aboutsummaryrefslogtreecommitdiff
path: root/lib/Target
AgeCommit message (Collapse)Author
2009-11-27add a deadargelim note.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@90009 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-27This testcase is actually only partially redundant, and requiresChris Lattner
the FIXME I added yesterday to be implemented. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@90008 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-27this (and probably several others) are now done.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89982 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-26Teach memdep to phi translate bitcasts. This allows us to compileChris Lattner
the example in GCC PR16799 to: LBB1_2: ## %bb1 movl %eax, %eax subq %rax, %rdi movq %rdi, (%rcx) movl (%rdi), %eax testl %eax, %eax je LBB1_2 instead of: LBB1_2: ## %bb1 movl (%rdi), %ecx subq %rcx, %rdi movq %rdi, (%rax) cmpl $0, (%rdi) je LBB1_2 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89978 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-26Teach basicaa that x|c == x+c when the c bits of x are clear. ThisChris Lattner
allows us to compile the example in readme.txt into: LBB1_1: ## %bb movl 4(%rdx,%rax), %ecx movl %ecx, %esi imull (%rdx,%rax), %esi imull %esi, %ecx movl %esi, 8(%rdx,%rax) imull %ecx, %esi movl %ecx, 12(%rdx,%rax) movl %esi, 16(%rdx,%rax) imull %ecx, %esi movl %esi, 20(%rdx,%rax) addq $16, %rax cmpq $4000, %rax jne LBB1_1 instead of: LBB1_1: movl (%rdx,%rax), %ecx imull 4(%rdx,%rax), %ecx movl %ecx, 8(%rdx,%rax) imull 4(%rdx,%rax), %ecx movl %ecx, 12(%rdx,%rax) imull 8(%rdx,%rax), %ecx movl %ecx, 16(%rdx,%rax) imull 12(%rdx,%rax), %ecx movl %ecx, 20(%rdx,%rax) addq $16, %rax cmpq $4000, %rax jne LBB1_1 GCC (4.2) doesn't seem to be able to eliminate the loads in this testcase either, it generates: L2: movl (%rdx), %eax imull 4(%rdx), %eax movl %eax, 8(%rdx) imull 4(%rdx), %eax movl %eax, 12(%rdx) imull 8(%rdx), %eax movl %eax, 16(%rdx) imull 12(%rdx), %eax movl %eax, 20(%rdx) addl $4, %ecx addq $16, %rdx cmpl $1002, %ecx jne L2 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89952 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-26teach basicaa that A[i] != A[i+1].Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89951 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-26update some notes slightlyChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89913 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Rollback changes r89516: Added two SubtargetFeatures::AddFeatures methods, ↵Viktor Kutuzov
which accept a comma-separated string or already parsed command line parameters as input, and some code re-factoring to use these new methods. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89893 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Tail duplicate indirect branches for PowerPC, too.Bob Wilson
With the testcase for pr3120, the "threaded interpreter" runtime decreases from 1788 to 1413 with this change. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89877 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Avoid some possibly unsafe uses of StringRef::data().Benjamin Kramer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89873 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Use StringRef (again) in DebugInfo interface.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89866 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Based on the testcase for pr3120, running on my MacPro with Xeon processors,Bob Wilson
it is definitely profitable to tail duplicate indirect branches for x86. This is likely to be true to various degrees for all modern x86 processors. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89865 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Support PIC loading of constant pool entriesBruno Cardoso Lopes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89863 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Sketch structure for X86 disassembler.Daniel Dunbar
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89850 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Use endianess dependent offsets for load/store of doubles whenBruno Cardoso Lopes
using two swc/lwc instead of sdc/ldc. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89826 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Fix compiler warnings.Dale Johannesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89824 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Only include in the callee saved regs the sub registers to avoidBruno Cardoso Lopes
unnecessary save/restore. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89823 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Add proper emission of load/store double to stack slots for mips1 targets!Bruno Cardoso Lopes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89821 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-25Revert r89803.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89819 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-24Refactor target hook for tail duplication as requested by Chris.Bob Wilson
Make tail duplication of indirect branches much more aggressive (for targets that indicate that it is profitable), based on further experience with this transformation. I compiled 3 large applications with and without this more aggressive tail duplication and measured minimal changes in code size. ("size" on Darwin seems to round the text size up to the nearest page boundary, so I can only say that any code size increase was less than one 4k page.) Radar 7421267. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89814 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-24Do not store R31 into the caller's link area on PPC.Dale Johannesen
This violates the ABI (that area is "reserved"), and while it is safe if all code is generated with current compilers, there is some very old code around that uses that slot for something else, and breaks if it is stored into. Adjust testcases looking for current behavior. I've verified that the stack frame size is right in all testcases, whether it changed or not. 7311323. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89811 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-24Enable debug info for ppc-darwin.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89803 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-24Enable predication of NEON instructions in Thumb2 mode.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89748 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-24Make capitalization of names starting "is" more consistent.Dale Johannesen
