port from the 64-bit port landing.Chris Lattner2006-09-131-2/+2
| | | | | | | | | | | | | | | | | | | | | We now compile CodeGen/X86/lea-2.ll into: _test: movl 4(%esp), %eax movl 8(%esp), %ecx leal -5(%ecx,%eax,4), %eax ret instead of: _test: movl 4(%esp), %eax leal (,%eax,4), %eax addl 8(%esp), %eax addl $4294967291, %eax ret llvm-svn: 30288
* new noteChris Lattner2006-09-131-0/+32
| | | | llvm-svn: 30286
* new noteChris Lattner2006-09-131-0/+13
| | | | llvm-svn: 30285
* Compile X << 1 (where X is a long-long) to:Chris Lattner2006-09-131-7/+24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | addl %ecx, %ecx adcl %eax, %eax instead of: movl %ecx, %edx addl %edx, %edx shrl $31, %ecx addl %eax, %eax orl %ecx, %eax and to: addc r5, r5, r5 adde r4, r4, r4 instead of: slwi r2,r9,1 srwi r0,r11,31 slwi r3,r11,1 or r2,r0,r2 on PPC. llvm-svn: 30284
* Compile X > -1 -> text X,X; js destChris Lattner2006-09-132-28/+23
| | | | | | This implements CodeGen/X86/jump_sign.ll. llvm-svn: 30283
* Initialize DontInternalize.Devang Patel2006-09-131-1/+2
| | | | llvm-svn: 30281
* Reflects MachineConstantPoolEntry changes.Evan Cheng2006-09-128-11/+12
| | | | llvm-svn: 30279
* Added support for machine specific constantpool values. These are useful forEvan Cheng2006-09-126-51/+152
| | | | | | representing expressions that can only be resolved at link time, etc. llvm-svn: 30278
* Reflect MachineConstantPoolEntry changes.Evan Cheng2006-09-121-3/+12
| | | | llvm-svn: 30277
* An sinkable instruction may exist with uses, if those uses are in dead blocks.Chris Lattner2006-09-121-0/+4
| | | | | | Handle this. This fixes PR908 and Transforms/LICM/2006-09-12-DeadUserOfSunkInstr.ll llvm-svn: 30275
* add a noteChris Lattner2006-09-121-0/+14
| | | | llvm-svn: 30271
* Testcase noticed from PR906Chris Lattner2006-09-111-0/+34
| | | | llvm-svn: 30269
* add compilable testcaseChris Lattner2006-09-111-1/+6
| | | | llvm-svn: 30268
* Fix PR905 and InstCombine/2006-09-11-EmptyStructCrash.llChris Lattner2006-09-111-1/+2
| | | | llvm-svn: 30266
* implement SRL and MULRafael Espindola2006-09-111-0/+9
| | | | llvm-svn: 30262
* add the correct fallback for ARMDAGToDAGISel::SelectAddrMode1Rafael Espindola2006-09-111-6/+3
| | | | llvm-svn: 30261
* partial implementation of the ARM Addressing Mode 1Rafael Espindola2006-09-115-37/+74
| | | | llvm-svn: 30252
* Skip the linear search if the answer is already known.Nick Lewycky2006-09-111-20/+22
| | | | llvm-svn: 30251
* call AsmPrinter::doInitialization in ARMAsmPrinter::doInitializationRafael Espindola2006-09-111-0/+1
| | | | llvm-svn: 30246
* Updates.Evan Cheng2006-09-112-154/+0
| | | | llvm-svn: 30245
* Update README file.Evan Cheng2006-09-111-98/+3
| | | | llvm-svn: 30244
* X86ISD::CMP now produces a chain as well as a flag. Make that the chainEvan Cheng2006-09-115-157/+131
| | | | | | | operand of a conditional branch to allow load folding into CMP / TEST instructions. llvm-svn: 30241
* Behold, more work on relocations. Things are looking pretty good now.Nate Begeman2006-09-104-79/+177
| | | | llvm-svn: 30240
* Removed unnecessary Mangler creation.Anton Korobeynikov2006-09-101-1/+0
| | | | llvm-svn: 30239
* Allow tail duplication in more cases, relaxing the previous restriction aChris Lattner2006-09-101-1/+12
| | | | | | bit. This fixes Regression/Transforms/TailDup/MergeTest.ll llvm-svn: 30237
* This code was trying too hard. By eliminating redundant edges in the CFGChris Lattner2006-09-101-6/+3
| | | | | | | | | due to switch cases going to the same place, it make #pred != #phi entries, breaking live interval analysis. This fixes 458.sjeng on x86 with llc. llvm-svn: 30236
* Replace EquivalenceClasses with a custom-built data structure. Many commonNick Lewycky2006-09-101-133/+259
| | | | | | | | | operations (like findProperties) should be faster, at the expense of unionSets being slower in cases that are rare in practise. Don't erase a dead Instruction. This fixes a memory corruption issue. llvm-svn: 30235
* Implement Transforms/InstCombine/hoist_instr.llChris Lattner2006-09-091-14/+54
| | | | llvm-svn: 30234
* Make inlining costs more accurate.Chris Lattner2006-09-091-3/+26
| | | | llvm-svn: 30231
* Turn div X, (Cond ? Y : 0) -> div X, YChris Lattner2006-09-091-19/+68
| | | | | | This implements select.ll::test18. llvm-svn: 30230
* Add cbe support for powiChris Lattner2006-09-091-0/+10
| | | | llvm-svn: 30226
* Implement the fpowi now by lowering to a libcallChris Lattner2006-09-093-1/+16
| | | | llvm-svn: 30225
* Allow targets to custom lower expanded BIT_CONVERT'sChris Lattner2006-09-091-2/+15
| | | | llvm-svn: 30217
OpenPOWER on IntegriCloud