No functional change. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89724 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-24Data type suffix must come after predicate.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89723 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-24Materialize global addresses via movt/movw pair, this is always betterAnton Korobeynikov
than doing the same via constpool: 1. Load from constpool costs 3 cycles on A9, movt/movw pair - just 2. 2. Load from constpool might stall up to 300 cycles due to cache miss. 3. Movt/movw does not use load/store unit. 4. Less constpool entries => better compiler performance. This is only enabled on ELF systems, since darwin does not have needed relocations (yet). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89720 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-2480 column violationsJim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89718 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23* Move stub allocation inside the JITEmitter, instead of exposing aJeffrey Yasskin
way for each TargetJITInfo subclass to allocate its own stubs. This means stubs aren't as exactly-sized anymore, but it lets us get rid of TargetJITInfo::emitFunctionStubAtAddr(), which lets ARM and PPC support the eager JIT, fixing http://llvm.org/PR4816. * Rename the JITEmitter's stub creation functions to describe the kind of stub they create. So far, all of them create lazy-compilation stubs, but they sometimes get used when far-call stubs are needed. Fixing http://llvm.org/PR5201 will involve fixing this. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89715 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23Remove ISD::DEBUG_LOC and ISD::DBG_LABEL, which are no longer used.Dan Gohman
Note that "hasDotLocAndDotFile"-style debug info was already broken; people wanting this functionality should implement it in the AsmPrinter/DwarfWriter code. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89711 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23Allow more than one stub to be being generated at the same time.Jeffrey Yasskin
It's probably better in the long run to replace the indirect-GlobalVariable system. That'll be done after a subsequent patch. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89708 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23Massive refactoring of NEON instructions. Separate opcode from data size ↵Evan Cheng
specifier suffix, move \t up stream to instruction format, and fix more 80 column violations. This fixes the NEON asm printing so the "predicate" field is printed between the opcode and the data type suffix. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89706 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23move fconst[sd] to UAL. <rdar://7414913>Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89700 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23Partially revert r84730 by removing N2VDup from ARMInstrFormats.td and modifyingJohnny Chen
VDUPLND and VDUPLNQ to derive from N2V instead of N2VDup. VDUPLND and VDUPLNQ now expect op19_18 and op17_16 as the first two args. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89699 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23fold immediate of a + Const into the user as a subtract if it can fit as a ↵Jim Grosbach
negated two-part immediate. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89694 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23Revert r84572 by removing N3VImm from ARMInstrFormats.td now that we can specifyJohnny Chen
{?,?,?,?} as op11_8 for VEXTd and VEXTq. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89693 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23Partially revert r89377 by removing NLdStLN class definition fromJohnny Chen
ARMInstrFormats.td and fixing VLD[234]LN* and VST[234]LN* to derive from NLdSt instead of NLdStLN. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89684 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23Make it clear that the index bit(s) of Vector Get Lane and Vector Set LaneJohnny Chen
should be left unspecified now that Bob Wilson has fixed pr5470. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89676 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-23Minor itinerary fixes for FP instructions.David Goodwin
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89672 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-22Move default FrameReg val to getFrameIndexReference(). Otherwise, debug info ↵Jim Grosbach
can get bogus values. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89618 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-22Generate more correct debug info for frame indices.Jim Grosbach
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89576 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-22Minor optimization: when doing eq/ne comparions and RHS is a constant - swap ↵Anton Korobeynikov
operands, this will allow us to fold imm into comparison. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89574 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-22Drop unsupported imm operandsAnton Korobeynikov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89573 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-22Use 2-byte alignment for functions. 4 bytes are clear overkill here.Anton Korobeynikov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89572 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-22Use semicolon as assembler comment stringAnton Korobeynikov
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89571 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-21Revert 89562. We're being sneakier than I was giving us credit for, and thisJim Grosbach
isn't necessary. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89568 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-21Darwin requires a frame pointer for all non-leaf functions to support correctJim Grosbach
backtraces. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89562 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-21Add predicate operand to NEON instructions. Fix lots (but not all) 80 col ↵Evan Cheng
violations in ARMInstrNEON.td. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89542 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-21We are not using DBG_STOPPOINT anymore.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89536 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-21Added two SubtargetFeatures::AddFeatures methods, which accept a ↵Viktor Kutuzov
comma-separated string or already parsed command line parameters as input, and some code re-factoring to use these new methods. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89516 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-20Fix a thinko that caused spurious @GOTOFFs.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@89509 91177308-0d34-0410-b5e6-96231b3b80d8